Lymphocyte Differentiation, Recognition, and Regulation

By David H. Katz

Lymphocyte Differentiation, Recognition, and Regulation provides an overview of the state of knowledge on cellular immunology. The focus is on animal work than on studies in man, although in certain areas human lymphocyte biology has been discussed in some detail. The book attempts to integrate information from diverse areas of cellular immunology, immunogenetics, and immunochemistry to form some cohesive concepts that can be perhaps utilized as a working foundation for students and investigators in various areas of immunology. The book begins with a general description of some of techniques and principles underlying the systems frequently employed in cellular immunology. This is followed by detailed analyses of lymphocyte differentiation, receptor function, and regulatory processes. The main points that emerge from such analyses are that the immune system is an infinitely complex and finely tuned network of cells, receptors, and molecules which interact with one another in a genetically controlled manner that is manifested ultimately in the process known as differentiation.

• Language: English
• Publisher: Academic Press
• Release date: Jun 28, 2014
• ISBN: 9781483274072

Book preview

Lymphocyte Differentiation, Recognition, and Regulation

DAVID H. KATZ, M.D.

Department of Cellular and Developmental Immunology, Scripps Clinic and Research Foundation, La Jolla, California

Table of Contents

Cover image

Title page

IMMUNOLOGY

Copyright

Dedication

Preface

Chapter 1: Introduction

Publisher Summary

THE TWO MAJOR CLASSES OF IMMUNOCOMPETENT LYMPHOCYTES

Chapter 2: Surface Antigens and Markers on T Lymphocytes

Publisher Summary

A ANTIGENS FOUND EXCLUSIVELY ON T CELLS

B ANTIGENS ON T LYMPHOCYTES AND ON OTHER NONLYMPHOID CELLS, BUT NOT PRESENT ON B LYMPHOCYTES

Chapter 3: Surface Antigens and Markers on B Lymphocytes

Publisher Summary

A ANTIGENS FOUND EXCLUSIVELY ON CELLS OF THE B LYMPHOCYTE LINEAGE

B ANTIGENS ON B LYMPHOCYTES AND ON OTHER NONLYMPHOID CELLS, BUT NOT PRESENT ON T LYMPHOCYTES

Chapter 4: Surface Antigens Present on T Lymphocytes and Also on B Lymphocytes

Publisher Summary

A THE MAJOR HISTOCOMPATIBILITY ANTIGENS

B THE I-REGION-ASSOCIATED (la) ANTIGENS

C THE Fc RECEPTOR (FcR)

D THE THYMUS–B CELL COMMON ANTIGEN (Th–B)

E THE LY-8 ANTIGEN

Chapter 5: Ontogeny of the Two Major Classes of Lymphocytes

A GENERAL CONSIDERATIONS

B ONTOGENY OF T LYMPHOCYTES

C ONTOGENY OF B LYMPHOCYTES

Chapter 6: Surface Immunoglobulin on Lymphocytes

Publisher Summary

A INTRODUCTION

B DETECTION OF SURFACE IMMUNOGLOBULIN

C QUANTITATION OF SURFACE IG

D TISSUE DISTRIBUTION OF SURFACE IG+ LYMPHOCYTES

E ONTOGENY OF IG+ LYMPHOCYTES

F IMMUNOGLOBULIN CLASS EXPRESSION OF SURFACE IG ON LYMPHOCYTES

G MOVEMENT AND REDISTRIBUTION OF SURFACE IMMUNOGLOBULIN MOLECULES ON B LYMPHOCYTES

H SURFACE IMMUNOGLOBULIN ON T LYMPHOCYTES

I PROPERTIES OF SURFACE IMMUNOGLOBULIN ISOLATED FROM B LYMPHOCYTES

Chapter 7: Immunological Specificity of Immunocompetent Lymphocytes

Publisher Summary

A FUNCTIONAL EVIDENCE FOR SPECIFICITY IN T AND B LYMPHOCYTES

B COMPARATIVE SPECIFICITIES OF THE T AND B CELL REPERTOIRES

C DETERMINANT SPECIFICITY OF T AND B CELL RECEPTORS

D ANTIGEN RECEPTORS ON B LYMPHOCYTES

E ANTIGEN RECEPTORS ON T LYMPHOCYTES

Chapter 8: Functional Properties of T Lymphocytes

Publisher Summary

A REGULATORY T LYMPHOCYTES

B EFFECTOR T LYMPHOCYTES

C DIFFERENTIAL SENSITIVITIES OF T CELL SUBPOPULATIONS TO CHEMICAL AND PHYSICAL MANIPULATIONS

Chapter 9: Functional Properties of B Lymphocytes

Publisher Summary

A ANTIBODY-FORMING CELL PRECURSORS (AFCP)

B FUNCTIONAL SUBPOPULATIONS OF B LYMPHOCYTES DISTINGUISHED BY RELATIVE DEPENDENCE ON T CELLS

C OTHER CRITERIA FOR FUNCTIONAL SUBPOPULATIONS OF B LYMPHOCYTES

D GENETIC DEFECTS IN B LYMPHOCYTE FUNCTION

Chapter 10: Regulatory Cellular Interactions in Immune Responses

Publisher Summary

A BASIC MODELS OF T–B CELL INTERACTIONS

B NATURE OF THE REGULATORY INFLUENCE OF ACTIVATED T CELLS ON ANTIBODY PRODUCTION

C SUPPRESSIVE EFFECTS OF T LYMPHOCYTES ON ANTIBODY PRODUCTION

D T CELL REGULATION OF IgE ANTIBODY SYNTHESIS

E BASIC MODELS OF T–T CELL INTERACTIONS

F SUPPRESSIVE EFFECTS OF T CELLS ON CELL-MEDIATED IMMUNE RESPONSES

Chapter 11: The Allogeneic Effect on Immune Responses

Publisher Summary

A ESSENTIAL FEATURES OF THE ALLOGENEIC EFFECT ON ANTIBODY RESPONSES

B MEDIATION OF THE ALLOGENEIC EFFECT VIA DEVELOPMENT OF THE GRAFT-VERSUS-HOST REACTION

C THE ALLOGENEIC EFFECT IN VITRO

D RELEVANCE OF THE ALLOGENEIC EFFECT TO THE MECHANISM OF REGULATORY PHYSIOLOGICAL INTERACTIONS BETWEEN T AND B LYMPHOCYTES IN IMMUNE RESPONSES

E PATHOPHYSIOLOGICAL SIGNIFICANCE OF THE ALLOGENEIC EFFECT

F CONCLUSIONS

Chapter 12: Genetic Control of Immune Responses and Cellular Interactions

Publisher Summary

A HISTOCOMPATIBILITY-LINKED Ir GENES

B HISTOCOMPATIBILITY-LINKED IMMUNE SUPPRESSION (Is) GENES

C THE ROLE OF PRODUCTS OF THE MAJOR HISTOCOMPATIBILITY COMPLEX IN CELLULAR INTERACTIONS WHICH REGULATE IMMUNE RESPONSES

D RELATIONSHIP OF HISTOCOMPATIBILITY GENE PRODUCTS TO EACH OTHER IN NATURE AND FUNCTION

E GENERAL CONCLUSIONS

Chapter 13: Mechanisms of Regulatory Cellular Interactions

Publisher Summary

A GENERAL CONSIDERATIONS

B PROPERTIES OF BIOLOGICALLY ACTIVE PRODUCTS OF ACTIVATED T CELLS AND MACROPHAGES MEDIATING HELPER OR SUPPRESSOR ACTIVITIES

C HYPOTHETICAL MODELS FOR CELLULAR INTERACTIONS

Chapter 14: Immunological Tolerance

Publisher Summary

A HISTORICAL PERSPECTIVE

B TARGET CELLS FOR TOLERANCE INDUCTION

C MECHANISMS OF TOLERANCE INDUCTION IN EITHER T OR B LYMPHOCYTES

D INTERRELATIONSHIPS AMONG THE ALTERNATIVE PATHWAYS TO UNRESPONSIVENESS

E RELATIONSHIPS OF THE ALTERNATIVE PATHWAYS TO INDUCTION AND MAINTENANCE OF SELF-TOLERANCE

F CONCLUDING REMARKS

Chapter 15: Concluding Remarks

Publisher Summary

Glossary

Bibliography

Index

IMMUNOLOGY

An International Series of Monographs and Treatises

EDITED BY

F. J. DIXON, JR.

Division of Experimental Pathology

Scripps Clinic and Research Foundation

La Jolla, California

HENRY G. KUNKEL

The Rockefeller University

New York, New York

G. J. V. Nossal and G. L. Ada, Antigens, Lymphoid Cells, and the Immune Response. 1971

Barry D. Kahan and Ralph A. Reisfeld, Transplantation Antigens: Markers of Biological Individuality. 1972

Alfred Nisonoff, John E. Hopper, and Susan B. Spring. The Antibody Molecule. 1975

David H. Katz, Lymphocyte Differentiation, Recognition, and Regulation. 1977

Norman Talal, Autoimmunity: Genetic, Immunologic, Virologic, and Clinical Aspects. 1978

Copyright

COPYRIGHT © 1977, BY ACADEMIC PRESS, INC.

ALL RIGHTS RESERVED.

NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOTOCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS, INC.

111 Fifth Avenue, New York, New York 10003

United Kingdom Edition published by

ACADEMIC PRESS, INC. (LONDON) LTD.

24/28 Oval Road. London NW1

Library of Congress Cataloging in Publication Data

Katz, David H

Lymphocyte differentiation, recognition, and regulation.

(Immunology: an international series)

Bibliography: p.

Includes index.

1. Lymphocytes. 2. Cellular control mechanisms. 2. Cellular control mechanisms. 3. Immune response. I. Title. [DNLM: 1. Lymphocytes. 2. Immunity, Cellular. WH200K 195L]

QR185.8.L9K38 599′.08′765 77-7952

ISBN 0-12-401640-5

PRINTED IN THE UNITED STATES OF AMERICA

Dedication

To Lee Lisa, and Danica

Preface

Basic immunology and, in particular, cellular immunology, has attracted an enormous amount of attention during the past decade. In fact, it seems clear that the great excitement generated in the 1960’s and early 1970’s about the structure and function of immunoglobulin molecules, which was the subject of another monograph in this series by Nisonoff, Hopper, and Spring, has translated itself into a similar level of excitement about the properties and functional characteristics of the cells comprising the immune system. An important difference, however, in the nature of the work concerning cell function and properties, which is the current focus of attention, and the studies on immunoglobulin structure and function of several years ago, is the character of the field itself. Hence, the number of active investigators and students in immunology has expanded severalfold during the past decade, a fact which should be a gratifying testament to the outstanding scientists who have worked so hard to lay the foundations and establish immunology as an important scientific discipline. This also, however, has created the inevitable consequence of opening many new subareas of active research and increasing the degree of specialization in which many of us find ourselves involved.

This monograph represents an attempt at broadening my own knowledge and perspective about modern cellular immunology. Quite frankly, I set out originally to update a review published in 1972 in Advances in Immunology on T and B cell regulatory interactions, and found that it was no longer feasible for me, or fair to a prospective reader, to approach the topic as an isolated one set apart from many other facets of lymphocyte biology. Indeed, I quickly realized that in many respects my own detailed knowledge of important, although at times peripheral, subjects was inadequate to perform the original task in a justifiable manner.

What appears in the 15 chapters of this book is, therefore, a product of my own learning exercise during the past months. It has covered many, but far from all, aspects of cellular immunology. It concentrates far more on animal work than on studies in man, although in certain areas human lymphocyte biology has been discussed in some detail. I have attempted to be as thorough as possible in most areas discussed and, in particular, to cite the published work of many investigators, but inevitably I am certain that inadvertent errors were made. Since the monograph is intended to be a detailed review and reference source of the topics covered for students and investigators actively working in immunology, it has not been written in textbook style. As a result, anyone wishing to obtain a more basic background in fundamentals of immunology will be disappointed as well as perhaps slightly overwhelmed by the complexities of the issues discussed herein.

I sincerely appreciate the efforts of Baruj Benacerraf, Harvey Cantor, Herman Eisen, George Schreiner, Barry Skidmore, and Emil Unanue who critically read selected chapters and, where appropriate, made constructive suggestions for improvement of them. Jerry Reicher and his staff at Arrco Medical Illustrations deserve considerable credit for the artwork. I also wish to thank the other members of my laboratory who bore with me during the preparation of this monograph. I cannot overstate the phenomenal efforts of Marsha Goldman and Charlene Small who typed the manuscript and helped me enormously in the organization of it. And, finally, I wish to express my gratitude to my wife, Lee, and our daughters, Lisa and Danica, who endured prolonged periods of my absence from home and graciously made other sacrifices that permitted me to finish this task on time.

DAVID H. KATZ, M.D.

I

Introduction

Publisher Summary

It is generally accepted that a class of bone marrow lymphocytes migrates to the thymus where the small lymphocytes adapt to certain specific immune functions by virtue of some crucial influence of the thymus. These thymus-derived lymphocytes, referred to as T cells, are responsible for the various phenomena of cell-mediated immunity, such as delayed hypersensitivity, transplantation reactivity, including cell-mediated cytotoxicity and mixed lymphocyte reactions, and cell-mediated resistance to infection; T cells are also the predominant cell type concerned with regulation of other lymphoid cells in the immune system. The second lymphocyte cell type, referred to as B cells, arises also in the bone marrow and settles ultimately in distinct anatomical sites in peripheral lymphoid tissues where they give rise to the precursors of antibody-secreting cells. Among the more helpful tools available to the cellular immunologist are experimental systems in which analyses can be made of the respective functions of T cells and B cells, either independently of one another or in an interdependent manner.

For the past 7 to 8 years, the phenomena of cell interactions in the development and regulation of immune responses have been extensively investigated. During this time, much has been learned about the importance of such cell interactions in regulating the immune system, and also about the possible mechanisms by which these interactions take place. During this same period, remarkable advances have been made in furthering our knowledge and understanding of cell surface membrane molecules, some of which perform receptor functions, either clonally restricted or nonclonal, and others which appear to be integrally associated with activation and differentiation events in lymphoid cells. Although it is true that much has yet to be determined, particularly at the molecular level, about the precise pathways involved in the development of immunocompetent cell functions and interactions, it seems fair to state that our insights on these events, derived to a great extent from phenomenological observations, have led to various working hypothetical models on which to proceed, eventually, and hopefully, to some concrete solutions.

Nearly five years ago, we undertook the task of reviewing the areas concerned with T cell regulation of antibody responses and the significance of cell interaction phenomena for the regulatory processes of the immune system (1051). Since then, there has continued to be an avalanche of new and exciting observations by many active investigators resulting in increasingly provocative modifications in certain basic concepts, and the strong indications that much of what is being analyzed in immunological systems may have considerably broader implications for developmental and molecular biology. Thus, the knowledge obtained about the fundamental regulatory control of lymphocyte and macrophage functions and the events initiating and perpetuating differentiation of such cells may now, or very shortly, be appropriately interpreted in the context of developmental functions of eukaryotic cells in general.

This monograph was prepared in an attempt to provide some insight on the systems and data on which the above statements are based. In so doing, I hope to integrate information from diverse areas of cellular immunology, immunogenetics, and immunochemistry to form some cohesive concepts that can be perhaps utilized as a working foundation for students and investigators in various areas of immunology. The main points that emerge from such analyses to be presented herein are, in my view, that the immune system is an infinitely complex and finely tuned network of cells, receptors, and molecules which interact with one another in a genetically controlled manner that is manifested ultimately in the process known as differentiation.

In order to develop meaningful insights into the immense complexity of the integrated systems governing immune defenses of the individual, it seems appropriate to consider in some detail certain aspects of lymphocyte differentiation, mechanisms underlying specific recognition by such cells and the processes concerned with regulation of lymphocyte function. The following chapters in this monograph are directed to these aspects of cellular immunology and are not intended to represent the entire spectrum of present day immunology. Notably absent, for example, are discussions of the complement system or details of immunoglobulin structure, various aspects of immunopathological processes or tumor immunology, among other things.

Before going into any detailed analysis of our current knowledge of lymphocyte differentiation, receptor function, and regulatory processes, it may be helpful to those readers who are not very familiar with some of the systems frequently employed in cellular immunology to provide at the outset a general description of these techniques and the principles underlying them. I will return to them in greater detail in appropriate subsequent chapters.

THE TWO MAJOR CLASSES OF IMMUNOCOMPETENT LYMPHOCYTES

One of the major advances in immunobiology in the past two decades has been the recognition of two pathways for the differentiation of antigen-reactive cells. It is generally accepted that a class of bone marrow lymphocytes migrates to the thymus where these small lymphocytes adapt to certain specific immune functions by virtue of some crucial influence of the thymus. These thymus-derived lymphocytes, referred to as T cells, are responsible for the various phenomena of cell-mediated immunity, such as delayed hypersensitivity, transplantation reactivity, including cell-mediated cytotoxicity and mixed lymphocyte reactions, and cell-mediated resistance to infection; T cells are also perhaps the predominant cell type concerned with regulation of other lymphoid cells in the immune system. The second lymphocyte cell type, referred to as B cells, arises also in the bone marrow and settles ultimately in distinct anatomical sites in peripheral lymphoid tissues where they give rise to the precursors of antibody-secreting cells (339, 448, 1051, 1381).

Among the more helpful tools available to the cellular immunologist are experimental systems in which analyses can be made of the respective functions of T cells and B cells, either independently of one another or in an interdependent manner. The introduction of defined haptenic determinants onto immunogenic carriers by Landsteiner (1201) has provided a most convenient method for the analysis of specific interactions between antigens and specific cells of the immune system. For many years it has been known that immunization with a hapten elicits anti-hapten antibody responses only when the hapten is coupled to a carrier substance which is itself immunogenic; nonimmunogenic substances serve only poorly, or not at all, as functional carriers for haptens. Moreover, as mentioned above, optimal hapten-specific secondary responses require challenge with the hapten–carrier conjugate used for primary immunization (i.e., the carrier effect). Since the anti-hapten antibodies produced by such immunizations were highly specific for the haptenic determinant employed (e.g., little, if any, contribution to binding energy was attributable to determinants on the carrier) and since the assumption was made that the specificity of antibody accurately expresses the specificity of the antigen-binding receptor molecules on the precursors of antibody-forming cells, then these observations suggested the operation of an additional recognition mechanism for the carrier molecule. Indeed, the demonstration that cooperative interactions between distinct lymphocytes respectively specific for carrier and haptenic determinants are essential for the development of anti-hapten immune responses validated this interpretation (reviewed in 1051; see Chapter X). Two basic in vivo experimental models have been employed to establish the latter point:

1. The adoptive secondary anti-hapten response following transfer of hapten-primed and carrier-primed cells into irradiated recipient mice.

2. The use of preimmunization or supplemental immunization with free carrier to enhance primary and secondary anti-hapten antibody responses in guinea pigs and rabbits.

The adoptive transfer experimental system is based on the use of a lethally or sublethally irradiated animal as a relatively immunologically inert recipient of primed or unprimed lymphoid cells whose functions is under analysis. Under usual circumstances inbred animals are employed in this technique, and the cell donors and recipients are syngeneic at the major histocompatibility locus. Modifications from this usual approach will be discussed in Chapters X and XII.

A schematic illustration of the basic approach for studying cooperative lymphocyte interactions in responses to hapten-protein (carrier) conjugates in mice is shown in Fig. I.1. In such systems, two types of donor syngeneic spleen cell populations are employed: One of these is obtained from donors previously immunized with a hapten–carrier conjugate, in this case, 2,4-dinitrophenyl (DNP)-keyhole limpet hemocyanin (KLH); the second donor population is obtained from donors previously immunized with an unrelated carrier alone, in this case bovine γ-globulin (BGG), or from normal unprimed donors as a control. Appropriate numbers of the respective spleen cell populations are mixed together and injected intravenously into unprimed, irradiated syngeneic recipients. These recipients are then challenged intraperitoneally with one of several DNP–carrier conjugates and bled at an appropriate time thereafter (7 days in Fig. I.1). The serum obtained from such recipients is then titrated for levels of anti-DNP antibodies; anti-carrier antibodies can also be titrated if desirable. Moreover, other variations of this scheme include repetitive sequential bleeding of recipients for antibody titrations and removing recipient spleens for analysis of DNP-(or protein)-specific antibody-secreting cells in a localized hemolysin-in-gel assay for plaque-forming cells (PFC).

Fig. I.1 See text for explanation.

The basic results obtained in this system, as shown in Fig. I.1, demonstrate that recipients of a mixture of DNP-KLH-primed and normal spleen cells develop secondary anti-DNP antibody responses when challenged with the homologous (immunizing) conjugate, DNP-KLH, but not when challenged with a heterologous (unrelated carrier) conjugate, DNP-BGG (cf. groups 1 and 2); however, when the second donor inoculum consists of BGG-primed spleen cells, DNP-BGG challenge elicits a secondary anti-DNP antibody response (group 3). The capacity of BGG-primed spleen cells to assist the DNP-KLH cells in responding to DNP-BGG is highly specific as indicated by the failure of such cells to assist in the development of a response to DNP coupled to a third unrelated carrier, ovalbumin (OVA, group 4).

As will be discussed in detail in Chapter X, the failure to obtain responses to DNP-BGG in group 2, the phenomenon known as the carrier effect, reflects the absence of carrier (BGG)-specific T cells necessary for regulating the response of the DNP-specific B lymphocyte precursors of anti-DNP antibody-secreting cells. Spleen cells from BGG-primed donors are one such source of BGG-specific helper T lymphocytes and, hence, enable recipients in group 3 to develop responses to DNP-BGG (but not to an unrelated carrier, DNP-OVA).

A similar type of adoptive transfer model can be used for analysis of responses of unprimed lymphocyte populations, mixtures of primed and unprimed lymphoid cells, or mixtures of lymphocyte populations originating in different lymphoid organs. For example, the classical experiments on T-B cell cooperation in responses to sheep red blood cell antigens (SRBC) were conducted with mixtures of unprimed thymus lymphocytes (thymocytes) and bone marrow lymphocytes in a system similar to that shown in Fig. I.1 (see Chapter X).

A somewhat different approach demonstrating the same cooperation phenomenon between carrier-specific T cells and hapten-specific B cells is that involving supplemental immunization of an intact animal with free carrier. This approach does not involve adoptive cell transfer into irradiated recipients and, hence, can be used in outbred as well as inbred animal populations such as rabbits, guinea pigs, rats, and mice. A schematic illustration of this approach in guinea pigs as studied in our own laboratory several years ago is presented in Fig. I.2. Guinea pigs primarily immunized with DNP-OVA will develop secondary anti-DNP antibody responses upon challenge with DNP-OVA, but not to secondary challenge with DNP-BGG (cf. groups 1 and 2). This carrier effect is circumvented, however, when DNP-OVA-primed animals are given a supplemental immunization with BGG. The higher magnitude of response obtained when such animals are challenged with DNP-BGG (group 3), as compared to the responses obtained in groups challenged with the homologous antigen, DNP-OVA (groups 1 and 4), reflects the mode of supplemental immunization which consists of BGG emulsified in complete Freund’s adjuvant (CFA); indeed, when DNP-OVA-primed guinea pigs are given a supplemental immunization of free OVA in CFA, anti-DNP antibody responses to DNP-OVA are considerably increased (group 6).

Fig. I.2 See text for explanation.

The results of group 5 illustrate that the supplemental immunization with OVA did not prepare such animals for a secondary response to DNP-BGG; this also points out that nonspecific stimulation due to mycobacteria in CFA is not itself responsible for circumventing the carrier effect in this way.

The effect of supplemental immunization with BGG as shown in Fig.I.2 is not related to circulating anti-BGG antibodies, as shown by the failure to duplicate this effect by passive transfer of anti-BGG serum (group 7), and, indeed, has been shown to be precisely comparable to the situation described in Fig. 1 in which a second donor inoculum of BGG-primed spleen cells assist DNP–KLH-primed cells to respond to DNP-BGG. In other words, supplemental immunization of guinea pigs with BGG (in this way) primes a second population of BGG-specific helper T lymphocytes which then cooperate with DNP-specific B lymphocytes in response to DNP-BGG. This phenomenon is not restricted to secondary responses, since under appropriate conditions of antigen dose and timing, animals which have been preimmunized with free carrier, e.g., BGG, manifest enhanced primary anti-DNP antibody responses following primary immunization with DNP-BGG. The kinetics as well as the magnitude of anti-DNP antibody production are sharply augmented under such conditions.

These experiments (Figs. I.1 and I.2) demonstrate that in hapten-specific antibody responses, an interaction of carrier-specific T cells with the hapten-carrier conjugate is required for optimal stimulation of the B cell precursors of anti-hapten antibody-producing cells. The complex regulatory role exerted by carrier-specific T cells in such antibody responses will be discussed at greater length in Chapter X. However, since such systems have been used in many studies directed to questions that will be discussed throughout this monograph, it is pertinent to establish the essential working vocabulary at this point. It should also be stated briefly here that a well-established property of carrier-primed T cells is that their capacity to function as regulatory helper cells is relatively radioresistant (see Chapter VIII). Accordingly, one not infrequently utilized modification of the adoptive transfer model in mice shown in Fig. I.1 is to employ carrier-primed animals as irradiated recipients of hapten-specific B cells. Thus, in contrast to what occurs in an unprimed, irradiated recipient, DNP-KLH-primed spleen cells will develop a secondary anti-DNP antibody response to DNP-BGG following adoptive transfer to an irradiated, BGG-primed recipient, without transfer of an additional inoculum of donor cells. Moreover, conditions have been established to perform similar types of experiments in vitro; for purposes of this introductory discussion, simply visualize the recipient mouse in Fig.I.l as a culture dish.

Another system frequently employed, and for which the reader should have an appreciation before proceeding, is the preparation of antigen-specific activated T cells (ATC). Since T cells stimulated by antigen respond by a clonal expansion and differentiation, on the one hand, and, on the other hand, by being activated to perform their specific function (i.e., helper cells, killer cells, etc.), we have elected to refer to the former as primed T cells, and the latter as activated T cells. As will be apparent throughout various sections of this monograph, activated T cells may result from stimulation by either specific antigen or agents other than specific antigen.

The preparation of antigen-specific ATC, as schematically depicted in Fig. I.3, is usually accomplished in the mouse by intravenous adoptive transfer of a suitable number of unprimed donor thymocytes into lethally or sublethally irradiated syngeneic recipients. Such recipient animals are then immunized with the desired antigen. With soluble proteins or erythrocyte antigens this is usually done intraperitoneally, and, in the former case, either emulsified in CFA or administered with another adjuvant [aluminum hydroxide gel (alum), Bordetella pertussis or both]; when ATC are prepared against alloantigens of a histoincompatible strain, the target cells are irradiated and usually administered intravenously. After a suitable interval (6-8 days), the spleens of recipient mice are removed and processed according to experimental design, e.g., for functional analysis in either in vivo or in vitro systems. The main point here is that a substantial proportion of the viable, mature donor T lymphocytes present in the recipient spleen consists of specifically primed ATC; it should be noted, however, that, although this is an enriched population of antigen-specific T cells, it is by no means an exclusive population.

Fig. I.3 See text for explanation.

Finally, mention must be made of the fact that throughout the remainder of this monograph, it will be apparent to the reader that a considerable number of uncertainties exist in all of the various areas of cellular immunology; indeed, it is perhaps more accurate to state that very few certainties can be cited with any comfortable degree of assurance that a given one will not be perceived as, or delineated to be, something different in a matter of time. This is not at all surprising for primarily two reasons: (1) The field is filled with an ever-growing body of active experimentalists and clinicians whose collective excitement about the seemingly expanding horizons of immunobiology into broader areas of biology and medicine has created an enormous momentum of inquiry into fundamental aspects of the system; inherent in this situation are the difficulties that arise as a consequence of our creative ideas running, at times, ahead of our current level of technological capabilities. (2) Most importantly, the immune system itself and, particularly, its cellular and molecular components, is so enormously complex that it defies any single answer as appropriate for explaining any part of its machinery; indeed, one of the lessons learned in reviewing the literature for preparing this monograph has been that evolution of the immune system has built into it an incredible degree of flexibility. Rarely, does it seem, has the system created a single pathway to an end with no alternative avenue to take when a biological detour becomes advantageous. Hence, one is almost safer to assume that only a few absolutes exist in the immune system.

I have attempted, therefore, to present both sides of the story in appropriate instances of debate, sometimes at the expense of redundancy in various spots. Also, because I am impressed by the overwhelming amount of detail that has arisen in recent years concerning cell surface markers of the lymphoid system and the confusion in many people’s minds of when such markers do or do not appear in ontogeny and/or functional stages of differentiation, an effort was made to synthesize the current body of knowledge in this area, realizing that it will change in many instances within a short time.

II

Surface Antigens and Markers on T Lymphocytes

Publisher Summary

The T cell has been the most thoroughly studied of all cells with respect to surface composition, and the antigens that can be detected on these lymphocytes and their precursors can be essentially divided into antigens expressed exclusively on the T cell, those that are present on other nonlymphoid cells, excluding B lymphocytes, in addition to C cells, and finally those antigens that are present on both T and B cells. There appear to be at least six distinct specificities that are detected solely on surface membranes of T lymphocytes or their precursors. These are the thymus leukemia (TL) antigen, the Ly-1, Ly-2/Ly-3, and Ly-5 antigens, and the mouse-specific lymphocyte antigen (MSLA). One of the interesting aspects of the TL system is the anomolous manner in which these antigens are expressed in leukemia cells of various mouse strains. Thus, irrespective of whether TL antigens are expressed in normal thymocytes of a given strain, all strains give rise to TL+ leukemias.

A Antigens Found Exclusively on T Cells

1. The TL Antigens

a. Definition and Genetics

b. Ontogenic and Functional Relationships

c. Modulation of TL Antigens

d. Associations between TL and H-2 Antigens

e. Biochemical Studies of TL Antigens

2. The Ly Antigen System

a. Definition and Genetics

b. Ontogenic and Functional Relationships

c. Biochemistry of Ly Antigens

3. The Mouse-Specific Lymphocyte Antigen (MSLA)

Definition and Genetics

B Antigens on T Lymphocytes and on Other Nonlymphoid Cells, but Not Present on B Lymphocytes

1. The θ Antigen (Thy-1)

a. Definition and Genetics

b. Tissue Distribution

c. Heteroantisera Specific for the θ Antigen

d. Ontogenic and Functional Relationships

e. Biochemistry of the θ Antigen

2. The GIX Antigens

a. Definition and Genetics

b. Biochemistry and Relationship to Viral Genes

c. Tissue Distribution and Ontogenic Relationship

Clearly, among the most significant achievements in cellular immunology, or perhaps more properly in mammalian cell biology, have been the studies designed to delineate the antigenic compositions of cell surfaces. Such studies have been largely conducted on lymphocytes, a logical choice in view of their accessibility and easily analyzed immunological functions, and, for the most part, in mice where the greatest flexibility and knowledge exists in terms of genetics and immunogenetics.

The T cell has been the most thoroughly studied of all cells with respect to surface composition, and the antigens that can be detected on these lymphocytes and their precursors can be essentially divided into antigens expressed exclusively on the T cell, those which are present on other nonlymphoid cells, excluding, of course, B lymphocytes, in addition to C cells, and finally those antigens which are present on both T and B cells; the first two categories will be discussed in this chapter, whereas the latter category will be discussed in Chapter IV.

A ANTIGENS FOUND EXCLUSIVELY ON T CELLS

There appear to be at least six distinct specificities which are detected solely on surface membranes of T lymphocytes or their precursors. These are the thymus leukemia (TL) antigen, the Ly-1, Ly-2/Ly-3, and Ly-5 antigens, and the mouse-specific lymphocyte antigen (MSLA). The first five of the aforementioned antigens are alloantigens, being detected by alloantisera of appropriate specificity. The latter antigen, MSLA, is an antigen detected by a heteroantiserum (usually made in rabbit) after appropriate absorption. We will consider each of these antigens individually, and the reader is referred to recent reviews which discuss one or more of these antigens in perhaps greater detail (227–229, 1660, 1818).

1 The TL Antigens

a Definition and Genetics

The TL (thymus leukemia) alloantigen system was first discovered in 1963 by Boyse and Old (230, 1514) during the course of studies initially designed to analyze the antigenic properties of an experimental radiation-induced leukemia. Antisera prepared by immunization of C57BL/6 mice with a radiation-induced leukemia from either A or (C57BL/6 × A)F1 donors were cytotoxic in vitro for cells of several C57BL/6 leukemias, but not for normal tissue of this donor haplotype. The critical finding was that, not only could the cytotoxic activity be absorbed from such antisera by C57BL/6 leukemia, but also with thymocytes derived from A, C58, or (C57BL/6 × A)F1 donors, but not with thymocytes of certain other mouse strains, including that of the C57BL/6 donor of the original leukemia cell preparation. Boyse and Old concluded, therefore, that thymocytes of normal mice of certain strains (A, C58) shared in common an antigen, which they termed TL, with one present on the C57BL/6 leukemia cells. The locus coding for the TL antigen is entitled Tla and has now been demonstrated to reside on the seventeenth chromosome of the mouse in close proximity to the D end of the H-2 gene complex (231). Although it initially appeared that different mouse strains could be simply classified as either TL-positive or TL-negative (232), it soon was demonstrated that TL was not a single antigen but, rather, a complex which consists of at least four distinct antigens, Tla. 1, Tla.2, Tla.3, and Tla.4 (237). The first three of these antigens can be detected on normal thymocytes, either expressed singly or in certain combinations in particular strains of mice. The fourth antigen has only been detected on leukemia cells, its absence in thymocytes of any strain indicating that the antigen is indeed leukemia-specific (227).

Analysis of the strain distribution of inbred mice carrying the TL antigens on normal thymocytes indicate that there exist three Tla haplotypes—Tlaa, Tlab, and Tlac—which behave as alleles at a single locus. Thus, the Tlaa haplotype consists of strains (prototype is A/J) expressing the thymocyte phenotype Tla. 1,2,3. Strains of the Tlab haplotype (C57BL/6 is the prototype) fail to express any known Tla antigens on their normal thymocytes, whereas mice of the Tlac haplotype (BALB/c as the prototype) express only Tla.2 on their surface. More extensive details of the complex genetics of the Tla locus are presented by Klein (1151).

b Ontogenic and Functional Relationships

One of the interesting aspects of the TL system is the anomolous manner in which these antigens are expressed in leukemia cells of various mouse strains. Thus, irrespective of whether TL antigens are expressed in normal thymocytes of a given strain, all strains give rise to TL+ leukemias. Moreover, with the single exception of the Tlaa haplotype, in no instance is the Tla phenotype of the leukemia identical to that of the normal thymocyte population. Thus, as indicated in Table II. 1, it is possible to observe leukemia cells expressing TL antigens not present on the thymocytes of the strain in which the leukemia arose. This anomalous appearance of TL antigens on leukemia versus normal thymocytes has been interpreted to indicate that existence of two types of genetic loci—structural and regulator—are involved in the expression of these antigens. As will be pointed out in greater detail below, it is clear that the expression of TL antigen on normal thymocytes is a reflection of maturational events in the T cell line. That is, early stem cells lack the TL antigen, whereas thymocytes developing within the thymus express the TL antigen (in TL+ strains), and, finally, this antigen is again lost from the cell surface of those cells leaving the thymus to go to peripheral lymphoid organs. Thus, it is conceivable that the regulator genes may act by repressing structural genes in the course of normal thymocyte development, and then act to derepress the structural genes during leukemogenesis. It is interesting to note that there is no a priori reason to require the presence of the Tla structural and regulator genes on the same chromosome, and if, indeed, they exist on different chromosomes then backcross data implies that it is the regulator gene that is linked to H-2, whereas the structural gene could reside anywhere within the mouse genome (1817). The provocative suggestion has been made that the Tla locus may be the integrated genome of a C-type RNA leukemia virus whose expression has become caught up in the differentiation program for T cells (227, 28, 1517).

TABLE II.1

Tla Phenotypes and Genotypes of Prototype Mouse Strains*

*Reproduced with permission, from E. A. Boyse and L. J. Old (1969). Some aspects of normal and abnormal cell surface genetics. Ann. Rev. Genet. 3:276. Copyright 1969 by Annual Reviews, Inc. All rights reserved.

†This corresponds to the presumed Tla genotype of these respective strains.

The interesting phenotypic changes with respect to expression of the TL antigen have made it perhaps the most useful marker for distinction between immature and mature T cells. Thus, presence of the TL antigen marker on a lymphocyte of a positive strain clearly marks that cell as an immature, nonfunctional thymocyte, whereas development into mature functioning T cells is accompanied by loss of detectable TL antigen. It is unlikely, however, that expression of the TL product itself is involved in some critical early step in establishing T cell function since the absence of the TL antigen expression in those mouse strains of the Tlab haplotype is not associated with any demonstrable T cell dysfunction. It is of interest, however, that those strains which fail to express the TL antigen during their thymocyte stage of development do compensate by expressing quantitatively more H-2d-region antigen.

In mice of the Tlaa and Tlac haplotypes, whose thymocytes display the TL antigen, there is a distinct absence of TL on bone marrow cells of such mice (232). Nevertheless, in TL+ mice some bone marrow lymphocytes possess the genetic information for synthesis of the TL antigen, as evidenced by the capacity of such cells to express the TL antigen upon entering the thymus of lethally irradiated recipients into which they have been transferred (1824). Moreover, only those bone marrow cells which migrate to the thymus of such recipients will express the TL antigens; bone marrow cells which settle out in either spleen or bone marrow of the adoptive recipients fail to express the TL antigen (1824). These results clearly implicate a crucial aspect of the thymic environment in the differentiation of these cells to express the TL antigen. It is pertinent to note that the influence of the thymus is not genetically conditioned in the sense that only TL+ strains possess a suitable thymic environment. Thus, Schlesinger et al. (1824) were able to show that bone marrow cells from TL+ mice were capable of acquiring the TL antigen upon migration to the thymus of irradiated TL− strain recipients. Moreover, bone marrow cells from TL″ mice failed to express the TL antigen, even within the thymic environment of TL+ irradiated recipients. Further support for the interpretation that acquisition of the TL antigen represents an event that, indeed, transpires within the thymic environment comes from the demonstration that prior exposure of bone marrow cells to anti-TL antibody plus complement either in vitro or in vivo failed to prevent the development of TL+ progeny from such bone marrow cells when transferred into irradiated recipients (232,1824).

It should be noted here that, as will be discussed later, phenotypic expression of the TL antigen, or other T cell antigens discussed here, is not absolutely dependent upon a thymic environment, since it has clearly been demonstrated that expression of such antigens can be induced in precursor cells with appropriate methods in vitro (1165,1166).

The TL+ thymocytes exist as a population of relatively small cells located in the thymic cortex and comprising the bulk of the total thymocyte population (1659,1660,1818). These same cells have a rather rich density of both the θ and Ly antigens, and also express H-2 and GIX antigens (see below). The MSLA antigen is also present on the thymocyte at this stage of differentiation. The simultaneous phenotypic expression of the Ly, θ, and TL antigens have been shown in thymocytes of irradiation chimeras (236) and in thymus grafts (1538,1817).

Further maturation and differentiation appears to result in loss of the TL antigen as the thymocyte moves from the cortical area into the thymic medulla. These cells, now medium in size, represent only around 5-10% of the total thymocyte population and are generally believed to be the immediate progenitors of the functionally mature peripheral T lymphocyte. The evidence for this is essentially circumstantial and is based upon the following observations. By injecting whole thymocyte suspensions from TL+ donor mice into lethally irradiated TL″ recipients, Lance et al. (1199) demonstrated that TL+ cells which could be detected transiently in the spleens and lymph nodes of such recipients rapidly changed into TL″ populations. Moreover, elimination of TL+ thymocytes by exposure to anti-TL antibody and complement in vitro leaves a TL″ subpopulation of thymocytes which is more efficient in recirculating and taking up residence in peripheral lymph nodes than the total thymocyte population (1659). The same population of TL″ thymocytes has also been shown to be functionally more active in terms of eliciting graft-versus-host reactions (1218). Additional circumstantial evidence stems from the demonstration that administration of cortisone to mice eliminates essentially the TL+ thymocyte population (1819), leaving behind a small subpopulation of medium size lymphocytes which are TL− and considerably more efficent in terms of mature T cell function (45,204,359) as well as in their recirculatory capacities (1198).

It should be pointed out that there is no direct evidence for the sequence suggested above, namely, that immature TL+ cells mature into a TL− cell that then migrates to the periphery. Thus, arguments have been made for the possibility that a TL− precursor cell remains TL− as functional maturation occurs [see Schlesinger (1818) for discussion]. It is, nevertheless, clear that TL+ cells can be derived from TL− precursors as demonstrated by the in vitro experiments described above, and it is most probable that the maturation sequence TL+ → TL− is correct. Further discussion on ontogenic relationships of TL antigens with other T cell markers and maturation of T cell function will be presented in Chapter V.

c Modulation of TL Antigens

A most interesting phenomenon, first described with the TL antigen system, is that of antigenic modulation. This phenomenon was discovered when it was observed that TL+ leukemias, which had been passaged in isogenic TL− mice which had been previously immunized against the TL antigen, lost their sensitivity to cytolysis by anti-TL antibodies plus complement in vitro, as well as their ability to absorb cytotoxic TL antibodies (230)—in other words, they had become phenotypically TL−. Further passage of such tumor cells into unimmunized hosts resulted in a complete reappearance of the TL antigen on such tumor cells. An elegant study of this phenomenon was reported by Boyse et al. (233) and demonstrated that antigenic modulation could be induced in both TL+ leukemia cells as well as normal TL+ thymocytes by in vivo passive transfer of anti-TL antiserum. Moreover, they were able to demonstrate the modulation of TL antigen in vitro by exposing leukemia cells or TL+ thymocytes to anti-TL antibodies, although, in this instance, differences were observed in which the kinetics of modulation were much faster with the rapidly turning-over leukemia cells than with thymocytes. Evidence was also presented for the requirement of active metabolism in the process of antigenic modulation, and appropriate controls were conducted to rule out the possibility that modulation was a result of masking of TL sites by blocking antibody (233). The process does not require the presence of complement (233) and can be induced by exposure of TL+ cells to monovalent Fab antibody fragments (1197). This latter point raises a crucial difference between the phenomenon of antigenic modulation as described for the TL system and that of antibody-induced antigen redistribution, extensively studied with surface immunoglobulins (Ig) and commonly referred to as capping (see Chapter VI). Thus, one could conclude that antigenic modulation of TL antigen does not require capping in the sense of the latter phenomenon.

This phenomenon was further explored as a means to delineate the topographical relationship of the cell surface antigens on the surface membrane (233,1197). In such studies it was found that antiserum directed to specificity Tla.3 would modulate not only Tla.3 but also Tla. 1 and Tla.2 The same was found to be true with an antiserum specific for Tla. 1; however, antibodies specific for Tla.2 were unable to modulate any of the Tla specificities and, moreover, appeared to block modulation of other Tla sites by anti-TL antiserum (237,1516). These results were interpreted to imply the close spatial relationship of the three Tla antigenic sites on the cell surface, resulting in a mutual interference among the respective antibodies. The precise explanation for the comodulation phenomenon is not at all clear, since the possibility that the TL antigens are carried by one molecule or membrane fragment is not consistent with the multiple gene locus hypothesis of the Tla complex.

Further studies designed to elucidate the interesting phenomenology of antigenic modulation in the TL system have been conducted in the last 2 years. Yu and Cohen (2286) studied the metabolism of TL antigens of ASL-1 leukemia cells (H-2a, Tla. 1,2,3) both in the absence and in the presence of anti-TL antiserum (i.e., nonmodulating versus modulating TL). In their system, TL antigens were prelabeled with radioactive amino acids by internal labeling techniques or by ¹²⁵I using surface labeling techniques. Yu and Cohen demonstrated that in both nonmodulating and in modulating cells, TL antigens were released spontaneously into the culture medium in equal quantities and, moreover, the biosynthesis of TL antigens was indistinguishable between the two cell types. Nevertheless, TL antigens prelabeled with radioisotope were found to disappear more rapidly from ASL-1 cells undergoing modulation than from the control nonmodulating cells. In an attempt to determine the fate of TL antigens during the course of modulation, immunofluorescent techniques were used to analyze the cell surface of modulating cells for the presence of residual TL antigen. Their observations, over a 10-20-hour incubation period, indicated that no evidence of cap formation could be detected, and the quantity of TL antibodies bound to the cell surface appeared to decrease with time during each incubation. However, when fresh anti-TL antiserum was added to such cells, followed by indirect immunofluorescent analysis, Yu and Cohen could observe detectable staining throughout the period of incubation, an observation which they interpreted to indicate the existence of contaminating antibodies reacting with non-TL antigens within the antiserum employed. It is worth noting that these investigators also analyzed the effects of TL modulation on the metabolism of H-2 antigens and failed to detect any appreciable consequence on the biosynthesis and metabolism of H-2. Thus, Yu and Cohen concluded that modulation of the TL antigens of ASL-1 cells reflected a faster rate of degradation of the TL antigens resulting in a gradual loss of the antigens from the external membranes of the cells (2286).

In a subsequent study, Liang and Cohen (1250) analyzed the phenomenon of antigenic modulation using a somatic hybrid of TL+ and TL− cells. Using Sendai virus, a somatic hybrid of the ASL-1 leukemia cells (H-2a, Tlaa, θ+) and a fibroblast line, LM (H-2k, TL−, θ−) was created. Such cells were shown to possess a hybrid karyotype and expressed H-2 antigens of both parental cells in approximately equivalent amounts. Using a variety of criteria, Liang and Cohen demonstrated that such hybrid cells possessed TL antigens but lacked the θ antigen. The provocative observation was that such cells, when exposed to anti-TL antiserum, failed to modulate the surface TL antigens in a normal manner. Thus, using two different anti-TL antisera, specific for Tla. 1,2,3 or Tla. 1,3, each of which was capable of inducing complete modulation of TL antigens on the ASL-1 parental cells within 10 hours, these investigators failed to modulate TL antigens on the hybrid cells despite incubation times as long as 30 hours in the presence of excess concentrations of the anti-TL antibodies. This was evidenced by the fact that, at the end of the incubation period, such cells were still susceptible to lysis by appropriate anti-TL antisera plus complement. Resistance to antigenic modulation was even observed when these investigators attempted to sandwich the TL antibodies with rabbit anti-mouse Ig antisera (1250).

These observations are extremely important in that the data indicate that the genetic mechanisms controlling the expression of TL antigens are quite distinct from those events responsible for their susceptibility to undergo modulation. Moreover, these findings imply that a cellular control mechanism more complicated than that known to be responsible for membrane antigen redistribution (capping) as a result of formation of antigen-antibody complexes on cell surfaces may be involved in the TL modulation phenomenon. Indeed, Liang and Cohen found that hybrid cells which had been exposed to the sandwich technique in attempts to modulate TL antigen demonstrated the presence of capping without endocytosis of the complexes on such cells as evidenced by binding of fluorescein-conjugated anti-mouse Ig.

Recently, Stackpole et al. (1964) examined modulation of TL antigens from mouse leukemia cells and thymocytes by cytotoxicity, immunofluorescence, and immunoelectron microscopy. These investigators reported that a considerable amount of antibody remained bound to the cell surface after modulation; in the case of bivalent antibody, patches and caps were observed to be displaced topographically, and, interestingly, such capping usually occurred over the pole of the cell opposite from the Golgi region where it occurs under normal capping circumstances. Some of the antibody was internalized, presumably by pinocytosis, but the bulk remained on the cell surface. When monovalent antibody was used, modulation occurred in the absence of patch or cap formation, thus confirming the previous notions of the lack of requirement for topographical redistribution of TL antigen–antibody complexes for modulation to occur.

Stackpole et al. (1964) observed an interesting difference between the sensitivity of modulated cells to rabbit complement versus guinea pig complement. Thus, tumor cells, which had undergone modulation in the classical sense as manifested by insensitivity to complement (guinea pig)-mediated cytotoxicity, were found to be sensitive to cytotoxicity by anti-TL antiserum when appropriately absorbed rabbit complement was employed. Moreover, these investigators stated that a heat-labile factor in the mouse anti-TL alloantisera was responsible for inhibiting the effects of guinea pig complement under normal circumstances, since it was possible to use guinea pig complement with anti-TL alloantisera (which had not been previously heat-inactivated) and demonstrate cytotoxicity on modulated tumor cells. It was concluded, therefore, that cells which have undergone modulation still possess TL antigens on their surface (1964).

The correctness of this last interpretation must be weighed in light of the aforementioned findings of other investigators which illustrate functional evidence for the absence of TL antigens on modulated cells. Is it possible, for example, to account for the very early studies of Boyse et al. (230), which demonstrated the loss of TL immunogenicity of TL+ tumor cells when passaged in hyperimmunized anti-TL antibody-producing TL− recipients, without invoking an antigen-antibody complex-mediated blocking phenomenon that would inhibit the response of such hosts? Perhaps more difficult to reconcile are the observations of Yu and Cohen (2286) who, using sensitive techniques for isolation of radiolabeled surface membrane TL antigens, demonstrated that the quantity of such antigens isolated by immunoprecipitation from modulated cells was significantly less than the corresponding amount detected in immunoprecipitates of nonmodulating control cells. Resolution of these differences will perhaps be forthcoming when further studies are conducted to elucidate the mechanisms involved in in vitro modulation. In this context, it may be perhaps pertinent to point out the observations of Lerner et al. (1234) in which the expression of Moloney leukemia virus on the surface of a viral-induced lymphoma cell, availability of the virus to anti-viral antibody, and the nature and extent of activation of the complement system during the cell cycle were studied in vitro. In these experiments, it was found that although viral antigen was present on the cell surface, making it accessible to antibody and capable of activating complement throughout all of the cellular growth phases, the susceptibility of cells to cytotoxicity by antibody-mediated complement activity was confined to the G1 phase of cell growth. During the logarithmic phase of cell growth, such cells, although expressing viral antigen capable of binding antibody and complement, were not themselves susceptible to being lysed by this mechanism (1234).

d Associations between TL and H-2 Antigens

Based on their earlier genetic studies demonstrating a close linkage between the Tla locus and the D end of the H-2 gene complex (231), Boyse et al. investigated the relationship of the quantity of surface H-2 antigens to the presence or absence of TL antigen (233). This was done using the C57BL/6 (H-2b) strain which is normally a TL″ (haplotype Tlab) and a congenic C57BL/6:TL+ strain derived by appropriate backcross breeding of a recombinant containing the Tlaa haplotype originating from strain A mice (H-2a). The latter congenic strain, therefore, is haplotype H-2b, Tlaa. Analysis with the appropriate anti-H-2b antisera demonstrated that the thymus cells of this congenic strain contained approximately 50% of the quantity of H-2b alloantigen as could be detected in the parent C57BL/6 strain (233). The conclusion that this decrease in surface H-2b antigen was, indeed, related to the presence of the TL antigen on such thymocytes was substantiated by the observation that lymph node lymphocytes of such congenic mice, which lack the TL antigen on their surface, contained equal quantities of H-2b alloantigen as could be found on lymph node lymphocytes of the parent C57BL/6,TL− strain. It was also demonstrated that the difference in content of H-2 antigen related to the TL phenotype did not extend to the θ isoantigen, which is not linked to the H-2/Tla complex. It is also noteworthy that depression of H-2 concentration occurred on cells of either of the two TL+ phenotypes, Tlaa or Tlac, indicating that the effects of Tla.2 alone on H-2 are similar to those of Tla. 1,2,3 (233). Moreover, antigenic modulation by anti-TL antibody causing loss of TL antigen from the cell surface resulted also in a compensatory increase in H-2 antigenic concentration (233,1197). These findings suggest that whatever the explanation for this interesting TL : H-2 interaction, it is the TL phenotype, rather than the Tla genotype, which is responsible for the influence on H-2.

Further analysis of the interesting association of TL and H-2 on the cell surface demonstrated the following interesting points (233). First, the presence of TL antigen on TL+ cells resulted in reduction of the H-2D antigens on such cells, but had no effect on antigens coded by the K end of H-2. Second, in a comparison between H-2 homozygous and H-2 heterozygous cell populations, the degree of suppression of H-2D antigens by TL was considerably greater in the H-2 heterozygote than in H-2 homozygotes; however, Boyse et al. concluded that the latter result probably reflected a technical artifact due to a somewhat lower absorptive capacity of H-2 heterozygote cells. Third, it was shown that H-2 is reduced by only half as much in TL heterozygous thymocytes as in TL homozygous thymocytes. Finally, the reduction of H-2 was similar for both cis- and trans-positions of Tla in relation to H-2 (233).

In addition, this investigation revealed an interesting extrachromosomal interaction between products of the Tla alleles (233). Thus, thymocytes of the progeny of Tla. 1,2,3 × Tla. 1,2,3 mating (haplotype Tlaa/Tlaa) exhibited double the amount of Tla. 1 and Tla.3 antigens present as could be detected in thymocytes of Tla. 1,2,3 × TL− progeny (haplotype Tlaa/Tlab), an observation consistent with the finding that the quantity of H-2 antigen in H-2 homozygotes is about double that in H-2 heterozygotes. The important implication of this finding, however, is that the inert Tla. 1 structural gene of Tl− mice is not activated by the presence of the Tlaa allele on the accompanying chromosome in the trans-position. On the other hand, thymocytes of the progeny of Tla. 1,2,3 × Tla.2 matings (haplotype Tlaa/Tlac) exhibited comparable quantities of Tla.1 and Tla.3 antigens, (but not of Tla.2 antigen) equal to that observed in homozygote Tlaa mice. This suggests that the Tla allele of Tla.2 strains can influence other Tla genes in the trans-position. The second example of interrelated Tla gene functions was their observation that thymocytes of homozygous Tla. 1,2,3/Tla. 1,2,3 individuals express a larger quantity of Tla.2 antigen than could be found on Tla.2/Tla.2 homozygous thymocytes (233). Boyse et al. concluded that such interactions apparently involve control of TL antigen synthesis within the cells.

e Biochemical Studies of TL Antigens

Attempts to define a chemical association between TL and H-2 by Davies et al. (453) demonstrated that, although TL activity was found in association with H-2 antigen activity in solubilized cell membrane preparations, the TL specificities could be clearly separated from H-3 antigen by chromatography on DEAE-Sephadex. More recently, conclusive demonstrations of this point were reported by Vitetta et al. (2150), Muramatsu et al. (1457), and Yu and Cohen (2285), using newer and more sophisticated immunobiochemical techniques. Vitetta et al. (2150) radioiodinated cell surface proteins of spleen and thymus cells with ¹²⁵I (external radiolabeling) and solubilized the surface membranes by lysis in a nonionic detergent. Specific anti-TL and anti-H-2 antisera were used for immunoprecipitation of the respective antigens which were then analyzed by polyacrylamide gel electrophoresis in sodium dodecyl sulfate (SDS). Using this approach, Vitetta et al. (2150) demonstrated that anti-TL antiserum precipitates a protein of molecular size 40,000-50,000 daltons from TL+ thymocytes, but not spleen cells, of TL+ mice. Specific anti-H-2 antiserum, however, precipitated H-2 antigens, also of the molecular size 40,000-50,000 daltons, from normal mouse spleen cells. Since the antisera employed in these studies were made in congenic strains of mice and, therefore, contained only anti-H-2 or anti-TL antibodies, respectively, the conclusion was reached that these results distinguished between the H-2 and TL antigenic molecular species (2150).

A more convincing demonstration of the nonassociation of TL and H-2 antigens was obtained by Yu and Cohen (2285), who utilized the same immunobiochemical techniques to isolate TL and H-2 antigens from ASL-1 leukemia cells (H-2a; Tlaa). Solubilized membranes from either internally (biosynthetically) or externally radiolabeled tumor cells were studied with appropriate anti-H-2 and anti-TL antisera. Using a sequential precipitation procedure, Yu and Cohen demonstrated that the complete removal of TL antigens by immunoprecipitation with excess anti-TL antisera had little effect upon subsequent immunoprecipitation of H-2a antigens by the appropriate antiserum. They also demonstrated a similarity in molecular size of H-2 and TL, i.e., approximately 40,000-45,000 daltons. Muramatsu et al. (1457) analyzed the TL antigens extracted from TL+ leukemia cells, which had been internally radiolabeled with radioactive [³H]amino acids and [¹⁴C]fucose, by solubilization of the membrane with papain and then isolated by indirect immunoprecipitation (1839). In this way, they were able to isolate the TL antigen as a doubly labeled glycoprotein fragment, which by chromotography on Sephadex G-150 in SDS was found to have approximately the same molecular weight (40,000-45,000 daltons) as H-2 antigens from the same cell. These chemical similarities in terms of glycoprotein nature and molecular size had been reported previously by Davies et al. (452,453). However, by extensive pronase digestion, Muramatsu et al. (1457) were able to demonstrate that the glycopeptide obtained from TL antigen (4500 daltons) differed in molecular size, as compared with the glycopeptide of H-2 (3500 daltons).

The most recent studies on the biochemistry of the TL antigens have demonstrated that these, like the H-2 alloantigens (see Chapter IV), have a subunit structure consisting of one heavy molecular weight entity of around 40,000-50,000 daltons and a smaller molecular weight subunit of around 12,000 daltons, which has been identified as β2-microglobulin (β2m) (1534,2158). Independent studies reported almost simultaneously by Vitetta et al. (2158) and Ostberg et al. (1534), using the technique of surface radiolabeling combined with detergent solubilization, immunoprecipitation, and SDS-polyaerylamide gel electrophoresis, confirmed earlier observations (1457,1839,2150,2285) that both H-2 and TL antigens exhibit similar molecular weight peaks of approximately 45,000 daltons. Furthermore, by extending their fraction analysis of the gels, both groups of investigators were able to detect a small molecular weight subunit of 12,000 daltons associated with both H-2 and TL antigens (1534,2158). The identity of the subunit structure in TL and H-2 antigens was demonstrated by Vitetta et al. by mixing specific precipitates of these two antigens before gel electrophoresis and observing that the 45,000 and 12,000 subunits from both H-2 and TL coelectrophoresed (2158). In addition, Ostberg et al. (1534) used an immunoadsorbent technique in which anti-human β2-microglobulin antibodies were used to react with thymus cell surface antigens, and demonstrated that the material bound to and eluted from such an immunoadsorbent contained both TL and H-2 antigens. Formal proof that