Advances in Immunology Volume 72

ADVANCES IN

Immunology VOLUME 72

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

VOLUME 72

ACADEMIC PRESS San Diego London Boston New York Sydney Tokyo Toronto

This book is printed on acid-free paper.

@

Copyright 0 1999 by ACADEMIC PRESS 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. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher's consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-1999 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0065-2776/99 $30.00

Academic Press a division of Harcourt Brace & Company 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.apnet.com

Academic Press 24-28 Oval Road, London NW1 7DX, UK

http://www.hbuk.co.uWap/

International Standard Book Number: 0-12-022472-0 PRINTED IN THE UNITED STATES OF AMERICA 98 99 0 0 0 1 02 03 EB 9 8 7 6

5

4

3 2 1

CONTENTS

ix

CONTRIBUTORS

The Function of Small GTPases in Signaling by Immune Recognition and Other Leukocyte Receptors

AMNONALTMANAND MARCELDECKERT

I. Introduction 11. The Function of Ras in IRR Signaling 111. The Function of Rho-Family GTPases in IRR Signaling IV. CD28 Signaling in T Cells: The Roles of Small GTPases V. The Function of Rab GTPases in Leukocytes VI. Small GTPases and Aberrant Leukoc e Functions VII. The Role of Small GTPases in Lymp ocyte Development References

i:

1 3 25 49 52 53 64 70

Function of the CD3 Subunits of the Pre-TCR and TCR Complexes during T Cell Development

BERNARD MALISSEN, LAURENCEARDOUIN,SHIH-YAO LIN,ANNEGILLET,AND MARIEMALISSEN I. Introduction

11. Mouse cup T Cell Development 111. The Pre-TCR Sensor

IV. The ap TCR Sensor V. Raison dEtre of the Pre-TCR VI. Are the Roles o f the Pre-TCR and TCR Complexes Limited to Ensure Cell Survival? VII. The Limits of Genetic Analysis: Redundancy and Adaptive Response to Certain Mutations VIII. Conclusions References V

103 106 109 128 131 131 138 140 141

vi

CONTENTS

Inhibitory Pathways Triggered by ITIM-ContainingReceptors

SILVIABOLLANDAND JEFFREY V. RAVETCH I. 11. 111. IV. V. VI. VII.

149 150 154 158 161 163 165 166

Introduction Inhibitory Receptors and Activatin Counterparts FcyRII-Mediated Inhibitory Signa Mechanism of Inhibition by SHIP Inhibition by KIR Receptors Lessons from SHIP, SHP-1, and SHP-2 Knockout Mice Conclusions References

P

ATM in Lymphoid Development and Tumorigenesis

YANGXu

I. Lymphoid Defects and Tumorigenesis in Ataxia-Telangiectasia Patients 11. The ATM Gene 111. Dissecting the Lymphoid Defects and Tumorigenesis in A-T Mouse Models IV. Future Perspectives References

179 180 181 185 186

Comparison of Intact Antibody Structures and the Implications for Effector Function

LISAJ. HARRIS, STEVEN B. LARSON, AND ALEXANDERMCPHERSON 191 192 193 196 198 199 203 205

I. Introduction 11. Hinge-Deleted Dob and Mcg 111. Partial Structure of Kol IV. Mab231 V. Mab61.1.3 VI. Biolo 'cal Implications and Effector Functions VII. Conc uding Remarks References

?i

Lymphocyte Trafficking and Regional Immunity

EUGENE c. BUTCHER, MARNA WILLIAMS, KENNETHYOUNGMAN, MICHAEL BRISKIN

LUSIJAH ROTI',

AND

I. Introduction 11. Interaction of Blood Lymphocytes with Endothelium Involves Multiple Steps That Control the Specificity of Lymphocyte Recruitment

209 211

CONTENTS

111. Molecules Involved in Lymphocyte Interactions with Intestinal Endothelium IV. Subset S ecificity and Mechanisms of Lymphocyte Targeting to IntestinaY Tissues in Vivo V. a4P7 and the Segregation of Intestinal from Systemic Memory VI. Clinical Sipficance and Therapeutic Opportunities References

vii

213 229 237 24 1 243

Dendritic Cells

DIANA BELL,JAMES W. YOUNG, AND JACQUES BANCHEREAU I. Introduction 11. Features of Dendritic Cells 111. Ontogeny of Dendritic Cells IV. Maturation of Dendritic Cells V. Interactions of Dendritic Cells with T Cells VI. Interactions of Dendritic Cells with B Lymphocytes VII. Dendritic Cells in Clinical Disease States VIII. Concluding Remarks References

255 257 274 279 28 1 285 29 1 305 305

lntegrins in the Immune System

Yojr SKIMIZU, DAWDM. ROSE,AND MARKH. GINSBERG Introduction Ligand-Bindin Sites in Integrins Integrin Ligan s in the Immune System Integrin Signaling Integrin Function and the Immune System Integrins in Lymphocyte Recirculation The Role of Integrins in Immune Responses and Inflammation: Two Case Studies VIII. Re ulation of Integrin Ligand Expression in Inflammation Re erences

325 325 330 335 337 344

INDEX CONTENTS OF RECENTVOLUMES

381 389

I. 11. III. IV. V. VI. VII.

5!

f

348 351 353

This Page Intentionally Left Blank

CONTRIBUTORS

Nunhers in parentheses indicate the pages on which the authors’ contributions begin

Amnon Altman (l),Division of Cell Biology, La Jolla Institute for Allergy and Immunology, San Diego, California 92121 Laurence Ardouin (103), Centre d’Immunologie, INSERM-CNRS de Marseille-Luminy, 13288 Marseille Cedex 9, France Jacques Banchereau (255), Baylor Institute for Immunology Research, Saminons Cancer Center, Dallas, Texas 75204 Dianas Bell (255), Baylor Institute for Immunology Research, Sammons Cancer Center, Dallas, Texas 75204 Silvia Bolland (149), The Rockefeller University, New York City, New York 10021 Michael Briskin (209),LeukoSite, Inc., Cambridge, Massachusetts 02142 Eugene C. Butcher (209), Laboratory of Immunology and Vascular Biology, Department of Pathology, Stanford University School of Medicine, Stanford, California 94305; and Center for Molecular Biology and Medicine, Veterans Affairs Palo Alto Health Care System, Palo Alto, California 94304 Marcel Deckert (l),Unite INSERM 343, HGpital de I’Archet, 06202 Nice Cedex 3, France Anne G a e t (103),Centre d’Immunologie, INSERM-CNRS de MarseilleLuminy, 13288 Marseille Cedex 9, France Mark H. Ginsberg (325), Department of Vascular Biology, Scripps Research Institute, La Jolla, California 92037 Lisa J. Harris (191), Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California 92697-3900 Steven B. Larson (191), Department of Molecular Biology and Biochernistry, University of California, Inine, Irvine, California 92697-3900 Shih-YaoLin(103),Centre dInimunologie, INSERM-CNRS de MarseilleLuminy, 13288 Marseille Cedex 9, France Bernard M a h e n (103), Centre d’Immunologie, INSERM-CNRS de Marseille-Luminy, 13288 Marseille Cedex 9, France ix

X

CONTRIBUTORS

Marie Malissen (103), Centre dImmunologie, INSERM-CNRS de Marseille-Luminy, 13288 Marseille Cedex 9, France Alexander McPherson (191), Department of Molecular Biology and Biochemistry, University of California, Irvine, Irvine, California 92697-3900 Jeffrey V. Ravetch (149), The Rockefeller University, New York City, New York 10021 David M. Rose (325), Department of Vascular Biology, Scripps Research Institute, La Jolla, California 92037 Lusijah Rott (209), Laboratory of Immunology and Vascular Biology, Department of Pathology, Stanford university School of Medicine, Stanford, California 94305; and Center for Molecular Biology and Medicine, Veterans Affairs Palo Alto Health Care System, Palo Alto, California 94304 Yoji Shimizu (325), Department of Laboratory Medicine and Pathology, Center for Immunology, Cancer Center, University of Minnesota Medical School, Minneapolis, Minnesota 55455-0392 Mama Williams (209), Laboratory of Immunology and Vascular Biology, Department of Pathology, Stanford University School of Medicine, Stanford, California 94305; and Center for Molecular Biology and Medicine, Veterans Affairs Palo Alto Health Care System, Palo Alto, California 94304 Yang Xu (179), Department of Biology, University of California, San Diego, La Jolla, California 92093-0322 JamesW. Young (255),Memorial Sloan Kettering Cancer Center, Cornell University Medical College, New York City, New York 10021 Kenneth Youngman (209), Laboratory of Immunology and Vascular Biology, Department of Pathology, Stanford University School of Medicine, Stanford, California 94305; and Center for Molecular Biology and Medicine, Veterans Affairs Palo Alto Health Care System, Palo Alto, California 94304

ADVANCES IN IMMUNOLOGY, VOL. 72

The Function of Small GTPases in Signaling by Immune Recognition and Other Leukocyte Receptors AMNON ALTMAN AND MARCEL DECKERT' Division of Cell Biology, Irr hllo Institub for A l l e w o d Immunology, San Diego, California 92 121; and 'Uniie INSEM 343, M p i d de I'Archet, 06202 Nice Cedex 3, France

1. Introduction

A. THEFUNCTION AND REGULATIONOF RASSUPERFAMILY GTPA~ES This article focuses on the role and regulation of small GTP-binding (G) proteins in leukocyte functions. The complex and rapidly evolving field of small G proteins has been extensively reviewed (Boguski and McCormick, 1993). Therefore, only a brief discussion of these proteins and their regulation is provided here in order to set the stage for the following review. The ras genes were first identified as the transforming principle of the Harvey and Kirsten strains of rat sarcoma viruses, and subsequently described as sites of somatic mutations in some 30% of human tumors. The Ras superfamily consists of -60 mammalian genes encoding small (20-29 kDa) membrane-associated G proteins. Their highly conserved structure in evolution, from yeast to humans, suggests basic and critical functions during development, growth, and differentiation. Despite their large number and ability to elicit diverse sets of responses in distinct cell types, the small GTPases all share two biochemical properties, i.e., guanine nucleotide binding and intrinsic GTPase activity. The products of this gene superfamily have been divided into several families. The major ones are the Ras, Rho, and Rab families, members of which play major roles in regulating cellular growth and differentiation, organization of the actin cytoskeleton, and intracellular membrane and vesicle trafficking, respectively. The other two families are Ran and Ad. GTPases cycle between an inactive, GDP-bound state and an active, GTPbound form. The activity of small GTPases is determined by the ratio of these two forms. In resting cells, GTPases are mostly inactive. When cells are stimulated via different receptors, including tyrosine kinase receptors, immune recognition receptors (see below), and cytokine receptors, bound GDPis exchanged for GTP, shiftingthe GTPaseto its active state. Oncogenic GTPases, the result of mutations at several residues, are constitutivelyin the GTP-bound state, because they are either not subject to negative regulation or undergo an accelerated rate of GDP dissociation that results in GTP binding. The activity of small GTPases is regulated by three functional classes of 1

Lopynglit 0 19Y9 hy Arddernic Press

. d I nghts of reprdorhon in m y tom

reserved 0065 277WSY $30 00

2

AMNON ALTMAN AND MARCEL DECKERT

proteins. First, hydrolysis of bound GTP is accelerated by GTPase-activating proteins (GAPS)that inactivate the GTPases. Conversely,exchange of bound GDP for GTP is mediated by guanine nucleotide exchange factors (GEFs), resultingin GTPase activation.A third group ofproteins, the guanine nucleotide dissociation inhibitors (GDIs),maintain some small GTPases in an inactive or active state by inhibitingthe dissociation of GDP or GTP, respectively, and/or the basal or GAP-stimulated GTPase activity. The most significant advances in the broad area of small GTPases and their role in signal transduction are probably represented by four sets of findings. First, small GTPases activate distinct serinelthreonine kinase cascades, known as mitogen-activated protein kinases (MAPKs), that ultimately lead to gene transcription (Treisman, 1996); second, “cross talk,” with important functional implications,occurs among Ras and Rho proteins (Khosravi Far et al., 1995; Prendergast et al., 1995; Qiu et al., 1995a,b, 1997); third, in addition to their “traditional” role as regulators of the cytoskeleton, Rho-family GTPases also regulate growth-signalingcascades that control cellular proliferation and/or programmed cell death (Hill and Treisman, 1995; Symons, 1995; Treisman, 1996); and, finally, small GTPases are each coupled to multiple effectors, thereby generating a large diversity in the potential outcomes of their activation (White et al., 1995; Joneson et al., 1996; Lamarche et al., 1996; Westwick et al., 1997). B. SIGNALTRANSDUCTION BY IMMUNE RECOGNITION RECEPTORS The family of immune recognition receptors ( IRRs) includes the antigenspecific T and B cell receptors (TCRs and BCRs, respectively) and receptors for the Fc fragment of immunoglobulin. IRRs evolved with unique strategies to transmit activation signals. They all constitute protein complexes consisting of 3-10 subunits. The tasks of ligand binding and signal transduction are divided among distinct subunits. In T cells, for example, the signaling subunits comprise the y , 6, and E chains of the CD3 complex and the TCR-associated 6 chain. The cytoplasmic tails of the signaling subunits in all IRRs share a motif, i.e., the immunoreceptor tyrosine-based activation motif (ITAM), critical for signal transduction. Based on current genetic and biochemical studies, it is believed that IRR-initiated hematopoietic cell activation results from the sequential activation of protein tyrosine kinases (PTKs) of the Src and Syk families, and their interaction with phosphorylated ITAMs. Src-family kinases (Lck and/or Fyn in T cells) are thought to induce the early tyrosine phosphorylation of the ITAMs, which leads to the recruitment and subsequent activation of Syk-family PTKs (Zap-70 and Syk) via their tandem Src-homology 2 (SH2) domains. The activation of receptor-coupled PTKs represents an obligatory and early event in IRR-mediated signaling cascades. The activated PTKs induce

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

3

increased cellular tyrosine phosphorylation, thereby modulating the activity and/or cellular localization of various enzymes and adapter proteins, and activating several signaling pathways, including phospholipase Cy (PLCy), various small GTPases, phosphatidylinositol 3-kinase (PIS-K), and cytoplasmic serine/threonine kinases such as MAPKs, which are divided into three families: extracellular-regulated kinases (ERKs), c-Jun N-terminal kinases (JNKs), and p38 kinases. This results in gene transcription, differentiation, and cellular proliferation. In T cells, induction of the interleukin2 (IL-2) gene serves as a hallmark and end point of cellular activation. The properties of IRRs and the pathways they use in order to transduce activation signals have been widely reviewed (Weiss and Littman, 1994; R a i n et al., 1995; Chan and Shaw, 1996; Wange and Samelson, 1996; Alberola-Ila et al., 1997; DeFranco, 1997; Kurosaki, 1997). To some extent, lymphocytes (and T cells in particular) occupy a special place in the extensively studied area of small GTPases and their biological functions. This is so because, historically, the first evidence for physiological Ras regulation came from the study of T cell activation by mitogenic antiCD3 antibodies or phorbol ester (Downward et al., 1990). Since then, stimulation of hematopoietic cells via other IRRs has also been found to activate Ras and other small GTPases, and their known downstream serine/ threonine kinases. The functions of Ras proteins (Cantrell, 1994; Cantrell et al., 1994; Izquierdo Pastor et al., 1995) and, to a lesser extent, of Rho and Rab proteins (Chavrier, 1993; Reif and Cantrell, 1998) in lymphocytes have been reviewed before. Nevertheless, both the rapid progress in this research field and the lack of comprehensive reviews that integrate distinct small GTPase families into a single scheme in the context of IRR-mediated signaling processes prompt us to revisit this fundamental problem. The aim of this review is, therefore, to summarize our current knowledge on the role of small GTPases, primarily of the Ras and Rho families, during lymphocyte activation and development. Although we focus on IRR signaling, the function of small GTPases in signal transduction by other leukocyte receptors is also reviewed, albeit in less detail. Because function and regulation of small GTPases have been analyzed more extensively in T lymphocytes than in any other cell type of the hematopoietic lineage, much of the information presented here is derived from the analysis of T cells. II. The Function of Ras in IRR Signaling

A. RAs ACTIVATIONBY IRRs As pointed out previously, analysis of TCR-stimulated T cells provided the first experimental evidence of physiological Ras activation (Downward et al., 1990). These studies demonstrated that stimulation of human leuke-

4

AMNON ALTMAN A N D MARCEL DECKERT

mic (Jurkat) or peripheral blood T cells with an anti-CD3 monoclonal antibody induced a marked increase in the cellular fraction of active, GTPbound Ras. All three forms of Ras, i.e., H-Ras, K-Ras, and N-Ras, were activated under these conditions. The observation that phorbol ester, a well-known protein kinase C (PKC) activator, can also activate Ras in the same cells suggested a role for PKC, and additional experiments indicated that phorbol ester-mediated Ras activation was due, at least in part, to inhibition of ras-GAP activity (Downward et al., 1990).The reported activation of Ras by the T cell accessory receptor CD2 (Graves et al., 1991) may reflect an effect of TCWCD3 ligation because, first, CD2 usually cannot signal in the absence of a functional TCWCD3 complex in mature T cells (Altman et al., 1990) and, second, a physical association between CD2 and TCWCD3 has been demonstrated (Beyers et al., 1992).Subsequent studies confirmed the ability of anti-CD3 antibodies or phorbol ester to activate Ras in T cells (Franklin et aZ., 1994; Ohtsuka et al., 1996),and demonstrated that BCR ligation by specific antigen, anti-Ig antibodies, or phorbol ester similarly activates Ras in B lymphocytes (Hanvood and Carnbier, 1993; Lazarus et aZ., 1993; Tordai et aZ., 1994; Sarmay et al., 1996; Tridandapani et al., 1997a).FcERI ligation on rat basophilic leukemia (RBL) cells ( Jabril Cuenod et aZ., 1996) or FcyRIII (CD16) cross-linking on natural killer (NK) cells (Galandrini et al., 1996)was also found to stimulate Ras activity. The physiological relevance of TCR-induced Ras activation was established by demonstrating that, first, constitutively active (CA) Ras mutants can synergize with Ca2+ionophore to activate the IL-2 gene in T cells (Baldari et al., 1992b; Rayter et al., 1992; Ohtsuka et aZ., 1996) or the IL3 and granulocyte/macrophage colony-stimulatingfactor (GM-CSF) genes in a mast cell line (Hahn et al., 1991) and, second, transient expression of a dominant-negative (DN) Ras mutant (N17Ras)inhibited TCR- or phorbol ester-induced IL-2 induction (Rayter et al., 1992; Baldari et al., 1993; Ohtsuka et al., 1996). Similarly, introduction of a neutralizing Ras-specific antibody into Jurkat T cells interfered with T cell activation (Werge et al., 1994). These studies indicate that Ras activation is necessary but not sufficient for stimulation of IL-2 production. Later studies have characterized the IL-2 promoter-associated transcription factors that are regulated by Ras in T cells (Section 11,D). The involvement of Ras in cytokine production is also evident from the finding that transformation of a mast cell line by an oncogenic mutant of ras is associated with constitutive IL3 production and an autocrine IL-3 response (Nair et al., 1992). However, it is clear that the IL-2 or other cytokine genes are not the only targets for Ras in hematopoietic cells. Thus, Ras function is also important for up-regulation of activation antigens such as CD69 (D’Ambrosioet al., 1994) and the IL-2 receptor a chain (Sirinian et al., 1993), and for transcription of

CTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

5

the human TCR-P gene (Wotton et al., 1993) in T cells. The targets of Ras are discussed in detail below. Figure 1presents a scheme of the IRRcoupled Ras signaling pathway, using the TCWCD3 complex as the model. A report by Boussiotis et al. (1997) indicated that Rapl, a member of the Ras family that is known to antagonize the activation of Ras (Hata et aZ., 1990; Kitayama et al., 1990), may play an important role in maintaining T cell anergy (Fig. 1). Thus, whereas antigen stimulation of control T cells led to a transient stimulation of Rapl, activated (GTP-bound) Rapl was constitutively present at a high level in anergic T cells, and this level was not affected by antigen stimulation (Boussiotis et al., 1997). The aberrant function of the Ras signaling pathway in anergic lymphocytes is discussed in more detail in Section VI,A. OF RASACTIVATION AND COUPLING TO IRRs B. REGULATION 1. The Role of PTKs versus PKC in Ras Activation The precise mechanism(s) by which IRRs activate Ras remains unclear (Fig. 1).Nevertheless, IRRs follow, at least in part, the paradigm established for growth factor receptors that possess intrinsic tyrosine kinase activity. Combined genetic, biochemical, and pharmacological approaches have established that Ras is activated by these receptors in a PTKdependent manner and that activated Ras couples receptor PTKs to downstream signaling cascades, leading to cellular growth and differentiation (Egan and Weinberg, 1993). Similarly, Ras activation by IRRs depends on intact PTK activity. This is evident from the findings that specific tyrosine ldnase inhibitors can prevent Ras activation induced by ligation of the TCR (Izquierdo et al., 1992a; Ohtsuka et al., 1996) or the BCR (Lazarus et aZ., 1993; Kawauchi et al., 1996). The contribution of PTKs to activation of the Ras pathway in lymphocytes is implicated by several other findings: first, the expression of CA Lck in thymoma cell lines derived from Lcktransgenic mice is associated with constitutive activation of the Ras/Raf1/ERK pathway (Lin et al., 1995); second, transient overexpression of Syk in an Lck-negative variant of Jurkat T cells (JCaM1) enabled the TCW CD3 complex to induce activation of the ERK pathway and the transcription factor nuclear factor of activated T cells (NFAT), which binds to the IL-2 gene promoter. This effect of Syk required its catalyhc activity, and it was blocked by a DN Ras mutant (Williams et al., 1997). As discussed below, adapter proteins, which either function as PTK substrates or bind PTK substrates via their SH2 domain, most likely represent the link between IRR-coupled activated PTKs and Ras. It is clear, however, that, in addition to the ubiquitous tyrosine kinasedependent Ras activation pathway shared by different receptors and cell

6

AMNON ALTMAN AND MARCEL DECKERT

FIG.1. TCR-induced Ras signaling pathways and their interactions with other GTPases. TCR-coupled PTKs include members of the Src (Lck, Fyn)and Syk (Zap-70, Syk) families. The scheme depicts two potential scenarios for Ras activation, i.e., via a LAT/Crb2/Sos complex (right) or via a ShdGrbWSos complex (left). The potential role of a Cbl/Crk/Rapl complex in T cell anergy (Section VI,A) is indicated. Rac is required (but not sufficient) either downstream of, or parallel to, Ras in a pathway leading to NFAT activation (Genot d al., 1996). Rac, calcineurin, and PKCO are involved in JNK activation in T cells (Werlen et aL, 1998; Ghaffari-Tabrizi et al., 1998), but the exact relationship between the three signals is unclear. Vavmac-mediated cytoskeletal reorganization is linked to TCR-derived growth signals (Holsinger et al., 1998). Known PTK substrates in T cells are lightly shaded, small GTPases are darkly shaded, and transcription factors are in black. DAG, Diacylglycerol; IP, inositol phosphates.

types, T and B cells display a Ras activation mechanism that is mediated by PKC. This is evident from the ability of PKC activators such as phorbol myristate acetate (PMA) to activate Ras in T (Downward et al., 1990;

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

7

Franklin et al., 1994; Ohtsuka et al., 1996) or B (Harwood and Cambier, 1993; Tordai et al., 1994) cells. PKC-mediated Ras activation is a lymphocyte-specific response not found in myeloid or mast cells (Downward et al., 1992; Izquierdo et al., 1992a), but this notion may need to be reexamined in view of the very recent finding that PKC can also lead to Ras activation in COS cells (Marais et al., 1998). The nature of the PKC-dependent Ras activation mechanism and its relationship to the IRR-mediated pathway of Ras activation have been addressed in a number of studies, and are only partially understood. As mentioned previously, activated PKC may stimulate Ras by inhibiting the activity of ras-GAP (Downward et al., 1990; Izquierdo et al., 1992a). Because PMA-mediated Ras activation is sensitive to specific PKC inhibitors but is not affected by selective PTK inhibitors (Hanvood and Cambier, 1993; Ohtsuka et al., 1996),PKC-dependent Ras activation does not appear to require tyrosine kinase activity. The use of a selective PKC inhibitor (Ro 31-8425) and transient expression of a CA Ras mutant indicated that PKC does not act downstream of Ras in the induction of NFAT and AP1 transcriptional activity and in the expression of IL-2 in Jurkat T cells (Williamset al., 1995).This conclusion is consistent with two other findings: first, an intracellularly introduced neutralizing anti-Ras antibody inhibited TCR-induced NFAT activation, but was much less efficient in inhibiting NFAT activation induced by direct PKC stimulation (i.e.,by PMA treatment) (Werge et al., 1994);second, Ro 31-8425 prevented ERK2 activation by PMA, but not by CA Ras (Izquierdo et al., 1994b). These data also suggest that PKC may have other effectors in addition to Ras. The contribution of PKC to TCR-mediated Ras activation was tested in permeabilized T cells under conditions in which TCR-induced PKC activation was blocked by using either a PKC pseudosubstrate peptide inhibitor or ionic conditions that prevent inositol phospholipid hydrolysis and, hence, diacylglycerol production and PKC stimulation (Izquierdo et al., 1992a). In the absence of PKC stimulation, TCR ligation still induced Ras activation, which correlated with ras-GAP inactivation. The selective PTK inhibitor, herbimycin A, prevented the non-PKC-mediated, TCRinduced stimulation of RAS (Izquierdo et nl., 1992a). These data suggest that the TCR is coupIed to Ras via a PKC-independent pathway that involves tyrosine kinases. However, another study reported that calphostin C, a specific PKC inhibitor, blocked not only PMA-induced but also TCRmediated accumulation of Ras-GTP and, furthermore, that PKC downregulation by prolonged PMA treatment severely impaired the activation of Ras in response to TCR stimulation (Ohtsuka et al., 1996). These results suggest that the activation of PKC is important for TCR-mediated Ras activation in Jurkat cells. The apparent discrepancy between these two

8

AMNON ALTMAN AND MARCEL D E C K E R T

conclusions may reflect the use of distinct T cells, which may depend to varying degrees on the PKC-mediated Ras activation pathway, i.e., human peripheral blood lymphoblasts (Izquierdo et al., 1992a) versus leukemic Jurkat cells (Ohtsuka et al., 1996). Thus, the functional relevance of PKC in coupling the TCR to Ras activation remains unclear. One potential problem with earlier studies that assessed the contribution of PKC to Ras activation in T cells stems from the fact that PMA was used to activate cellular PKC. Phorbol esters nonselectively stimulate the conventional, Ca2'-dependent (a,/3, y ) and novel, Ca'+-independent (6, E , 7,6, p ) PKC enzymes. Therefore, interpretation of the effects of PMA on Ras activation is compounded because PKC isoforms manifest specificity in terms of their ability to translocate to different cellular compartments, differentiallyphosphorylate cellular substrates, and regulate distinct signaling pathways. Thus, at present it is unknown which, if any, of the T cellexpressed PKC isoforms can activate Ras.

2. The Role of Shc and Grb2/Sos in Bas Activation The mechanisms that couple receptor PTKs to Ras activation have been a subject of intense research, and two coupling mechanisms have been established (Fig. 1).First, the adapter protein Grb2 (Lowenstein et al., 1992) directly binds, via its SH2 domain, to phosphorylated tyrosine residue(s) in the cytoplasmic tail of activated receptor PTKs (Buday and Downward, 1993; Egan et al., 1993; Li et al., 1993; Rozakis Adcock et al., 1993). Second, Grb2 can indirectly associate with growth factor receptors via an intermediate adapter protein, Shc (Pelicci et al., 1992). Shc contains an SH2 domain that binds directly to the phosphorylated receptor, and is itself a substrate of receptor and nonreceptor PTKs. Phosphorylated tyrosine residue(s) of Shc bind the SH2 domain of Grb2 (Rozakis Adcock et al., 1992; Skolnik et al., 1993a,b; de Vries Smits et al., 1995; Pronk et al., 1994). In either case, the recruitment of Grb2 to the activated receptor allows Ras activation, because Grb2 is constitutively associated with the ubiquitous Ras GEF, Sos (Chardin et al., 1993), thereby localizing Sos to the membrane where its target (i.e.,Ras) is localized. The finding in one (Graziadei et al., 1990),but not another (Boyeret al., 1994),study that BCR cross-linking induces cocapping of Ras with the BCR indirectly supports the existence of such a recruitment mechanism in lymphocytes. The notion that targeting of Sos to the plasma membrane in the vicinity of Ras is a primary mechanism leading to Ras activation is supported by the finding that enforced localization of Sos to the plasma membrane via the addition of membrane-targeting sequences is sufficient for activating the Ras signaling pathway (Aronheim et at., 1994; Quilliam et al., 1994).

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

9

Potential mechanisms that couple IRRs to Ras could include the stimulation and/or membrane recruitment of a Ras GEF, and/or the inhibition of GAP activity. As noted previously, Ras activation by its ubiquitous GEF, Sos, does not reflect an increase in the specific activity of Sos but, rather, recruitment of Sos by Grb2 and/or Shc to the vicinity of Ras in the membrane. Nevertleless, one group reported that TCR stimulation of T hybridoina cells increased the guanine nucleotide exchange activity of Sos immunoprecipitates and induced Fyn-Shc, Shc-Grb2, and Grb2-mSOS complex formation (B. Li et al., 1996). Although the physiological GEF that may activate Ras in lymphocytes has not been identified, a role for Sos is implicated by the finding that membrane localization of Sos mimicked activated Ras and synergized with a Ca'+-dependent signal to induce NFAT activation in T cells; furthermore, this effect of Sos was strongly inhibited by DN Ras (Holsinger et al., 1995). Because physiological Sos recruitment to the vicinity of Ras in the membrane (albeit not its exchange activity per se) requires association with Grb2 and/or formation of a trimolecular phospho-ShdGrbWSos complex, a number of studies have attempted to elucidate the mechanisms through which Shc and/or Grb2 can be localized to the membrane following the triggering of IRRs (reviewed in Koretzky, 1997) (Fig. 1).One clue came from a study demonstrating that TCR stiinulation induces tyrosine phosphorylation of Shc and subsequent formation of a Shc/GrbWSos complex; moreover, the SH2 domain of Shc directly interacted with the phosphorylated f chain of the TCWCDS complex (Ravichandran et al., 1993). Additional studies by the same group demonstrated that TCR ligation increased Sos-Grb2 association. Shc was implicated as a regulator of this association because a Shc-based phosphopeptide that displaces Shc from Grb2 abolished the enhanced association between Grb2 and Sos, and addition of phosphorylated Shc to unactivated T cell lysates was sufficient to enhance the interaction of Grb2 with Sos (Ravichandran et al., 1995).These findings potentially account for the apparent increase in the exchange activity of Sos observed in T cells overexpressing activated Fyn (B. Li et al., 1996). The TCR-induced, Shc-mediated up-regulation of the Grb2/Sos complex in T cells is different from the situation found in fibroblasts, wherein Grb2 and Sos are constitutively associated and the level of the Grb&/Soscomplex is not affected by growth factor stimulation. Other studies also demonstrated that Shc is a substrate for TCR-stimulated PTKs (Baldari et al., 1995a; B. Li et al., 1996; Lin and Abraham, 1997),where it forms a complex with Grb2 (Lin and Abraham, 1997). Similarly, BCR, FcERI, FcyRI, or FcyRIII (CD16) ligation on B cells, mast cells, monocytes, and NK cells, respectively, also induces tyrosine phosphorylation of Shc, its association with phospho-ITAMs, and formation

10

AMNON ALTMAN AND MARCEL DECKERT

of complexes containing phospho-Shc, Grb2, Sos, and additional tyrosinephosphorylated proteins (Saxton et al., 1994; Smit et al., 1994,1996; Kumar et al., 1995; Galandrini et al., 1996, 1997b; Jabril Cuenod et al., 1996; Kimura et al., 1996b; Park et al., 1996; Teramoto et al., 1997; Tridandapani et al., 1997a). This response was dependent on intact Syk kinase activity (Jabril Cuenod et al., 1996). Taken together, these findings suggest that Shc, on becoming phosphorylated by receptor-associated PTKs, couples IRR activation to the Ras signaling pathway. However, this conclusion has been called into question by other studies suggesting that Shc does not play an important role in Ras activation or in GrbWSos recruitment to the membrane in T cells. First, TCR or FceRI stimulation failed to induce an increase in the constitutive tyrosine phosphorylation of Shc in T cells (Gupta et al., 1994) or in mast cells (Turner et al., 1995), respectively, or formation of an Shc/GrbWSos complex in Jurkat T cells (Buday et al., 1994; Gupta et al., 1994), despite the fact that Grb2, together with additional tyrosine-phosphorylated proteins, translocated from the cytosol to the membrane under similar conditions of TCR stimulation (Nel et al., 1995). Second, the interaction of Shc with phosphohas a much lower dfinity compared to 6 interaction with the Zap-70 or Fyn tyrosine kinases (Osman et al., 1995).Finally, transfection of the Grb2 SH2 domain (Northrop et al., 1996),but not the Shc SH2 domain (Baldari et al., 1995b; Northrop et al., 1996), inhibited T cell activation. These findings, along with other reports documenting the membrane localization of Grb2 in TCR-stimulated T cells (Buday et al., 1994; Sieh et al., 1994; Nel et al., 1995), suggest that although Grb2 plays an important role in coupling Ras activation to stimulated IRRs, Shc is not critical for this process. A potential resolution to the controversial role of Shc in coupling TCR signals to Ras activation comes from studies that addressed the relative contribution of TCWCD3 versus CD4 signals to the activation of NFAT, a known Ras-dependent event in T cells (Rao et al., 1997). Whereas DN Ras inhibited NFAT activation induced by either CD3 or CD4 crosslinking, a DN Shc mutant consisting of its isolated SH2 domain only inhibited the CD4-mediated NFAT activation (Baldari et al., 1995b).Moreover, Lck and tyrosine kinase activity coimmunoprecipitated with Shc following CD4, but not CD3, triggering (Baldari et al., 1995a), and CD4 cross-linking by monoclonal antibodies recognizing distinct CD4 epitopes invariably led to increased tyrosine phosphorylation of Shc (Baldari et al., 1995a). However, only some of these antibodies caused NFAT activation, consistent with the notion that tyrosine phosphorylation of Shc is necessary, but not sufficient, for NFAT (and, hence, Ras) activation. These findings raise the possibility that the primary role of Shc is to couple the CD4 (and

r

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

11

CD8?) coreceptor to Ras activation, whereas the TCR is coupled to the Ras pathway via an alternative mechanism. This model and the controversial findings regarding the role of Shc in TCR-mediated Ras activation are consistent with the fact that the relative requirement of CD4-dependent signals for optimal T cell activation varies among different T cells, and depends on the strength of the TCR signal (Viola et al., 1997). 3. Linker for Activation of T Cells Evidence reviewed above suggests that a Shc-independent mechanism may mediate the recruitment of the Grb2/Sos complex to activated IRRs. A Grb2-associated tyrosine-phosphorylated protein of 36-38 kDa is a prime candidate for mediating this coupling in T cells. p36-38 is the major PTK substrate to associate with Grb2 in activated T cells, and has been found to associate with several other well-known signaling proteins, including Sos, PLCyl, and PI3-K (Buday et al., 1994; Motto et al., 1994; Reif et al., 1994; Sieh et al., 1994; Lahesmaa et al., 1995; Nel et al., 1995; Ingham et aE., 1996). The inducible association of Grb2 with p36-38 (Buday et aE., 1994; Nunes et al., 1994; Reif et al., 1994; Sieh et al., 1994), the finding that formation of a p36-38/Grb2/Sos complex occurs rapidly and correlates with Ras activation (Nunes et al., 1994), and the exclusive localization of p36-38 in the T cell plasma membrane (Buday et al., 1994; Sieh et at., 1994) are consistent with a putative role as a coupling element between the activated TCR complex and Ras. The association of p36-38 with Grb2 is mediated by the SH2 domain of Grb2 (Koretzky, 1997) and, in this regard, it is of interest that although various signaling proteins coimmunoprecipitate with p36-38, Shc is not one of them. This would perhaps be expected if phosphorylated p36-38 and Shc compete for binding to the Grb2 SH2 domain. The importance of p36-38 in TCR signaling is suggested by the finding that expression of a chimeric receptor with a cytoplasmic tail consisting of the Grb2 SH2 domain (which binds p36-38) and the catalytic domain of CD45 selectively reduced the TCR-induced tyrosine phosphorylation of p36-38, and this correlated with inhibition of inositol phosphate production and Ca2+mobilization;surprisingly,however, Ras and ERK2 activation remained intact under the same conditions (Motto et al., 1996a). The finding that p36-38 associates with the SH2 domain of Grb2 more stably than Shc (Osman et al., 1995) also supports the notion that a p36-38/Grb2/Sos complex (rather than a Shc/Grb2/Sos complex) is likely to be more relevant in TCR-induced Ras activation. Considerable effort has been invested in identifylng p36-38. Initially, this effort led to the cloning of Lnk, a 38-kDa protein that, like p36-38, becomes phosphorylated on tyrosine in activated T cells and associates with the SH2 domains of Grb2, PLCyl, and PI3-K (Huang et al., 1995).

12

AMNON ALTMAN AND MARCEL DECKERT

However, subsequent analysis of transgenic mice expressing a Ink transgene in the thymus suggested that Lnk does not play an important role in TCR signaling. Thus, although TCR stimulation induced tyrosine phosphorylation of Lnk in the transgenic T cells, Lnk did not play a limiting role in TCR signaling and, furthermore, it could be biochemically distinguished from the more prominent tyrosine-phosphorylated p36 in activated T cells (Takaki et al., 1997). It has been reported that p36-38 almost certainly corresponds to a novel protein, linker for gctivation of T cells (LAT), which was isolated by a combination of biochemical purification, partial peptide sequencing, and cDNA cloning (Zhang et al., 1998). The deduced amino acid sequence of LAT identified an integral membrane protein containing multiple potential tyrosine phosphorylation sites. LAT was phosphorylated by Zap-7O/Syk and associated with Grb2, PLCyl, the regulatory subunit (p85)of PI3-K, and several other tyrosine-phosphorylated proteins in activated T cells. The functional relevance of LAT in T cell activation and its potential connection to the Ras signaling pathway are indicated by two findings: first, mutation of two tyrosine residues in LAT abolished its association with Grb2 and p85, and greatly reduced PLCyl binding, in activated T cells; second, transient transfection of Jurkat cells with the tyrosine-mutated (but not wild-type) LAT inhibited the anti-CD3-induced transcriptional activity of AP-1 and NFAT (Zhang et al., 1998), which is known to require intact Ras function (Foletta et al., 1998; Rao et al., 1997; Su et al., 1994). Based on these findings, LAT appears to be a critical membrane linker that couples activated Zap-7O/Syk to the stimulation of the Ras, PLCyl, and PI3-K signaling pathways in T cells (Fig. 1).The isolation of LAT will certainly lead to intense efforts aimed at characterizing its function and regulation in T cells. Furthermore, because LAT is selectively expressed in T cells, mast cells, and NK cells (Zhang et al., 1998), it would be interesting to determine whether a structurally and/or functionally related linker is expressed in other hematopoietic cells, e.g., B lymphocytes and monocytes. 4. The Role of Other Grb2-Associated Proteins in Ras Activation

The Ras-specific GEF, Sos, is the most obvious link between Grb2 and Ras activation. However, a number of Grb2-associated proteins that could potentially affect upstream regulators or downstream targets of Ras have been isolated, and their role in TCR-mediated T cell activation has been studied in detail (reviewed in Koretzky, 1997).Although a complete survey of Grb2-associated proteins is beyond the scope of this review, three of these proteins, i.e., SLP-76, Cbl, andVav, have been implicated in signaling pathways mediated by Ras- or Rho-family GTPases.

GTF'ases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

13

An SH2 domain-containing leukocyte protein of 76 kDa (SLP-76) was originally described by several groups as a 76-kDa tyrosine-phosphorylated protein associated with Grb2 in activated T cells (Buday and Downward, 1993; Motto et al., 1994; Reif et al., 1994; Sieh et al., 1994). Affinity purification of the protein using antiphosphotyrosine antibodies and a glutathione S-transferase (GST)-Grb2 fusion protein and peptide microsequencing led to cloning of the corresponding cDNA (Jackman et nl., 1995). Expression of SLP-76 mRNA is restricted to T cells, B cells, and monocytes. The protein contains an N-terminal acidic region with several tyrosine phosphorylation sites that are involved in binding the SH2 domain of Vav (Wu et al., 1996), a central proline-rich region that binds Grb2 (Jackman et nl., 1995), and a C-terminal SH2 domain that binds two tyrosine-phosphorylated protein of -62 and -130 kDa (Motto et al., 1996b). Although the exact function of SLP-76 in TCR signaling is unknown, several findings indicate an important regulatory role with potential connection to the Ras pathway. First, transient overexpression of SLP-76 enhanced TCR-mediated NFAT, AP-1, and IL-2 promoter activities, and this activity required intact phosphorylation of SLP-76 by Zap-70 (Fang et al., 1996; Motto et al., 199613;Wardenburget al., 1996; Musci et al., 1997). Second, SLP-76 interacts with Vav, and the two synergize to augment IL2 promoter activity (Wu et al., 1996), but this interaction per se is not essential for TCR-induced IL-2 production in all T cells (Raab et al., 1997). Finally, transient SLP-76 overexpression does not affect calcium signaling, but augments TCR stimulation of ERK (Musci et al., 1997). Because AP1, NFAT, and ERK activation by the TCR depends, at least in part, on Ras function, a link between SLP-76 and the Ras pathway is implicated. Cbl, a 120-kDa protein that was originally isolated in its oncogenic form, is a prominent PTK substrate in activated T cells (Donovan et al., 1994). Cbl consists of an amino-terminal transforming region, a zinc ring finger, multiple proline-rich stretches that mediate constitutive association with Grb2 (Donovan et al., 1994; Fukazawa et al., 1995; Meisner et al., 1995), and several tyrosine phosphorylation sites. It is rapidly phosphorylated on tyrosine and serine residues in response to stimulation of many cell surface receptors and, as a result, becomes inducibly associated with a number of intracellular signaling proteins such as PTKs, PIS-K, Crk, and 14-3-3 through different protein-interacting modules, leading to the formation of multimolecular signaling complexes (reviewed in Liu and Altman, 1998). The exact biological function of this adapter protein remains largely unknown. Nevertheless, Cbl and its transforming mutants have been shown to display both negative and positive regulatory activities, including in PTK- and Ras-mediated signaling pathways.

14

AMNON ALTMAN AND MARCEL DECKERT

Several observations suggest that Cbl negatively regulates the Ras signaling pathway. First, genetic studies on vulval development in the nematode Caenorhabditis elegans revealed activating mutations in sli-1 (the C. elegans homolog of c-Cbl ) that restore normal development in nematodes expressing inactive Ras (LET-60) mutants (Yoon et al., 1995); second, a transforming mutant of Cbl, 70Z3, activates NFAT in a Ras-dependent manner in T cells (Liu et al., 1997);third, as discussed in more detail in Section VI,A, hyperphosphorylation of Cbl on tyrosine may indirectly lead to activation of a Ras-related protein, Rap-1, which functions as a suppressor of Ras and may play an important role in the induction of T cell anergy (Boussiotis et al., 1997);finally, the association of Cblwith Zap-70, Syk, or Fyn (Lupher et al., 1996; Deckert et al., 1998), and its ability to suppress Syk activation and mast cell degradation on FcERI triggering (Ota and Samelson, 1997), suggest that it may indirectly affect Ras activation by modulating the activity of IRR-coupled PTKs required for Ras activation. However, another study showed that Cbl overexpressionor inhibition of Cbl expression by antisense oligonucleotidesdid not have any detectable effect on EGF receptor signals leading to activation of the Ras pathway, although these manipulations affected a signaling pathway involving the Janus kinases ( JAKs) and signal transducers and activators of transcription (STATs) (Ueno et al., 1997). Because Cbl interacts with many signaling proteins, it probably displays versatile regulatory activities, the exact nature of which is dictated by the specific cellular context and the spectrum of the proteins that associate with it. Thus, the details of the functional interactions between Cbl and the Ras signalingpathway in lymphocytes and other cells remain to be elucidated. The role of Vav and its Grb2 association in regulating the activity of small GTPases is discussed in detail in Section III,D,l. 5. GAPS GAPSinhibit the activity of small GTPases by accelerating their intrinsic GTPase activity, thereby promoting the hydrolysis of GTP to GDP and, hence, inactivation of the GTPase. Two mammalian ras-GAPS, pl20-GAP and neurofibromin (NFl), have been isolated (reviewed in Boguski and McCormick, 1993), and both are expressed in lymphocytes (Boyer et al., 1994; Downward et al., 1992). pl20-GAP becomes phosphorylated on tyrosine in response to stimulation of various receptors, including IRRs (Gold et al., 1993; Lazarus et al., 1993; H. L. Li et aI., 1997), and it represents a binding partner and a substrate of the T cell-expressed kinase Lck (Amrein et al., 1992). However, the functional relevance of this phosphorylation event remains unclear. In addition, pl20-GAP was found to be associated with two tyrosine-phosphorylated proteins of -62 and

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

15

-190 kDa in activated B cells in one (Gold et al., 1993),but not in another (Lazarus et al., 1993), study. These two proteins correspond to ~ 6 (Carpino et al., 1997; Yamanashi and Baltimore, 1997) and a rho-GAP (Settleman et al., 1992a,b), respectively. Inhibition of GAP activity by IRR-, CD2-, or PKC-mediated signals has been described in both T (Downward et al., 1990; Graves et al., 1991; Izquierdo et al., 1992a) and B (Lazarus et al., 1993) cells, but its precise mechanism is unknown. In addition to PKC, PTK activity has been implicated in regulating ras-GAP activity in T cells (Izquierdo et al., 1992a). However, the inhibited GAP activity in B cells does not appear to be regulated by tyrosine phosphorylation because GAP activity following BCR triggering remained inhibited at a time when the inducible tyrosine phosphorylation of GAP was no longer detectable (Lazarus et al., 1993). It is not known which of the two ras-GAPS is inhibited by antigen receptor triggering or PMA treatment. The most compelling evidence that p12O-GAP plays an important role in T cell activation comes from a study demonstrating that overexpression of this protein in Jurkat cells inhibited TCR-mediated NFAT activation. The inhibition was overcome by expressing a CA Ras mutant (Baldari et al., 1994). In addition, pervanadate stimulation of T hybridoma cells overexpressing an active form of Lck was reported to induce the formation of a complex containing CD45, Lck, p12O-GAP, Grb2, and Sos (Lee et al., 1996).A study by Boyer et al., (1994) suggested a potential role of rasGAP in regulating BCR signaling. Thus, analysis of the distribution of p120GAP and NF1 in splenic B lymphocytes by immunofluorescent staining indicated that BCR cross-linking induced the redistribution of NF1, but not p12O-GAP. NF1 colocalized both spatially and temporally with the BCR, and this translocation was inhibited by cytoskeletal disrupting agents. However, cocapping of N F 1 with the BCR was independent of the Ras redistribution (Boyer et al., 1994). These findings indicate that NF1 and pl20-GAP can be differentially regulated in B cells and suggest that NF1 is a component of the signaling pathway initiated by BCR cross-linking (Boyer et al., 1994). C. RASEFFECTORS IN IRR SIGNALING Activated (GTP-bound) Ras regulates independent signaling cascades (White et al., 1995) via its ability to bind different partners. The highly conserved effector domain of Ras mediates this binding, and a number of yeast, C . elegans, Drosophila, and mammalian proteins, collectively termed Ras effectors, have been found to bind activated Ras directly (Marshall, 1996). Some of the known mammalian Ras effectors include GAPS (p120GAP and NF1); Raf kinases (Raf-1, A-Raf, and B-RaO, which function as

2

~

16

AMNON ALTMAN AND MARCEL DECKERT

MAPK kinase kinases in the ERK pathway; the lipid kinase PI3-K; RalGDS, a GEF for the Ras-related protein Ral; PKCG and MEKK-1, a MAPK kinase kinase in the JNK pathway. These effectors initiate signaling pathways consisting primarily of serinehhreonine kinase cascades, the stimulation of which ultimately leads to the activation of transcription factors in a receptor- and cell type-specific manner. The Raf-1 kinase and its downstream serinekhreonine kinase cascade represent the most extensively characterized Ras effector pathway (Fig. 1).In this cascade, membrane-localized activated Ras directly binds Raf1,leading to its enzymatic activation. Activated Raf-1 stimulates two MAPK kinases, MEK-1/2, which in turn activate the MAPKs ERK1/2. A number of studies have demonstrated that IRR stimulation or IRR-coupled PTKs of the Src and Syk families induce the association of active Ras with Raf1 (Finney et al., 1993), and stimulate all members of this cascade, i.e., Raf-1, MEKs, ERKs, or the ribosomal S6 kinase p90rskin T cells (Nel et al., 1990;Siegel et al., 1990, 1993; Izquierdo et al., 1993, 1994a,b; Franklin et al., 1994; Gupta et al., 1994; Lin and Abraham, 1997; Williams et al., 1997; Dumont et al., 1998), B cells (Tordai et al., 1994; Kumar et al., 1995; Kawauchi et al., 1996; H. L. Li et al., 1997; Tridandapani et al., 1997a), mast cells (Fukamachi et al., 1993; Hirasawa et al., 1995a,b; Rider et al., 1996; Ishizuka et al., 1997; Teramoto et al., 1997; Turner and CantreU, 1997; Zhang et al., 1997), monocytes (Durden et al., 1995; Taylor et al., 1997), and neutrophils (Alonso et al., 1996). CA Raf mimics the effect of Ras and synergizes with Ca2' signals to induce the IL-2 gene (Owaki et al., 1993)or activate ERK2 (Izquierdo et al., 1994a), and stable overexpression of ERKl in Jurkat T cells enhances expression of IL-2, IL-3, and GM-CSF mRNA (Park and Levitt, 1993). Finally, a signaling complex containing Vav (Section III,D,l), Grb2, Raf-1, and ERK2 was found to associate with FceRI in RBL cells (Song et al., 1996). The importance of Ras/Raf-1 in coupling IRRs to downstream activation events is indicated by the findings that DN Ras (Izquierdo et al., 1993; Williams et al., 1997), DN Raf-1 (Owaki et al., 1993; Wotton et al., 1993; Izquierdo et al., 1994a), or PD90859, a specific pharmacological inhibitor of MEK-1 (Dumont et al., 1998; Zhang et al., 1997),were found to inhibit IRR-mediated activation of the downstream serinehhreonine kinase cascade or cytokine production. Cytokine genes are not the only targets for the Raf-l/MEK/ERK cascade. Thus, by using DN Raf-1 mutants, it has been shown that Raf-1 is also required for the TCR-induced up-regulation of CD69 (Taylor Fishwick and Siegel, 1995) and the TCR-/3 chain (Wotton et al., 1993). The mechanisms leading to Raf-1 or ERK activation in T cells have been addressed in a number of studies. Generally, Raf-1 is activated as a

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

17

result of its binding to Ras-GTP (Zhang et al., 1993) and its subsequent membrane localization (Leevers et al., 1994). The reported association of Raf-1 with the 6 and y chains of the CD3 complex (Loh et al., 1994) may facilitate this localization in T cells. However, full Raf-1 activation by at least some receptors, including the IL-2 receptor (Tuner et al., 1991, 1993), appears to also require its tyrosine phosphorylation by Src-family PTKs (Morrison et al., 1989; Marais et al., 1995; Jelinek et al., 1996; Stokoe and McCormick, 1997). In addition, PKC can also directly activate Raf-1 (Sozeri et al., 1992; Kolch et al., 1993). Similarly, PKC-dependent ERK activation pathways have also been described, including in T cells (Izquierdo et nl., 1993, 199413; Gupta et al., 1994). The finding that, in comparison to phorbol ester-induced ERKZ activation, ERK2 activation induced by TCR stimulation was considerably more sensitive to inhibition by DN Ras (Izquierdo et al., 1993) led to the conclusion that activation of the Raf-1IERK2 cascade by PKC is, at least in part, Ras independent. However, this notion has to be reexamined in view of findings that have shed new light on the role of Ras in PKC-mediated Raf-1 activation (Marais et al., 1998). Thus, although DN Ras did not block PMA (i.e., PKC)mediated ERK and Raf-1 activation in COS cells (Howe et al., 1992; Marais et al., 1998), PMA stimulation caused Ras activation and formation of Ras/ Raf-1 complexes containing active Raf-1; moreover, a Raf-1 mutation that prevented its ability to associate with active Ras, or microinjection of a neutralizing anti-Ras antibody (Y13-259),blocked ERK or Raf-1 activation by PMA stimulation (Marais et al., 1998).These findings demonstrate that the absence of an effect of DN Ras does not necessarily indicate that Ras is not part of a given signaling pathway, and that Ras activation is, in fact, a component of PKC signaling. In T cells, either TCR or PMA stimulation activates Raf-1 (Siegel et nl., 1990, 1993). TCR-activated Raf-1 was found to be phosphorylated on serine but not on tyrosine and, furthermore, PKC depletion by prolonged PMA treatment abrogated TCR-induced Raf-1 activation (Siegel et at., 1990) suggesting that tyrosine phosphorylation is, at best, a minor part of the mechanism utilized by the TCR to activate Raf-1. Rather, TCRstimulated PKC plays a predominant, if not exclusive, role in Raf-1 activation in T cells. Nevertheless, a role for T cell-expressed PTKs in Raf-1 activation cannot be excluded. It has recently become clear that in addition to their role in regulating growth factor-induced cell proliferation, Ras and Raf-1 are also activated during mitosis (Laird et al., 1995; Taylor and Shalloway, 1996; Downward, 1997). Activation of Raf-1 during mitosis was also described in T cells, and its dependence on Lck activity suggests that in this case, Raf-1 is regulated by tyrosine phosphorylation (Pathan et al., 1996). In addition, binding of purified HIV-1 or its envelope glycoprotein,

18

A M N O N ALTMAN A N D M A R C E L D E C K E R T

gpl20, to CD4 on T cells resulted in association, tyrosine phosphorylation, and activation of Lck and Raf-1 (Popik and Pitha, 1996). Under the same conditions, Ras activation or its association with Raf-1 were undetectable, suggesting that Lck activates Raf-1 via a CD4-mediated, Ras-independent activation pathway. Studies conducted in T cells have made it clear that, as in other cell types, IRR-induced Ras activation leading to the induction of cytokine genes and other targets does not operate exclusively through the Raf-l/ MEUERK cascade, but also stimulates additional Ras effectors and their downstream signaling pathways (Fig. 2), which coordinately act to stimulate gene transcription. This is supported by several findings: first, although a DN MEK transgene can mimic Ras and inhibit positive selection of thymocytes (Section VII,A), it does not inhibit the TCR-induced, Ras-dependent proliferation of thymocytes (Alberola-Ila et al., 1995); second, CA MEK1 fully activates ERK2, but does not substitute for activated Ras and synergize with Ca2+signals to induce NFAT in T cells (Genot et al., 1996); third, functional Rac, which is clearly in a pathway distinct from the Raf1/MEWERK pathway, is required for TCR-induced, Ras-dependent T cell activation (Genot et al., 1996); finally, the use of CA or DN mutants of Ras or MEKK-1, an activator of the JNK pathway, demonstrated that Ras regulates TCR-induced activation of the MEKK-UJNK pathway (Fans et al., 1996). Analysis of the role of Ras effectors other than Raf-1 in IRR signaling is still in its infancy, and is likely to be a productive area of future research. D. TRANSCRIPTION FACTORS AS TARGETS OF THE RAS SIGNALING PATHWAY Induction of the IL-2 gene and production of its protein product represent the hallmark of T cell activation by the combination of TCR-derived

FIG.2. Ras effectors and their known functions (Marshall, 1996). Effectors that have been found to be involved in leukocyte functions are boxed.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

19

and costimulatory signals. The promoter region of the IL-2 gene has been extensively characterized, and is now known to contain a collection of response elements that bind die transcription factors AP-1, NF-KB, Oct1,and NFAT (Crabtree and Clipstone, 1994).Efficient IL-2 gene transcription depends on higher order assembly and synergistic interactions among the various transcription factor complexes that bind to the promoter (Garrity et al., 1994). The finding that the combination of phorbol ester and Ca" ionophore can mimic the TCR signal and lead to full T cell activation, including IL-2 production and concomitant proliferation (Truneh et al., 1985), established a fundamental paradigm for TCR-mediated signaling. Subsequent studies attempted to define the physiological signals that are mimicked by these two pharmacophores, and the transcription factors that represent the targets of these signals and regulate expression of the IL-2 gene. It is now well established that Ras and the Ca"/calmodulin-dependent protein phosphatase 2B (calcineurin) are the targets of the phorbol ester and Ca" signals, respectively (Crabtree and Clipstone, 1994; Weiss and Littman, 1994; Su and Karin, 1996; Alberola-Ila et al., 1997; Rao et at., 1997). Thus, CA Ras can synergize with Ca2+ionophore or CA calcineurin to activate the IL-2 gene or its NFAT element (Baldari et al., 1992b; Rayteretal., 1992;Woodrowet al., 1993a;Ohtsukaetal., 1996).Conversely, DN Ras blocks the induction of these transcriptional activation events by the TCR (Rayter et al., 1992; Baldari et nl., 1993; Ohtsuka et al., 1996). As a corollary of the requirement for Ras, MEK-1 and ERKs (which are downstream targets of Ras) are also required for the stimulation of IL-2 gene transcription in T cells (Whitehurst and Geppert, 1996). At the transcriptional level, the AP-1 (Foletta et al., 1998) and NFAT (Rao et al., 1997) transcription factors are the targets for the Ras- and calcineurin-mediated signals, respectively. Ca" ionophore represents a sufficient signal for NFAT activation as measured by NFAT dephosphorylation, nuclear translocation, and the increase in its DNA binding affinity (Rao et al., 1997). Similarly, phorbol ester or CA Ras alone activates AP1, including in T cells (Rayter et al., 1992; Williams et al., 1995), and regulates both of its components, Jun and Fos, at the transcriptional and posttranscriptional levels (Foletta et al., 1998). Nevertheless, it is well established that optimal NFAT activation requires a Ras signal in addition to the Ca2+/calcineurinsignal. For example, CA Ras synergizes with Ca2+ ionophore or with CA calcineurin to activate NFAT (Woodrow et al., 1993a) and, conversely, DN Ras inhibits TCR-induced NFAT activation (Woodrow et al., 199313). The requirement of Ras for optimal NFAT activation reflects the fact that NFAT proteins interact with the AP-1 complex and bind cooperatively to the composite NFAT/AP-1 site in the promoter of the IL-2 and other cytokine genes (Raoet al., 1997).Therefore,

20

AMNON ALTMAN AND MARCEL DECKERT

the contribution of Ras to NFAT activation reflects its role in AP-1 regulation. A similar picture emerges from the characterization of NFAT complexes in B, NK, or mast cells. Stimulation via the BCR or CD40 on B cells (Yaseen et al., 1993, 1994; Choi et al., 1994; Venkataraman et al., 1994), FcyRIIIA (CD16) on NK cells (Aramburu et al., 1995), or FcERI on mast cells (Baumruker et al., 1997; Hutchinson and McCloskey, 1995; Prieschl et al., 1995; Turner and Cantrell, 1997; Weiss et al., 1996) induces Ca2'dependent, cyclosporin A-sensitive NFAT activity. Furthermore, like T cells, these complexes contain JudFos heterodimers, i.e., active AP-1. The presence of AP-1 in the NFAT complexes observed in B, NK, and mast cells implicates a role for Ras in their activation. However, in only two of these studies was the role of Ras directly addressed. Thus, Ras function was required for the FceRI-mediated activation of the NFAT complex that binds to the NFAT response element in the IL-4 gene promoter (Turner and Cantrell, 1997), or to the promoter region of the genes encoding IL-5 and the chemokine MARC (Prieschl et al., 1995);however, NFAT action on the MARC promoter occurred in the absence of AP-1, suggesting that NFAT cooperates in this case with another, yet to be characterized factor (Prieschl et al., 1995).The finding in both of these studies that PKC was not required for this transcriptional activation event (Prieschl et al., 1995; Turner and Cantrell, 1997) suggests that Ras activation was PKC independent, consistent with the report that PMA does not activate Ras in mast cells (Izquierdo et al., 1992b). PKC almost certainly represents the link between the phorbol ester signal and Ras activation leading to AP-1 induction in T cells. It acts either upstream of, or in parallel to, Ras, but it is not a Ras effector in T cells (Williams et al., 1995). As discussed earlier, the functional relationship between PKC and Ras activation in T cells has been a subject of intense study, and has not been completely resolved. Although it is reasonably clear that PKC can activate Ras in T (Downward et al., 1990; Franklin et al., 1994; Ohtsuka et al., 1996), B (Harwood and Cambier, 1993; Tordai et al., 1994), and COS (Marais et al., 1998) cells, the mechanism of this activation is unknown. Direct assessment of the role of PKC in Ras activation will require genetic approaches, e.g., by testing the ability of CA mutants of distinct PKC isoforms to activate Ras. This approach has not been implemented to date, although PKC mutants were used to address indirectly the relationship between Ras and PKC in signaling pathways leading to transcriptional activation of the IL-2 gene and its isolated response elements. In one study (Baier Bitterlich et al., 1996), transient overexpression of wild-type PKC 6, but not PKCa, in murine EL4 leukemic T cells increased

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

21

the PMA-induced transcriptional activation of AP-1. Expression of a CA PKCB (but not PKCa) mutant was sufficient to activate AP-1 in the absence of PMA stimulation. Conversely, a catalytically inactive PKCO (but not PKCa) mutant abrogated PMA-mediated AP-1 activation. The ability of a DN Ras mutant to block the PKCB-induced AP-1 activation indicates that intact Ras function was required for the PKCB effect (Baier Bitterlich et al., 1996). However, it is not clear whether PKCB acts upstream of, or in parallel to, Ras in the signaling cascade leading to AP-1 activation. Another study (Genot et aZ., 1995) demonstrated that PKCE and, to a lesser extent, PKCa (but not PKCC) can activate the transcription factors AP-1 and NFAT-1. PKCE mimicked the stimulatory effect of CA Ras in this regard. Unlike Ras, however, none of the activated PKC mutants upregulated CD69 expression, a known Ras-dependent event (D’Ambrosio et al., 1994). Another indication of the potential importance of Ras in PKC-dependent signaling events is provided by the finding that, at sufficiently high levels of overexpression, DN Ras can inhibit ERK2 activation mediated not only by TCR, but also by phorbol ester stimulation (Izquierdo et al., 1993). It has been found that transient transfection of Jurkat T cells with CA PKCB, but not with other PKC isoforms (aor E ) , cooperated with CA calcineurin to activate the IL-2 promoter (Werlen et al., 1998).This combination was as effective as, or even more potent than, the combination of PMA and Ca2+ionophore, which generally stimulates maximal induction of the IL-2 promoter and is routinely used as a positive control in IL-2 gene induction studies. The PKC specificity of this induction operates at the level of JNWc-Jun activation because PKC 8 also specifically synergized with CA calcineurin to stimulate JNK and c-Jun transcriptional activity via a Rac-dependent pathway; in contrast, ERK was activated nonselectively by several PKC isoforms (Werlen et al., 1998). Finally, expression of DN PKCB, but not PKCa, inhibited the activation of JNK by PMA plus ionomycin. These findings are consistent with the selective ability of PKCB to activate AP-1 (Baier Bitterlich et al., 1996) and stimulate JNK (GhaffariTabrizi et aZ., 1998) in T cells, and with the demonstration that the specific MEK-l/JNK activator, MEKK-1, regulates the IL-2 promoter and is regulated, in turn, by Ras (Fans et al., 1996). Thus, PKCB, which is selectively expressed in T cells (Baier et al., 1993), and specifically colocalizes with the TCR to the contact area between antigen-specific T cells and antigenpresenting cells (APCs) (Monks et al., 1997), becomes an attractive candidate to mediate the PMA effect on Ras activation and IL-2 gene induction in T cells. It will be equally important to determine whether, in addition to its potential ability to activate Ras, PKCB also regulates other effectors involved in JNK andlor IL-2 gene activation.

22

AMNON ALTMAN AND MARCEL DECKERT

It is not clear whether Ras regulates transcription factors other than AP1 and NFAT in the IL-2 promoter. Reporter gene experiments demonstrated that NF-KB transcriptional activity is not affected by expression of activated Ras (Williams et al., 1995). However, TCR-induced NF-KB activation was found to require intact Raf-1 kinase function and a Ca2'/ calcineurin signal, suggesting a functional synergy between Raf-1 and calcineurin (Kanno and Siebenlist, 1996). In the face of the apparent lack of role for Ras (Williams et al., 1995), the Raf-1 requirement for NF-KB activation may reflect activation of Raf-1 by a Ras-independent mechanism, e.g., by PTK-mediated phosphorylation similar to what has been observed in CD4-ligated T cells (Popik and Pitha, 1996). Other transcription factors whose activation appears to require intact Ras function are Elk-1 (Turner and Cantrell, 1997) and Egr-1 (McMahon and Monroe, 1995). Elk-1 is a member of the ternary complex factor (TCF) family, members of which form complexes with serum response factor (SRF). These complexes are important for activation of the serum response element (SRE), which regulates the transcription of immediateearly genes such as c-fos and egr-1 (Treisman, 1994). Elk-1 can be phosphorylated and activated by various members of the MAPK family, i.e., ERK, JNK, and p38 kinases (Treisman, 1994). FceRI ligation or PMA (but not ionomycin) stimulation of RBL cells induced activation of Elk-1. Elk-1 was also induced by CA mutants of Ras or Raf-1 and, conversely, its FceRI-, active Ras-, or active Raf-l-mediated activation was markedly inhibited by DN Ras or DN Raf-1, as well as by PD90859, a specific MEK1 inhibitor (Turner and Cantrell, 1997). These findings indicate that Elk1 is activated by the Ras/Raf-l/MEK/ERK cascade in response to FceRI ligation on mast cells. The primary response gene egr-1 encodes a sequence-specific transcription factor whose expression is necessary for antigen receptor-stimulated activation of B lymphocytes (Monroe et al., 1993). Expression of activated Ras resulted in egr-1 induction similar to that induced by BCR crosslinking. Conversely, DN mutants of Ras and Raf-1 inhibited BCR-induced egr-1 induction (McMahon and Monroe, 1995). Although Ras is generally considered as a positive regulator in IRRmediated signaling pathways, a recent study (Chen et al., 1996) indicated that it may also exert negative regulatory influences. Expression of activated Ras inhibited induction of the immediate-early genes egr-1, c-fos, and cjun by Ca2' ionophore in Jurkat T cells. This inhibition was reversed by treatment with cyclosporin A, suggesting the involvement of calcineurin. A later reflection of this inhibitory effect was down-regulation of AP-1 activity and subsequent coordinate reduction in IL-2 mRNA and protein expression (Chen et al., 1996).These results suggest that Ras is an essential

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

23

mediator not only in positive but also in negative modulatory mechanisms controlling the competence of T cells in response to inductive stimuli (Chen et al., 1996). The ability of Ras to regulate negatively Ca'+/calcineurindependent immediate-early gene induction in T cells may be related to the late phase of NFAT deactivation found to be induced in T cells by TCR ligation (Loh et al., 1996). Because this deactivation was facilitated by PMA treatment (Loh et al., 1996), it may also involve PKC and/or Ras activation. The role of Ras in IRR signaling was analyzed almost exclusively in the context of the Raf-1IMEWERK effector pathway. However, with the recent realization that Ras activates different effector pathways (RodriguezViciana et al., 1994; White et al., 1995; Marshall, 1996), the role of other Ras effectors in NFAT activation has been addressed in T cells (Genot et al., 1996) and in mast cells (Turner and Cantrell, 1997). Although CA MEK-1 f d y activated ERK2, it did not substitute for activated Ras and synergize with Ca2+/calcineurinsignals to induce NFAT. Expression of DN Rac also prevented TCR- and Ras-mediated activation of NFAT, but did not inhibit ERK2 activation. Similarly, the induction of AP-1 by Ras also required Rac-1 function. Activated Rac-1 mimicked active Ras to induce AP-1, but not NFAT. Moreover, the combination of activated MEK1 and Rac-1 could not substitute for activated Ras and synergize with Ca" signals to induce NFAT. Thus, Ras regulation of NFAT in T cells requires the activity of multiple effector pathways, including those regulated by MEK-lERK2 and Rac-1 (Genot et al., 1996). This is consistent with the role of Rac in the PKCB- and calcineurin-mediated signaling pathway leading to JNK activation in T cells (Werlen et al., 1998). A similar conclusion was drawn from the analysis of NFAT activation (using an IL-4 promoter element) by FcERI ligation in mast cells (Turner and Cantrell, 1997). In this case it was found that Ras-dependent NFAT activation did not involve the Raf-lIMEWERK pathway but, rather, a Racassociated pathway. These studies indicate that, as in other cells, IRR triggering also stimulates multiple Ras effector pathways that coordinately act to regulate cytokine gene induction. Much remains to be learned in this important area. E. NEGATIVE SIGNALING AND THE RASPATHWAY The majority of studies on signal transduction pathways in various cell types have focused on receptors that, upon ligation, lead to cellular activation. However, it has recently become clear that many cell types express inhibitory receptors. Triggering of these receptors initiates a negative signal that intercepts activation cascades at defined points and most likely acts as a negative regulatory mechanism to terminate activation responses (Cam-

24

AMNON ALTMAN AND MARCEL DECKERT

bier, 1997). Examples of inhibitory receptors on lymphocytes are provided by the FcyRIIbl receptor on B lymphocytes (Tridandapani et al., 1997b), killer cell inhibitory receptors (KIRs) on NK cells (Leibson, 1995; Yokoyama, 1995),and KIRs or CTLA-4 (Bluestone,1997;Thompson and Allison, 1997) on T cells. Essentially all known inhibitory receptors share the property of an immunoreceptor tyrosine-based inhibitory motif ( ITIM) present in their cytoplasmic domain. Receptor ligation induces tyrosine phosphorylation of the ITIM motif, thereby creating a docking site for the SH2 domain of two types of enzymes, i.e., the structurally related SH2containing phosphotyrosine phophatases 1 and 2 (SHP-1 and SHP-2, respectively) and SH2-containing inositol polyphosphate 5-phosphatase (SHIP). Activation of SHP-112 or SHIP generates negative signals that attenuate tyrosine phosphorylation or elevations in inositol phosphate and intracellular Ca2+concentrations (Scharenberg and Kinet, 1996). Several recent reports indicate that the Ras signaling pathway may represent one target of inhibitory receptors. Co-cross-linking of the BCR and FcyRIIbl on B cells has been known to inhibit B cell proliferation and antibody production, and probably serves as physiological mechanism to prevent excess antibody production (Sinclair and Panoskaltsis, 1989). This inhibition is mediated by recruitment of SHIP to the phosphorylated ITIM in FcyRII (On0 et al., 1996).Biochemical analysis of signaling events induced by BCR cross-linking alone versus BCWFcyRII coligation in human (Sarmay et al., 1996)or murine (Tridandapani et al., 1997a) B cells established that, under coligation conditions, the activation of Ras, Raf-1, and ERK was inhibited. A clue for the mechanism of this inhibition comes from the findings that, first, Shc is hypophosphorylated on tyrosine and coimmunoprecipitateswith FcyRII in the coligated B cells (Sarmay et al., 1996) and, second, the BCR-induced association of Shc and Grb2 was abrogated under negative signaling conditions; instead, Shc associated with a tyrosine-phosphorylated 145-kDa protein (Tridandapani et al., 1997a), previously identified as SHIP (Damen et al., 1996), only under conditions of negative signaling. Based on these findings, it was suggested that SHIP mediates Fcy RII-induced inhibition of B cell activation by competing with Grb2 for binding phospho-Shc (Tridandapani et al., 1997b). The resulting dissociation of the ShdGrb2 complex would disrupt Ras activation because it would prevent recruitment of the GrbYSos complex to the membrane. This attractive hypothesis does not take into account, however, the role of the catalytic activity of SHIP in inhibiting cellular activation. It is clear that the enzymatic activity of SHIP is also important for its inhibitory action (Deuter-Reinhard et al., 1997; Bolland et al., 1998) and, therefore, it is possible that the adapter

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

25

and catalpc functions of SHIP are both necessary for optimal inhibition under physiological conditions. The role of the Ras signaling pathway as a potential target for other inhibitory receptors has not been explored in detail. However, it was reported that KIR recognition of major histocompatibility complex (MHC) class I ligands inhibits distal signaling events and ultimately NK cell cytotoxicity by blocking the association of an adapter protein, pp36 (which is most likely LAT), with PLCyl in NK cells; furthermore, tyrosine-phosphorylated pp36 was a substrate in vitro for the KIR-associated tyrosine phosphatase SHP-1 (Valiante et al., 1996). Thus, it is possible that in KIR-ligated NK cells, hypophosphorylated LAT is also deficient in its ability to bind the GrbUSos complex, a situation that would be expected to prevent or reduce Ras activation. F. RASIN CVTOKINE RECEPTOR SIGNALING Although this review focuses primarily on the function of small GTPases in IRR signaling, it is well established that Ras is also activated by most cytokine receptors (Satoh et al., 1991; Graves et al., 1992; Izquierdo et al., 1992b; Nakafuku et al., 1992; Izquierdo and Cantrell, 1993; Miura et al., 1994). In T cells, PTKs, but not PKC, have been found to couple the IL2 receptor to Ras activation (Izquierdo and CantreU, 1993). The Raf-1 kinase is also activated by IL-2 (Turner et al., 1991, 1993), IL-3, and GM-CSF (Carroll et al., 1990). The Ras signaling pathway is one of two independent cascades that cooperate to mediate cytokine receptor signaling, the other one involving JAK-family kinases and STAT transcription factors. It is generally accepted that the mechanism by which cytokine receptors induce Ras activation involves tyrosine phosphorylation and receptor recruitment of a ShdGrbUSos complex. As an example, stimulation of the IL-2 (Ravichandran and Burakoff, 1994; Ravichandran et al., 1996) or GM-CSF (Pratt et al., 1996) receptors induces tyrosine phosphorylation of Shc and its binding to the phosphorylated receptor. This scenario allows Shc to interact simultaneously with phosphorylated tyrosine residue(s) in the cytoplasmic domain of the activated receptor via its phosphotyrosinebinding (PTB) domain, and with the SH2 domain of Grb2 via its own phosphotyrosine(s). Other details of signal transduction by cytokine receptors and the role of Ras in these events are discussed in reviews by Ihle et al. (1995) and Karnitz and Abraham (1995). 111. The Function of Rho-Family GTPases in IRR Signaling

A. INTRODUCTION Small GTPases of the Rho family (RhoA, RhoB, RhoC, RhoE, RhoC, Racl, Rac2, Cdc42, and TC10) were until recently considered to be primar-

26

AMNON ALTMAN AND MARCEL DECKERT

ily involved in the organization of the actin cytoskeleton. Cdc42, Rac, and Rho independently induce the formation of filopodia, lamellipodia and membrane ruffles, or stress fibers and focal adhesion complexes, respectively (Nobes and Hall, 1995).However, these GTPases are also linked in a linear cascade in which Cdc42 stimulates Rac, which in turn activates Rho (Nobes and Hall, 1995) (Fig. 3). The major discoveries regarding Rho GTPases were made through the study of their functions in fibroblasts. However, there is now compelling evidence for a similar role for these

FIG. 3. The cascade of Rho-family GTPases in fibroblasts and their role in cytoskeleton organization. Rho GTPases are regulated by GEFs, GAPS,or GDIs (Section 1,A).Extracellular stimuli activate Cdc42, Rac, and Rho, leading to the formation of filopodia, lamellipodia, and stress fibers, respectively. These GTPases are also linked in a linear functional cascade whereby Cdc42 activates Rac, which in turn activates Rho. Furthermore, the Ras GTPase can “talk to the Rho GTPases cascade by activating Rac. The mechanisms that connect these GTPases remain largely unclear. Moreover, the functional implications of such a cascade in lymphocyte activation have not been addressed. LPA, Lysophosphatidic acid.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

27

proteins in additional actin-dependent cell structures, e.g., tight junction regulation by Rho in polarized epithelia; control of axonal outgrowth by RacKdc42 in mammalian neuronal cells and Drosophila; and regulation of cell polarization by Cdc42 in the budding yeast. Through their ability to establish different types of adhesive structures including focal complexes, adherens junctions, and tight junctions, Rho-family GTPases provide a molecular basis for many of the morphological and motility changes that cells undergo during differentiation, migration, and mitosis (Chant and Stowers, 1995; Ridley, 1995; Symons, 1995; Tapon and Hall, 1997; van Aelst and D’Souza-Schorey, 1997; Hall, 1998). In the past 3 years or so it has become clear that Rho GTPases also have critical functions in the control of cellular proliferation initiated by multiple receptor signals or, under aberrant conditions, in malignant transformation. Significant progress has been made toward characterizing the signaling cascades that couple signals emanating from Rho-family GTPases to the nucleus. As a result, it is now clear that Rho-family GTPases activate members of the JNKs and p38 families of stress-activated protein kinases (SAPKs) (Minden et al., 1995; Coso et al., 1995; Olson et al., 1995; Vojtek and Cooper, 1995), and deliver signals that regulate transcriptional activation by SRF and induce cell cycle progression through the GI phase (Hill and Treisman, 1995; Hill et al., 1995; Olson et al., 1995; Symons, 1995; Treisman, 1996) (Fig. 4).The functions of Rho GTPases in regulating the cytoskeleton versus cell growth appear to be independent and to be mediated by distinct effectors (White et al., 1995;Joneson et al., 1996; Lamarche et al., 1996; van Aelst and D’Souza-Schorey, 1997; Westwick et al., 1997; Hall, 1998). Ras interacts with the Rho GTPases cooperatively to regulate these signaling events (Khosravi Far et al, 1995; Prendergast et al., 1995; Qiu et al., 1995a,b, 1997). Mostly indirect evidence has also accumulated implicating an important role for Rho-family GTPases in coupling IRRs and other receptors in the reorganization of the actin cytoskeleton in leukocytes. The use of genetic and biochemical approaches has also started to yield important clues on the role of these small G proteins in IRR-initiated signaling pathways leading to proliferation and cytokine production by T cells and mast cells. These studies are reviewed below.

B. RHO-FAMILY GTPASESAND THE CYTOSKELETON IN LEUKOCYTES The morphological changes occurring during lymphocyte polarization induced by chemokines, adhesion molecules, or antigenic stimulation were described by Haston et al. (1982). The migrating lymphocyte extends a cytoplasmic pseudopod-like protrusion called a uropod, which is involved in adhesion to endothelial or extracellular matrix (ECM) proteins, motility,

28

AMNON ALTMAN AND MARCEL DECKERT

FIG.4. Effectors of Rho-family GTPases and their known functions. Cdc42 or Rac effectors, which share a CddiWRac interactive binding (CRIB) motif, are shaded, and Rhofamily effectors which have been imphated in leukocyte functions are boxed. Rho kinase (p160) was shown to phosphorylate the myosin-binding subunit (MBS) of myosin phosphatase (myosin PPase) and this, in turn, inhibits the phosphatase activity (Kimura et al., 1996a).

and activation. In addition, important regulatory and effector cell-cell contacts occur among cells of the immune system, accompanied by secretion of various soluble products. Polarized secretion is a common property of many cell types and lymphocytes are no exception. Two major T cell interactions involve vectorial secretion, which requires cytoskeleton reorganization and polarization: contact between T helper cells and APCs, which results in cytokine production; and cytotoxic T lymphocyte (CTL)-target cell interactions accompanied by cytotoxic granule release and target cell lysis (Kupfer et al., 1987; Po0 et al., 1988; Kupfer and Singer, 1989; Podack and Kupfer, 1991). For example, conjugate formation between cytotoxic

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

29

effector lymphocytes and target cells is accompanied by capping of lymphocyte function antigen-1 (LFA-1) to the contact area, followed by talin colocalization (Podack and Kupfer, 1991). It is therefore not surprising that small GTPases of the Rho family have been implicated in the regulation of lymphocyte adhesion, polarization, motility, and effector functions (Chavrier et al., 1993; Quinn, 1995; Dharmawardhane and Bokoch, 1997; Reif and Cantrell, 1998). Direct evidence for the activation of Rho-family GTPases by IRRs and other leukocyte receptors, or morphological evidence for their distinct roles in inducing defined cytoskeletal structures in leukocytes, is missing. The difficulty in demonstrating Rho-family GTPase activation is not unique to leukocytes. It reflects the fact that GTPases of this family display a much higher intrinsic GTPase activity compared to Ras, thus making it technically more demanding to isolate and detect their active (GTP-bound) form. Morphological analysis of actin cytoskeleton structures in leukocytes is very difficult because these cells either grow in suspension and/or contain relatively little cytoplasm. Nevertheless, indirect evidence based on morphological, pharmacological, biochemical, or genetic approaches implicates important roles for Rho-family GTPases in reorganization of the cytoskeleton in leukocytes (Fig. 5). Many studies have demonstrated that triggering of IRRs (TCR, BCR, FcsRI, FcyR) leads to actin cytoskeleton reorganization in T cells (DeBell et al., 1992; Donnadieu et al., 1992, 1994; Pardi et al., 1992; Parsey and Lewis, 1993; Phatak and Packman, 1994; Negulescu et al., 1996), B cells (Melamed et al., 1991a,b; Cox et al., 1996), and mast cells (Apgar, 1991; Norman et aZ., 1994, 1996; Pfeiffer and Oliver, 1994; Barker et al., 1995; Prepens et al., 1996; Guillemot et al., 1997). Reorganization of the cytoskeleton is manifested by an increase in the content of filamentous actin (Factin), formation of membrane ruffles and focal adhesions (most likely reflecting Rac and Rho activation, respectively), spreading and adhesion to the substrate, uropod extension and cell elongation or rounding, and changes in motility, pinocytosis, and phagocytosis. In activated T cells, the redistributed F-actin colocalized and was physically associated with the adhesion receptors LFA-1 (CDllaICD18) and VLA-4 (Pardi et al., 1992). These changes are accompanied by Ca2+oscillations (Donnadieu et al., 1992, 1994; Valitutti et al., 1995a; Negulescu et al., 1996). Pharmacological approaches based on the use of agents that disrupt the cytoskeleton, e.g., cytochalasins or phalloidin, demonstrated that reorganization of the actin cytoskeleton is tightly coupled to secretion in mast cells (Norman et al., 1994), T cell shape changes induced by contact with antigen-pulsed APCs (Donnadieu et al., 1992), TCR internalization (DeBell et al., 1992), and B cell proliferation (Melamed et al., 1991a).However,

30

AMNON ALTMAN AND MARCEL DECKERT

FIG.5. The role of Rho-family GTPases in leukocyte signaling. The TCFVCD3 complex is used as an example. Integrins and chemokine receptors activate Rho-family GTPases and can provide a costimulatory signal whose exact biochemical nature is unclear. Known PTK substrates in activated T cells are lightly shaded, and small GTPases are in black. For details, see Section 111.

these agents do not seem to inhibit degranulation in mast cells (Prepens et al., 1996).Similarly, overexpressionof transfected mutant small GTPases, which interefere with cytoskeletal reorganization, does not inhibit mast cell secretion (Norman et al., 1996)or IL-2 production by T cells (Stowers et al., 1995).Importantly, these findings do not exclude the likely possibility that reorganization of the cytoskeleton is essential for a physiologically relevant polarized secretion, e.g., localized lymphokine secretion at the site of contact between antigen-specific T cells and AF'Cs (Po0 et al., 1988; Stowers et al., 1995). The stimulus-coupled reorganization of the actin cytoskeleton is not only important for changes in cell shape, adhesion, and motility. Rather, this reorganization is also essential for the proper transmission of growth signals.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

31

This is not surprising because intact architecture of the cytoskeleton is most likely essential for recruitment of signaling complexes to the phosphorylated ITAM motifs in IRRs, including those that contribute to the activation of Ras, i.e., Shc, LAT, Grb2, and Sos. Studies on the role of the actin cytoskeleton in antigen-specific T cell activation demonstrated that agents that disrupt the actin cytoskeleton prevent not only the cell shape changes occurring on contact with APCs but, in addition, also interfere with Ca" mobilization, which is known to regulate gene transcription (Dolmetsch et al., 1997; Timmerman et al., 1997), and with interferon-? production (Valitutti et al., 1995a), as well as with TCFUCD28-induced proliferation and NFAT activation (Holsinger et al., 1998). These results indicate that the actin cytoskeleton drives TCR cross-linking and sustains signal transduction, and offer an explanation for the well-known phenomenon that T cells, despite being equipped with a low-affinity TCR, can be triggered by very few MHC/peptide complexes (Demotz et al., 1990) that serially engage many TCRs (Valitutti et al., 199513). Cytoskeletal changes induced in T cells by antigen/MHC binding, which are most likely mediated by Rho GTPases, may also be essential for the ligand-specific oligomerization of TCFUMHClpeptide ternary complexes on the surface of T cells (Reich et al., 1997). The quantity and/or quality of cytoskeleton assembly and the degree of oligomerization may, in turn, be important determinants in differential TCR signaling that can lead to distinct functional outcomes, i.e., activation and proliferation, anergy, or apoptosis (Alberola-Ila et al., 1997). The exact mechanisms through which cytoskeleton reorganization regulates PTK-mediated growth signals in T cells are unknown. However, several clues have emerged. First, T or B cell activation was found to induce increased tyrosine phosphorylation of a-tubulin (Ley et al., 1994; Mane Cardine et al., 1995; Peters et al., 1996), most likely mediated by Syk-family tyrosine kinases (Huby et al., 1995; Peters et al., 1996). The presence of phosphorylated a-tubulin in the unpolymerized soluble fraction of T (Ley et al., 1994) and B (Peters et al., 1996)cells, and its association with Zap-70 (Huby et al., 1995) or Syk (Peters et al., 1996), suggest that tyrosine phosphorylation of a-tubulin may inhibit its polymerization into microtubules (Ley et al., 1994) and/or serve to dissociate Syk from the BCR, thereby allowing it to phosphorylate cytosolic proteins (Peters et al., 1996). Phosphorylated a-tubulin was also found to associate with Vav in intact T cells (Huby et al., 1995), and with the SH2 domain of Fyn in vitro (Marie Cardine et al., 1995). The association of a-tubulin with Vav and Zap-70 in unactivated T cells (Huby et al., 1995) most likely reflects the constitutive tyrosine phosphorylation occurring in the transformed

32

AMNON ALTMAN AND MARCEL DECKERT

Jurkat cells. Indeed, a-tubulin was not phosphorylated on tyrosine in untransformed resting human T lymphocytes (Marie Cardine et al., 1995). TCR-induced reorganization of the cytoskeleton may also regulate growth signaling pathways by affecting a population of phospho-( chains (and their associated TCR complexes) residing in the cytoskeleton. Thus, two groups have reported that cell surface-expressed TCR-J is associated with the actin cytoskeleton (Caplan and Baniyash, 1995, 1996; Caplan et al., 1995; Rozdzial et al., 1995). A fraction of cell surface-expressed ( was found to be associated with the cytoskeleton in resting T cells and in transfected COS cells, and a unique 16-kDa tyrosine-phosphorylated species of [was detected only in the cytoskeletal fraction (Caplan et d., 1995). The phosphorylated species of ( in the cytosolic and cytoskeletal fractions were distinct in both nonactivated and activated lymphocytes, and assembly of the CD3 subunits with cytoskeleton-associated (was necessary for their maximal localization to the cytoskeleton (Caplan and Baniyash, 1996). It was further demonstrated that TCR ligation induced increased association of phospho-(with the cytoskeleton (Caplan et al., 1995), and that cytochalasin treatment disrupted it (Caplan et al., 1995; Rozdzial et al., 1995). The importance of this association in growth signaling is implicated by the finding that it correlated with IL-2 production by activated T cells, and was not detected in immature thymocytes (Rozdzialet al., 1995).Anchorage of cell surface-expressed ( chain to the cytoskeleton in T cells may, therefore, facilitate recycling of receptor complexes and/or allow the transduction of TCR signals into the cell. Studies have addressed the role of Ca” or different enzymes in cytoskeleton reorganization using either transfection of PTKs such as Syk (Cox et al., 1996), or selective pharmacological inhibitorshnducers that modulate Ca2+mobilization or the activities of PKC and P13-K. The conclusions drawn were that changes in the cytoskeleton are dependent on tyrosine kinase (Melamed et al., 1991b; Pfeiffer and Oliver, 1994; Cox et al., 1996) and PKC (Phatak et al., 1988; Apgar, 1991; Melamed et al., 1991b; Par& et al., 1992; Pfeiffer and Oliver, 1994) activity. Ca2+also modulates these events (Apgar, 1991; Donnadieu et ul., 1992,1994; Negulescu et al., 1996), apparently via a calcineurin-independent mechanism (Stowers et al., 1995; Negulescu et al., 1996), although an increase in intracellular Ca2+concentration is neither necessary nor sufficient for actin polymerization (Apgar, 1991). Selective inhibition of PIS-K activity in intact cells indicated that PI3-K is required for the polarization of antigen-specific T cells toward APCs, possibly via a Cdc42-mediated pathway (Stowers et al., 1995), and for membrane ruffling and fluid pinocytosis in FcsRI-stimulated RBL cells (Barker et al., 1995). However, PI3-K apparently was not required for actin

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

33

polymerization, receptor internalization, spreading, and adhesion plaque formation (Barker et al., 1995). Because Cdc42, Rac, and Rho are all involved to varying degrees in cytoskeleton reorganization and, furthermore, these GTPases as well as Ras are linked in a signaling cascade, it is impossible to conclude from studies assessing cell morphology or the effects of pharmacological agents which Rho-family GTPase regulates a particular aspect of cytoskeleton reorganization. This question was addressed more directly by pharmacological approaches based on the use of bacterial toxins that selectively inhibit defined GTPases, or genetic approaches relying on transfection with wildtype or mutated forms of Rho-family GTPases. This analysis allowed more precise conclusions. In mast cells, transient transfection with CA mutants of RhoA and Racl (V14RhoA and V12Rac1, respectively) demonstrated that although Rho was responsible for de novo actin polymerization occurring in the cell interior, presumably from a membrane-bound monomeric actin pool, Racl was required for entrapment of the released cortical filaments following activation by several stimuli (Norman et al., 1994); furthermore, both of these active GTPases enhanced the stimulated secretion by increasing the proportion of cells responding to the stimulus (Price et al., 1995) in a cytochalasin-resistant manner (Norman et al., 1996). Inhibition of Rac and Rho by a DN Rac mutant (N17Racl) or Clostridium botulinurn C3 exotoxin (which ADP-ribosylates and inactivates Rho), respectively, reduced the stimulus-induced secretory response of the cells (Price et al., 1995). Expression of DN mutants in RBL cells also had distinct effects on mast cells. DN Cdc42 decreased cell adhesion, interfered with FcsRI-induced actin plaque assembly, and reduced the recruitment of vinculin at the cell-substratum interface, whereas the inhibitory Racl mutant abolished FcsRI-mediated membrane ruffling. Both mutants significantly inhibited antigen-induced degranulation (Guillemot et al., 1997). Similar approaches were also used in T cells. Introduction of recombinant C3 exotoxin into electropermeabilized NK cells or CTLs led to a dosedependent inhibition of their cytolpc function; the only substrate efficiently ADP-ribosylated by C3 was RhoA, which was localized in the cytosol (Lang et al., 1992). A role for Rho in T cell activation is also suggested by the finding that lysophosphatidic acid (LPA), a selective Rho activator, induced Ca2+mobilization and increased IL-2 production in Jurkat T cells (Xu et al., 1995). Similarly, when a B lymphoblastoid cell line was treated with C3, only RhoA was ADP-ribosylated in situ in a time- and concentration-dependent manner, and this correlated with inhibition of PMAinduced LFA-lhntercehlar adhesion molecule (1CAM)-l-dependent cell aggregation (Tominaga d al., 1993). This result indicates that RhoA functions downstream of PKC to induce LFA-l-dependent cell aggregation.

34

AMNON ALTMAN AND MARCEL DECKERT

It is not clear, however, whether C3 inhibits aggregation by interfering with cell polarization or by preventing Rho-dependent adhesion of ICAM1to LFA-1, because LFA-1 is known to undergo changes in ligand affinity and to inetarct with the actin cytoskeleton on activation (Hynes, 1992). Expression of CA or DN Cdc42 in antigen-specific T cells inhibited the polarization of both actin and microtubules toward APCs (Stowers et al., 1995). Finally, Rho-family GTPases also regulate the invasiveness of malignant T cells, most likely by controlling cell shape, adhesion, and/or motility. Thus, transfection of Tiam-1, a Rac-specific GEF, or active Racl induced invasiveness in the murine T lymphoma line BW5147 (Michiels et al., 1995). Pretreatment of several T lymphoma cell lines with C3 exotoxin inhibited the characteristic shape changes resulting from extension and retraction of pseudopodia and, concomitantly, the invasion of the cells through a fibroblast monolayer. These effects correlated with the absence of F-actin in the pseudopodia of the treated (but not control) cells (Verschueren et al., 1997). The role of Rho in FcyR-mediated phagocytosis was investigated by microinjecting 3774 macrophages with C3 exotoxin (Hackam et al., 1997). C3 induced retraction of filopodia, disappearance of focal complexes, and a global decrease in the F-actin content. In addition, the cells exhibited increased spreading and formation of vacuolar structures. Importantly, inactivation of Rho completely abrogated phagocytosis of opsonized particles. The same effects were also observed in FcyRIIA-transfected COS cells. Moreover, Rho was essential for the accumulation of tyrosinephosphorylated proteins and F-actin around phagocybc cups and for F v R mediated Ca2+signaling. The effect of the toxin was specific, because clustering and internalization of transfenin receptors were unaffected by C3 microinjection. These data i d e n q a role for Rho in FcyR-mediated phagocytosis in leukocytes (Hackam et al., 1997). C. RHO-FAMILY GTPASESCONTROL GROWTH SIGNALS IN LYMPHOCYTES The role of Rho-family GTPases in controlling growth signaling pathways (Symons, 1995; Vojtek and Cooper, 1995; Treisman, 1996)has just begun to be explored in lymphocytes (Genot et al., 1996; Turner et al., 1997; Reif and Cantrell, 1998) (Fig. 5). The limited studies conducted to date indicated that expression of a DN Racl mutant (N17Racl) prevents TCRand Ras-mediated activation of NFAT and AP-1, but does not interfere with ERK2 activation; conversely, activated Racl mimicked active Ras to induce AP-1, but not NFAT (Genot et al., 1996). Similarly, Racl was necessary, but not sufficient, for FceRI-mediated NFAT activation (and, hence, cytokine expression), acting either in parallel to, or downstream of,

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

35

Ras (Turner and Cantrell, 1997). Research in this important area is likely to expand in the near future. OF RHO-FAMILY GTPASESIN HEMATOPOIETIC CELLS D. REGULATION 1. Vau-An Integrator of Rho-Family and Ras Signals in Hematopoietic Cells The Dbl family of proteins consists of some -15 mammalian members that share a so-called Dbl-homology (DH) domain followed by a pleckstrinhomology (PH) domain. Dbl-family proteins function as physiological activators of Rho family GTPases, and display GEF activity for these GTPases (Cerione and Zheng, 1996). Among members of the Dbl family, special attention has been given to Vav because of its unique structural characteristics and specific expression in hematopoietic cells. Vav was originally isolated as the product of a transforming gene in fibroblasts, the result of a deletion of the amino-terminal 67 residues from the protooncogene product (Katzav et al., 1989). Vav is a 95-kDa protein that contains, in addition to its DH and PH domain, a cysteine-rich zinc-binding domain similar to that found in members of the PKC family, and an SH2 domain flanked by two SH3 domains. These domains mediate physical interactions with many unrelated proteins and with lipid second messengers (Romero and Fischer, 1996; Collins et al., 1997). It is, therefore, evident that, aside from its intrinsic enzymatic (GEF) activity, Vav is likely to serve as an adapter or docking molecule to help convey signals from membrane-associated signaling complexes. Severalkey findings have demonstrated the importance of Vav to hematopoietic cell function: first, Vav is phosphorylated on tyrosine in response to stimulation of a large variety of hematopoietic cell receptors and interacts with PTKs of the Src, Syk, Tec, and JAK families, suggesting a broad involvement in cellular responses; second, Vav-deficient mice display defective T and B lymphocyte development and responsiveness to antigen receptor-mediated activation signals (Fischer et al., 1995; Tarakhovsky et al., 1995; Turner et al., 1997; Zhang et al., 1995); third, overexpression of protooncogenic Vav in T lymphocytes activates transcription factors involved in IL-2 production, including NFAT, and potentiates TCRmediated activation of the IL-2 gene (Holsinger d al., 1995; Wu et al., 1995; Deckert et al., 1996);finally, Vav-deficient T cells were very recently found to display severe defects in cytoskeletal reorganization and actin-cap formation in response to TCR stimulation ( Fischer et al., 1998;Holsinger et al., 1998).The severe developmental and activation defects in lymphocytes of Vav-deficient mice represent perhaps the most compelling indirect evidence for a critical function of Rho-family GTPases in IRR signaling.

36

AMNON ALTMAN AND MARCEL DECKERT

Although Vav was originally reported to be an activator of Ras (Gulbins et al., 1993), more recent studies have shown that Vav functions as a GEF for Rho GTPases (Crespo et al., 1997; Han et al., 1997). Tyrosine phosphorylation mediated by Lck (Crespo et al., 1997; Han et al., 1997) and, possibly, by Syk-family kinases (Deckert et al., 1996) stimulates the exchange activity of Vav. The B cell negative regulatory receptor, CD22, may be linked to a signaling pathway that selectively inhibits the tyrosine phosphorylation of Vav and/or accelerates its dephosphorylation, thereby potentially down-regulating its exchange activity (Sat0 et al., 1997). Substrates or products of PI3-K were more recently found also to regulate the activity of Vav in a negative or a positive manner, respectively (Han et al., 1998). Dual regulation of Vav by PTKs and PI3-K-related lipids appears to be similar to the dual regulatory mechanism of Bruton tyrosine kinase (BTK) (Z. Li et al., 1997). The specificity of Vav for members of the Rho family is not fully established. Although one study demonstrated specific (or at least preferential) exchange activity for Rac (Crespo et al., 1997), two other studies based on in vitro exchange assays (Han et al., 1997) or morphological analysis of Vav-transfected COS cells (Olson et al., 1996) suggested that Vav regulates Rac, Cdc42, and Rho. Similar to Rac and Cdc42 (Coso et al., 1995; Minden et al., 1995; Olson et at., 1995; Teramoto et al., 1996), Vav overexpression in COS cells (Crespo et al., 1996; Olson et al., 1996) or in mast cells (Teramoto et al., 1997) leads to JNK activation. However, most of these studies utilized the oncogenic form of Vav, which is inert in T cells with regard to inducing transcriptional activation of the IL-2 gene (Wu et al., 1995). Thus, an important question that remains unsettled is whether, in the context of IRR signaling in hematopoietic cells, Vav is also a relevant physiological activator of the JNK cascade. In fact, findings indicate that TCWCDZ8-induced JNK activation remains intact in Vav-deficient T cells (Fischer et al., 1998; Holsinger et al., 1998). The positioning of Vav in lymphocyte activation pathways makes it an attractive candidate for coupling Ras- and Rho-family signaling pathways in hematopoietic cells (Fig. 5). It has recently become clear that these two families of small GTPases “talk’ to each other (Khosravi Far et al., 1995; Prendergast et al., 1995; Qiu et al., 1995a,b, 1997). Vav could fulfill the important function of integrating both pathways in hematopoietic cells. Such a function would be analogous to that of another Dbl-family member isolated from yeast, Scdl (Chang et al., 1994). Biochemical and genetic analysis has demonstrated that Scdl forms a complex in a cooperative manner with two small GTPases of the Ras and Rho families, Rasl and Cdc42sp, respectively. Although the functional relationship between Scdl and Cdc42sp has not been elucidated, it was determined that an active

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

37

(i.e., GTP-bound) Rasl acts upstream of Scdl to stabilize its interaction with Cdc42sp, respectively (Chang et al., 1994). As reviewed above, the link of Vav to Rho-family GTPases is evident from the demonstration of its in witro GEF activity for these GTPases; its ability to activate JNK via a Rac-dependent pathway, including in mast cells (Teramoto et al., 1997); the cytoskeletal changes induced in Vavtransfected cells; and the cytoskeletal defects in Vav-deficient T cells. However, it is clear that Vav also interacts with the Ras signaling pathway. First, Vav-mediated transformation of fibroblasts (Katzav et al., 1995) or NFAT activation in T cells (Wu et al., 1995) can be blocked by DN Ras. Conversely, DN Vav mutants also block fibroblast transformation by oncogenic Ras (Katzav et al., 1995) or Ras-dependent NFAT/AP-1 activation in T cells (Deckert et al., manuscript in preparation). Second, integrin receptor stimulation of neutrophils activated Ras and induced the association of Ras with tyrosine-phosphorylated Vav; moreover, a selective tyrosine kinase inhibitor prevented in parallel Ras activation, tyrosine phosphorylation of Vav, and the association between these two proteins (Zheng et al., 1996). Third, in mast cells, FcERI was found to associate with an active signaling complex that included Vav, Grb2, Raf-1, and ERK2 (Song et al., 1996). Finally, a phosphorylated Vav/SLP-76/Grb2 complex that also contained 36- to 38-kDa (most likely LAT) and 116-kDa (Cbl?) tyrosinephosphorylated proteins was observed in TCWCD28-activated antigenspecific T cells (Tuosto et al., 1996). At least three potential mechanisms through which Vav may be linked to Ras activation can be considered. The first two are related to Grb2. A proline-rich sequence in the amino-terminal Vav SH3 domain binds to the carboxy-terminal SH3 domain of Grb2 (Ye and Baltimore, 1994; RamosMorales et al., 1995), providing two potential mechanisms for linking Vav to Shc and Grb2 and, thus, potentially to Sos, the ubiquitous Ras GEF. First, a Shc/Grb2 complex may bind to Vav SH3. This, however, is unlikely to mediate interaction with Sos, because the SH3 domains of Grb2 have been shown to bind cooperatively to Sos (Li et al., 1993) and, therefore, would compete for association with Vav. Indeed, in some studies it has been difficult to detect a Vav/Grb2 complex in lymphocytes (Nel et al., 1995).The observed Grb2Nav association in intact activated T cells (Tuosto et al., 1996) or in in vitro “pull-down” experiments using Grb2 fusion proteins (Ramos-Morales et al., 1994) may be indirect and mediated by an adapter protein that binds both simultaneously, e.g., SLP-76 (Tuosto et al., 1996; Koretzky, 1997). Second, in activated cells, Shc could simultaneously bind phosphorylated Vav and Grb2 SH2 via its own SH2 domain or phosphorylated tyrosine residue(s), respectively. In this complex, the Grb2 SH3 domains would mediate association with Sos. Formation of such

38

AMNON ALTMAN AND MARCEL DECKERT

a Vav/Shc/Grb2/Sos complex could offer a potential mechanism for the integration of signals from the Ras pathway and a signaling pathway involving Rho-family small G proteins. A third potential coupling mechanism could be mediated by lipids that regulate Vav activity. Because the products of PI3-K up-regulate the exchange activity of Vav (Han et al., 1998), and PI3-K is an immediate effector of Ras (Rodriguez-Vicianaet al., 1994), a pathway could be envisaged in which IRR-stimulated Ras would activate PI3-K, and the lipids generated by activated PI3-K would, in turn,cooperate with tyrosine kinases to stimulate the exchange activity of Vav toward Rho-family GTPases. 2. Other Regulators Rho-family GTPases expressed in leukocytes are most likely regulated by the same kinds of regulators that have been characterized in other cell types, i.e., GEFs, GAPS, and GDIs. A comprehensive review of these regulators is beyond the scope of this article, but some pertinent studies are of interest. In general, the expression of small GTPases is not regulated at the transcriptional level. One exception is Race, which was found to be expressed in peripheral blood lymphocytes, purified B and T cells, thymus, and several B and T cell lines, but not in other tissues analyzed, including liver, brain, lung, heart, and kidney (Reibel et al., 1991). After 24 hr of in vitro stimulation with phytohemagglutinin A, 30- to 50-fold accumulation of Rac2 mRNA was observed in peripheral blood lymphocytes and in purified T lymphocytes. These findings suggest that Rac2 fulfills specific roles in leukocyte activation. One such role may be related to the NADPH oxidase system (Section III,F,4). This plasma membrane-associated multimolecular enzyme complex generates superoxide anions that serve as bactericidal agents in phagocytes. NADPH oxidase is known to be regulated by Rac (Bokoch, 1994). Despite the 92% homology between Racl and Rac2, the latter interacted 6-fold better with p67phox (Dorseuil et al., 1996). A posttranslational mechanism regulating the activity of Rho GTPases has been described in NK cells (Lang et al., 1996)and leukocytes (Laudanna et al., 1997). The phosphorylation of RhoA on a single serine residue (Ser108)by CAMP-dependentprotein kinase (PKA)increased the affinity of the active, GTP-loaded form of RhoA for its GDI, resulting in a translocation of RhoA from the membrane to the cytoplasm (Lang et al., 1996) and inhibition of its guanine nucleotide exchange (Laudanna et al., 1997). As a result, the cytotoxic activity of NK cells (Lang et al., 1996) or the chemoattractanttriggered integrin-dependent leukocyte adhesion of neutrophils or an IL8 receptor-transfected lymphoid cell line (Laudanna et al., 1997) were inhibited.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

39

IN IRR SIGNALING E. RHO-FAMILY EFFECTORS The evolving interest in the role of Rho-family GTPases as regulators of growth signals has been accompanied by an intense search for effectors of these GTPases. This search has led to the isolation of various proteins that bind to the activated (GTP-bound) forms of Rac, Cdc42, or Rho (Amano et al., 1996; Ishizaki et al., 1996; Joneson et al., 1996; Kolluri et al., 1996; Quilliam et al., 1996; Symons et al., 1996; Van Aelst et al., 1996; Watanabe et al., 1996; 1997; Aspenstrom, 1997; reviewed in Narumiya et al., 1997; van Aelst and D'Souza-Schorey, 1997; Hall, 1998) (Fig. 4). A detailed analysis of nearly all of these immediate effectors, with the exception of one (Section VI,C), in leukocyte function has not been conducted. However, studies have addressed the role of stress-activated protein kinases of the JNK and p38 families, which are downstream targets of Rac and Cdc42 (Coso et al., 1995; Minden et al., 1995; Olson et al., 1995), in IRR signaling in T cells, B cells, and mast cells. The pathway leading from Rad Cdc42 to JNWp38 activation in nonhematopoietic cells involves the p21activated protein kinase (PAK) family as an intermediate (Bagrodia et al., 1995; S. Zhang et al., 1995; Frost et al., 1996), but the role of PAKs in lymphocyte activation is less clear. TCR-mediated activation of T cells stimulates JNK (Su d al., 1994), its upstream activating kinase MEKK-1 (Faris et al., 1996), and its immediate activating kinases MKK4 (SEK1) and MKK7 (Matsuda et al., 1998), as well as activation of p38 (Salmon et al., 1997; Matsuda d al., 1998), its immediate activator MKK6 (Matsuda et al., 1998), and its downstream target MAPKAP kinase-2 (Salmon et al., 1997) (Fig. 5). Unlike ERK activation, which can be induced by TCR or phorbol ester stimulation alone (Su et al., 1994; Matsuda et al., 1998), MEKK-l/MKKUMKK7/ JNK or MKK6/p38 activation required CD28 costimulation because it was undetectable or very weak in the absence of CD28 (Su et al., 1994; Faris et al., 1996; Salmon et al., 1997; Matsuda et al., 1998). The two-signal requirement for JNKlp38 activation is also evident from the finding that it can be induced by a combination of phorbol ester and Ca" ionophore, but not by either of these signals alone (Su et al., 1994; Matsuda et al., 1998). The requirement for a Ca" signal reflects the important role of calcineurin, because activation of JNK and p38, as well as their upstream kinases, was sensitive to cyclosporin A (Su et nl., 1994; Matsuda et a]., 1998). Indeed, CA calcineurin was found to synergize with a PKC signal to activate JNK and the IL-2 promoter in T cells via a Rac-dependent pathway (Werlen et al., 1998). The important role of JNK and p38 in induction of IL-2 production is implicated by the finding that blocking of p38 by a specific pharmacological inhibitor (SB203580)or by a DN MKK6,

40

AMNON ALTMAN AND MARCEL DECKERT

as well as inhibition of JNK by expression of DN MKK7, abrogated transcriptional activation of the IL-2 promoter or NFAT (Matsudaet al., 1998). Further clues regarding the signals leading to JNK activation in T cells come from studies demonstrating that combinations of a CA form of Rac plus Syk tyrosine kinase (Jacinto et aL, 1998), or Vav plus CA calcineurin (Villalba-Gonzales and Altman, 1998), cooperate to activate JNK. In contrast to Syk, Lck did not cooperate with Rac to activate JNK. It has been suggested that Syk functions in a pathway parallel to Rac, which regulates PKC and calcineurin (Jacinto et al., 1998). The activation of various MAPKs in the murine immature B cell lymphoma, WEHI-231, was also analyzed in some studies. These cells undergo apoptosis in response to BCR ligation, which is prevented by coligation of the B cell costimulatory receptor CD40. BCR or CD40 cross-linking on freshly isolated or LPS-activated splenic B cells, or on WEHI-231 cells, resulted in activation of p38 and MAPKAP kinase-2 (Salmon et al., 1997). Inhibition of p38 activity by pretreating intact WEHI-231 cells with SB203580 had no effect on either BCR-induced apoptosis or anti-CD40mediated suppression of apoptosis. In another study, BCR ligation caused strong activation of ERK2, but only a weak or modest activation of p38 or JNK, respectively, in WEHI-231 cells; however, CD40 was a potent activator of JNK, p38, and MAPKAP kinase-2 in these B cells (Sutherland et al., 1996). These results suggest that, in this system, ERK2 activation correlates with B cell apoptosis, but that p38/MAPKAP kinase-2 activation is not required for BCR-mediated apoptosis. However, a different conclusion regarding the role of p38 in B cell apoptosis was reached in another study using a human B cell lymphoma line (B104). These cells undergo apoptosis in response to anti-IgM, but not anti-IgD, stimulation. In parallel, IgM, but not IgD, cross-linking led to activation of JNK and p38 in these cells. Similar activationwas induced by ionomycin, and ionomycin- or IgMinduced JNWp38 activation and apoptosis were inhibited in parallel by cyclosporin A (Graves et al., 1996).The difference between the two studies serves to emphasize the critical importance of the cellular context and developmental stage in determining the contribution of distinct signaling pathways to a particular biological outcome. This point is also evident from the finding that ligation of the incomplete BCR complex expressed on proB cells (which consists of the ITAM-containing IgdIgP signaling subunits bzlacks surface IgM), activates ERK2, but not JNK or p38 (Nagata et al., 1997). Several studies analyzed the activation of JNWp38 in mast cells. FcsRI ligation was found to lead to activation of MEKK-1 (Ishizuka et al., 1996, 1997), JNK (Ishizuka et. al., 1996, 1997; Kaga et al., 1997; Kawakami et al., 1997; Teramoto et al., 1997), and p38 (Ishizuka et al., 1997; Kawakami

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

41

et al., 1997; Zhang et al., 1997). The activation of JNK in mast cells was dependent on intact PI3-K activity (Ishizuka et al., 1996, 1997), and was mediated by a Vav/Rac-dependent pathway (Teramoto et al., 1997),consistent with the requirement for Rac in FceRI induction of NFAT in the same cells (Turner and Cantrell, 1997). Analysis of cultured bone marrowderived mast cells from Btk tyrosine kinase-deficient mice indicated that the FceRI-mediated activation of JNK and, to a lesser extent, of p38, but not ERK, was compromised, suggesting that Btk positively regulates JNK (and p38) activation (Kawakami et al., 1997). Different conclusions were reached in two studies with regard to the role of distinct MAP kinases in FceRI-stimulated TNF-a production by mast cells. Based on the use of specific pharmacological inhibitors of MEWERK (PD98059) or p38 (SB203580),or MEKK-1/JNK mutants, one study concluded that JNK, but not ERK, regulated TNF-m production by stimulated mast cells (Ishizuka et a/., 1997). In contrast, the other study (Zhang et al., 1997) demonstrated that ERK (but not p38) contributed to the FceRI-stimulated production of TNF-a and the release of arachidonic acid in these cells. Neither kinase, however, was essential for FceRI-mediated degranulation or constitutive production of TNF-0. The latter study also revealed that p38 negatively regulated activation of ERK and the responses mediated by this kinase (Zhang et al., 1997). Activation of neutrophils by FcyR cross-linkingor by trimeric G proteincoupled receptors was found to induce activation of PAK, a known downstream target of RacKdc42. However, only the FcyR-induced activation was dependent on cytoskeleton reorganization and intact PI3-K activity as revealed by the use of pharmacological inhibitors (Jones et al., 1998).

FUNCTIONS F. RHOGTPASESAND OTHERLEUKOCYTE Rho-family GTPases are coupled to leukocyte receptors other than IRRs, and regulate additional leukocyte functions such as cell adhesion and rolling mediated by integrins and selectins (reviewed in Springer, 1990, 1994; Hynes, 1992, 1996; Dunon et al., 1996), responses to chemoattractants and chemokines initiated by seven transmembrane-spanning and other receptors (reviewed in Bokoch, 1995; Downey et al., 1995), and the phagocyte NADPH oxidase system (reviewed in Bokoch and Knaus, 1994; Bokoch, 1995; Quinn, 1995; Dharmawardhane and Bokoch, 1997). In addition, these GTPases also regulate the process of programmed cell death, or apoptosis. These functions are briefly reviewed below. 1. Integrins and Cell Adhesion

Integrins are heterodimeric transmembrane receptors that couple components of the ECM to the actin cytoskeleton. Each integrin consists of

42

AMNON ALTMAN AND MARCEL DECKERT

an a and a p subunit, and, in mammals, 16 distinct a subunits combine with 8 p subunits to form over 22 receptors with distinct cellular distribution and biological functions. Binding of integrins to ECM proteins causes formation of complex protein structures, termed focal adhesions, which anchor the actin cytoskeleton to the plasma membrane. Integrin-mediated signals regulate various processes during embryogenesis and development, cell growth, motility, and survival and, in malignant cells, uncontrolled growth and metastasis. The best characterized signaling pathways triggered by integrins include activation of PTKs and Rho-family GTPases (reviewed in Clark et al., 1994; Schwartz et al., 1995; Dedhar and Hannigan, 1996; Hotchin and Hall, 1996; Parsons, 1996; Yamada and Geiger, 1997) (Fig. 6). Focal adhesion kinase (FAK) represents the major tyrosine kinase associated with focal adhesions (Zachary and Rozengurt, 1992; Parsons, 1996). FAK can couple ECM signals to the Ras signaling pathway via its association with the GrbWSos complex (Schlaepfer et al., 1994), leading to activation of Ras-dependent ERK kinases (Schlaepfer et al., 1994; Renshaw et al., 1996; Fig. 6). Although the integrin-mediated signaling pathways in lymphocytes remain largely unclear, integrin ligation by natural ligands can stimulate the Ras and Rho pathways in lymphocytes (Kapron-

FIG.6. The involvement of small GTPases in signaling by integrins and chemoattractant receptors. For details see Sections II,F,l and III,F,3.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

43

Bras et al., 1993; Schwartz et al., 1995). Integrin-dependent adhesion has been shown to generate additional signals in adherent cells, including turnover of phosphoinositides and Rho-dependent activation of its effector, phosphatidylinositol 4-phosphate 5-kinase (PIP5-K) (Chong et al., 1994). is a One major product of PIPS-K, phosphatidylinositol4,5-bisphosphate, substrate for both PLCy and PISK, two enzymes involved in the generation of critical lipid second messengers in IRR-stimulated leukocytes. Integrin-mediated adhesion to the extracellular matrix or APCs plays an important role in regulating lymphocyte homing and recruitment to inflammatory sites. Integrin ligation also generates a costimulatory signal for T lymphocyte proliferation (Van Seventer et al., 1991;Croft and Dubey, 1997). As an example, integrin a 4 p l and the TCR synergized to increase tyrosine phosphorylation of PLCy and calcium mobilization in T cells (Kanner et al., 1993). Conversely, IRR stimulation leads to an increase in the avidity of integrins for their ligands via an “inside-out” signalingpathway (springer, 1990; Schwartz,et al., 1995; Dedhar and Hannigan, 1996).Thus, FcyR cross-linking on human neutrophils was found to increase the avidity of a M P 2 integrin for its ligand via a pathway that required PI3-K activity and an intact cytoskeleton (Jones et al., 1998). The signals delivered by integrins are integrated with those originating from antigen or inflammatory cytokine receptors in order to activate reorganization of the cytoskeleton, gene expression, secretion, differentiation, and proliferation. The integration of the integrin-associated costimulatory signal with IRR signals may involve the association of the tyrosine-phosphorylated 6 subunit with polymerized actin (Caplan and Baniyash, 1995, 1996; Caplan et al., 1995; Rozdzial et al., 1995; Caplan and Baniyash, 1996) or the ITAM-associated Syklzap-70 kinases, because, for example, Syk was found to be required for BCR-dependent actin assembly in B lymphocytes (Coxet al., 1996).Two additional elements that could represent important points of integration between IRRs and integrins are Vav, which can undergo tyrosine phosphorylation (and, presumably, enzymatic activation) in response to ligation of both IRRs and integrins (Cichowski et al., 1996; Zheng et al., 1996; Gotoh et al., 1997), and FAK, which is phosphorylated and activated by integrin, including in leukocytes (T. H. Lin et al., 1995), or IRR (e.g., Haimovich et al., 1996; Berg and Ostergaard, 1997) stimulation. T helper cell-APC and CTL-target ceIl initial contact involves T cell-expressed /32 integrins (Springer, 1994), and PZ integrin (LFA-1)-dependent aggregation of lymphocytes (via ICAM-1 binding) is blocked by the Rho-inactivating C3 transferase exotoxin (Tominaga et al., 1993). 2. Selectins Regulation of T lymphocyte cell shape has to be considered as a critical process because the circulating T cell adheres to cellular and extracellular

44

AMNON ALTMAN AND MARCEL DECKERT

matrices, migrates through blood vessel endothelia, and interacts with APCs in the crowded environment of the lymph nodes. Lymphocytes express another family of adhesion receptors, the selectins, which are involved in the rolling process, the initial step of leukocyte adhesion to endothelial cells (Springer, 1994; Dunon et al., 1996; Rosen et al., 1997). This process may play an important role in the recruitment of T cells to inflammatory sites, as indicated by the finding that P- and E-selectin mediate recruitment of T helper-1 but not T helper-2 cells into inflamed tissues (Austrup et al., 1997). T cell adhesion to P-selectin was found to induce tyrosine phosphorylation of FAK and other substrates (Haller et al., 1997).L-Selectin ligation has been shown to trigger a signaling cascade involving activation of the Src-kinase Lck and association of GrbWSos with L-selectin in Jurkat T cells; this correlated with Ras, ERK, and Rac2 activation, and with a transient increase of reactive oxygen intermediates (Brenner et al., 1996). L-Selectin also induced cytoskeleton reorganization in the cells as demonstrated by a marked increase in F-actin content, which was dependent on intact Ras and Rac2 function (Brenner et al., 1997). 3. Chemoattractant Responses In addition to selectins, integrin-mediated cell arrest and adhesion can be triggered when seven transmembrane-spanning, trimeric G proteincoupled receptors are stimulated by their ligands, e.g., during stimulation of neutrophils by formyl peptides, leukotriene B4, or IL-8 (Fig. 6). These chemoattractants are released by endothelid cells during inflammation and generate signals that activate leukocytes, induce their integrin-mediated adhesion to the endothelial surface, and, finally, stimulate transmigration. In addition, cytokines coupled to tyrosine kinase signaling pathways, e.g., GM-CSF in neutrophils (Coffer et al., 1998), or colony-stimulating factor1(CSF-1) in macrophages (Allen et al., 1997),also act as chemoattractants. The precise chemoattractant signaling mechanisms are not known. However, the “conventional” model that implicated trimeric G proteins as central players has been updated to include the important contribution of tyrosine kinases. For example, formyl peptide stimulation of human neutrophils was found to stimulate tyrosine phosphorylation and activation of the Lyn kinase and its association with tyrosine-phosphorylated Shc and PI3-K (Ptasznik et al., 1995). Phosphorylated Shc and activated PIS-K might lead, in turn, to recruitment of GrbWSos and subsequent Ras activation (see Section II,B,2) or activation of Rac, respectively. In addition, activated Ras may stimulate PI3-K activity (Rodriguez-Vicianaet al., 1994) and other effectors. Indeed, chemotactic peptides activate Ras, Raf-1, BRaf, MEK-1, and ERKs in human neutrophils (Buhl et al., 1994; Fialkow et al., 1994; Grinstein et al., 1994; Worthen et al., 1994). In addition, a

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

45

tyrosine kinase (Lyn?)-dependent pathway is required for chemoattractantstimulated PI3-K activation in human neutrophils (Ptasznik et al., 1996). Numerous studies have shown that small GTPases mediate important signaling functions in neutrophil chemoattractant signaling (reviewed in Bokoch, 1995; Downey et al., 1995) (Fig. 6). Phospholipase At (PLA,), which is activated by phosphorylation and mediates the production of lipid second messengers (arachidonic acid and, subsequently, eicosanoids and leukotrienes), appears to represent a major target for the chemoattractantstimulated Ras signaling pathway. One important target for Rho GTPases is PIP5-K, which, on activation, generates PIP2. This is required for the halting of lymphocyte rolling and a firm integrin-mediated adhesion (Campbell et al., 1996). Therefore, as also discussed above, lipid metabolism enzymes and intermediates appear to serve important regulatory functions in Ras and Rho GTPase-mediated cytoskeletal reorganization (Hartwig et al., 1995) and lymphocyte adhesion, and this property is shared by integrins, selectins, and chemoattractants. Agonist stimulation was reported to stimulate within seconds guanine nucleotide exchange (i.e., activation) on RhoA in lymphoid cells transfected with formyl peptide or IL-8 receptors. Inactivation of Rho by C3 transferase exoenzyme blocked agonist-induced lymphocyte a 4 p l integrin adhesion to vascular cell adhesion molecule-1 (VCAM-1)and neutrophil P2 integrin adhesion to fibrinogen. These findings suggest that Rho participates in signals from chemoattractant receptors, which trigger rapid adhesion in leukocytes (Laudanna et al., 1996). Selective inhibitors of the MEWERK pathway (PD98059) or P13-K (wortmannin and LY294002) were used to investigate the roles of these kinases in the regulation of neutrophil effector functions. GM-CSF, platelet-activating factor (PAF), or a formyl peptide were capable of activating ERK1/2 and PI3-K in human neutrophils. ERK activation correlated with the stimulation of Ras by both tyrosine kinase and G protein-coupled receptors. PI3-K inhibition interfered, to various degrees, with superoxide generation, neutrophil migration, and PAF release. PD098059 treatment, however, inhibited only the PAF release stimulated by serum-treated zymosan. This demonstrates that, although MEUERK kinases are not involved in the activation of respiratory burst or neutrophil migration, they have a potential role in the activation of cytosolic phospholipase A2. PI3-K, however, seems to have a much wider role in regulating neutrophil function (Coffer et al., 1998). The requirement for Rho-family members in cytoskeletal events mediated by structurally diverse chemoattractant receptors was examined in RAW 264.7 monocytic cells transfected with the human chemotactic peptide receptor and stimulated with formyl peptide, CSF-1, IgG-coated parti-

46

AMNON ALTMAN AND MARCEL DECKERT

cles, or PMA. Expression of DN mutants of Racl or Cdc42 inhibited cytoskeletal responses to FMLP and CSF-1, blocked phagocytosis, and partially inhibited accumulation of F-actin-rich phagocytic cups. The finding that PMA-induced ruffling was not inhibited by expression of DN Racl, but was blocked by DN Cdc42, indicated that these GTPases acted via nonoverlapping pathways (Cox et al., 1997). The role of Rho-family GTPases in the chemotactic response of a CSFl-dependent murine macrophage cell line (Bac1.2F5) to CSF-1 was also analyzed by expression of DN or CA Racl, Cdc42, or RhoA mutants (Allen et at., 1997). In addition to stimulating the proliferation and motility of these cells, CSF-1 acts as a chemoattractant. CSF-1 rapidly induced actin reorganization in Bacl cells, evidenced by formation of filopodia, lamellipodia, and membrane ruffles at the plasma membrane, as well as the appearance of fine actin cables inside the cell. Microinjection of CA Racl stimulated lamellipodia formation and membrane ruffling and, conversely, a DN Rac mutant inhibited CSF-l-stimulated lamellipodia formation and induced cell rounding. CA Cdc42 induced the formation of long filopodia, whereas the DN Cdc42 mutant prevented CSF-l-induced formation of filopodia but not lamellipodia. Finally, CA RhoA stimulated actin cable assembly and cell contraction, but the Rho inhibitor, C3 transferase, caused a loss of these cables (Allen et al., 1997). Thus, Cdc42, Rac, and Rho regulate the formation of distinct actin filament-based structures during the chemotactic response of Bacl macrophages to CSF-1. 4. NADPH Oxidase

The chemotactic response of leukocytes at inflammatory sites is tightly coupled to activation of the NADPH oxidase system in phagocytic leukocytes. These cells use this specialized enzyme to generate reactive oxygen metabolites that kill microorganisms engulfed by the phagocyte (reviewed in Bokoch, 1994; Bokoch and Knaus, 1994). The active membraneassociated NADPH oxidase consists of four components: flavocytochrome b558, SH3-containing p47ph""and p67ph", and Rac. Rac is absolutely necessary for reconstitution of NADPH oxidase activity in purified protein preparations (Abo et al., 1991; Knaus et al., 1991). In resting neutrophils, Rac is associated in the cytosol with a Rho-GDI, which negatively regulates its activity by inhibiting guanine nucleotide exchange and membrane translocation. NADPH activation is accompanied by translocation of a Raclp47ph"/ p67ph" complex, in which Rac-GTP directly binds p67p'" (Diekmann et al., 1994), from the cytoplasm to the membrane. The Ras-related small GTPase Rapl is also associated with the complex, but is not necessary for activity. Rapl may negatively regulate the activity of NADPH oxidase

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

47

because PKA-elevating agents (e.g., CAMP) inhibit oxidant production, and Rap1 was found to be a substrate for PKA (Quilliam et al., 1991). 5. Phospholipase D The phospholipase D (PLD) family includes several isoenzymes that hydrolyze phospholipids to generate phosphatidic acid (PA).PLD enzymes are stimulated by both tyrosine kinase receptors and G protein-coupled receptors. As an example, T cell activation by either anti-CD3 antibodies (Mollinedo et al., 1994; Reid et al., 1997) or chemokines (Bacon et al., 1998) stimulates PLD activity. The hydrolysis product of PLD, PA, acts as a second messenger in various growth and differentiation signaling pathways, and it can be metabolized to form other intercellular and intracelMar lipid messengers (reviewed in Boarder, 1994; English, 1996; Olson arid Lambeth, 1996; Spiegel et al., 1996). Selective PLD stimulation has been found to induce actin stress fiber formation (Cross et al., 1996; Colley et al., 1997). In the past few years it has been established that, in addition to its regulation by tyrosine kinases (Natarajan et ul., 1996), PLD activity is also regulated by two families of small GTPases, i.e., the Rho and ADPribosylation factor (Arf)families (reviewed in Frohman and Morris, 1996; Kanaho et al., 1996; Olson and Lambeth, 1996; Wakelam et al., 1997). Members of the Arf family, which are located in the Golgi apparatus and represent major components of non-clathrin-coated vesicles, regulate intracellular protein transport (Nuoffer and Balch, 1994; Boman and Kahn, 1995). 6. Apoptosis

The process of programmed cell death, or apoptosis, mediated by interaction of Fas (CD95) with its ligand (FasL), is also influenced by small GTPases. Although a comprehensive review of this subject cannot be provided here, a few examples are noteworthy. The association of Ras with positive signals leading to cell growth would intuitively suggest that Ras does not participate in apoptosis. However, experimental evidence indicates that Ras can contribute to apoptosis in T cells, and that distinct signaling pathways cooperating with Ras determine whether the outcome of Ras stimulation will be growth or apoptosis. Thus, Fas ligation activated Ras in one (Gulbins et al., 1995), but not another (Wilson et al., 1996), study, and a DN Ras mutant or introduction of neutralizing anti-Ras antibodies into intact cells by electroporation inhibited Fas-mediated apoptosis in T cells (Gulbins et ul., 1995).Because Fas ligation activates the sphingomyelinase pathway that results in ceramide production, but does not cause PKC activation (Gulbins et al., 1995), it appears that one outcome of the cooperation between the Ras and sphingomyelinase signaling pathways is

48

AMNON ALTMAN AND MARCEL DECKERT

apoptosis; conversely, Ras cooperates with PKC to induce TCR-stimulated cell growth (Downward et al., 1990; Izquierdo et al., 1992a). This view is consistent with the findings that, under conditions wherein cellular PKC activity is inhibited, the expression of an activated Ras mutant caused apoptosis in T cells, which was antagonized by Bcl-2 (Chen and Faller, 1995, 1996). Along a similar line, PKC activation by PMA treatment was found to attenuate early signaling events induced by Fas ligation, i.e., cleavage of the CPP32 protease and its substrate poly(ADP-ribose) polymerase (Ruiz-Ruiz et al., 1997). Fas-mediated apoptosis is associated with the activation of JNK (Toyoshima et al., 1997; Wilson et al., 1996) and its activating kinase MKK7 (Toyoshimaet al., 1997),as well as activation of the MKKGIp38IMAPKAP kinase-2 pathway (Salmon et al., 1997; Toyoshima et al., 1997); the activation of these kinase pathways did not require CPP32-like proteases (Toyoshima et al., 1997). However, whether JNK or p38 activation is required for apoptosis is an unresolved question. In fact, another group reported that Fas-induced JNK activation occurred late, and that expression of a DN mutant of a JNK-activating kinase, MKK4, blocked Fas-induced JNK activation but had no effect on apoptosis. In addition, "-1 was not activated under these conditions (Lenczowskiet al., 1997). Similarly, pharmacological inhibition of p38IMAPKAP kinase-2 by SB203580 did not affect Fas-mediated apoptosis, indicating that p38 activation is probably not required for this event (Salmon et al., 1997). The Ras and JNK pathways can also affect apoptosis in T cells via regulation of FasL expression. Using a luciferase reporter construct containing elements of the FasL promoter in transient transfection assays, TCR-stimulated activation of the Ras signaling pathway was found to be required for optimal induction of the FasL (Latinis et at., 1997). In another study, inducible expression of CA MEKK1 (which activates the JNK pathway) resulted in apoptosis of Jurkat T cells in conjunction with prolonged JNK activation and induction of FasL expression (Faris et al., 1997). Ras is also involved in the apoptosis that cytokine-dependent cell lines undergo when deprived of the respective growth factor. By expressing a DN Ras mutant controlled by a tetracycline promoter in an IL-2AL-4dependent T cell line, it was found that Ras has a crucial role in both proliferation and prevention of apoptosis through the IL-2 receptor, whereas IL-4 promoted proliferation and inhibited apoptosis by Ras-independent signals (Gomez et al., 1996). Furthermore, Ras activation under conditions of IL-2 deprivation led to apoptotic cell death; in contrast, Ras enhanced proliferation in the presence of IL-2 (Gomez et al., 1997). Other studies have also implicated a role for Cdc42 or Rho in apoptosis. First, an infectious recombinant Sindbis virus expressing C3 exoenzyme

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

49

was used to infect EL4 T lymphoma cells. This resulted in modification and inactivation of virtually all endogenous Rho and, in parallel, in formation of multinucleate cells (likely by inhibiting the actin microfilament-dependent step of cytokinesis) and apoptosis (Moorman et al., 1996). Apoptosis occurred even when multinucleate cell formation was blocked by 5-fluorouracil, which induces cell cycle arrest. Second, an activated form of Cdc42 induced apoptosis in Jurkat T cells. This response was mediated by activation of a protein kinase cascade leading to JNK stimulation, and it was inhibited by expressing DN components of the JNK cascade or by caspase inhibitors (Chuang et al., 1997). The same group later demonstrated that Fas-mediated JNK activation and apoptosis were blocked by expression of a DN PAK mutant and, furthermore, that expression of the catalytically active C-terminal region of PAK in Jurkat cells, which is generated in situ during Fas-mediated apoptosis (Rudel and Bokoch, 1997), induced apoptosis (Rudel et al., 1998). IV. CD28 Signaling in T Cells: The Roles of Small GTPases

A. RASAND CD28 COSTIMULATION In addition to TCWCDS-derived signals, efficient T cell activation requires a second signal, which can be provided by a variety of costimulatory receptors expressed on T cells (reviewed in Van Seventer et al., 1991; Croft and Dubey, 1997). Among these, CD28 plays a critical role in IL2 production, autoimmunity, tumor immunity, and anergy (reviewed in Lenschow et al., 1996; Rudd, 1996; Ward, 1996; Chambers and Allison, 1997). The mechanisms of CD28 signaling are still incompletely understood, but the findings that CD28 becomes phosphorylated on tyrosine, and is associated with three defined signaling proteins, i.e., PI3-K (August et al., 1994; Prasad et al., 1994; Truitt et al., 1994), Grb2 (Schneider et al., 1995b; Kim et al., 1998), and the Itk/Emt tyrosine kinase (August et al., 1994; Marengere et al., 1997), represent important advances in this area (Fig. 7 ) . Although CD28 and the TCWCDS complex share some signaling properties, e.g., the ability to activate PIS-K, the two receptors clearly differ in several regards. In contrast to the TCR, CD28 cross-linking by its physiological ligand, CD80 (B7-1), fa& to stimulate Ras, Raf-1, and ERK2, or to induce tyrosine phosphorylation of LAT and SLP-76 (however, anti-CD28 monoclonal antibodies do induce these events); under the same conditions, a potent and prolonged phosphorylation of Vav on tyrosine is induced (Nunes et al., 1994). Conversely, CD28 but not TCWCDS crosslinking induces tyrosine phosphorylation of a 62-kDa protein (Nunes et al., 1996), which is most likely identical to ~ 6 (Carpino 2 ~ et ~al., 1997; Yamanashi and Baltimore, 1997).

50

AMNON ALTMAN AND MARCEL DECKERT

FIG.7 . The integration of CD28 costimulatory signals and TCR signah leading to T cell activation via small GTPases. Although the key elements integrating these signals remain largely unclear, they appear to operate on at least two levels. First, productive activation of PTKs and optimal phosphorylation of their substrates require the two sets of signals. Second, both the TCWCDS complex and CD28 are linked to Ras and Rho-family GTPases, and to their regulatory proteins (e.g., Sos, Vav, GAPS).Lack of the CD28 costimulatory signal results in TCR-induced anergy. Known PTK substrates in activated T cells are lightly shaded, and small GTPases are in black. In addition to the TCR, CD28 cross-linking was also found to induce cytoskeleton reorganization in T cells (Kaga et al., 1997).For details, see Sections TV and VI,A.

The association of CD28 with the GrbYSos complex (Schneider et al., 1995b; Kim et al., 1998) provides a potential link to Ras activation (Fig. 7 ) .However, it is more likely that CD28 ligation by its physiological ligand does not activate Ras in itself (Nunes et al., 1994), but rather potentiates Ras activation by TCWCDS triggering. This view is consistent with the finding that CD28 costimulation converges with TCWCDS signals to activate JNKs, AP-1, and the IL-2 promoter in T cells (Su et al., 1994; Fans et al., 1996). The CD28 costimulatory signal is essential because TCW CD3 ligation alone does not activate JNK (Su et al., 1994) or its upstream activating kinase, MEKK-1 (Gupta et al., 1994), in T cells (although it can activate Raf-1 and ERK). The ability of DN Ras to block JNK activation by TCR plus CD28 cross-linking (Fans et al., 1996) indicates that Ras regulates JNK activation by these two receptors.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

51

The inducible association of CD28 with PI3-K provides an additional potential mechanism through which CD28 could regulate Ras activation (Fig. 7), because PI3-K can activate the Ras signaling pathway in NIH 3T3 fibroblasts and Xenopus laevis oocytes (Hu et al., 1995), or function as an immediate Ras effector (Rodriguez-Vicianaet al., 1994). However, no definite information exists regarding a function of this putative mechanism in T cells. Although Grb2 and PI3-K (p85) bind to the same site in the phosphorylated cytoplasmic tail of CD28 (and, therefore, should compete for CD28 binding), both signalingproteins coimmunoprecipitate with CD28 from activated T cells. Thus, the relative contribution of PI3-K versus Grb2 association to CD28-mediated costimulatory signals remains to be determined. B. CD28 COSTIMUJATION Is LINKED TO RHO-FAMILY GTPASES Recent studies have connected CD28 signaling events to small GTPases of the Rho family (Fig. 7). First, CD28 cross-linking by CD80 induces a rapid and sustained increase in the tyrosine phosphorylation of Vav in the absence of TCR ligation (Nunes et al., 1994). Because the exchange activity of Vav is stimulated by tyrosine phosphorylation (Crespo et al., 1997; Han et al., 1997), this finding suggests that CD28 ligation would stimulate the guanine nucleotide exchange activity of Vav toward member(s) of the Rho family. This is consistent with the finding that stimulation of T cells with the CD28 ligand B7-2 promoted formation of focal adhesion-like plaques where Rho-family small G proteins accumulated (Kaga et al., 1997). However, it remains to be formally demonstrated that CD28-induced tyrosine phosphorylation of Vav enhances its exchange activity. Second, as mentioned earlier, CD28 costimulation is required for activation of JNK (Su et al., 1994) and MEKK-1 (Faris et al., 1996) in T cells. Because JNK activation is coupled to Rac and Cdc42 in fibroblasts (Coso et al., 1995; Minden et al., 1995; Olson et al., 1995) and in T cells (Villalba-Gonzales and Altman, 1998), and the oncogenic form of Vav was also found to induce JNK activation in COS cells (Crespo et al., 1996; Olson et al., 1996) and in mast cells (Teramoto et al., 1997), this finding suggests that CD28dependent MEKK-UJNK activation is mediated by a Vav/Rac (and/or Vav/ Cdc42) pathway. However, it is not known whether wild-type Vav alone activatesJNK in hematopoietic cells. Another T cell costimulatory receptor, CD5, was also found to stimulate a PI3-K-dependent signaling pathway regulated by Vav and Rac (Gringhuis et al., 1998).Similar to CD28 (Nunes et al., 1994), the B cell costimulatory receptor CD19 was found to induce potent tyrosine phosphorylation of Vav in the absence of BCR ligation (Sat0 et al., 1997).Thus, in terms of the underlying signaling events, CD19 coupling to Vav and PI3-K (Weng et al., 1994) and, potentially, to Rho-

52

AMNON ALTMAN AND MARCEL DECKERT

family GTPases, may represent an analogous costimulatory pathway to the one mediated by CD28 in T cells. Finally, CD28 (but not TCR) ligation leads to tyrosine phosphorylation of a 62-kDa adapter protein (Nunes et al., 1996), now known as ~ 6 (Carpino et al., 1997; Yamanashi and Baltimore, 1997). This PH domaincontaining protein forms a complex with p120 ras-GAP and p190 rho-GAP (Fig. 7). Tyrosine phosphorylation of ~ 6 is 2thought ~ to ~ modulate the cellular distribution of the associated GAPSand sequester them away from their activated Rac and Ras targets, thereby prolonging their activated to regulate state (Reif and Cantrell, 1998). Thus, ~ 6 has2 the~potential ~ both Ras and Rho GTPases. An additional possibility is that via its coupling to CD28 (but also to TCR), PI3-K could couple immune receptors to Rad Rho signaling pathways. This notion is based on the finding that PI3-K products can stimulate Rac- and Rho-mediated cytoskeletal responses in fibroblasts (Reif et al., 1996). V. The Function of Rab GTPases in Leukocytes

The Rab family of small GTPases includes more than 40 mammalian members that localize in a highly restricted manner in distinct membrane compartments and regulate different vesicular transport steps along the endocytic and secretory pathways. Cycles of GTP binding and hydrolysis by Rab proteins are linked to the recruitment of specific effector molecules on cellular membranes (Nuoffer and Balch, 1994; Olkkonen and Stenmark, 1997). Because hematopoietic cell activation triggers production and secretion of different soluble mediators (cytokines, chemokines, antibodies, cytotoxic granules, inflammatory mediators, etc.), and leukocytes undergo phagocytosis, it is reasonable to expect that Rab proteins regulate some of these processes. In addition, the functional dichotomy between MHC class I and class I1 molecules is reflected by the distinct intracellular processing pathways of MHC molecules and antigens in APCs. Therefore, Rab proteins, by controlling distinct transport steps along these routes, are likely to play important roles in antigen processing. Because different Rab proteins are associated with specific subcellular compartments, they could potentially serve as ideal markers to identif) distinct compartments associated with antigen processing (Chavrier et al., 1993). However, this potentially productive area has not been addressed. Levels of several Rab proteins were shown to increase during the differentiation of precursor cell lines into macrophage- or neutrophil-like cells (Bokoch, 1995). A few studies have addressed the function of Rab proteins in exocytic membrane fusion events leading to mast cell degranulation. Intracellular perfusion of mast cells with a nonhydrolyzable GTP analog, GTPyS (which functions as a nonspecific activator of trimeric and small

2

~

~

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

53

GTPases), is sufficient to trigger complete mast cell degranulation. This process could be mimicked by synthetic peptides corresponding to the effector domain of Rab3 but not other Rab proteins (Law et al., 1993; Oberhauser et al., 1992).A more direct approach for evaluating the role of Rab proteins in FcERI-stimulated mast cells was undertaken by transfecting RBL cells with wild-type or mutant forms of Rab3 proteins (Roa et al., 1997; Smith et aE., 1997). One of these studies led to the conclusion that active Rab3a functions as a negative regulator of degranulation by inhibiting an early stage of granule targeting to the membrane, but it does not regulate granule fusion with the plasma membrane (Smith et al., 1997). However, another study reported that RabSd, but not RabSa, negatively regulates mast cell exocytosis (Roaet al., 1997).Taken together, these studies indicate that Rab3 proteins selectively regulate mast cell secretion. One group reported that peripheral blood mononuclear cells from patients with SBzary syndrome and other lymphoid and myeloid malignancies overexpress Rab2 at the mRNA and protein level (Culine et al., 1992). This overexpression was restricted to nonmalignant CD2+ peripheral lymphocytes (Culine et al., 1993) and its significance is unclear. Another member of the Rab family, Rab5, which has been implicated in the regulation of early steps in the endocytic pathway, appears to play a role in the down-regulation of the TCWCD3 complex induced by TCR ligation (Andre et al., 1997). This modulation, which is mediated by receptor endocytosis, probably serves to attenuate T cell activation by depleting antigen-binding receptors from the cell surface, thereby regulating the magnitude and/or duration of the activation response. Analysis of transgenic mice expressing a DN form of Rab5 in their T cells revealed that mature thymocytes developed normally, but the absolute number of CD4+CD8+ double-positive thymocytes was reduced. Fluid-phase endocytosis was severely impaired in the transgenic thymocytes. In peripheral T cells, the kinetics and rate of ligand-induced TCR down-modulation were delayed and reduced. These effects were correlated with enhanced early and late signaling responses (Andre et al., 1997). These findings suggest that TCR endocytosis is an important regulatory component of TCR signahng and that defects in this regulation can result in prolonged signaling and altered thymic development. The effects of the Rab5 transgene lend potential physiologic relevance to the finding that PMA stimulation of peripheral blood neutrophils induced translocation of Rab5a from the cytosol to the membrane (Vita et al., 1996). W. Small GTPases and Aberrant Leukocyb Functions

A. RAs SIGNALING AND T CELLANERGY Clonal anergy is a well-established mechanism for the maintenance of peripheral T or B cell tolerance, which plays a key role in preventing

54

AMNON ALTMAN AND MARCEL DECKERT

reactivity to self antigens in the immune system (Goodnow, 1997; Schwartz, 1997). T cell anergy is usually induced when antigen receptor stimulation occurs in the absence of a costimulatory signal. The hallmark of anergic T cells is their inability to produce IL-2 or to proliferate in response to antigen stimulation. Ample evidence indicates that clonal anergy does not simply reflect a global failure of antigen receptor signal transduction but, rather, selective and aberrant triggering of a s’ubset of the signaling pathways that are normally induced by functional antigen receptor agonists (Alberola-Ilaet al., 1997), as reflected, for example, by the ability of anergic T cells to activate NFAT-1 and generate a normal increase in their intracelMar calcium concentration following antigenic stimulation (W. Li et al., 1996; Mondino et al., 1996; Schwartz, 1997). The obvious clinical implications of antigen-specific anergy have generated considerable interest in understanding the molecular basis of the aberrant signaling responses associated with this phenomenon. Several studies have implicated Ras (and, hence, its downstream elements) as a key target for the induction of T cell anergy (Fig. 1). The first clue that a defect in Ras activation may underlie T cell anergy came from a study in which it was demonstrated that activation of an IL2 promoter-reporter gene construct and, more specifically, activation of an AP-1 reporter and its DNA-binding activity were severely reduced in anergic T cells (Kang et al., 1992). A similar defect has since been documented by others (Mondino et al., 1996; Sundstedt et al., 1996). The defective AP-1 activation in anergic T cells results from the reduced nuclear expression of several inducible members of the Fos and Jun families of transcription factors that together combine to form AP-1 heterodimers (Foletta et al., 1998), i.e., c-Fos, FosB, and JunB (Mondino et al., 1996). The anergic state was not associated, however, with a global impairment of IL-2 gene induction based on the fact that other elements in the IL-2 gene promoter, i.e., NFAT-, NF-KB-, or Oct-binding sites, were not affected (Kang et al., 1992). Other studies reported, however, either a reduction in the transcriptional activation of reporter constructs containing multimerized NFAT-binding sites (Mondino et al., 1996) or a decrease in Fos/ Jun-containing NFAT complexes (Sundstedt et al., 1996). This reflects the fact that the AP-1 complex associates with NFAT and cooperatively binds to the NFAT site in the IL-2 promoter (Rao et al., 1997; Foletta et al., 1998), thereby stabilizing DNA association and transcriptional activation. Indeed, when NFAT activity is measured by more direct assays, i.e., dephosphorylation of NFAT and its nuclear translocation, it remains unaffected in anergic T cells (W. Li et al., 1996; Mondino et al., 1996),consistent with the intact TCR-induced Ca2+response in these cells (Mondino et al., 1996; Schwartz, 1997).

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

55

Inasmuch as AP-1 activation in T cells depends on functional Ras, reflecting primarily the Ras-dependent phosphorylation of c-Jun and the transcriptional upregulation of the c-jun gene by JNK (Su et al., 1994; Su and Karin, 1996), the AP-1 defect in anergic T cells suggested that Ras activity may have been affected. This notion was directly confirmed by demonstrating that stimulation of anergic T cells with antigen-pulsed APCs failed to activate Ras as measured by the increased level of GTP-bound Ras (Fields et al., 1996a,b). In accordance with this finding, the activation of ERK1/2, which is known to require functional Ras (Izquierdo et al., 1993), was also suppressed in the anergic cells (Fields et al., 1996a,b). Deficient ERK activation was confirmed by other studies that additionally documented defective activation of JNK (which depends on a combination of Ras and calcineurin-dependent signals for optimal activation in T cells) (Su et al., 1994; Su and Karin, 1996), as well as the p38 stress-activated protein kinase, in anergic T cells (DeSilva et al., 1996, 1997; W. Li et al., 1996). Although the activation of p38 in T cells has not been formally shown to require Ras, the ability of PMA alone to activate p38 in T cells (DeSilva et al., 1997) suggests that this may be the case. The defect in Ras activation did not reflect an intrinsic inability of Ras to become activated because Ras activation was restored in anergic T cells treated with a combination of PMA and calcium ionophore (Fields et al., 1996a,b). This observation is most likely a reflection of the ability of PMA to bypass TCR-mediated signals and activate Ras via a PKC-dependent pathway in T cells (Downward et al., 1990, 1992; Izquierdo et al., 1992a). It is also in agreement with the findings that PMA partially or completely restores antigen-induced proliferation and IL-2 production in anergic T cells (Bhandoola et al., 1993; Fields et al., 1996a,b; W. Li d al., 1996). However, the restoration of IL-2 production or even ERK activation in anergic T cells by PMA was not observed in other studies (DeSilva et al., 1996; Schwartz, 1997), perhaps indicating that, in addition to Ras, other signaling pathways are also affected during T cell anergy. This is also implied by the findings that a defect in Ca2+mobilization occurs in at least some in vivo models of T cell anergy (e.g., Blackman et al., 1991). Some of the observed discrepancies may reflect differences in the protocols used to induce anergy, e.g., induction by antigen-pulsed fixed APCs, polyclonal mitogens, immobilized anti-CD3 antibodies, or superantigen using either in vitro or in vivo systems. Perhaps more important, however, is the possibility that these differences reflect the ability of distinct biochemical pathways to induce anergy. Whether defective Ras activation is common to all pathways leading to T cell anergy remains to be determined. Defects in the Ras signaling pathway, reflected by reduced levels of GTP-bound Ras and tyrosine-phosphorylated (and, thus, active) ERK, as

56

AMNON ALTMAN AND MARCEL DECKERT

well as defective PKC activity, were also reported in another model of T cell anergy, i.e., in nonobese diabetic (NOD) mice (Rapoport et al., 1993b). The mature thymic or peripheral T cells of these mice are hyporesponsive to TCR-initiated activation signals, as measured by proliferation and IL2 or IL-4 production, and the development of hyporesponsiveness correlates with the onset of insulitis. This fact, as well as the finding that hyporesponsiveness and diabetic disease can be reversed in parallel by IL4 both in vitro and in vivo (Rapoport et al., 1993a), suggests a causal relationship between T cell hyporesponsiveness and autoimmune disease development (Rapoport et al., 1993a). The relationship of this anergic model to more commonly studied models of T cell anergy (Schwartz, 1997) is not clear, particularly because IL-4 (rather than IL-2) is effective in reversing NOD T cell hyporesponsiveness (Rapoport et al., 1993a). Nevertheless, the similarity between this model and others in terms of inhibited Ras and ERK activation implies a defect in Ras activation as a common underlying mechanism in different types of T cell anergy. Interestingly, a similar T cell hyporesponsiveness has been documented in human type I diabetes (De Maria et al., 1994), where it is associated with defective PKC activation and CD6Qexpression. As in most other anergy models, however, the anergic human T cells display a normal increase in their intracellular Ca2’ concentration following anti-CD3 stimulation. Because PKC can lead to Ras activation in T cells (Downward et al., 1990, 1992; Izquierdo et al., 1992a), and CD69 expression in T cells has been found to depend on functional Ras (D’Ambrosioet al., 1994),the defect in CD69 up-regulation suggests that here, too, Ras activation is adversely affected. Ras is clearly a critical target in the induction of T cell anergy, but much less is known regarding the molecular basis for the defect in Ras activation. As discussed earlier, Ras activation in response to antigen receptor ligation is mediated by PTK-dependent or a PKC-dependent (but PTKindependent) pathways (Downward et al., 1990, 1992; Izquierdo et al., 1992a; Hanvood and Cambier, 1993).Thus, each of the intervening events that couple receptor ligation to Ras activation is potentially a candidate target for anergy induction. Several studies addressed the contribution of upstream elements in the Ras activation cascade to the anergic state. The tyrosine phosphorylation of Shc, as well as the expression of a phosphoShc/Grb2/Soscomplex, which represents a major mechanism for Ras activation by growth factor receptors (Rozakis Adcock et al., 1992; Skolnik et al., 1993a,b; Pronk et al., 1994; de Vries Smits et al., 1995), were found to be intact in anergic T cells (Fields et al., 1996a,b). However, this study did not determine whether this complex (or a LAT/GrbYSos complex) is properly recruited to the T cell plasma membrane, a step necessary for Sos-mediated Ras activation (Aronheim et al., 1994; Quilliam et al., 1994;

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

57

Holsinger et al., 1995). Indeed, it has recently been reported that in the NOD mouse anergy model, TCR stimulation leads to deficient membrane translocation of Sos, and Sos (as well as PLCyl) was excluded from the phospho-4'-associated (and, hence, membrane-localized) G r b m T E a p 70 signaling complex (Salojin et al., 1997), implying that this complex is incompetent to activate Ras. Although one TCR-induced Ras activation pathway is mediated by PTKs (Izquierdo Pastor et al., 1995), it is unlikely that a global defect in the activation of TCR-coupled PTKs underlies the deficient Ras activation in anergic T cells, because, as mentioned earlier, the TCR-induced Ca2+ response, which also depends on TCR-coupled PTKs (June et nl., 1990; Mustelin et al., 1990),is usually intact in these cells (Mondino et al., 1996; Schwartz, 1997). Nevertheless, several studies reported abnormal tyrosine phosphorylation profiles in anergic T cells, including the lack of (Migita et al., 1995), or differential (Boussiotis et al., 1996), tyrosine phosphorylation of the TCR-associated f chain, deficient tyrosine phosphorylation of 38- and 74-kDa proteins (Bhandoola et al., 1993; Cho et al., 1993), and hyperphosphorylation of Fyn and/or concomitant increase in its catalyhc activity (Gajewski et al., 1995; Boussiotis et al., 1996; Salojin et al., 1997). In parallel, a reduction in the expression of Lck and, conversely, an increase in the level of Fyn (Quill et al., 1992), deficient recruitment of Zap-70 to 4' [which would be expected if f tyrosine phosphorylation was deficient (Migita et al., 1995; Salojin et al., 1997)], and decreased levels of Lck associated with the CD3/@ap-70 complex (Boussiotis et al., 1996) were also observed in anergic T cells. Consistent with the finding that, in the absence of CD28-mediated costimulation, TCR engagement alone leads to anergy (Schneider et al., 1995a; Schwartz, 1997),CD28 coimmunoprecipitated with the phospho-&/Lck/Zap-70complex in productively activated, but not in anergized, T cells (Boussiotiset al., 1996).Taken together, these findings suggest the following scenario: in anergic T cells, TCR engagement in the absence of CD28 costimulation leads to deficient Lck activation and hyperstimulation of Fyn. Consequently, the CD3 subunits and 4'are hypophosphorylated or abnormally phosphorylated in a manner that makes them incompetent to recruit Zap-70 via its tandem SH2 domains and activate it, a situation perhaps akin to the aberrant phosphorylation in T cells stimulated by altered peptide ligands (Sloan-Lancaster et al., 1994; Madrenas et al., 1995). As a result, Zap-70 and/or Lck cannot phosphorylate and recruit to the membrane adapters (or enzymes) that are critical for the activation of the Ras signaling pathway, such as Shc, LAT, Grb2, or Sos (Fig. 1).The deficient membrane translocation and Grb2 association of the Ras activator Sos (Salojinet al., 1997) are consistent with this scenario.

<

58

AMNON ALTMAN AND MARCEL DECKERT

What, on the other hand, are the consequences of Fyn hyperactivity in anergic T cells? One important outcome may be an increase in the tyrosine phosphorylation of Cbl (Boussiotis et al., 1997; Salojin et al., 1997) Cbl is an important adapter protein and a prominent PTK substrate whose exact role in regulating diverse receptor signaling pathways is still unclear (Liu and Altman, 1998). The recruitment of hyperphosphorylated Cbl to the TCR-associated signaling complex in anergic T cells may block Ras activation by at least two independent mechanisms: first, it may directly inhibit the activity of Syk-family PTKs (Syk or Zap-70) ( Fig. l),which are required for Ras activation, as recently documented in FcsRI-stimulated mast cells (Ota and Samelson, 1997); second, hyperphosphorylated Cbl may recruit to the membrane increased levels of a complex consisting of CrkL and C3G (Liu and Altman, 1998) (Fig. 1). C3G functions as a GEF that activates Rapl (Gotoh et al., 1995), a Ras-related small GTPase known to antagonize the activation of Ras (Hata et al., 1990; Kitayama et al., 1990). The increased Fyn/Cbl/CrkL/CSG complex expression and the selective activation of the l/FynlCbl pathway (Salojin et al., 1997) in anergic T cells, as well as the inhibition of TCR-induced Ras activation by transiently overexpressed Rapl in Jurkat T cells (Boussiotis et al., 1997),are consistent with this model. Furthermore, Raf-1 activation and its association with Ras are not detectable in the anergic T cells that display high constitutive Rapl activity;instead, Raf-1 is associatedwith Rapl (Boussiotiset al., 1997). Finally, the lack of CD28-derived costimulation, which is a hallmark of T cell anergy (Schneider et al., 1995a; Schwartz, 1997), may represent another potential mechanism underlying the Ras activation defect in anergic T cells (Fig. 7). Ligation of CD28 leads to tyrosine phosphorylation of its cytoplasmic tail and, as a result, to recruitment of two SH2-containing signaling proteins, i.e., PI3-K and Grb2 (Schneider et al., 1995a,b; Kim et al., 1998). Because the Grb2-Sos complex plays an important role in Ras activation (Buday and Downward, 1993; Egan et al., 1993; Li et al., 1993; Rozakis Adcock et al., 1993),and PISK was found to either activate Ras (Hu et al., 1995) or act as a Ras effector (Rodriguez-Vicianaet al., 1994), Ras activation may not properly occur in the absence of Grb2 and/or PI3K recruitment by CD28. This notion is consistent with the finding that CD28 cooperates with the TCWCD3 complex to activate Ras and JNK in T cells (Su et al., 1994; Faris et al., 1996), or can activate Ras by itself (Nunes et al., 1994). The significance of the latter response is unclear, however, because it is induced only by cross-linking anti-CD28 antibodies, but not by physiological CD28 ligands (Nunes et al., 1994). The association of anergy with aberrant signaling responses in B cells has been addressed much less extensively than in T cells. Nevertheless, a recent study investigated this issue by analyzing several signaling events

GTPaves IN IMMUNE RECOGNITION RECEPTOR SIGNALING

59

in B cells from mice expressing a hen egg lysozyme (HEL)-specific immunoglobulin transgene (Healy et al., 1997). When B cells of these mice are chronically exposed to HEL during their development, they become selftolerant (i.e., anergic) to HEL challenge. Comparison of these cells with naive B cells responding to a foreign antigen revealed that although stimulation of NFAT transcriptional activity and activation of the ERK or pp90nk kinases were intact in the anergic B cells, Ca2+oscillations were reduced, and the activation of NF-KB or JNK was absent. This behavior is different from anergic T cell clones in which the Ca2+ response usually remains intact but, conversely, ERK activation is deficient. However, as mentioned above, several in vivo models of T cell anergy also display a deficiency in Ca2+ mobilization (Blackman et al., 1991). The intact ERK activity in anergic B cells indicates that, at least in this system, the Ras signaling pathway is not a target for anergy induction. Additional studies on B cells will be required in order to determine whether a fundamental difference exists between T and B lymphocytes with regard to the signaling pathways affected by anergy. B. HIV INFECTION The ability of human immunodeficiency virus type 1(HIV-1) and several of its defined gene products to modulate the function of T cells via diverse pathways is well established (Miedema et al., 1990; Feinberg and Greene, 1992; Shearer and Clerici, 1992). Among these interactions, the ligation of the CD4 coreceptor by thesoluble or antibody-complexed HIV envelope glycoprotein (gpl20)was found to induce an inhibitory signal that interferes with productive T cell activation induced by subsequent stimulation of the TCFUCDS complex (Capon and Ward, 1991; Chirmule and Pahwa, 1996). This inhibitory response is manifested by T cell apoptosis (Banda et al., 1992; Groux et al., 1992) or anergy (Diamond et al., 1988; Linette et al., 1988; Goldman et al., 1997),phenomena that can also be induced in CD4+ cells when CD4 is cross-linked by specific antibodies prior to TCFUCDS engagement (Rosoff et al., 1987; Haughn et al., 1992). This mechanism may contribute in an important way to the apoptosis and/or anergy commonly found in T cells of HIV-1' individuals (Miedema et al., 1990; Feinberg and Greene, 1992; Shearer and Clerici, 1992; Ameisen et al., 1995). The anergic state induced by the interaction of gpl20 with CD4 appears to be more profound and to differ from the more conventional clonal T cell anergy systems discussed earlier, in that it also involves a pronounced defect in Ca" mobilization in response to TCWCD3 stimulation (Linette et al., 1988; Jabado et al., 199%). However, in agreement with other experimental models of T cell anergy, the Ras signaling pathway also

60

AMNON ALTMAN AND MARCEL DECKERT

appears to be a target. The connection between HIV-1 infection and the Ras pathway represents a bidirectional relationship. On one hand, activated forms of Ras (Baldari et al., 1992a) or one of its immediate effectors, Raf1 (Popik and Pitha, 1996), were found to stimulate HIV-1 replication or transcriptional activation of the HIV-1 long terminal repeat (LTR) in T cells. This is compatible with the fact that T cell activation promotes HIV1replication on one hand (Feinberg and Greene, 1992), and the evidence reviewed earlier that the Ras pathway is required for productive TCRinduced T cell activation. Thus, it is conceivable that activation of the Ras pathway induced by antigen stimulation plays an important role in rendering the TCR-ligated cells permissive for HIV-1 infection. On the other hand, binding of the HIV-1 envelope glycoprotein to CD4, in particular when it leads to CD4 cross-linking (as would be expected when the intact virus or gp1201160 immune complexes bind to CD4' cells), generates signaling events that can involve components of the Ras signaling pathway. The nature of CDCgenerated activation signals is somewhat controversial. Earlier studes indicated that the activation signal generated by CD4 cross-linking is limited to activation of the associated Lck (Veillette et al., 1989) and possibly, as a result, activation of ERK (Ettehadieh et al., 1992). However, some anti-CD4 monoclonal antibodies were later found to induce independently complete T cell activation as measured by proliferation and IL-2 production (Carrel et al., 1988). Others have documented additional anti-CD4-induced responses such as tumor necrosis factor a and interferon y (IFN-y), but not IL-2 or IL-4, production (Oyaizu et al., 1994), Ca2' mobilization (Baldari et al., 1995a; Milia et al., 1997), increase in the tyrosine phosphorylation of Shc (Baldari et al., 1995a; Milia et al., 1997) or ezrin (Thuillier et al., 1994), and activation of Fyn (Baldari et al., 1995a), AP-1 (Chirmule et al., 1995), and NFAT (Baldari et al., 1995a,b). Inasmuch as AP-1 and NFAT activation in T cells is Ras-dependent, these findings suggest that Ras was also activated by CD4 cross-linking, although this was not directly examined. Discrepancies among various studies with regard to the range of activation responses induced by CD4 cross-linking most likely reflect differences among the epitopes being recognized by the antibodies, the degree of cross-linking, and/or the nature and activation state of the T cells under analysis (e.g., Baldari et al., 1995a). Similar to anti-CD4 antibodies, gp120/160 or inactivated HIV-1 was found to induce: Ca" mobilization in some (Milia et al., 1997), but not other (Baldari et al., 1995a), studies; increased tyrosine phosphorylation of cellular proteins (Schmid Antomarchiet al., 1996),including Shc (Baldari et al., 1995a; Milia et al., 1997), Vav (Tamma et al., 1997b), and Lck and PIS-K (Briand et al., 1997); activation of Ras (Tamma et al., 199%) and the kinases Fyn (Baldari et al., 1995a) Zap-70 (Tamma et al., 1997b), Raf-

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

61

1(Popik and Pitha, 1996),phosphatidylinositol4-kinase, and ERK (Schmid Antomarchi et al., 1996); and stimulation of the transcription factors AP1 (Chirmule et al., 1995; Briant et al., 1996) or NF-KB (Benkirane et al., 1994; Chirmule et al., 1994; Briant et al., 1996), but not NFAT (Baldari et al., 1995a). HIV-1 infection also activates NF-KB in monocytes (Folgueira et d.,1996).The essential role of the Ras cascade in these processes is implicated by the findings that, first, DN Ras or Raf mutants can inhibit the HIV-1- or gp120/160-induced stimulation of TNF-a production (Tamma et al., 1997a) or NF-KB activity (Folgueira et al., 1996), and, second, DN MEK-1 or ERK2 inhibited AP-1 or NF-KB activation induced by inactivated HIV-1 (Briant et al., 1998). Some controversy exists among the different studies with regard to Ras activation and Ras-Raf coupling. Thus, HIV-l-induced Ras activation could not be detected in one study (Popik and Pitha, 1996). However, the same study documented the activation of Raf-1 and, furthermore, demonstrated that Raf-1 was not associated with Ras under the same conditions, but rather with activated Lck, leading to the conclusion that a Ras-independent, LcWRaf-dependent pathway is involved in NF-KB activation and HIV-1 replication (Popik and Pitha, 1996). The mechanism mediating the HIV-l-induced Ras activation is unknown, but could involve the tyrosine phosphorylation of Shc (Baldari et al., 1995a; Milia et al., 1997) and its subsequent recruitment to the activated TCR complex, presumably as a complexwith GrbYSos (Ravichandran et al., 1993). According to current dogma, the inhibition of TClUCD3-mediated activation by prior ligation of CD4 (by anti-CD4 antibodies or HIV-1 binding) reflects the fact that CD4 ligation sequesters the CD4/Lck complex and prevents it from gaining access to the activated TCR complex. Assembly of a productive signaling complex requires association of the engaged TCR and its associated PTKs (i.e., Zap-70, Syk, and/or Fyn) with the CD4/Lck (or CDWLck) complex, presumably reflecting the essential role of Lck in phosphorylating ITAM motifs in a manner that allows them to recruit Zap70 and/or Syk. In the absence of such association, the ITAM motifs in the TCWCD3 will not be properly phosphorylated and Zap-70 (and perhaps Syk) will not be activated. As a result, the TClUCD3 complex will not generate the signals required for complete activation. This scheme is compatible with several observations: first, CD4 (or CD8) physically associates with the TCR during productive T cell activation (Rivas et al., 1988; Mittler et al., 1989; Rojo et al., 1989) and cross-linking of CD4 or CD8 to the TCR complex markedly increases T cell activation (Sleckman et al., 1987); second, Lck is present in phospho-t immune complexes from productively activated, but not from anergic, T cells (Boussiotiset al., 1996);and, finally, Fyn leads to a different pattern of { tyrosine phosphorylation compared

62

AMNON ALTMAN AND MARCEL DECKERT

to Lck, and this is associated with a failure of phospho-%to recruit and activate Zap-70 (Boussiotis et al.,1996). In agreement with this scheme, CD4 ligation by gp120/160 was found to prevent several discrete signaling events induced by subsequent TCR ligation. Of interest, one study reported that human CD4' T cells from early HIV infection states are characterized by high levels of Fyn activity and low levels of Lck activity (Cayota et al., 1994), a situation highly reminiscent of anergic T cells studied in other models (Quill et al., 1992; Gajewski et al., 1995; Boussiotis et al., 1996; Salojin et al., 1997). The reduced Lck activity may correlate with the abnormal redistribution of Lck from the detergent-soluble to the detergent-insoluble cellular fraction following binding of gp120 immune complexes to CD4' T cells (Goldman et al., 1997).Other aberrant responses include, in addition to the defective IL-2 production and Ca" mobilization mentioned earlier (Linette et al., 1988; Jabado et al., 1997b), a reduction in the tyrosine phosphorylation and/or activities of PLCyl (Tamma et al., 1997a), several kinases such as Fyn, Raf-1, Erk, and JNK, and the transcription factors NFAT, AP-1, and NF-KB, and decreased TNF-a production (Jabado et al., 1997a). The reduced activation of several Ras effectors (Raf-1, ERK, and JNK) implicates Ras as a target, a notion directly confirmed in one of these studies (Tamma et al., 1997a).Thus, Ras is one of several targets for the inhibitory action of HIV-l-mediated CD4 ligation. HIV-l-derived proteins can affect the activity and/or localization of small GTPases in other ways. For example, gp120 was found to bind members of the ezridradixidmoesin (ERM) family of proteins (Hecker et al., 1997), which link the plasma membrane to the cytoskeleton (Tsukita and Yonemura, 1997). Because Rho regulates the association of ERM proteins with the plasma membrane and/or cytoskeleton (Hirao et al., 1996; Kotani et al., 1997), gpl20 binding may indirectly affect Rho activity and its interaction with its effectors. Finally, the Nef protein, which is known to have different immunomodulatory activities, represents another HIV-1 component that can modulate cellular functions via its interactions with small GTPases. The most clearcut example is provided by the recent finding that Nef physically associates with, and activates, a cellular serinelthreonine kinase activity termed Nefassociated kinase (NAK) (Lu et al., 1996; Nunn and Marsh, 1996), which is functionally and structurally related to the PAK family of RadCdc42 effectors. A role for Cdc4YRac in these activities of Nef is implicated by the findings that DN Cdc42LRac mutants inhibited the activation of NAK by Nef and, conversely, CA forms of Cdc4YRac potentiated the activation of NAK (Lu et al., 1996; Nunn and Marsh, 1996).The biological relevance of these findings is evident from the observation that DN Cdc4YRac or

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

63

PAK decreased the level of HIV-1 production in infected cells to those of virus from which the nefgene has been deleted (Lu et al., 1996). These findings implicate Rac/Cdc42 in regulating HIV replication. Nef may also indirectly affect the Ras signaling pathway via two potential pathways. First, because Nef was found to associate with PKCB (Smith et al., 1996), and Ras functionally interacts with a PKCB-dependent pathway leading to AP-1 and JNK activation (Baier Bitterlich et al., 1996; GhaffariTabrizi et al., 1998; Werlen et al., 1998), Nef could potentially modulate the function of Ras. Second, Nef was found to interfere selectively with a platelet-derived growth factor-mediated PI3-K signaling pathway (Graziani et al., 1996). The documented interactions between PI3-K and Ras (Rodriguez-Vicianaet al., 1994; Hu et al., 1995) raise the possibility that this inhibition of PI3-K signaling may also involve Ras or downstream targets of the Ras signaling pathway. C. WISKOTT-ALDRICH SYNDROME Wiskott-AIdrich syndrome (WAS) is a rare X-linked recessive immunodeficiency characterized by eczema, thrombocytopenia, recurrent infections, and cytoskeletal abnormalities in hematopietic cells (Kirchausen and Rosen, 1996). The most severe pathology occurs in platelets and T lymphocytes. T and B Iymphocytes of WAS patients display a defective response to antigen receptor stimulation that is manifested at the level of proliferation, IL-2 production, up-regulation of activation antigens, and actin polymerization (Molina et al., 1993; Henriquez et al., 1994; Gallego et al., 1997). However, tyrosine phosphorylation and Ca" mobilization are intact (Molina et al., 1993; Henriquez et al., 1994), indicating that the primary defect is restricted and occurs at a point downstream of PTK activation. The mutated gene in WAS, isolated in 1994,was found to encode a 501-amino acid proline-rich protein (WASp) expressed in lymphocytes, spleen, and thymus (Derry et al., 1994). A significant breakthrough in understanding the molecular basis of WAS occurred in 1996 when three groups independently found that WASP is a direct effector of Cdc42, i.e., it selectively associated with the active, GTP-bound form of this small GTPase (Kolluri et al., 1996; Symons et al., 1996; Aspenstrom, 1997). This finding suggested that complex formation between active Cdc42 and WASP plays an important role in reorganization of the actin cytoskeleton and, more specifically, in the formation of filopodia, which are known to be induced by activated Cdc42 (Nobes and Hall, 1995).The cytoskeletal abnormalities observed in hematopoietic cells of WAS patients (Kirchausen and Rosen, 1996),the colocalization of WASP with actin, and the induction of ectopic actin polymerization by WASP overexpression (Symons, 1995) strongly support such a role for WASp. As

64

AMNON ALTMAN AND MARCEL DECKERT

reviewed earlier, Cdc42 mutants have been shown to interfere with the polarization of antigen-specific T cells toward APCs (Stowers et al., 1995), and aberrant Cdc42-WASP interactions may also contribute to this defect. It remains unclear, however, whether the cytoskeletal defects caused by the WAS mutation represent the primary underlying cause of the deficient antigen receptor-induced lymphocyte proliferation in WAS patients. It is now well established that Rho-family small GTPases can stimulate independent signaling pathways leading to cellular growth or cytoskeleton reorganization (Hill and Treisman, 1995; Olson et al., 1995; Symons, 1995; Joneson et al., 1996; Treisman, 1996; Westwick et al., 1997), thus it is possible that mutated WASP directly interferes with Cdc42-mediated growth signals independently of its effects on the cytoskeleton. Indeed, WASP or its neural homolog, N-WASP (Miki et al., 1996), are known to associate, via their proline-rich region, with the SH3 domains of several signalingproteins that couple receptors to growth signals. These include adapters such as Nck (River0 Lezcano et al., 1995) and Grb2 (Cory et al., 1996; Miki et al., 1997; She et al., 1997), and lymphocyte-expressed kinases, i.e., the Tec-family members Itk/Emt and Btk (Bunnell et al., 1996; Cory et al., 1996), or Fyn (Banin et al., 1996). The WASp-Grb2 association also raises the possibility that WASP may indirectly regulate Ras activation by the GrbWSos complex. It is not known, however,whether a trimolecular WASP/ GrbWSos complex exists in intact cells and, if so, whether WASP association can modulate the exchange activity of Sos for Ras. MI. The Role of Small GTPases in Lymphocyte Development

A. RAs PROTEINS AND THEIREFFECTORS The development of mature T and B lymphocytes from hematopoietic precursors in the thymus or the bone marrow, respectively, is a complex and highly regulated process that has been extensively studied (reviewed in Anderson et al., 1996; Aguila et al., 1997; Kearney et al., 1997; Kipps, 1997). In order to generate mature T and B lymphocytes expressing highly diverse repertoires of antigen receptors and, at the same time, avoid potentially harmful recognition of self-antigens, developing lymphocytes are subjected to a series of positive and negative selection steps, during which they undergo maturation, programmed cell death, and cellular proliferation. These processes are initiated by TCR or BCR engagement in combination with costimulatory signals provided by cytokines and adhesion receptors. The ensuing signals are relayed within the cell through the assembly of signaling complexes that are similar to those operating in mature lymphocytes, e.g., various nonreceptor tyrosine kinases (Perlmutter, 1995). A role of Ras- and Rho-family small GTPases during these processes is therefore

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

65

not surprising. The use of genetic approaches in the mouse has been illuminating in generating data that establish the role of small GTPases during lymphocyte development, particularly with regard to T cells. Several reports have recently demonstrated the essential role of Ras and some of its effectors in controkng T cell development. The most direct evidence comes from studies in which a DN Ras (RasNl7) transgene was expressed under control of the proximal lck promoter, either in otherwise normal mice or in TCR-transgenic mice (Swan et al., 1995; AlberolaIla et al., 1996). The positive selection of double-positive (DP) CD4'8' thymocytes was severely compromised in these mice, whereas differentiation of the double-negative (DN) CD4-8- thymocytes into DP thymocytes, negative selection, tyrosine phosphorylation, and Ca2+mobilization proceeded normally in the thymocytes. As expected, expression of the DN Ras transgene prevented the anti-CD3-induced activation of ERK and blocked proliferation. Similar findings were obtained in mice expressing DN MEK1, the immediate activator of ERK kinases (Alberola-Ila et al., 1995, 1996). The generation of T cells expressing the yi3 TCR complex and of immature CD4-8- TCR-a/3+ T cells was normal despite blockade of the MAPK cascade. These findings suggest that activation of the MAPK cascade by Ras is necessary for positive selection, but is irrelevant for negative selection. In contrast to the above conclusions, analysis of the effect of retrovirally transduced DN or CA MEKl in fetal thymic organ cultures revealed that an intact MAPK signaling pathway is required for the differentiation of DN thymocytes into DP thymocytes (Crompton et al., 1996).The discrepancy may reflect qualitative differences in the signals regulating this transition between fetal (Crompton et al., 1996) and mature (Alberola-Ilaet al., 1995,1996) thymocytes, the use of different DN MEKl mutants, and/or differences in the expression systems used. The importance of Ras and its effectors in T cell development is also apparent from other studies. Using the RAG-2-deficient blastocyst complementation system (Chen et al., 1993), transgenic expression of CA Ras in the thymus induced the differentiation of DN prothymocytes into DP cortical thymocytes with accompanying expansion to normal thymocyte numbers. However, activated Ras was not sufficient for promoting proliferation or maturation of the DP thymocytes to CD4+8- or CD4-8+ thymocytes, suggesting that additional signaling pathways are required (Swat et al., 1996).Second, consistent with its function as an immediate Ras effector, expression of a DN Raf-1 transgene in the thymus also inhibited positive selection and TCR-induced proliferation and, conversely, CA Raf-1 enhanced the differentiation of DP thymocytes into mature, single-positive thymocytes (O'Shea et al., 1996). Finally, ERK kinases also appear to play a role in regulating the differentiation of DP thymocytes into mature

66

AMNON ALTMAN A N D MARCEL DECKERT

CD4' versus CD8' cells. Thus, a dominant gain-of-function ERK2 mutant expressed in T cells of transgenic mice reduced thymic cellularity and enhanced differentiation of DP thymocytes into the CD4 lineage (Sharp et aZ., 1997). Conversely, treatment of fetal thymic organ cultures from normal mice with PD90859, a MEK1-specific pharmacological inhibitor (which, therefore, blocks the activation of ERKs) favored in vitro differentiation into mature CD8' thymocytes (Sharp et al., 1997). The contribution of upstream signals to the activation of the Ras signaling pathway in thymocytes was addressed in a study that analyzed biochemical signaling events in thymoina cell lines derived from transgenic mice expressing CA Lck in the thymus (Lin and Abraham, 1997). Earlier studies documented the important role of Lck in T cell development and activation, and demonstrated that mice expressing this Lck transgene develop thymic Iymphomas (Perlmutter, 1995). The cell lines derived from these tumors displayed constitutive activation of the Ras/Raf/MAPK pathway as well as an increase in the tyrosine phosphorylation ofVav and Shc and a constitutive phospho-Shc/Grb2 complex (Lin and Abraham, 1997). It is conceivable that the CA Lck expressed in these cell lines is responsible for the malignant transformation of the cells via activation of the Ras and other signaling pathways. Using a novel transcriptional element consisting of the E p enhancer and the Zck proximal promoter, Iritani et al. examined the role of Ras in B lymphocyte development by generating transgenic mice that selectively expressed a DN Ras mutant in B lymphocyte progenitors. In these mice, the development of B cells was arrested at a very early stage preceding formation of the pre-B cell receptor (Iritani et al., 1997). Raf-1 was involved in this process downstream of Ras, because an activated form of Raf-1 could rescue development of progenitors expressing DN Ras (Iritani et al., 1997). Thus, to the limited extent that it has been analyzed, the Ras signaling pathway also appears to be essential for B lymphocyte development.

B. RHO,RAC,AND FOCAL ADHESIONKINASE Less is known about the role of Rho-family GTPases in lymphocyte development. T cell development requires contact with selecting epithelial and/or dendritic cells that deliver costimulatory signals, including those mediated by integrins. Therefore, it is conceivable that Rho-family GTPases regulate the actin cystoskeleton reorganization that is likely to occur on such contact, and are necessary during integrin receptor signaling events that are known to be coupled to these GTPases (Clark and Brugge, 1995; Schwartz et aZ., 1995; Parsons, 1996).

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

67

Two reports by Cantrell and co-workers (Galandrini et al., 1997a; Henning et al., 1997) have demonstrated that the GTPases Rho has a critical regulatory role in early thymocyte development. Targeting a transgene encoding C3 exotoxin, which is a selective Rho inhibitor, to the thymus caused a severe defect in fetal and adult thymopoiesis by comparison with normal littermates. Mice lacking thymic function of Rho were deficient in CD44+CD25' pro-T cells and CD44-CD25' early pre-T cells because Rho function was required for survival but not G1/Sphase cell cycle progression in these populations. The selective apoptosis defect in Rho prothymocytes could be rescued by expression of a bcl-2 transgene. A second function for Rho was observed in CD44-CD25- late pre-T cells in which Rho regulated cell cycle progression but not survival. As a result, the C3transgenic mice displayed a striking decrease in thymic cellularity paralleled by a decrease in the number of peripheral T cells. Interestingly, the differentiation of immature thymocytes into mature T cells and both positive and negative T cell selection appeared to be intact. Thus, Rho appears to mediate two different functions in prothymocyte and late pre-T thymic development, and to be required for the proliferative expansion of thymocytes (Galandrini et al., 1997a; Henning et al., 1997).The thymic phenotype in C3-transgenic mice is similar to the one caused by a loss in components of the IL-7 receptor complex. IL-7 regulates thymocyte survival by controlling Bcl-2 expression level, and bcl-2 transgene expression can rescue the Rhodeficient phenotype; thus, a potential IL-7-WRho/Bcl-2 connection is implicated. The FAK tyrosine kinase is an important component of Rho-regulated focal adhesions (Zachary and Rozengurt, 1992; Parsons, 1996). To examine the role of FAK in thymocyte development, a heterozygousfak mutation was introduced into homozygous Fyn-deficient mice. Mice carrying the double mutation, but neither Fyn deficiency nor FAK heterozygosity alone, displayed impaired development of DP (CD4'CD8+) thymocytes with atrophy of the thymic cortex, suggesting a unique cooperation between Fyn and FAK in DP thymocyte development. Another study examined the role of Rho-family member expressed selectively in hematopoietic cells, i.e., Rac2 (Chavrier et al., 1993), in T cell development. The analysis was conducted in transgenic mice that expressed wild-type or two CA Rac2 mutants (V12 or L61) in the thymus (Lores et al., 1997). Although wild-type Rac2 expression lacked any noticeable effect, mice expressing even low levels of the active Rac2 mutants displayed severe thymic atrophy that correlated with a decrease in the numbers of DP or single-positive thymocytes. This deficiency was due to an increase in thymocyte apoptosis (Lores et al., 1997).

68

AMNON ALTMAN AND MARCEL DECKERT

C. RHO-FAMILY EFFECTOR PATHWAYS: SEKl AND ~ 3 8

SEKl (JNKWMKK4) is a MAPK kinase that is activated by MEKKl and, in turn, activates JNK in response to environmental stress or mitogenic signals (Sanchez et al., 1994; Derijard et al., 1995; A. Lin et al., 1995). The role of SEKl in T cell development and apoptosis was investigated by homologous recombination-mediated deletion of the corresponding gene in embryonic stem (ES) cells (Nishina et al., 1997). The resulting sekl-I- X RAG-2-I- chimeric mice had normal numbers of mature T cells but fewer immature CD4+CD8+DP thymocytes. The sekl mutation did not affect the induction of apoptosis in response to environmental stress in ES and T cells. Instead, SEKl was found to protect thymocytes from apoptosis because SEK1-deficient thymocytes were considerably more susceptible to apoptosis induced by CD95 (Fas) or CD3 ligation (Nishina et aZ., 1997). These data indicate that SEKl mediates survival signals in T-cell development. However, a more recent study reached a different conclusion.Thus, using the same RAG-2-deficientblastocyst complementation assay, Swat et a2. found that SEK1-deficient ES cells supported unimpaired T and B lymphocyte development in the chimeric mice and, furthermore, that JNK activation was intact in lymphocytes from these mice (Swat et al., 1998). However, aging SEK1-deficient mice frequently developed lymphadenopathy and polyclonal expansion of T and B cells, leading to the conclusion that SEKl is not required for lymphocyte development but, rather, for maintaining peripheral lymphoid homeostasis. The reasons for the different findings between the two groups are unclear. Indirect evidence for the role of another branch of the MAPK superfamily, i.e., p38 stress-activated protein kinases, in T cell development and/or survival comes from the analysis of p38 kinase activity in murine thymocytes. Thus, p38 was found to be highly activated in response to intrathymic signals in vivo (Sen et al., 1996). OF SMALL GTPASES D. REGULATORS A number of studies reviewed earlier examined the role of Vav in antigen receptor signaling of mature lymphocytes (Fischer et al., 1995,1998;Tarakhovsky et aZ., 1995; R. Zhang et al., 1995; Holsinger et al., 1998). The importance of Vav in lymphocyte development was also assessed in some of these studies. Although an earlier study indicated that Vav deficiency is lethal at an early stage of embryogenesis (Zmuidzinas et al., 1995), this was most likely due to an experimental artifact because it was later established that vav-'- mice are viable, healthy, and fertile (Turner et al., 1997; Fischer et aZ., 1998). Using the RAG-2-deficient blastocyst complementation assay, three groups have demonstrated the critical role of Vav

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

69

in T and B cell development (Fischer et al., 1995; Tarakhovsky et al., 1995; R. Zhang et al., 1995).The vav-’- X RAG-2-‘- chimeras displayed thymic atrophy and peripheral T lymphopenia. In addition, only the positive selection of the immature CD4+CD8+Tcell precursors seemed to be affected in these mice. These results are consistent with a more recent study of Vav-deficient mice generated by germ-line transmission. These mice displayed a profound defect in the positive selection of both MHC class I- and class II-restricted T cells; however, Vav was not essential for negative selection, although this process was less effective in its absence (Turner et al., 1997). B cell cellularity in the vav-’- mice was normal (R. Zhang et al., 1995) or reduced (Tarakhovsky et al., 1995),but a subset of peritoneal B-1 (CD5+)B cells was missing (Tarakhovsky et al., 1995; R. Zhang et al., 1995). The simplest interpretation of these results is that the deficient development of T cells (and, to a lesser extent, B cells) reflects the inability of Rho, Rac, and/or Cdc42 [which represent direct targets for the exchange activity of Vav (Crespo et al., 1996, 1997; Olson et al., 1996; Han et al., 1997)] to become activated by antigen receptor-mediated signals in the absence of Vav. However, the evidence reviewed earlier raises uncertainties regarding the identity of GTPases that are activated by Vav in intact cells (particularly in cells that normally express Vav, i.e., hematopoietic cells) and, moreover, suggests that Vav-dependent signahng pathways in lymphocytes are distinct from those identified in ectopic expression systems.Therefore, it cannot be excluded that the developmental abnormalities of vau-’ - lymphocytes reflect interaction with targets other than, or in addition to, Rho-family GTPases. In that regard, it would be interesting to determine whether constutively active Rho-family transgenes can reconstitute normal lymphocyte development in the Vav-deficient mice. Finally, targeted inactivation in mice of Ly-GDI, a hematopoietic cellspecific Rho GDI, did not cause major abnormalities of lymphocyte development or activation. However, mice lacking Ly-GDI exhibited deregulated T and B cell interactions, leading to overexpansion of B lymphocytes in vitro, and increased resistance to apoptosis induced by IL-2 withdrawal (Yin et al., 1997).Others studied the role of Ly-GDI in hematopoiesis and phagocytosis by generating ES cells with a targeted disruption of both LyGDI alleles (Guillemot et al., 1996). The mutant ES cells developed in a normal manner into colonies that contained heterogeneous populations of progenitor cells and differentiated erythromyeloid cells. Although phagocytosis by the differentiated macrophages was normal, they demonstrated a slight but consistent reduction in superoxide production (Guillemot et al., 1996).The biological significance of the relatively modest effects observed in both studies is unclear, and it is possible that functions normally regulated

70

AMNON ALTMAN AND MARCEL DECKERT

by Ly-GDI are partially complemented by other Rho regulatory proteins present in the mutant cells. ACKNOWLEDGMENTS This is Publication No. 247 from the La Jolla Institute for Allergy and Immunology. The studies in the authors' laboratory were supported by National Institutes of Health Grants CA35299 and GM50819 (to AA), and an INSERM Fellowship and a Leukemia Society of America, Inc. Special Fellowship (to MD).

REFERENCES Abo, A., Pick, E., Hall, A,, Totty, N., Teahan, C., and Segd, A. (1991). Activation of the NADPH oxidase involves the small GTP-binding protein p21"'. Nature 353, 668-670. Aguila, H. L., Akashi, K., Domen, J,, Gandy, K. L., Lagasse, E., Mebius, R. E., Momson, S. I., Shizuru, J,, Strober, S., Uchida, N., Wright, D. E., and Weissman, I. L. (1997). From stem cells to lymphocytes: Biology and transplantation. Immunol. Rev. 157,13-40. Alberola-Ila, J., Forbush, K. A., Seger, R., Krebs, E. G., and Perlmutter, R. M. (1995). Selective requirement for MAP kinase activation in thymocyte differentiation. Nature 373, 620-623. Alberola-Ila, J,, Hogquist, K. A., Swan, K. A., Bevan, M. J., and Perlmutter, R. M. (1996). Positive and negative selection invoke distinct signaling pathways.J. Exp. Med. 184,9-18. Alberola-Ila,J., Takaki, S., Kerner, J. D., and Perlmutter, R. M. (1997).Differential signahng by lymphocyte antigen receptors. Annu. Rev. Immunol. 15, 125-154. Allen, W. E., Jones, G. E., Pollard, J. W., and Ridley, A. J. (1997). Rho, Rac and Cdc42 regulate actin organization and ceIl adhesion in macrophages.J. CeZZ Sci. 110, 707-720. Alonso, A,, Bayon, Y., and Crespo, M. S. (1996). The expression of cytokine-induced neutrophil chemoattractants (CINC-1 and CINC-2) in rat peritoneal macrophages is triggered by Fc gamma receptor activation: Study of the signaling mechanism. Eur. J. Immunol. 26, 2165-2171. Altman, A., Coggeshall, K. M., and Mustelin, T. (1990). Molecular events mediating T cell activation. Adv. Immunol. 48, 227-360. Amano, M., Mukai. H., Ono, Y., Chihara, K., Matsui, T., Ilamajima, Y., Okawa, K., Iwamatsu, A., and Kaibuchi, K. (1996). Identification of a putative target for Rho as the serinethreonine kinase protein kinase N. Science 271, 648-650. Arneisen, J. C., Estaquier, J,, Idziorek, T., and De Bels, F. (1995). Programmed cell death and AIDS pathogenesis: Significance and potential mechanisms. C u m Top.Microbid. Immunol. 200, 195-211. Amrein, K. E., Flint, N., Panholzer, B., and Bum, P. (1992). Ras GTPase-activatingprotein: A substrate and a potential binding protein of the protein-tyrosine kinase ~ 5 6 ~Proc. '~. Natl. Acad. Sci. U.S.A. 89, 3343-3346. Anderson, G., Moore, N. C., Owen, J. J., and Jenkinson, E. J. (1996). Cellular interactions in thyrnocyte development. Annu. Rev. Immunol. 14, 73-99. Andre, P., Boretto, J., Hueber, A. O., Regnier Vigourow, A., Gorvel, J. P., Ferrier, P., and Chavrier, P. (1997). A dominant-negative mutant of the Rab5 GTPase enhances T cell signaling by interfering with TCR down-modulation in transgenic mice. J. ImmunoZ. 159, 5253-5263. Apgar, J. R. (1991). Regulation of the antigen-induced F-actin response in rat basophilic leukemia cells by protein kinase C. J. CeZZ. Biol. 112, 1157-1163.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

71

Aramburu, J., Azzoni, L., Rao, A,, and Perussia, B. (1995).Activation and expression of the nuclear factors of activated T cells, NFATp and NFATc, in human natural killer cells: Regulation upon CD16 ligand binding. J. Exp. Med. 182,801-810. Aronheim, A,, Engelberg, D., Li, N., al-Alawi, N., Schlessinger, J., and Karin, M. (1994). Membrane targeting of the nucleotide exchange factor Sos is sufficient for activating the Ras signaling pathway. Cell 78,949-961. Aspenstrom, P. (1997). A Cdc42 target protein with homology to the non-kinase domain of FER has a potential role in regulating the actin cytoskeleton. Cum. Biol. 7,479-487. August, A., Gibson, S., Kawakami, Y.,Kawakami, T., Mills, G. B., and Dupont, B. (1994). CD28 is associatedwith and induces the immediatetyrosine phosphorylationand activation of the Tec family kinase ITK/EMT in the human Jurkat leukemic T-cell line. Proc. Nutl. Acud. Sci. U.S.A. 91, 9347-9351. Austrup, F., Vestweber, D., Borges, E., Lijhning, M., Brauer, R., Hen, U., Renz, H., Hallmann, R., Scheffold, A., Radbruch, A., and Hamman, A. (1997). P- and E-selectin mediate recruitment of T-helper-1 but not T-helper-2 cells into inflamed tissues. Nature 385, 81-83. Bacon, K. B., Schall, T. J., and Dairaghi, D. J. (1998).RANTES activation of phospholipase D in Jurkat T cells: Requirement of GTP-binding proteins ARF and RhoA. J. Immunol. 160, 1894-1900. Bagrodia, S., Derijard, B., Davis, R. J., and Cerione, R. A. (1995).Cdc42 and PAK-mediated signaling leads to Jun kinase and p38 mitogen-activatedprotein kinase activation.J. Biol. Chem. 270,27995-27998. Baier, G., Telford, D., Giampa, L., Coggeshall, K. M., Baier-Bitterlich, G., Isakov, N., and Altman, A. (1993). Molecular cloning and characterization of PKCB, a human novel member of the protein b a s e C (PKC)gene family expressedpredominantlyin hematopoietic cells. J. B i d . Chem. 268,4997-5004. Baier Bitterlich, G., Uberall, F., Bauer, B., Fresser, F., Wachter, H., Grunicke, H., Utermann, G., Altman, A., and Baier, G. (1996). Protein kinase C-theta isoenzyme selective stimulation of the transcription factor complex AP-1 in T lymphocytes. Mol. Cell. Biol. 16, 1842-1850. Baldari, C. T., Macchia, G., Massone, A., and Telford, J. L. (1992a).p21" contributes to HIV-1 activation in T-cells. FEBS Lett. 304,261-264. Baldari, C. T., Macchia, G., and Telford, J. L. (1992b). Interleukin-2 promoter activation in T-cells expressing activated Ha-ras. J. Biol. Chem. 267, 4289-4291. Baldari, C. T., Heguy, A., and Telford, J. L. (1993). Ras protein activity is essential for Tcell antigen receptor signal transduction. I. Biol. Chem. 268, 2693-2698. Baldari, C. T., Heguy, A., Di Somma, M. M., and Telford, J. L. (1994). Inhibition of T-cell antigen receptor signaling by overexpression of pl20GAP. Cell Growth Difer. 5, 95-98. Baldari, C. T., Milia, E., Di, S. M., Baldoni, F., Valitutti, S., and Telford, J. L. (1995a). Distinct signaling properties identifyfunctionally different CD4 epitopes.Eur. J. Immunol. 25, 1843-1850. Baldari, C. T., Pelicci, G., Di Somma, M. M., Milia, E., Giuli, S., Pelicci, P. G., and Telford, J. L. (1995b). Inhibition of CD4/~56'~ signaling by a dominant negative mutant of the Shc adaptor protein. Oncogene 10, 1141-1147. Banda, N. K., Bernier, J,, Kurahara, D. K., Kurrle, R., Haigwood, N., Sekaly, R. P., and Finkel, T. H. (1992).Crosslinking CD4 by human immunodeficiencyvirus gp120 primes T cells for activation-induced apoptosis. J. Exp. Med. 176, 1099-1106. Banin, S., Tmong, O., Katz, D. R., Waterfield, M. D., Brickell, P. M., and Gout, I. (1996). Wiskott-Aldrich syndrome protein (WASP)is a binding partner for c-Src family proteintyrosine kinases. Cum. B i d . 6, 981-988.

72

AMNON ALTMAN A N D MARCEL DECKERT

Barker, S. A,, Caldwell, K. K., H d , A,, Martinez, A. M., Pfeiffer, J. R., Oliver, J. M., and Wilson, B. S. (1995). Wortmannin blocks lipid and protein kinase activities associated with the PI 3-kinase and inhibits a subset of responses induced by FceRl cross-linking. Mol. Biol. Cell 6, 1145-1158. Baumruker, T., Pendl, G. G., and Prieschl, E. E. (1997). Gene regulation after FcsRI stimulation in the murine mast cell line CPII. lnt. Arch. Allergy Immunol. 113, 39-41. Benkirane, M., Jeang, K. T., and Devaux, C. (1994).The cytoplasmic domain of CD4 plays a critical role during the early stages of HIV infection in T-cells. EMBOJ. 13,5559-5569. Berg, N. N., and Ostergaard, H. L. (1997). T cell receptor engagement induces tyrosine phosphorylation of FAK and Pyk2 and their association with Lck.]. lrnrnunol. 159,17531757. Beyers, A. D., Spruyt, L. L., and Williams, A. F. (1992). Molecular associations between the T-lymphocyte antigen receptor complex and the surface antigens CD2, CD4 or CD8 and CD5. Proc. Natl. Acad. Sci. U.S.A. 89, 2945-2949. Bhandoola, A., Cho, E. A., Yui, K., Saragovi, H. U., Greene, M. I., and Quill, H. (1993). Reduced CD3-mediated protein tyrosine phosphorylation in anergic CD4' and CD8+ T cells. J. Immunol. 151, 2355-2367. Blackman, M. A,, Finkel, T. H., Kappler, J., Cambier, J.. and Marrack, P. (1991). Altered antigen receptor signaling in anergic T cells from self-tolerant T-cell receptor &chain transgenic mice. Proc. Natl. Acad. Sci. U.S.A. 88, 6682-6686. Bluestone, J. A. (1997). Is CTLA-4 a master switch for peripheral T cell tolerance? J. lmmunol. 158, 1989-1993. Boarder, M. R. (1994). A role for phospholipase D in control of mitogenesis. Trends Phannacol. Sci. 15, 57-62. Boguski, M. S., and McCormick, F. (1993). Proteins regulating Ras and its relatives. Nature 399, 643-654. Bokoch, G. M. (1994). Regulation of the human neutrophil NADPH oxidase by the Rac GTP-binding proteins. Cum. Opin. Cell Biol. 6, 212-218. Bokoch, G. M. (1995). Chemoattractant signaling and leukocyte activation. J. Am. SOC. Hematol. 86, 1649-1660. Bokoch, G. M., and Knaus, U. G. (1994).The role of small GTP-bindingproteins in leukocyte function. Cum. Opin. lmrnunol. 6,98-105. Bolland, S., Pearse, R. N., Kurosaki, T., and Ravetch, J. V. (1998).SHIP modulates immune receptor responses by regulating membrane association of Btk. Immunity 8,509-516. Boman, A. L., and Kahn, R. A. (1995).Arfproteins: The membrane traffic police? Trends Biochem. Sci. 20, 147-150. Boussiotis, V. A,, Barber, D. L., Lee, B. J,, Gribben, J. G., Freeman, G. J., and Nadler, L. M. (1996).Differential association of protein tyrosine kinases with the T cell receptor is linked to the induction of anergy and its prevention by B7 family-mediated costimulation. J. Exp. Med. 184, 365-376. Boussiotis, V. A., Freeman, G. J., Berezovskaya,A., Barber, D. L., and Nadler, L. M. (1997). Maintenance of human T cell anergy: Blocking of IL-2 gene transcription by activated Rap 1. Science 278, 124-128. Boyer, M. J., Gutmann, D. H., Collins, F. S., and Bar Sagi, D. (1994).Crosshnkingof the surface immunoglobulin receptor in B lymphocytes induces a redistribution of neurofibromin but not p12O-GAP. Oncogene 9,349-357. Brenner, B., Gulbins, E., Schlottmann, K., Koppenhoefer, U., Busch, G. L., Walzog, B., Steinhausen, M., Coggeshall, K. M., Linderkamp, O., and Lang, F. (1996). L-Selectin activates the Ras pathway via the tyrosine lcinase ~ 5 6 ' ~Proc. . Natl. Acad. Sci. U.S.A. 93, 15376-15381.

GTPases IN IMMUNE RECOGNITION RECEFTOR SIGNALING

73

Brenner, B., Gulbins, E., Busch, G. L., Koppenhoefer, U., Lang, F., and Linderkamp, 0. (1997). L-Selectin regulates actin polymerisation via activation of the small G-protein Race. Biochem. Biophys. Res. Commun. 231, 802-807. Briand, G., Barbeau, B., and Tremblay, M. (1997). Binding of HIV-1 to its receptor induces tyrosine phosphorylation of several CD4-associated proteins, including the phosphatidylinositol 3-kinase. Virology 228, 171-179. Briant, L., Coudronniere, N., Robert Hebmann, V., Benkirane, M., and Devaux, C. (1996). Binding of HIV-1 virions or gp120-anti-gp120 immune complexes to HIV-1-infected quiescent peripheral blood mononuclear cells reveals latent infection. J. Immunol. 156, 3994-4004. Briant, L., Robert Hebmann, V., Sivan, V., Brunet, A,, Pouyssegur, J., and Devaux, C. (1998). Involvement of extracellular signal-regulated kinase module in HIV-mediated CD4 signals controlling activation of nuclear factor-kappa B and AP-1 transcription factors. J. Immunol. 160, 1875-1885. Buday, L., and Downward, J. (1993). Epidermal growth factor regulates p21" through the formation of a complex of receptor, Grb2 adapter protein, and Sos nucleotide exchange factor. CeZZ 73, 611-620. Buday, L., Egan, S. E., Rodriguez Viciana, P., Cantrell, D. A,, and Downward, J. (1994). A complex of Grb2 adaptor protein, Sos exchange factor, and a 36-kDa membranebound tyrosine phosphoprotein is implicated in ras activation in T cells. J. Biol. Chem. 269,9019-9023. Buhl, A. M., Avdi, N., Worthen, G. S., and Johnson, G. L. (1994). Mapping of the C5a receptor signal transduction network in human neutrophils. Proc. Natl. Acad. Sci. U.S.A. 91,9190-9194. Bimnell, S. C., Henry, P. A., Kolluri, R., Kirchhausen, T., Rickles, R. J., and Berg, L. J. (1996). Identification of Itk/Tsk Src homology 3 domain ligands.]. Biol. Chem. 271,25646-25656. Cambier, J. C. (1997). Inhibitory receptors abound? Proc. Natl. Acad. Sci. U.S.A.94,59935995. Campbell, J. J., Bacon, K. B., Mackay, C. R., and Butcher, E. C. (1996). The biology of chemokine and classical chemoattractant receptors: differential requirements for adhesion-triggeringvs. chemotactic responses in lymphoid cells.]. Cell Biol. 134,255-266. Cantrell, D. (1994). G proteins in lymphocyte signalling. Cum. Opin. Immunol. 6,380-384. Cantrell, D., Pastor, M. I., and Woodrow, M. (1994). The regulation and function of p21" in signal transduction by the T cell antigen receptor. Adv. Exp. Med. Biol. 365, 73-79. Caplan, S., and Baniyash, M. (1995). Multisubunit receptors in the immune system and their association with the cytoskeleton: In search of functional significance. Immunol. Rex 14, 98-118. Caplan, S., and Baniyash, M. (1996). Normal T cells express two T cell antigen receptor populations, one of which is linked to the cytoskeleton via zeta chain and displays a unique activation-dependent phosphorylation pattern. 1.Biol. Chem. 271,20705-20712. Caplan, S., Zeliger, S., Wang, L., and Baniyash, M. (1995). Cell-surface-expressed T-cell antigen-receptor zeta chain is associated with the cytoskeleton. Proc. Natl. Acad. Sci. U.S.A. 92, 4768-4772. Capon, D. J., and Ward, R. H. R. (1991).The CD4-gp120 interaction and AIDS pathogenesis. Annu. Ret;. Immunol. 9,649-678. C q i n o , N., Wisniewski, D., Strife, A,, Marshak, D., Kobayashi, R., Stillman, B., and Clarkson, B. (1997). ~ 6 2A ~constitutively ~ : tyrosine-phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells. Cell 88, 197-204. Carrel, S., Moretta, A,, Pantaleo, G., Tambussi, G., Isler, P., Perussia, B., and Cerottini, J.4.(1988). Stimulation and proliferation of CD4' peripheral blood T lymphocytes induced by anti-CD4 monoclonal antibody. Eur. J. Immunol. 18, 333-339.

74

AMNON ALTMAN AND MARCEL DECKERT

Carroll, M. P., Clark-Lewis, I., Rapp, U. R., and May, W. S. (1990). Interleukin-3 and granulocyte-macrophage colony stimulating factor mediate rapid phosphorylation and activation of cytosolic c-Raf. J. Biol. Chem. 265, 812-817. Cayota, A., Vuillier, F., Siciliano, J., and Dighiero, G. (1994). Defective protein tyrosine phosphorylation and altered levels of p59fy" and ~ 5 6 in ' ~CD4 T cells from HN-1 infected patients. Int. Immunol. 6, 611-621. Cerione, R. A., and Zheng, Y. (1996). The Dbl family of oncogenes. Cum. Opin. Cell Biol. 8,216-222. Chambers, C. A., and Allison, J. P. (1997). Co-stimulation in T cell responses. Cum. Opin. lmmunol. 9,396-404. Chan, A. C., and Shaw, A. S . (1996). Regulation of antigen receptor signal transduction by protein tyrosine kinases. Cum. Opin. Immunol. 8, 394-401. Chang, E. C., Barr, M., Wang, Y., Jung, V., Xu, H.-P., and Wigler, M. H. (1994).Cooperative interaction of S. pombe proteins required for mating and morphogenesis. Cell 79,131-141. Chant, J., and Stowers, L. (1995). GTPase cascades choreographing cellular behavior: Movement, morphogenesis and more. Cell 81, 1-4. Chardin, P., Camonis, J. H., Gale, N. W., van Aelst, L., Schlessinger, J,, Wigler, M. H., and Bar Sagi, D. (1993). Human Sosl: A guanine nucleotide exchange factor for Ras that binds to GRB2. Science 260, 1338-1343. Chavrier, P., Gorvel, J. P., and Bertoglio, J. (1993).An immunologist's look at the Rho and Rab GTP-binding proteins. Immunol. Today 14,440-444. Chen, C. Y., and Faller, D. V. (1995). Direction of pel"-generated signals towards cell growth or apoptosisis determined byproteinkinase C and Bcl-2. Oncogene 11,1487-1498. Chen, C. Y., and Faller, D. V. (1996). Phosphorylation of Bcl-2 protein and association with p2lRas in Ras-induced apoptosis. J. Biol. Chem. 271, 2376-2379. Chen, J., Lansford, R., Stewart, V., Young, F., and Alt, F. W. (1993). RAG-2-deficient blastocyst complementation: An assay of gene function in lymphocyte development. Proc. Natl. Acad. Sci. U.S.A.90,4528-4532. Chen, C. Y., Forman, L. W., and FalIer, D. V. (1996). Calcium-dependent immediateearly gene induction in lymphocytes is negatively regulated by ~21~"". Mol. Cell. Biol. 16,6582-6592. Chirmule, N., and Pahwa, S . (1996). Envelope glycoproteins of human immunodeficiency virus type 1: Profound influences on immune functions. Microbiol. Rev. 60, 386-406. Chirmule, N., Kalyanaraman, V. S., and Pahwa, S. (1994). Signals transduced through the CD4 molecule on T lymphocytes activate NF-kappa B. Biochem. Biophys. Res. Commun. 203,498-505. Chirmule, N., Goonewardena, H., Pahwa, S., Pasieka, R., Kalyanaraman, V. S., and Pahwa, S. (1995). HN-1 envelope glycoproteins induce activation of activated protein-1 in CD4+ T cells. J. Biol. Chem. 270, 19364-19369. Cho, E. A., Riley, M. P., Sillman, A. L., and Quill, H. (1993). Altered protein tyrosine phosphorylation in anergic Thl cells. 1. Immunol. 151, 20-28. Choi, M. S., Brines, R. D., Holman, M. J., and Klaus, G. G. (1994). Induction of NF-AT in normal B cells by anti-immunoglobulin or CD40 ligand in conjunction with IL-4. Immunity 1, 179-187. Chong, L. D., Traynor-Kaplan, A., Bokoch, G. M., and Schwartz, M. A. (1994). The small GTP-binding protein Rho regulates a phosphatidylinositol4-phosphate5 - b a s e in mammalian cells. Cell 79,507-513. Chuang, T.-H., Hahn, K. M., Lee, J.-D., Danley, D. E., and Bokoch, G. M. (1997). The small GTPase Cdc42 initiates an apoptotic signaling pathway in Jurkat T lymphocytes. Mol. Biol. Cell 8, 1687-1698.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

75

Cichowski, K., Brugge, J. S., and Brass, L. F. (1996). Thrombin receptor activation and integrin engagement stimulate tyrosine phosphorylation of the proto-oncogene product, p9SW, in platelets. 1.Biol. Chem. 271, 754-7550. Clark, E. A., and Brugge, J. S. (1995). Integrins and signal transduction pathways: The road taken. Science 268, 233-239. Clark, E. A., Shattil, S. J., and Brugge, J. S. (1994). Regulation of protein tyrosine kinases in platelets. Trends Biochem. Sci. 19, 464-469. Coffer, P. J., Geijsen, N., M'Rabet, L., Schweizer, R. C., Maikoe, T., Raaijmakers, J. A., Lammers, J. W., and Koenderman, L. (1998). Comparison of the roles of mitogenactivated protein ldnase and phosphatidylinositol3-kinase signal transduction in neutrophil effector function. Biochem. J 329, 121-130. Colley, W. C., Sung, T. C., Roll, R., Jenco, J., Hammond, S. M., Altshuller, Y., Bar-Sagi, D., Morris, A. J., and Frohman, M. A. (1997). Phospholipase D2, a distinct phospholipase D isoform with novel regulatory properties that provokes cytoskeletal reorganization. Curr. Biol. 7, 191-201. Collins, T., Deckert, M., and Altman, A. (1997).Views on Vav. Immunol. Today 18,221-225. Cory, G. O., MacCarthy Morrogh, L., Banin, S., Gout, I., Brickell, P. M., Levinsky, R. J., Kinnon, C., and Lovering, R. C. (1996). Evidence that the Wiskott-Aldrich syndrome protein may be involved in lymphoid cell signalingpathways.f. Immunol. 157,3791-3795. Coso, 0. A,, Chiariello, M., Yu, J. C., Teramoto, H., Crespo, P., Xu, N., Miki, T., and Gutkind, J. S. (1995). The small GTP-binding proteins Racl and Cdc42 regulate the activity of the JNWSAPK signaling pathway. Cell 81, 1137-1146. Cox, D., Chang, P., Kurosaki, T., and Greenberg, S. (1996). Syk tyrosine kinase is required for immunoreceptor tyrosine activation motif-dependent actin assembly. J. Biol. Chem. 271, 16597-16602. Cox, D., Chang, P., Zhang, Q., Reddy, P. G., Bokoch, G. M., and Greenberg, S. (1997). Requirements for both Racl and Cdc42 in membrane rufAing and phagocytosis in leukocytes. J. Exp. Med. 186, 1487-1494. Crabtree, G. R., and Clipstone, N. A. (1994). Signal transmission between the plasma membrane and nucleus of T lymphocytes. Annu. Reu. Biochem. 63, 1045-1083. Crespo, P., Bustelo, X. R., Aaronson, D. S., Coso, 0. A., Lopez, B. M., Barbacid, M., and Gutkind, J. S. (1996). Rac-1 dependent stimulation of the JNWSAPK signaling pathway by Vav. Oncogene 13,455-460. Crespo, P., Schuebel, K. E., Ostrom, A. A., Gutkind, J. S., and Bustelo, X. R. (1997). Phosphotyrosine-dependent activation of Rac-1 GDP/GTP exchange by the uan protooncogene product. Nature 385, 169-172. Croft, M., and Dubey, C. (1997). Accessory molecule and costimulation requirements for CD4 T cell response. Crit. Reu. Immunol. 17, 89-118. Crompton, T., Gilmour, K. C., and Owen, M. J. (1996). The MAP kinase pathway controls differentiation from double-negative to double-positive thymocyte. Cell 86, 243-251. Cross, M. J., Roberts, S., Ridley, A. J., Hodgkin, M. N., Stewart, A., Claesson Welsh, L., and Wakelam, M. J. 0. (1996). Stimulation of actin stress fibre formation mediated by activation of phospholipase D. C u m Biol. 6,588-597. Culine, S., Honore, N., Tavitian, A,, and Olofsson, B. (1992). Overexpression of the rasrelated rab2 gene product in peripheral blood mononuclear cells from patients with hematological and solid neoplasms. Cancer Res 52, 3083-3088. Culine, S., Honore, N., Closson, V., Lang, P., Bertoglio, J., Laroche, L., Bussel, A., Racadot, E., Hovnanian, A,, Dubertret, L., etal. (1993).A CD2' subset of non-malignant peripheral blood lymphocytes from patients with Sezary syndromes overexpress the low-molecularweight GTP-binding protein Rab2. Leukemia 7, 601-608.

76

AMNON ALTMAN AND MARCEL DECKERT

D’Ambrosio, D., Cantrell, D. A., Frati, L., Santoni, A,, and Testi, R. (1994). Involvement of p2lras activation in T cell CD69 expression. Eur. J. Immunol. 24, 616-620. Damen, J. E., Liu, L., Rosten, P., Humphries, R. K., Jefferson, A. B., Majerus, P. W., and Krystal, G. (1996). The 145-kDa protein induced to associate with Shc by multiple cytokines is an inositol tetraphosphate and phosphatidylinositol 3,4,5-triphosphate 5phosphatase. Proc. Natl. Acad. Sci. U.S.A. 93, 1689-1693. DeBell, K. E., Conti, A., dava, M. A., Hoffman, T., and Bonvini, E. (1992). Microfilament assembly modulates phospholipase C-mediated signal transduction by the TCWCD3 in murine T helper lymphocytes. J. Immunol. 149,2271-2280. 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. Deckert, M., Elly, C., Altman, A., and Liu, Y.-C. (1998). Coordinated regulation of the tyrosine phosphorylation of Cbl by Fyn and Syk tyrosine kinases.J. Biol. Chem. 273,88678874. Dedhar, S., and Hannigan, G. E. (1996). lntegrin cytoplasmic interactions and bidirectional transmembrane signaIling. CUP. @in. CeZZ Bid. 8, 657-669. DeFranco, A. L. (1997). The complexity of signaling pathways activated by the BCR. Cuw. Opdn. Immunol. 9, 296-308. De Maria, R., Todaro, M., Stassi, G., Di Blasi, F., Giordano, M., Gduzzo, A,, and Giordano, C. (1994). Defective T cell receptorKD3 complex signaling in human type I diabetes. Eur. J. Immunol. 24, 999-1002. Demotz, S., Grey, H. M., and Sette, A. (1990). The minimal number of class I1 MHCantigen complexes needed for T cell activation. Science 249, 1028-1030. Derijard, B., Raingeaud, J.. Barrett, T., Wu, I. H., Han, J., Ulevitch, R. J., and Davis, R. J. (1995). Independent human MAP-kinase signal transduction pathways defined by MEK and MKK isoforms. Science 267, 682-685. Derry, J. M., Ochs, H. D., and Francke, U. (1994). Isolation of a novel gene mutated in Wiskott-Aldrich syndrome. Cell 78, 633-644. DeSilva, D. R., Feeser, W. S., Tancula, E. J., and Scherle, P. A. (1996). Anergic T cells are defective in both jun NHt-terminal kinase and mitogen-activated protein kinase signaling pathways. J. Exp. Med. 183, 2017-2023. DeSilva, D. R., Jones, E. A., Feeser, W. S., Manos, E. J., and Scherle, P. A. (1997). The p38 mitogen-activated protein kinase pathway in activated and anergic Thl cells. Cell Immunol. 180,116-123. Deuter-Reinhard, M., Apell, G., Pot, D., Klippel, A., Williams, L. T., and Kavanaugh, W. M. (1997). SIP/SHIP inhibits Xenopus oocyte maturation induced by insulin and phosphatidylinositol 3-kinase. Mol. Cell. Biol. 17, 2559-2565. de Vries Smits, A. M., Pronk, G. J., Medema, J. P., Burgering, B. M., and Bos, J. L. (1995). Shc associates with an unphosphorylated form of the p21m guanine nucleotide exchange factor mSOS. Oncogene 10, 919-925. Dharmawardhane, S., and Bokoch, G. M. (1997). Rho GTPases and leukocyte cytoskeIetaJ regulation. Curr. Opin. Hematol. 4, 12-18. Diamond, D. C., Sleckman, B. P., Gregory, T., Lasky, L. A,, Greenstein, J. L., and Burakoff, S.J. (1988). Inhibition of CD4+ T cell function by the HIV envelope glycoprotein, gp120. J. Immunol. 141,3715-3717. Diehann, D., Abo, A., Johnston, C., Segal, A. W., and Hall, A. (1994). Interaction of Rac with p67p’” and regulation of phagocytic NADPH oxidase activity. Science 265,531-533. Dolmetsch, R. E., Lewis, R. S., Goodnow, C. C., and Healy, J. I. (1997). Differential activation of transcription factors induced by Ca” response amplitude and duration. Nature 386, 855-858.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

77

Donnadieu, E., Cefai, D., Tan, Y. P., Paresys, G., Bismuth, G., and Trautmann, A. (1992). Imaging early steps of human T cell activation by antigen-presenting cells. J . Immunol. 148,2643-2653. Donnadieu, E., Bismuth, G., and Trautmann, A. (1994). Antigen recognition by helper T cells elicits a sequence of distinct changes of their shape and intracellular calcium. Cum. Biol. 4, 584-595. Donovan, J. A,, Wange, R. L., Langdon, W. Y., and Samelson, L. E. (1994). The protein product of the c-cbl protooncogene is the 120-kDa tyrosine-phosphorylated protein in Jurkat cells activated via the T cell antigen receptor. J. B i d . Chem. 269, 22921-22924. Dorseuil, O., Reibel, L., Bokoch, G. M., Camonis, J., and Gacon, G. (1996).The Rac target NADPH oxidase p67ph" interacts preferentially with Race rather than Rac1.J. B i d . Chem. 271,83-88. Downey, G. P., Fukushima, T., and Fialkow, L. (1995). Signaling mechanisms in human neutrophils. Cum. Opin. Hematol. 2, 76-88. Downward, J. (1997). Cell cycle: Routine role for Ras. Cum. Biol. 7 , R258-R260. Downward, J., Graves, J. D., Warne, P. H., Rayter, S., and Cantrell, D. A. (1990).Stimulation of ~ 2 1 "upon ~ T-cell activation. Nature 346, 719-723. Downward, J., Graves, J., and Cantrell, D. (1992). The regulation and function of p21" in T cells. Immunol. Today 13, 89-92. Dumont, F. J., Staruch, M. J., Fischer, P., DaSilva, C., and Camacho, R. (1998). Inhibition of T cell activation by pharmacologic disruption of the MEKl/ERK MAP kinase or calcineurin signaling pathways results in differential modulation of cytokine production. 1.Immunol. 160,2579-2589. Dunon, D., Piali, L., and Imhof, B. A. (1996). To stick or not to stick: the new leukocyte homing paradigm. Cum. Opin. Cell B i d . 8, 714-723. Durden, D. L., Kim, H. M., Calore, B., and Liu, Y. (1995). The Fc gamma RI receptor signals through the activation of hck and MAP kinase. J. Immunol. 154,4039-4047. Egan, S . E., and Weinberg, R. A. (1993). The pathway to signal achievement. Nature 365, 781-783. Egan, S. E., Giddings, B. W., Brooks, M. W., Buday, L., Sizeland, A. M., and Weinberg, R. A. (1993). Association of Sos Ras exchange protein with Grb2 is implicated in tyrosine kinase signal transduction and transformation. Nature 363, 45-51. English, D. (1996). Phosphatidic acid: A lipid messenger involved in intracellular and extracellular signalling. Cell Signal 8, 341-347. Ettehadieh, E., Sanghera, J. S., Pelech, S. L., Hess Bienz, D., Watts, J., Shastri, N., and Aebersold, R. (1992). Tyrosyl pbosphorylation and activation of MAP kinases by ~ 5 6 ' ~ . Science 255, 853-855. Fang, N., Motto, D. G., Ross, S. E., and Koretzky, G. A. (1996). Tyrosines 113, 128, and 145 of SLP-76 are required for optimal augmentation of NFAT promoter activity. 1.Immunol. 157,3769-3773. Fans, M., Kokot, N., Lee, L., and Nel, A. E. (1996). Regulation ofinterleukin-2 transcription by inducible stable expression of dominant negative and dominant active mitogen-activated protein kinase kinase kinase in jurkat T cells. Evidence for the importance of Ras in a pathway that is controlled by dual receptor stimulation.J. Biol. Chem. 271,27366-27373. Fans, M., Kokot, N., Latinis, K., Kasibhatla, S., Green, D. R., Koretzky, G. A., and Nel, A. (1997). The c-Jun N-terminal kinase cascade plays a role in stress-induced apoptosis in Jurkat cells by up-regulating Fas ligand expression. J. Immunol. 160, 134-144. Feinberg, M. B., and Greene, W. C. (1992). Molecular insights into human immunodeficiency virus type 1 pathogenesis. CUT. Opin. Immunol. 4, 466-474.

78

AMNON ALTMAN AND MARCEL DECKERT

Fialkow, L., Chan, C. K., Rotin, D., Grinstein, S., and Downey, G. P. (1994). Activation of the mitogen-activatedprotein kinase signaling pathway in neutrophils. Role of oxidants. J. Biol. Chem. 269, 31234-31242. Fields, P., Fitch, F. W., and Gajewski, T. F. (1996a). Control of T lymphocyte signal transduction through clonal anergy. J. Mol. Med. 74, 673-683. Fields, P. E., Gajewski, T. F., and Fitch, F. W. (1996b).Blocked Ras activation in anergic CD4+ T cells. Science 271, 1276-1278. Finney, R. E., Robbins, S. M., and Bishop, J. M. (1993).Association of pRas and pRaf-1 in a complex correlates with activation of a signal transduetion pathway. Cum Bid. 3,805-812. Fischer, K. D., Zmuldzinas, A., Gardner, S., Barbacid, M., Bernstein, A., and Guidos, C. (1995).Defective T-cell receptor signalling and positive selection of Vav-deficient CD4' CD8+ thymocytes. Nature 374, 474-477. Fischer, K.-D., Kong, Y.-Y., Nishina, H., Tedford, K., Marengere, L. E. M., Kozieradzki, I., Sasaki, T., Starr, M., Chan, G., Gardener, S., Nghiem, M. P., Bouchard, D., Barbacid, M., Bernstein, A,, and Penninger, J. M. (1998).Vavis a regulator of cytoskeletal reorganization mediated by the T-cell receptor. Cum Biol. 8, 554-562. Foletta, V. C., Segal, D. H., and Cohen, D. R. (1998). Transcriptional regulation in the immune system: All roads lead to AP-1. J. Leuk. Biol. 63, 139-152. Folgueira, L., Algeciras, A., MacMorran, W. S., Bren, G. D., and Paya, C. V. (1996).The Ras-Raf pathway is activated in human immunodeficiency virus-infected monocytes and participates in the activation of NF-KB.J. Virol. 70, 2332-2338. Franklin, R. A., Tordai, A., Patel, H., Gardner, A. M., Johnson, G. L., and Gelfand, E. W. (1994). Ligation of the T cell receptor complex results in activation of the Rasmaf-l/ MEWMAPK cascade in human T lymphocytes. J. Clin. Inuest. 93, 2134-2140. Frohman, M. A., and Morris, A. J. (1996). Phospholipid signalling: Rho is only ARF the story. CUP. Biol. 6, 945-947. Frost, J. A., Xu, S., Hutchison, M. R., Marcus, S., and Cobb, M. H. (1996).Actions of Rho family small G proteins and pel-activated protein kinases on mitogen-activatedprotein kinase family members. Mol. Cell. Biol. 16, 3707-3713. Fukamachi, H., Takei, M., and Kawakami, T. (1993). Activation of multiple protein kinases including a MAP kinase upon FcERI cross-linking. Int. Arch. Alhriy immunol. 102, 15-25. Fukazawa, T., Reedquist, K. A., Trub, T., Soltoff, S., Panchamoorthy, G., Druker, B., Cantley, L., Shoelson, S. E., and Band, H. (1995).The SH3 domain-bindingT cell tyrosyl phosphoprotein p120. Demonstration of its identity with the c-cbl protooncogene product and in uiuo complexes with Fyn, Grb2, and phosphatidylinositol 3-kinase.J. Biol. Chem. 270,19141-19150. Gajewski, T. F., Fields, P., and Fitch, F. W. (1995).Induction of the increased Fyn kinase activity in anergic T helper type 1 clones requires calcium and protein synthesis and is sensitive to cyclosporin A. Eur. J. Immunol. 25, 1836-1842. Galandrini, R., Palmieri, G., Piccoli, M., Frati, L., and Santoni, A. (1996).CD16-mediated p21" activation is associated with Shc and p36 phosphorylation and their binding with Grb2 in human natural killer cells. J. Exp. Med. 183, 179-186. Galandrini, R., Henning, S. W., and Cantrell, D. A. (1997a). Different functions of the GTPase Rho in prothymocytes and late pre-T cells. Immunity 7 , 163-174. Galandrini, R., Palmieri, G., Paolini, R., Piccoli, M., Frati, L., and Santoni, A. (1997b). Selective binding of Shc-SH2 domain to tyrosine-phosphorylated5 but not y-chain upon CD16 ligation on human NK cells. J. Immunol. 159,3767-3773.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

79

Gallego, M. D., Santamaria, M., Pena, J., and Molina, I. J. (1997). Defective actin reorganization and polymerization of Wiskott-Aldrich T cells in response to CD3-mediated stimulation. Blood 90, 3089-3097. Gamty, P. A., Chen, D., and Rothenberg, E. V. (1994). Interleukin-2 transcription in vivo is regulated at the level of coordinate binding of both constitutive and regulated factors. Mol. Cell. Biol. 14, 2159-2169. Genot, E. M., Parker, P. J.. and Cantrell, D. A. (1995).Analysis of the role ofprotein kinase C-a, - E , and -5 in T cell activation. 1. Biol. Chem. 270, 9833-9839. Genot, E., Cleverley, S., Henning, S., and Cantrell, D. (1996). Multiple p21m effector pathways regulate nuclear factor of activated T cells. EMBOJ. 15, 3923-3933. Ghaffari-Tabrid, N., Bauer, B., Baier-Bitterlich, G., Altman, A,, Utermann, G., Uberall, F., and Baier, G . (1998). PKCB, a selective upstream regulator of JNWSAPK in Jurkat T-cells. Eur. 1.lmmunol. 28, 1-11. Gold, M. R., Crowley, M. T., Martin, G. A,, McCormick, F., and DeFranco, A. L. (1993). Targets of B lymphocyte antigen receptor signal transduction include the p21" GTPaseactivating protein (GAP) and two GAP-associated proteins. J. lmmunol. 150,377-386. Goldman, F., Crabtree, J., Hollenback, C., and Koretzky, G. (1997). Sequestration of ~56'"" by gp120, a model for TCR desensitization. J. Immunol. 158, 2017-2024. Gomez, J.. Martinez, A. C., Fernandez, B., Garcia, A,, and Rebollo, A. (1996). Critical role of Ras in the proliferation and prevention of apoptosis mediated by IL-2.1. Immunol. 157,2272-2281. Gomez, J., Martinez, C., Fernandez, B., Garcia, A,, and Rebollo, A. (1997). Ras activation leads to cell proliferation or apoptotic cell death upon interleukin-2 stimulation or lymphokine deprivation, respectively. Eur. 1.Immunol. 27, 1610-1618. Goodnow, C. C. (1997). Balancing immunity, autoimmunity, and self-tolerance. Ann. N.Y. Acud. Sci. 815, 55-66. Gotoh, T., Hattori, S., Nakamura, S., Kitayama, H., Noda, M., Takai, Y.,Kaibuchi, K., Matsui, H., Hatase, O., Takahashi, H., et al. (1995). Identification of Rap1 as a target for the Crk SH3 domain-binding guanine nucleotide-releasing factor C3G. Mol. Cell. Biol. 15, 6746-5753. Gotoh, A., Takahira, H., Gealden, R. L., and Broxmeyer, H. E. (1997). Cross-linking of integrins induces tyrosine phosphorylation of the proto-oncogene product Vav and the protein tyrosine kinase Syk in human factor-dependent myeloid cells. Cell Growth DijJ 8, 721-729. Graves, J. D., Downward, J., Rayter, S., Warne, P., Tutt, A. L., Glennie, M., and Cantrek D. A. (1991).CD2 antigen mediated activation of the guanine nucleotide binding proteins p21" in human T lymphocytes.1. lmmunol. 146,3709-3712. Graves, J. D., Downward, J., Izquierdo Pastor, M., Rayter, S., Warne, P. H., and Cantrell, D. A. (1992). The growth factor IL-2 activates p21" proteins in normd human T lymphocytes. 1.Immunol. 148, 2417-2422. Graves, J. D., Draves, K. E., Craxton, A,, Saklatvala, J., Krebs, E. G., and Clark, E. A. (1996). Involvement of stress-activated protein kinase and p38 mitogen-activated protein kinase in mIgM-induced apoptosis of human B lymphocytes. Proc. Nutl. Acud. Sci. U.S.A. 93, 13814-13818. Graziadei, L., Riabowol, K., and Bar-Sagi, D. (1990).Co-capping of ras proteins with surface immunoglobulins in B lymphocytes. Nature 347,396-400. Graziani, A,, Galimi, F., Medico, E., Cottone, E., Gramaglia, D., Boccaccio, C., and Comoglio, P. M. (1996). The HIV-1 nef protein interferes with phosphatidylinositol 3-kinase activation.J. Biol. Chem. 271, 6590-6593.

80

AMNON ALTMAN A N D MARCEL DECKERT

Gringhuis, S. I., de Leij, L. F., Coffer, P. J., and Vellenga, E. (1998). Signaling through CD5 activates a pathway involving phosphatidylinositol3-kinase,Vav, and Racl in human mature T lymphocytes. Mol. Cell. Biol. 18, 1725-1735. Grinstein, S., Butler, J. R., Furuya, W., LAllemain, G., and Downey, G. P. (1994).Chemotactic peptides induce phosphorylation and activation of MEK-1 in human neutrophils. J. Biol. Chem. 269, 19313-19320. Grow, H., Torpier, G., Monte, D., Mouton, Y.,Capron, A., and Ameisen, J. C. (1992). Activation-induced death by apoptosis in CD4' T cells from human immunodeficiency virus-infected asymptomatic individuals.J. Erp. Med. 175, 331-340. Guillemot, J. C., Kruskal, B. A,, Adra, C. N., Zhu, S., KO, J. L., Burch, P., Nocka, K., Seetoo, K., Simons, E., and Lim, B. (1996).Targeted disruption of guanosine diphosphatedissociation inhibitor for Rho-related proteins, GDID4: Normal hematopoietic differentiation but subtle defect in superoxide production by macrophages derived from in vitro embryonal stem cell differentiation. Blood 88,2722-2731. Guillemot, J. C., Montcoumer, P., Vivier, E., Davoust, J., and Chavrier, P. (1997). Selective control of membrane ruffling and actin plaque assembly by the Rho GTPases Racl and CDC42 in FceRI-activated rat basophilic leukemia (RBL-2H3) cells. J. Cell Sci. 110,2215-2225. Gulbins, E., Coggeshall, K. M., Baier, G., Katzav, S., Bum, P., and Altman, A. (1993). Tyrosine kinase-stimulated guanine nucleotide exchange activity of Vav in T cell activation. Science 260, 822-825. Gulbins, E., Bissonnette, R., Mahboubi, A., Martin, S., Nishioka, W., Brunner, T., Baier, G., Baier Bitterlich, G., Byrd, C., Lang, F., et al. (1995). FAS-induced apoptosis is mediated via a ceramide-initiated RAS signaling pathway. Immunity 2, 341-351. Gupta, S., Weiss, A., G., K., S., W., and Nel, A. (1994). The T-cell antigen receptor utilizes Lck, Raf-1, and MEK-1 for activating mitogen-activated protein kinase. Evidence for the existence of a second protein kinase C-dependent pathway in an Lck-negative Jurkat cell mutant. J. Biol. Chem. 269, 17349-17357. Hackam, D. J., Rotstein, 0. D., Schreiber, A., Zhang, W., and Grinstein, S.(1997). Rho is required for the initiation of calcium signaling and phagocytosis by Fcy receptors in macrophages. J. Exp. Med. 186, 955-966. Hahn, S., Wodnar-Filipowicz, A., Nair, A. P. K., and Moroni, C. (1991). Ras oncogenes amplify lymphokine (interleukin 3, granulocyte-macrophage colony-stimulating factor) induction by calcium ionophore. Oncogene 6, 2327-2332. Haimovich, B., Regan, C., DiFazio, L., Ginalis, E., Ji, P., Purohit, U., Rowley, R. B., Bolen, phosphorylation in J., and Greco, R. (1996). The FcyRlI receptor triggers pp12SFAK platelets. J. BioZ. Chem. 271, 16332-16337. Hall, A. (1998). Rho GTPases and the actin cytoskeleton. Science 279,509-514. Haller, H., Kunzendorf, U., Sacherer, K., Lindschau, C., Walz, G., Distler, A,, and Luft, F. C. (1997). T cell adhesion to P-selectin induces tyrosine phosphorylation of pp125 focal adhesion b a s e and other substrates. J. Immunol. 158, 1061-1067. 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). Lck regulates Vav activation of members of the Rho family of GTPases. Mol. Cell. Biol. 17, 1346-1353. Han, J., Lubu-Phelps, K., Das, B., Shu, X., Xia, Y., Mosteller, R. D., Krishna, U. M., Falck, J. R., White, M. A., and Broek, D. (1998). Role of substrates and products of PI 3kinase in regulating activation of rac-related guanosine triphosphatases by Vav. Science 279, 558-560. Hartwig, J. H., Bokoch, G. M., Carpenter, C. L., Janmey, P. A,, Taylor, L. A,, Toker, A,, and Stossel, T. P. (1995). Thrombin receptor ligation and activated Rac uncap actin

GTPdses IN IMMUNE RECOGNITION RECEPTOR SIGNALING

81

filament barbed ends through phosphoinositide synthesis in permeabilized human platelets. Cell 82, 643-653. Hanvood, A. E., and Cambier, J. C. (1993). B cell antigen receptor cross-linking triggers rapid protein kinase C-independent activation of p21". J. lmmunol. 151, 4513-4522. Haston, W. S., Shields, J. M., and Wilkinson, P. C. (1982). Lymphocyte locomotion and attachment on two-dimensional surfaces and in three-dimensional matrices. J. Cell Biol. 92, 747-752. Hata, Y., Kikuchi, A., Sasaki,T., Schaber, M. D., Gibbs, J. B., and Takai, Y.(1990).Inhibition of the ras p21 GTPase-activating protein-stimulated GTPase activity of c-Ha-ras p21 by smg p21 having the same putative effector domain as ras p21s. J. Biol. Chem. 265,71047107. Haughn, L., Gratton, S., Caron, L., Sekaly, R. P., Veillette, A., and Julius, M. (1992). Association of tyrosine kinase p561ck with CD4 inhibits the induction of growth through the alpha beta T-cell receptor. Nature 358, 328-331. Healy, J. I., Dolmetsch, R. E., Timmerman, L. A,, Cyster, J. G., Thomas, M. L., Crabtree, G. R., Lewis, R. S., and Goodnow, C. C. (1997). Different nuclear signals are activated by the B cell receptor during positive versus negative signaling. Immunity 6, 419-428. Hecker, C., Weise, C., Schneider Schaulies, J., Holmes, H. C., and ter Meulen, V. (1997). Specific binding of HIV-1 envelope protein gp120 to the structural membrane proteins ezrin and moesin. Virus Res. 49, 215-223. Henning, S. W., Galandrini, R., Hall, A,, and Cantrell, D. A. (1997). The GTPase Rho has a critical regulatory role in thymus development. EMBO J . 16, 2397-2407. Henriquez, N. V., Rijkers, G. T., and Zegers, B. J. M. (1994). Antigen receptor-mediated transmembrane signaling in Wiskott-Aldrich syndrome. J. lmmunol. 153, 395-399. Hill, C. S., and Treisman, R. (1995). Transcriptional regulation by extracellular signals: Mechanisms and specificity. Cell 80, 199-211. Hill, C. S., Wynne, J., and Treisman, R. (1995). The Rho Family GTPases RhoA, Racl, and CDC42Hs regulate transcriptional activation by SRF. Cell 81, 1159-1170. Hirao, M., Sato, N., Kondo, T., Yonemura, S., Monden, M., Sasaki, T., Takai, Y., Tsukita, S., and Tsukita, S. (1996). Regulation mechanism of ERM (ezrin/radixin/moesi) protein/ plasma membrane association: possible involvement of phosphatidylinositol turnover and Rho-dependent signaling pathway. J. Cell Biol. 135, 37-51. Hirasawa, N., Santini, F., and Beaven, M. A. (1995a). Activation of the mitogen-activated protein kinase/cytosolic phospholipase Az pathway in a rat mast cell line. Indications of different pathways for release of arachidonic acid and secretory granules. J. lmmunol. 154,5391-5402. Hirasawa, N., Scharenberg, A,, Yamamura, H., Beaven, M. A,, and Kinet, J. P. (1995b). A requirement for Syk in the activation of the microtubule-associated protein kinasel phospholipase A2 pathway by FcERI is not shared by a G protein-coupled receptor. J. Biol. Chern. 270, 10960-10967. Holsinger, L. J., Spencer, D. M., Austin, D. J., Schreiber, S. L., and Crabtree, C. R. (1995). Signal transduction in T lymphocytes using a conditional allele of Sos. Proc. Natl. Acad. Sci. U.S.A. 92, 9810-9814. 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). Defects in actin-cap formation in Vavdeficient mice implicate an actin requirement for lymphocyte signal transduction. C u m Biol. 8, 563-572. Hotchin, N. A,, and Hall, A. (1996). Regulation of the actin cytoskeleton, integrins and cell growth by the Rho family of small GTPases. Cancer Sum. 27, 311-322.

82

AMNON ALTMAN AND MARCEL DECKERT

Howe, L. R., Leevers, S. J., Gomez, N., Nakiehy, S., Cohen, P., and Marshall, C. J. (1992). Activation of the MAP kinase pathway by the protein kinase Raf. Cell 71, 335-342. Hu, Q., Klippel, A,, Muslin, A. J., Fantl, W. J., and Williams, L. T. (1995). Ras-dependent induction of cellular responses by constitutively active phosphatidylinositol-3kinase. Science 268, 100-102. Huang, X., Li, Y., Tanaka, K., Moore, K. G., and Hayashi, J. I. (1995). Cloning and characterization of Lnk, a signal transduction protein that links T-cell receptor activation signal to phospholipase C gamma 1, Grb2, and phosphatidylinositol 3-kinase. Proc. Nutl. Acud. Sci. U.S.A. 92, 11618-11622. Huby, R. D., Carlile, G. W., and Ley, S. C. (1995). Interactions between the proteintyrosine kinase ZAP-70, the proto-oncoprotein Vav, and tubulin in Jurkat T cells. ]. Biol. Chem. 270,30241-30244. Hutchinson, L. E., and McCloskey, M. A. (1995). Fc epsilon RI-mediated induction of nuclear factor of activated T-cells. 1.Biol. Chem. 270, 16333-16338. Hynes, R. 0. (1992). Integrins: versatility, modulation and signaling in cell adhesion. Cell 69, 11-25. Hynes, R. 0. (1996). Targeted mutations in cell adhesion genes: what have we learned from them? Dev. Biol. 180, 402-412. Ihle, J. N., Witthuhn, B. A., Quelle, F. W., Yamamoto, K., and Silvennoinen, 0. (1995). Signaling through the hematopoietic cytokine receptors. Annu. Rev. Immunol. 13, 369-398. Ingham, R. J., Krebs, D. L., Barbazuk, S. M., Turck, C. W., Hirai, H., Matsuda, M., and Gold, M. R. (1996). B cell antigen receptor signaling induces the formation of complexes containing the Crk adapter proteins. J. Biol. Chem. 271, 32306-32314. Iritani, B. M., Forbush, K. A., Farrar, M. A., and Perlmutter, R. M. (1997). Control of B cell development by Ras-mediated activation of Raf. E M B O 1, 16, 7019-7031. Ishizaki, T., Maekawa, M., Fujisawa, K., Okawa, K., Iwamatsu, A,, Fujita, A., Watanabe, N., Saito, Y., Kakizuka, A,, Morii, N., and Narumiya, S. (1996). The small GTP-binding protein Rho binds to and activates a 160 kDa SerRlu protein kinase homologous to myotonic dystrophy kinase. E M B O f. 15, 1885-1893. Ishizuka, T., Oshiba, A., Sakata, N., Terada, N., Johnson, G. L., and Gelfand, E. W. (1996). Aggregation of the FcepsilonRI on mast cells stimulates c-Jun amino-terminal kinase activity. A response inhibited by wortmannin. 1.Biol. Chem. 271, 12762-12766. Ishizuka, T., Terada, N., Gerwins, P., Hamelmann, E., Oshiba, A., Fanger, G. R., Johnson, G. L., and Gelfand, E. W. (1997). Mast cell tumor necrosis factor alpha production is regulated by MEK kinases. Proc. Nutl. Acud. Sci. U.S.A. 94, 6358-6363. Izquierdo, M., and Cantrell, D. A. (1993). Protein tyrosine kinases couple the interleukin2 receptor to p21". Ear. f. fmmetnol. 23, 131-135. Izquierdo, M., Downward, J., Graves, J. D., and Cantrell, D. A. (1992a). Role of protein kinase C in T-cell antigen receptor regulation of p21Y Evidence that two p21m regulatory pathways coexist in T cells. Mol. Cell. Biol. 12, 3305-3312. Izquierdo, M., Downward, J., Otani, H., Leonard, W. J., and Cantrell, D. A. (1992b). Interleukin (IL)-2 activation of p21m in murine myeloid cells transfected with human IL-2 receptor p chain. Eur. 1.Immunol. 22, 817-821. Izquierdo, M., Leevers, S. J., Marshall, C. J., and Cantrell, D. (1993). p21" couples the T cell antigen receptor to extracellular signal-regulated kinase 2 in T lymphocytes. 1.Exp. Med. 178, 1199-1208. Izquierdo, M., Bowden, S., and Cantrell, D. (1994a). The role of Raf-1 in the regulation of extracellular signal-regulated kinase 2 by the T cell antigen receptor. f. Erp. Med. 180,401-406.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

83

Izquierdo, M., Leevers, S. J., Williams, D. H., Marshall, C. J., Weiss, A,, and Cantrell, D. (1994b). The role of protein kinase C in the regulation of extracellular signal-regulated kinase by the T cell antigen receptor. Eur. J. Immunol. 24, 2462-2468. Izquierdo Pastor, M., Reif, K., and Cantrell, D. (1995). The regulation and function of p21" during T-cell activation and growth. Immunol. Today 16, 159-164. Jabado, N., Pdlier, A., Jauliac, S., Fischer, A., and Hivroz, C. (1997a). gp160 of HIV or anti-CD4 monoclonal antibody ligation of CD4 induces inhibition of JNK and ERK-2 activities in human peripheral CD4+ T lymphocytes. Eur. J. Immuml. 27,397-404. Jabado, N., Palher, A., Le Deist, F., Bernard, F., Fischer, A., and Hivroz, C. (199%). CD4 ligands inhibit the formation of multifunctional transduction complexes involved in T cell activation. 1.Immunol. 158, 94-103. Jabril Cuenod, B., Zhang, C., Scharenberg, A. M., Paolini, R., Numerof, R., Beaven, M. A,, and Kinet, J. P. (1996). Syk-dependent phosphorylation of Shc. A potential link between FceRI and the Radmitogen-activated protein kinase signaling pathway through SOS and Grb2. J. Biol. Chem. 271, 16268-16272. Jacinto, E., Werlen, G., and Karin, M. (1998). Cooperation behveen Syk and Racl leads to synergistic JNK activation in T lymphocytes. Immunity 8,31-41. Jackman, J. K., Motto, D. G., Sun, Q., Tanemoto, M., Turck, C. W., Peltz, G. A., Koretzky, G. A,, and Findell, P. R. (1995). Molecular cloning of SLP-76, a 76-kDa tyrosine phosphoprotein associated with Grb2 in T cells. 1.Biol. Chem. 270, 7029-7032. Jelinek, T., Dent, P., Sturgill, T. W., and Weber, M. J. (1996). Ras-induced activation of Raf-1 is dependent on tyrosine phosphorylation. Mol. Cell. Biol. 16, 1027-1034. Jones, S. L., Knaus, U. C.,bokoch, G. M., and Brown, E. J. (1998).Two signalingmechanisms avidityin polymorphonuclear neutrophiIs.1. Biol. Chem. 273,10556for activation of aMf12 10566. Joneson, T., McDonough, M., Bar-Sagi, D., and Van Aelst, L. (1996). RAC regulation of actin polymerization and proliferation by a pathway distinct from Jun b a s e . Science 274, 1374-1376. June, C. H., Fletcher, M. C., Ledbetter, J. A., Schieven, G. L., Siegel, J. N., Phillips, A. F., and Samelson, L. E. (1990). Inhibition of tyrosine phosphorylation prevents T cell receptor-mediated signal transduction. Proc. Natl. Acad. Sci. U.S.A. 87, 7722-7726. Kaga, S., R a g , S., Rogers, K. A., and Ochi, A. (1997). Cutting edge: Stimulation of CD28 with B7-2 promotes focal adhesion-like contacts where Rho family small G proteins accumulate in T cells. J. lmmunol. 160, 24-27. Kanaho, Y.,Yokozeki, T., and Kuribara, H. (1996). Regulation of phospholipase D by low molecular weight GTP-binding proteins. J. Lipid Mediat. Cell Signal 14, 223-227. Kang, S.-M., Beverly, B., Tran, A.-C., Brorson, K., Schwartz, R. H., and Lenardo, M. J. (1992). Transactivation by AP-1 is a molecular target of T cell clonal anergy. Science 257, 1134-1138. Kanner, S. B., Grosmaire, L. S., Ledbetter, J. A,, and Damle, N. K. (1993). p2-integrin LFA-1 signding through phospholipase C-y 1 activation. Proc. Natl. Acud. Sci. U.S.A. 90,7099-7103. Kanno, T., and Siebenlist, U. (1996). Activation of nuclear factor-rcB via T cell receptor requires a Raf kinase and Ca2' influx. Functional synergy between Raf and cdcineurin. J. lmmunol. 157,5277-5283. Kapron-Bras, C., FitzGibbon, L., Jeevaratnam, P., Wilkins, J., and Dedhar, S. (1993). Stimulation of tyrosine phosphorylation and accumulation of GTP-bound p21" upon antibody-mediated a2bl integrin activation in T-lymphoblastic cells. J. Biol. Chem. 268,20701-20704.

84

AMNON ALTMAN A N D MARCEL DECKERT

Kamitz, L. M., and Abraham, R. T. (1995). Cytokine receptor signaling mechanisms. Cum Opin. Immunol. 7 , 320-326. Katzav, S., Packham, G., Sutherland, M., Aroca, P., Santos, E., and Cleveland, J. L. (1995). Vav and Ras induce fibroblast transformation by overlapping signaling pathways which require c-Myc function. Oncogene 11, 1079-1088. Katzav, S., Martin-Zanca, D., and Babacid, M. (1989).uau, a novel human oncogene derived from a locus ubiquitously expressed in hematopoietic cells. EMBO 18, 2283-2290. Kawakami, Y., Miura, T., Bissonnette, R., Hata, D., Khan, W. N., Kitamura, T., Maeda Yamamoto, M., Hartman, S. E., Yao, L., Alt, F. W., and Kawakami, T. (1997). Bruton's tyrosine kinase regulates apoptosis and JNWSAPK kinase activity. Proc. Natl. Acud. Sci. U.S.A. 94,3938-3942, Kawauchi, K., Lazarus, A. H., Sanghera, J. S., Man, G. L., Pelech, S. L., and Delovitch, T. L. (1996). Regulation of BCR- and PCK/Ca'+-mediated activation of the Rafl/MEK/ MAPK pathway by protein-tyrosine kinase and -tyrosine phosphatase activities. Mol. Immunol. 33, 287-296. Keamey, J. F., Won, W. J., Benedict, C., Moratz, C., Zimmer, P., Oliver, A., Martin, F., and Shu, F. (1997). B cell development in mice. Int. Rev. Immunol. 15, 207-241. Khosravi Far, R. S. P., Clark, G . J., Kinch, M. S., and Der, C. J. (1995). Activation of Rac, RhoA and mitogen activated protein kinases is required for Ras transformation. Mol. Cell. Bid. 15, 6443-6453. Kim, H. H., Tharayil, M., and Rudd, C. E. (1998). Growth factor receptor-bound protein 2 SHUSH3 domain binding to CD28 and its role in co-signaling. /. Bid. Chem. 273, 296-301. Kimura, K., Ito, M., Amano, M., Chihara, K., Fukata, Y., Nakafuku, M., Yamamori, B., Feng, J., Nakano, T., Okawa, K., Iwamatsu, A,, and Kaibuchi, K. (1996a). Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science 273, 245-248. Kimura, T., Kihara, H., Bhattachawa, S., Sakamoto, H., Appella, E., and Siraganian, R. P. (1996b). Downstream signaling molecules bind to different phosphorylated immunoreceptor tyrosine-based activation motif (ITAM) peptides of the high affinity IgE receptor. J. Biol. Chem. 271, 27962-27968. Kipps, T. J. (1997). Human B cell biology. Znt. Rev. Immunol. 15, 243-264. Kirchausen, T., and Rosen, F. S. (1996). Disease mechanism: Unravelling Wiskott-Aldrich syndrome. Cum. B i d . 6 , 676-678. Kitayama, H., Matsuzaki, T., Ikawa, Y., and Noda, M. (1990). Genetic analysis of the Kirsten-ras-revertant 1gene: Potentiation of its tumor suppressor activity by specific point mutations. Proc. Natl. Acad. Sci. U.S.A. 87, 4284-4288. Knaus, U., Heyworth, P., Evans, T., Curnutte, J., and Bokoch, G. (1991). Regulation of phagocyte oxygen radical production by the GTP-binding protein Race. Science 254, 1512-1515. Kolch, W., Heidecker, G., Kochs, G., Hummel, R., Vahidi, H., Mischak, H., Findenzeller, G., Marme, D., and Rapp, U. R. (1993). Protein kinase C a activates Raf-1 by direct phophorylation. Nature 364,249-252. Kolluri, R., Tohas, K. F., Carpenter, C. L., Rosen, F. S., and Kirchhausen, T. (1996). Direct interaction of the Wiskott-Aldrich syndrome protein with the GTPase Cdc42. Proc. Natl. Acad. Sci. U.S.A. 93,5615-5618. Koretzky, G. A. (1997). The role of Grb2-associated proteins in T-cell activation. Immunol. Today 18,401-406. Kotani, H., Takaishi, K., Sasaki, T., and Takai, Y. (1997). Rho regulates association of both the ERM family and vinculin with the plasma membrane in MDCK cells. Oncogene 14, 1705-1713.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

85

Kumar, G., Wang. S., Gupta, S., and Nel, A. (1995). The membrane immunoglobulin receptor utilizes a ShdGrbZhSOS complex for activation of the mitogen-activated protein kinase cascade in a B-cell line. Biochem. J. 307,215-223. Kupfer, A,, Swain, S. L., and Singer, S. J. (1987). The specific direct interaction of helper T cells and antigen-presenting B cells. 11. Reorientation of the microtubule organizing center and reorganization of the membrane-associated cytoskeleton inside the bound helper T cells. J. Exp. Med. 165, 1565-1580. Kupfer, A., and Singer, S. J. (1989). Cell biology of cytotoxic and helper T cell functions: Immunofluorescence microscopic studies of single cells and cell couples. Annu. Rev. Immunol. 7,309-337. Kurosaki, T. (1997). Molecular mechanisms in B cell antigen receptor signaling. Cuw. Opin. lmmunol. 9,309-318. Lahesmaa, R., Allsup, A., Soderberg, C., Jackman, J., Findell, P., and Peltz, G. (1995). Modulation of the Grb2-associated protein complex in human CD4+ T cells by receptor activation.J. lmmunol. 155,3815-3822. Laird, A. D., Taylor, S. J., Oberst, M., and Shalloway, D. (1995). Raf-1 is activated during mitosis. J. Biol. Chem. 270, 26742-2674s. Lamarche, N., Tapon, N., Stowers, L., Burbelo, P. D., Aspenstrom, P., Bridges, T., Changt, J., and Hall, A. (1996). Rac and Cdc42 induce actin polymerization and G1 cell cycle progression independently of ~ 6 5 and ' ~ the ~ JNUSAPK MAPkinase cascade. Cell 87, 519-529. Lang, P., Guizani, L., Vitte Mony, I., Stancou, R., Dorseuil, O., Gacon, G., and Bertoglio, J. (1992). ADP-ribosylation of the Ras-related, GTP-binding protein RhoA inhibits lymphocyte-mediated cytotoxicity.J. Biol. Chem. 267, 11677- 11680. Lang, P., Gesbert, F., Delespine-Carmagnat, M., Stancou, R., Pouchelet, M., and Bertoglio, J. (1996). Protein kinase A phosphorylation of RhoA mediates the morphological and functional effects of cyclic AMP in cytotoxic lymphocytes. EMBO 1. 15, 510-519. Latinis, K. M., Carr, L. L., Peterson, E. J., Norian, L. A,, Eliason, S. L., and Koretzky, C. A. (1997). Regulation of CD95 (Fas) ligand expression by TCR-mediated signaling events. J. lmmunol. 158,4602-4611. Laudanna, C., Campbell, J. J., and Butcher, E. C. (1996). Role of Rho in chemoattractantactivated leukocyte adhesion through integrins. Science 271, 981-983. Laudanna, C., Campbell, J. J.. and Butcher, E. C. (1997). Elevation of intracellular CAMP inhibits RhoA activation and integrin-dependent leukocyte adhesion induced by chemoattractants. J. Biol. Chem. 272, 24141-24144. Law, G. J., Northrop, A. J., and Mason, W. T. (1993). rab3-peptide stimulates exocytosis from mast cells via a pertussis toxin-sensitive mechanism. FEBS Lett. 333,56-60. Lazarus, A. H., Kawauchi, K., Rapoport, M. J., and Delovitch, T. L. (1993). Antigen-induced B lymphocyte activation involves the p2lras and ras GAP signaling pathway.J. Exp. Med. 178, 1765-1769. Lee, J. M., Fournel, M., Veillette, A., and Branton, P. E. (1996). Association of CD45 with Lck and components of the Ras signalling pathway in pervanadate-treated mouse T-cell lines. Oncogene 12, 253-263. Leevers, S. J., Paterson, H. F., and Marshall, C. J. (1994). Requirement for Ras in Raf activation is overcome by targeting Raf to the plasma membrane. Nature 369,411-414. Leibson, P. J. (1995). MHC-recognizing receptors: They're not just for T cells anymore. lmmunity 3, 5-8. Lenczowski,J. M., Dominguez, L., eder, A. M., King, L. B., Zacharchuk, C. M., and Ashwell, J. D. (1997). Lack of role for Jun kinase and AP-1 in Fas-induced apoptosis. Mol. Cell. Biol. 17, 170-181.

86

AMNON ALTMAN A N D MARCEL DECKERT

Lenschow, D. J., Walunas, T. L., and Bluestone, J. A. (1996). CD28B7 system of T cell costimulation.Annu. Reu. lmmunol. 14, 233-258. Ley, S. C., Verbi, W., Pappin, D. J., Druker, B., Davies, A. A., and Crumpton, M. J. (1994).Tyrosine phosphorylationof (Y tubulin in human T lymphocytes. Eur. 1.lmmunol. 24, 99-106. Li, B., Subleski, M., Fusaki, N., Yamamoto, T., Copeland, T., Princler, G. L., Kung, H., and Kamata, T. (1996). Catalytic activity of the mouse guanine nucleotide exchanger mSOS is activated by Fyn tyrosine protein kinase and the T-cell antigen receptor in T cells. Proc. Natl. Acad. Sci. U.S.A. 93, 1001-1005. Li, H. L., Forman, M. S., Kurosaki, T., and Pure, E. (1997). Syk is required for BCRmediated activation of p9Pk, but not p7OSa, via a mitogen-activated protein kinaseindependent pathway in B cells. 1.Bio2. Chem. 272, 18200-18208. Li, N., Baker, A., Daly, R., Yajnik, V., Skolnik, E., Chardin, P., Bar Sagi, D., Margolis, B., and Schlessinger,J. (1993).Guanine-nucleotide-releasingfactor hSosl binds to Grb2 and links receptor tyrosine kinases to Ras signalling. Nature 363, 85-88. Li, W., Whaley, C. D., Mondino, A., and Mueller, D. L. (1996).Blocked signal transduction to the ERK and JNK protein kinases in anergic CD4+ T cells. Science 271, 1272-1276. Li, Z., Wahl, M. I., Eguinoa, A., Stephens, L. R., Hawkins, P. T., and Witte, 0. N. (1997). Phosphatidylinositol 3-kinase-gamma activates Bruton’s tyrosine kinase in concert with Src family kinases. Proc. Natl. Acad. Sci. U.S.A. 94, 13820-13825. Lin, A., Minden, A., Martinetto, H., Claret, F. X., Lange, C. C., Mercurio, F., Johnson, G. L., and Karin, M. (1995). Identification of a dual specificity ldnase that activates the Jun kinases and p38-Mpk2. Science 268,286-290. Lin, K., and Abraham, K. M. (1997). Targets of ~ 5 activity 6 ~ in ~immature thymoblasts: Stimulation of the Ras/RdMAPK pathway. lnt. Immuaol. 9,291-306. Lin, T. H., Rosales, C., Mondal, K., Bolen, J. B., Haskill, S., and Jdano, R. L. (1995). Integrin-mediatedtyrosine phosphorylationand cytokine message induction in monocytic cells. Apossible signalingroleforthe Syktyrosinekinase.]. Biol. Chem. 270,16189-16197. Linette, G . P., Hartzman, R. J., Ledbetter, J. A,, and June, C. H. (1988). HN-1-infected T cells show a selective signaling defect after perturbation of CD3/antigen receptor. Science 241, 573-576. Liu, Y.-C., and Altman, A. (1998). Cbl: Complex formation and functional implications. Cell Signal 10, 377-385. Liu, Y. C., Elly, C., Langdon,W. Y., and Altman, A. (1997).Ras-dependent,Ca2+-stimulated activation of nuclear factor of activated T cells by a constitutively active Cbl mutant in T cells. 1.Biol. Chern. 272, 168-173. Loh, C., Romeo, C., Seed, B., Bruder, J. T., Rapp, U., and Rao, A. (1994).Association of Raf with the CD3 6 and y chains of the T cell receptor-CD3 complex. 1.Biol. Chem. 269,8817-8825. Loh, C., Carew, J. A., Kim, J., Hogan, P. G., and Rao, A. (1996).T-cell receptor stimulation elicits an early phase of activation and a later phase of deactivation of the transcription factor NFAT1. Mol. Cell. Biol. 16, 3945-3954. Lores, P., Morin, L., Luna, R., and Gacon, G. (1997).Enhanced apoptosis in the thymus of transgenic mice expressing constitutively activated forms of human Race GTPase. Oncogene 15,601-605. Lowenstein, E. J., Daly, R. J., Baker, A. G., Li, W., Margolis, B., Lammers, R., Ullrich, A., Skolnik, E. Y., Bar Sagi, D., and Schlessinger,J. (1992).The SH2 and SH3 domaincontainingprotein GRB2 links receptor tyrosine kinases to ras signaling. Cell 70,431-442. Lu, X., Wu, X., Plemenitas, A., Yu, H., Sawai, E. T., Abo, A,, and Peterlin, B. M. (1996). CDC42 and Racl are implicated in the activation of the Nef-associated kinase and replication of HIV. Cum Biol. 6, 1677-1684.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

87

Lupher, M. L., Jr., Reedquist, K. A,, Miyake, S., Langdon, W. Y., and Band, H. (1996). A novel phosphovosine-binding domain in the N-terminal transforming region of Cbl interacts directly and selectivelywith ZAP-70 in T cells.J. Biol. Chem. 271,24063-24068. Madrenas, J., Wange, R. L., Wang, J. L., Isakov, N., Samelson, J. F., and Germain, R. N. (1995). 4' phosphorylation without ZAP-70 activation induced by TCR antagonists or partial agonists. Science 267,515-518. Marais, R., Light, Y., Paterson, H. F., and Marshall, C. J. (1995). Ras recruits Raf-1 to the plasma membrane for activation by tyrosine phosphorylation. EMBO J. 14,3136-3145. Marais, R., Light, Y., Mason, C., Paterson, H., Olson, M. F.. and Marshall, C. J. (1998). Requirement of Ras-GTP-Raf complexes for activation of Raf-1 by protein kinase C. Science 280, 109-112. Marengere, L. E., Okkenhaug, K., Clavreul, A., Couez, D., Gibson, S., Mills, G. B., Mak, T. W., and Rottapel, R. (1997).The SH3 domain ofItk/Emt binds to proline-rich sequences in the cytoplasmic domain of the T cell costimulatory receptor CD28. I. Immunol. 159,3220-3229. Marie Cardine, A., Kirchgessner, H., Eckerskorn, C., Meuer, S. C., and Schraven, B. (1995). Human T lymphocyte activation induces tyrosine phosphorylation of alpha-tubulin and its association with the SH2 domain of the p59fyn protein tyrosine kinase. Eur.1.Immunol. 25,3290-3297. Marshall, C. J. (1996). Ras effectors. Cum @in. Cell Biol. 8, 197-204. Matsuda, S., Moriguchi, T., Koyasu, S., and Nishida, E. (1998). T lymphocyte activation signals for interleukin-2 production involve activation of MKK6-p38 and MKK7-SAPW JNK signaling pathways sensitive to cyclosporin A. J. Biol. Chem. 273, 12378-12382. McMahon, S. B., and Monroe, J. G. (1995).Activation of the p21" pathway couples antigen receptor stimulation to induction of the primary response gene egr-1 in B lymphocytes. J. Exp. Med. 181,417-422. Meisner, H., Conway, B. R., Hartley, D., and Czech, M. P. (1995). Interactions of Cbl with Grb2 and phosphatidylinositol 3'-kinase in activated Jurkat cells. Mol. Cell. B i d . 15,3571-3578. Melamed, I., Downey, G. P., Aktories, K., and Roifman, C. M. (1991a). Microfilament assembly is required for antigen-receptor-mediated activation of human B lymphocytes. J. Immunol. 147, 1139-1146. Melamed, I., Downey, G. P., and Roifman, C. M. (1991b). Tyrosine phosphorylation is essential for microfilament assembly in B lymphocytes. Biochem. Biophys. Rex Commun. 176,1424-1429. Michiels, F., Habets, G. G. M., Stam, J. C., van der Kammen, R. A., and Collard, J. G. (1995). A role for Rac in Tiaml-induced membrane ruffling and invasion. Nature 375,338-340. Miedema, F., Tersmette, M., and van Lier, R. A. W. (1990).AIDS pathogenesis: A dynamic interaction between HIV and the immune system. Immunol. Today 11,293-297. Migita, K., Eguchi, K., Kawabe, Y., Tsukada, T., Ichinose, Y., and Nagataki, S. (1995). Defective TCR-mediated signaling in anergic T cells. J. Immunol. 155, 5083-5087. Miki, H., Miura, K., and Takenawa, T. (1996). N-WASP, a novel actin-depolymerizing protein, regulates the cortical cytoskeletal rearrangement in a PIP2-dependent manner downstream of tyrosine kinases. EMBO J. 15, 5326-5335. Miki, H., Nonoyama, S., Zhu, Q., Aruffo, A,, Ochs, H. D., and Takenawa, T. (1997).Tyrosine kinase signaling regulates Wiskott-Aldrich syndrome protein function, which is essential for megakaryocyte differentiation. Cell Growth DzZ 8, 195-202. Milia, E., Di Somma, M. M., Majolini, M. B., Ulivieri, C., Somma, F., Piccolella, E., Telford, J. L., and Baldari, C. T. (1997).Gene activatingand proapoptotic potential are independent properties of different CD4 epitopes. Mol. Immunol. 34, 287-296.

88

AMNON ALTMAN A N D MARCEL DECKERT

Minden, A., Lin, A,, Claret, F. X., Abo, A., and Karin, M. (1995). Selective activation of the JNK signaling cascade and c-Jun transcriptional activity by the s m d GTPases Rac and Cdc42Hs. Cell 81, 1147-1157. Mittler, R. S., Coldman, S. J., Spitalny, G. L., and Burakoff, S. J. (1989). T-cell receptorCD4 physical association in a murine T-cell hybridoma: Induction by antigen receptor ligation. Proc. Nutl. Acud. Sci. U.S.A. 86, 8531-8535. Miura, Y., Miura, O., Ihle, J. N., and Aoki, N. (1994). Activation of the mitogen-activated protein b a s e pathway by the erythropoietin receptor. J. Biol. Chem. 266,29962-29969. Molina, I. J., Sancho, J., Terhorst, C., Rosen, F. S., and Remold-O’Donnell, E. (1993). T cells of patients with the Wiskott-Aldrich syndrome have a restricted defect in proliferative responses. J. Immunol. 151,4383-4390. Mollinedo, F., Gajate, C., and Flores, I. (1994). Involvement of phospholipase D in the activation of transcription factor AP-1 in human T lymphoid Jurkat cells. J. Immunol. 153,2457-2469. Mondino, A., Whaley, C. D., DeSilva, D. R., Li, W., Jenkins, M. K., and Mueller, D. L. (1996). Defective transcription of the IL-2 gene is associated with impaired expression of c-Fos, FosB, and JunB in anergic T helper 1 cells. J. Immunol. 157,2048-2057. Monks, C . R. F., Kupfer, H., Tamir, I., Barlow, A,, and Kupfer, A. (1997). Selective modulation of protein b a s e C-8 during T-cell activation. Nature 385, 83-86. Monroe, J. G., Yellen-Shaw, A. J,, and Seyfert, V. L. (1993). Molecular basis for unresponsiveness and tolerance induction in immature stage B lymphocytes. Adv. Mol. Cell. Immunol. lB, 1-32. Moorman, J. P., Bobak, D. A,, and Hahn, C. S. (1996). Inactivation of the small GTP binding protein Rho induces multinucleate cell formation and apoptosis in murine T lymphoma ELI. J. Immunol. 156,4146-4153. Morrison, D. K., Kaplan, D. R., Escobedo, J. A., Rapp, U. R., Roberts, T. M., and Williams, L. T. (1989). Direct activation of serine-threonine kinase activity of the protooncogene Raf-1 through tyrosine phosphorylation by the PDGF-/3 receptor. Cell 58, 649-657. Motto, D. G., Ross, S. E., Jackman, J. K., Sun, Q., Olson, A. L., Findell, P. R., and Koretzky, G. A. (1994). In vivo association of Grb2 with pp116, a substrate of the T cell antigen receptor-activated protein tyrosine kinase. J. Biol. Chem. 269, 21608-21613. Motto, D. G., Musci, M. A,, Ross, S. E., and Koretzky, G. A. (1996a).Tyrosine phosphorylation of Grb2-associated protein correlates with phospholipase Cyl activation in T cells. Mol. Cell. B i d . 16, 2823-2829. Motto, D. G., Ross, S. E., Wu, J., Hendricks Taylor, L. R., and Koretzky, G. A. (1996b). Implication of the GRB2-associated phosphoprotein SLP-76 in T cell receptor-mediated interleuldn 2 production. J. Exp. Med. 183, 1937-1943. Musci, M. A., Motto, D. G., Ross, S. E., Fang, N., and Koretzky, G. A. (1997). Three domains of SLP-76 are required for its optimal function in a T cell line. J. Immunol. 159, 1639-1647. Mustelin, T., Coggeshall, K. M., Isakov, N., and Altman, A. (1990).Tyrosine phosphorylation is required for T cell receptor-mediated activation of phospholipase C. Science 247,15841587. Nagata, K., Nakamura, T., Kitamura, F., Kuramochi, S., Taki, S., Campbell, K. S., and Karasuyama, H. (1997). The Ig alphagbeta heterodimer on mu-negative proB cells is competent for transducing signals to induce early B cell differentiation. Immunity 7,559-570. Nair, A. P., Hirsch, H. H., and Moroni, C. (1992). Mast cells sensitive to v-H-ras transformation are hyperinducible for interleukin 3 expression and have lost tumor-suppressor activity. Oncogene 7 , 1963-1972.

GTPasrs IN IMMUNE RECOGNITION RECEPTOR SIGNALING

89

Nakafuku, M., Satoh, T., and Kaziro, Y. (1992). Differentiation factors, including nerve growth factor, fibroblast growth factor, and interleukin-6, induce accumulation of an active Ras.GTP complex in rat pheochromocytoina PC12 cells.J. Biol. Chem. 267,19448-19454. Narumiya, S., Ishizaki, T., and Watanabe, N. (1997). Rho effectors and reorganization of actin cytoskeleton. FEBS Lett. 410, 68-72. Natarajan, V., Scribner, W. M., and Vepa, S. (1996). Regulation of phospholipase D by tyrosine kinases. Chem. Phys. Lipids 80, 103-116. Negulescu, P. A,, Krasieva, T. B., Kerschbaum, H. H., and Cahalan, M. D. (1996). Polarity of T cell shape, motility, and sensitivity to antigen. Immunity 4, 421-430. Nel, A. E., Hanekom, C., Rheeder, A,, Williams, K., Pollack, S., Katz, R., and Landreth, G . (1990). Stimulation of MAP-2 b a s e activity in T lymphocytes by anti-CD3 or antiTi monoclonal antibody is partially dependent on protein kinase C.J. Zrnrnunol. 144,26832689. Nel, A. E., Gupta, S., Lee, L., Ledbetter, J. A,, and Kanner, S. B. (1995). Ligation of the T-cell antigen receptor (TCR) induces association of hSosl, ZAP-70, phospholipase Cy l , and other phosphoproteins with Grb2 and the kchain of the TCR. I. Biol. Chem. 270, 18428-18436. Nishina, H., Fischer, K. D., Radvanyi, L., Shahinian, A,, Hakem, R., Rubie, E. A,, Bernstein, A,, Mak, T. W., Woodget, J. R., and Penninger, J. M. (1997). Stress-signallingkinase Sekl protects thymocytes from apoptosis mediated by CD95 and CD3. Nature 385,350-353. Nobes, C. D., and Hall, A. (1995). Rho, Rac, and Cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia. Cell 81, 53-62. Norman, J. C., Price, L. S., Ridley, A. J., Hall, A,, and Koffer, A. (1994). Actin filament organization in activated mast cells is regulated by heterotrimeric and small GTP-binding proteins. I. Cell Biol. 126, 1005-1015. Norman, J. C., Price, L. S., Ridley, A. J., and Koffer, A. (1996). The small CTP-binding proteins, Rac and Rho, regulate cytoskeletd organization and exocytosis in mast cells by parallel pathways. Mol. B i d . Cell 7, 1429-1442. Northrop, J. P., Pustelnik, M. J., Lu, A. T., and Grove, J. R. (1996). Characterization of the roles of SH2 domain-containingproteins in T-lymphocyteactivatioii by using dominant negative SH2 domains. Mol. Cell. Biol. 16, 2255-2263. Nunes, J. A., Collette, Y., Truneh, A,, Olive, D., and Cantrell, D. A. (1994). The role of p21” in CD28 signal transduction: Triggering of CD28 with antibodies, but not the ligand B7-1, activates p21“. I. Exp. Med. 180, 1067-1076. Nunes, J. A., Truneh, A,, Olive, D., and Cantrell, D. A. (1996). Signal transduction by CD28 costimulatory receptor on T cells. B7-1 and B7-2 regulation of tyrosine kinase adaptor molecules. J. B i d . Chem. 271, 1591-1598. Nunn, M. F., and Marsh, J. W. (1996).Human immunodeficiencyvirus type 1Nef associates with a member of the pel-activated kinase family. J . Virol. 70, 6157-6161. Nuoffer, C., and Balch, W. E. (1994). CTPases: Multifunctional molecular switches regulating vesicular traffic. Annu. Rezj. Biochem. 63, 949-990. Oberhauser, A. F., Monck, J. R., Balch, W. E., and Fernandez, J. M. (1992). ExocFc fusion is activated by Rab3a peptides. Nature 360, 270-273. Ohtsuka, T., Kaziro, Y., and Satoh, T. (1996). Analysis of the T-cell activation signding pathway mediated by tyrosine kinases, protein kinase C, and Ras protein, which is modulated by intracellular cyclic AMP. Biochim. Biuphys. Actu. 1310, 223-232. Olkkonen, V. M., and Stenmark, H. (1997). Role of Rab GTPases in membrane traffic. Znt. Rev. Cytol. 176, 1-85.

90

AMNON ALTMAN AND MARCEL DECKERT

Olson, S. C., and Lambeth, J. D. (1996). Biochemistry and cell biology of phospholipase D in human neutrophils. Chem. Phys. Lipids 80,3-19. Olson, M. F., Ashworth, A., and Hall, A. (1995).An essential role for Rho, Rac, and Cdc42 GTPases in cell cycle progression through G1. Science 269, 1270-1272. Olson, M. F., Pasteris, N. G., Gorski, J. L., and Hall, A. (1996). Faciogenital dysplasia protein (FGD1)and Vav, two related protein required for normal embryonicdevelopment, are upstream regulators of Rho GTPases. Cum. Biol. 6, 1628-1633. Ono, M., Bolland, S., Tempst, P., and Ravetch, J. V. (1996).Role of the inositol phosphatase SHIP in negative regulation of the immune system by the receptor FqRIIB. Nature 383,263-266. OShea, C. C., Crompton, T., Rosewell, I. R., Hayday, A. C., and Owen, M. J. (1996). Raf regulates positive selection. Eur. J. lmmunol. 26, 2350-2355. Osman, N., Lucas, S. C., Turner, H., and Cantrell, D. (1995).Acomparison ofthe interaction of Shc and the tyrosine kinase ZAf-70 with the T cell antigen receptor zeta chain tyrosinebased activation motif. J. Biol. C h a . 270, 13981-13986. Ota, Y., and Samelson, L. E. (1997).The product of the proto-oncogene c-cbl: A negative regulator of the Syk tyrosine kinase. Science 276, 418-420. Ow&, H., Vanna, R., Gillis, B., Bruder, J. T., Rapp, U. R., Davis, L. S., and Geppert, T. D. (1993). Raf-1 is required for T cell IL2 production. EMBO J. 12, 4367-4373. Oyaizu, N., McCloskey, T. W., Than, S., Hu, R., Kalyanaraman, V. S., and Pahwa, S. (1994). Cross-linking of CD4 molecules upregulates Fas antigen expression in lymphocytes by inducing interferon-gammaand tumor necrosis factor-a secretion. Blood 84,2622-2631. Pardi, R., Inverardi, L., Rugarli, C., and Bender, J. R. (1992). Antigen-receptor complex stimulation triggers protein kinase C-dependent CDlldCD18-cytoskeleton association in T lymphocytes. J. Cell Biol. 116, 1211-1220. Park, J. H., and Levitt, L. (1993).Overexpression of mitogen-activatedprotein kinase (ERK1) enhances T-cell cytokine gene expression: Role of AP1, NF-AT, and NF-KB. Blood 82,2470-2477. Park, R. K., Liu, Y., and Durden, D. L. (1996).A role for Shc, Grb2, and Raf-1 in FcyRI signal relay. J. Biol. Chem. 271, 13342-13348. Parsey, M. V., and Lewis, G. K. (1993).Actin polymerization and pseudopod reorganization accompany anti-CD3-inducedgrowth arrest in JurkatT ce1ls.J. lmmunol. 151,1881-1893. Parsons, J. T. (1996). Integrin-mediated signalling: Regulation by protein tyrosine kinases and small GTP-binding proteins. C u m @in. Cell Biol. 8, 146-152. Pathan, N. I., Ashendel, C. L., Geahlen, R. L., and Harrison, M. L. (1996). Activation of T cell Raf-1 at mitosis requires the protein-tyrosinekinase Lck.1.Biol. Chem. 271,3031530317. Pelicci, G., Lanfrancone, L., Grignani, F., McGlade, J., Cavallo, F., Forni, G., Nicoletti, F., Grignani, T., Pawson, T., and Pelicci, P. G. (1992). A novel transforming protein (SHC)with an SH2 domain is implicatedin mitogenicsignal transduction. Cell 70,93-104. Perlmutter, R. M. (1995). Control of T cell development by non-receptor protein tyrosine kinases. Cancer Sum. 22, 85-95. Peters, J. D., Furlong, M. T., Asai, D. J., Harrison, M. L., and Geahlen, R. L. (1996). Syk, activated by cross-linking the B-cell antigen receptor, localizes to the cytosol where it interacts with and phosphorylates a-tub& on tyrosine.]. Biol. C h a . 271,4755-4762. Pfeiffer, J. R., and Oliver, J. M. (1994). Tyrosine kinase -dependent assembly of actin plaques linking FcsRI cross-linking to increased cell substrate adhesion in RBL-2H3 tumor mast cells. J. lmmunol. 152, 270-279. Phatak, P. D., Paclanan, C. H., and Lichtman, M. A. (1988).Protein kinase C modulates actin conformation in human T lymphocytes. J. lmmunol. 141, 2929-2934.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

91

Phatak, P. D., and Packman, C. H. (1994). Engagement of the T-cell antigen receptor by anti-CD3 monoclonal antibody causes a rapid increase in lymphocyte F-actin. J. Cell Physiol. 159, 365-370. Podack, E. R., and Kupfer, A. (1991). T-cell effector functions: mechanisms for delivery of cytotoxicity and help. Annu. Reu. Cell Biol. 7 , 479-504. Poo, W. J., Conrad, L., and Janeway, C. A,, Jr. (1988). Receptor-directed focusing of lymphokine release by helper T cells. Nature 332, 378-380. Popik, W., and Pitha, P. M. (1996). Binding of human immunodeficiency virus type 1 to CD4 induces association of k k and Raf-1 and activates Raf-1 by a Ras-independent pathway. Mol. Cell. Biol. 16, 6532-6541. Prasad, K. V., Cai, Y. C., Raab, M., Duckworth, B., Cantley, L., Shoelson, S. E., and Rudd, C. E. (1994). T-cell antigen CD28 interacts with the lipid b a s e phosphatidylinositol3kinase by a cytoplasmicTyr(P)-Met-Xaa-Metmotif. Proc. Nutl. Acud. Sci. U.S.A. 91,28342838. Pratt, J. C., Weiss, M., Sieff, C. A., Shoelson, S. E., Burakoff, S. J., and Ravichandran, K. S. (1996). Evidence for a physical association between the Shc-PTB domain and the beta c chain of the granulocyte-macrophage colony-stimulating factor receptor. J. Biol. Chem. 271, 12137-12140. Prendergast, G. C., Khosravi, F. R., Solski, P. A,, Kurzawa, H., Lebowitz, P. F., and Der, C. J. (1995). Critical role of Rho in cell transformation by oncogenic Ras. Oncogene 10,2289-2296. Prepens, U., Just, I., von Eichel-Streiber, C., and Aktories, K. (1996). Inhibition of FcsRImediated activation of rat basophilic leukemia cells by Clostridium dificile toxin B. J. Biol. Chem. 271, 7324-7329. Price, L. S., Norman, J. C., Ridley, A. J., and Koffer, A. (1995). The small GTPases Rac and Rho as regulators of secretion in mast cells. Cum. Biol. 5, 68-73. Prieschl, E. E., Pendl, G. G., Harrer, N. E., and Baumruker, T. (1995). p2lW links FcsRI to NF-AT family member in mast cells. The AP3-like factor in this cell type is an NFAT family member. J. Immunol. 155,4963-4970. Pronk, G. J., de Vries Smits, A. M., Buday, L., Downward, J., Maassen, J. A., Medema, R. H., and Bos, J. L. (1994). Involvement of Shc in insulin- and epidermal growth factorinduced activation of p2lras. Mol. Cell. Biol. 14, 1575-1581. Ptasznik, A., Traynor-Kaplan,A., and Bokoch, G. M. (1995). G protein-coupled chemoattractant receptors regulate Lyn tyrosine kinase. Shc adapter protein signaling complexes. J. Biol. Chem. 270, 19969-19973. Ptasznik, A,, Prossnitz, E. R., Yoshikawa, D., Smrcka, A,, Traynor-Kaplan, A,, and Bokoch, G. M. (1996).A tyrosine ldnase signalingpathway accounts for the majority ofphosphatidylinositol3,4,5-trisphosphateformation in chemoattractant-stimulated human neutrophils. J. Biol. Chem. 271,25204-25207. Qiu, R. G., Chen, J., Kim, D., McCormick, F., and Symons, M. (1995a). An essential role for Rac in Ras transformation. Nature 374, 457-459. Qiu, R. G., Chen, J,, McCormick, F., and Symons, M. (1995b). A role for Rho in Ras transformation. Proc. Nutl. Acad. Sci. U.S.A.92, 11781-11785. Qiu, R.-G., Abo, A., McCormick, F., and Symons, M. (1997). Cdc42 regdates anchorageindependent growth and is necessary for Ras transformation. Mol. Cell. Biol. 17, 34493458. Quill, H., Riley, M. P., Cho, E. A., Casnellie, J. E., Reed, J. C., and Torigoe, T. (1992). Anergic Thl cells express altered levels of the protein tyrosine ldnases pSSkkand p5*. 1.Immunol. 149,2887-2893.

92

AMNON ALTMAN AND MARCEL DECKERT

Quilliam, L. A,, Mueller, H., Bohl, B. P., Prossnitz, V., Sklar, L. A., Der, C. J., and Bokoch, G. M. (1991). RaplA is a substrate for CAMP-dependentprotein kinase in human neutrophils. I. Immunol. 147, 1628-1635, Quilliam, L. A,, Huff, S. Y., Rabun, K. M., Wei, W., Park, W., Broek, D., and Der, C. J. (1994). Membrane-targeting potentiates guanine nucleotide exchange factor CDC25 and SOSl activation of Ras transforming activity. Proc. Nutl. Acud. Sci. U.S.A.91,8512-8516. Quilliam, L. A., Lambert, Q. T., Mickelson Young, L. A., Westwick, J. K., Sparks, A. B., Kay, B. K., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., and Der, C. J. (1996). Isolation of a NCK-associated kinase. PRK2, an SH3-binding protein and potential effector of Rho protein signaling. J. Biol. Chem. 271, 28772-28776. Quinn, M. T. (1995). Low-molecular-weight GTP-binding proteins and leukocyte signal transduction. J. Leuk. Biol. 58,263-276. Raab, M., Da Silva, A. J.. Findell, P. R., and Rudd, C. E. (1997). Regulation of VavSLP-76 binding by ZAP-70 and its relevance to TCR gCD3 induction of interleukin-2. Immunity 6, 155-164. Ramos-Morales, F., Druker, B. J., and Fischer, S. (1994). Vav binds to several SHUSH3 containing proteins in activated lymphocytes. Oncogene 9, 1917- 1923. Ramos-Morales, F., Romeo, F., Schweighoffer, F., Bismuth, G., Camonis, J., Tortolero, M., and Fischer, S. (1995). The proline-rich region of Vav binds to Grb2 and Grb3-3. Oncogene 11, 1665-1669. Rao, A,, Luo, C., and Hogan, P. G. (1997). Transcription factors of the NFAT family: Regulation and function. Annu. Rev. Immunol. 15, 707-747. Rapoport, M. J.. Jaramillo, A., Zipris, D., Lazarus, A. H., Serreze, D. V., Leiter, E. H., Cyopick, P., Danska, J. S., and Delovitch, T. L. (1993a). Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. /. Exp. Med. 178, 87-99. Rapoport, M. J., Lazarus, A. H., Jaramillo, A,, Speck, E., and Delovitch, T. L. (1993b). Thymic T cell anergy in autoimmune nonobese diabetic mice is mediated by deficient T cell receptor regulation of the pathway ofp2lraSactivation.]. Exp. Med. 177,1221-1226. Ravichandran, K. S., and Burakoff, S. J. (1994). The adapter protein Shc interacts with the interleukin-2 (IL-2) receptor upon IL-2 stimulation. J . Bid. Chern. 269, 1599-1602. Ravichandran, K. S., Lee, K. K., Songyang, Z., Cantley, L. C., Bum, P., and Burakoff, S. J. (1993). Interaction of Shc with the 5 chain of the T cell receptor upon T cell activation. Science 262, 902-905. Ravichandran, K. S., Lorenz, U., Shoelson, S. E., and Burakoff, S. J. (1995). Interaction of Shc with Grb2 regulates association of Grb2 with mSOS. MoZ. Cell. B i d . 15, 593-600. Ravichandran, K. S., Igras, V., Shoelson, S. E., Fesik, S. W., and Burakoff, S. J. (1996). Evidence for a role for the phosphotyrosine-binding domain of Shc in interleukin 2 signaling. Proc. Natl. Acad. Sd. U.S.A. 93,5275-5280. Rayter, S. I., Woodrow, M., Lucas, S. C., Cantrell, D. A,, and Downward, J. (1992). p2lras mediates control of IL-2 gene promoter function in T cell activation. EMBO J . 11,4549-4556. Razin, E., Pecht, I., and Rivera, J. (1995). Signal transduction in the activation of mast cells and basophils. Immunol. Today 16,370-373. Reibel, L., Dorseuil, O., Stancou, A,,Bertoglio, J., and Gacon, G. (1991). A hemopoietic specific gene encoding a small GTP binding protein is overexpressed during T cell activation. Biochem. Biuphys. Res. Commun. 175, 451-458. Reich, Z., Boniface, J. J., Lyons, D. S., Borochov, N., Wachtel, E. J., and Davis, M. M. (1997). Ligand-specific oligomerization of T-cell receptor molecules. Nuture 387, 617-620.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

93

Reid, P. A., Gardner, S. D., Williams, D. M., and Harnett, M. M. (1997). The antigen receptors on mature and immature T lymphocytes are coupled to phosphatidylcholinespecific phospholipase D activation. Immunology 90, 250-256. Reif, K., and Cantrell, D. A. (1998). Networking Rho family GTPases in lymphocytes. Immunity 8, 395-401. Reif, K., Buday, L., Downward, J., and Cantrell, D. A. (1994). SH3 domains of the adapter molecule Grb2 complex with two proteins in T cells: The guanine nucleotide exchange protein Sos and a 75-kDa protein that is a substrate for T cell antigen receptor-activated tyrosine kinases. 1.Biol. Chem. 269, 14081-14087. Reif, K., Nobes, C. D., Thomas, G., Hall, A., and Cantrell, D. A. (1996). Phosphatidylinositol 3-kinase signals activate a selective subset of Rac/Rho-dependent effector pathways. Curr. Biol. 6, 1445-1455. Renshaw, M. W., Toksoz, D., and Schwartz, M. A. (1996). Involvement of the small GTPase rho in integrin-mediated activation of mitogen-activated protein kinase. /. Biol. Chem. 271,21691-21694. Rider, L. G., Hirasawa, N., Santini, F., and Beaven, M. A. (1996).Activation of the mitogenactivated protein kinase cascade is suppressed by low concentrations of dexamethasone in mast cells. I. Immunol. 157, 2374-2380. Ridley, A. J. (1995). Rho-related proteins: Actin cytoskeleton and cell cycle. Cum: q i n . Genet. Deu. 5,24-30. Rivas, A,, Takada, S., Koide, J., Sonderstrupp-McDevitt, G., and Engelman, E. G. (1988). CD4 molecules are associated with the antigen receptor complex on activated but not resting T cells. J. Immunol. 140, 2912-2918. Rivero Lezcano, 0. M., Marcilla, A., Sameshima, J. H., and Robbins, K. C. (1995).WiskottAldrich syndrome protein physically associates with Nck through Src homology 3 domains. Mul. Cell. B i d . 15, 5725-5731. Roa, M., Paumet, F., Le Mao, J., David, B., and Blank, U. (1997). Involvement of the raslike GTPase rab3d in RBL-2H3 mast cell exocytosis following stimulation via high affinity IgE receptors (FcsRI).J Immunol. 159, 2815-2823. Rodriguez-Viciana, P., warne, P. H., Dhand, R., Vanhaesebroeck, B., Gout, I., Fry, M. J., Waterfield, M. D., and Downward, J. (1994). Phosphatidylinositol-3-OHkinase as a direct target of Ras. Nature 370, 527-532. Rojo, J. M., Saizawa, K., and Janeway, C. A. J. (1989). Physical association of CD4 and the T-cell receptor can be induced by anti-T-cell receptor antibodies. Proc. Natl. Acad. Sci. U.S.A. 86, 3311-3315. Romero, F., and Fischer, S. (1996). Structure and function of Vav. Cell Signal 8,545-553. Rosen, S. D., Hwang, S. T., Giblin, P. A,, and Singer, M. S . (1997). High-endothelial-venule ligands for L-selectin: Identification and functions. Biochem. SOC. Trans. 25, 428-433. Rosoff, P. M., Burakoff, S. J., and Greenstein, J. L. (1987). The role of the L3T4 molecule in mitogen and antigen-activated signal transduction. Cell 49, 845-853. Rozakis Adcock, M., McGlade, J., Mbamalu, G., Pelicci, G., Daly, R., Li, W., Batzer, A., Thomas, S., Brugge, J., Pelicci, P. G., Sclessinger, J., and Pawson, T. (1992). Association of the Shc and GrbWSem5 SH2-containing proteins is implicated in activation of the Ras pathway by tyrosine kinases. Nature 360,689-692. Rozakis Adcock, M., Fernley, R., Wade, J., Pawson, T., and Bowtell, D. (1993). The SH2 and SH3 domains of mammalian Grb2 couple the EGF receptor to the Ras activator mSosl. Nature 363, 83-85. Rozdzial, M. M., Malissen, B., and Finkel, T. H. (1995). Tyrosine-phosphorylated T cell receptor 5 chain associates with the actin cytoskeleton upon activation of mature T lymphocytes. Immunity 3, 623-633.

94

AMNON ALTMAN AND MARCEL DECKERT

Rudd, C. E. (1996).Upstream-downstream: CD28 cosignaling pathways and T cell function. Immunity 4,527-534. Rudel, T., and Bokoch, G. M. (1997). Membrane and morphological changes in apoptotic cells regulated by caspase-mediated activation of PAK2. Science 276, 1571-1574. Rudel, T., Zenke, F. T., Chuang, T. H., and Bokoch, C. M. (1998). pel-activated kinase (PAK) is required for Fas-induced JNK activation in Jurkat cells.J. Immunol. 160,7-11. Ruiz-Ruiz, M. C., Izquierdo, M., de Murcia, C., and Lopez-Rivas, A. (1997). Activation of protein b a s e C attenuates early signals in Fas-mediated apoptosis. Eur. J. Immunol. 27, 1442-1450. Salmon, R. A., Foltz, I. N., Young, P. R., and Schrader, J. W. (1997). The p38 mitogenactivated protein kinase is activated by ligation of the T or B lymphocyte antigen receptors, Fas or CD40, but suppression of kinase activity does not inhibit apoptosis induced by antigen receptors. J. Immunol. 159, 5309-5317. Salojin, K., Zhang, J., Cameron, M., Gill, B., Arreaza, G., Ochi, A,, and Delovitch, T. L. (1997). Impaired plasma membrane targeting of Grb2-murine son of sevenless (mSOS) complex and differential activation of the Fyn-T cell receptor (TCR)-l-Cbl pathway mediate the T cell hyporesponsiveness in autoimmune nonobese diabetic mice. J. Enrp. Med. 186,887-897. Sanchez, I., Hughes, R. T., Mayer, B. J,, Yee, K., Woodgett, J. R., Avruch, J., Kyriakis, J. M., and Zon, L. I. (1994). Role of SAPWERK kinase-1 in the stress-activated pathway regulating transcription factor c-Jun. Nature 372, 794-798. Sarmay, G., Koncz, G., and Gergely, J. (1996). Human type I1 Fcy receptors inhibit B cell activation by interacting with the p2lrW-dependentpathway. J. Biol. Chem. 271,3049930504. Sato, S., Jansen, P. J., and Tedder, T. F. (1997). CD19 and CD22 expression reciprocally regulates tyrosine phosphorylation of Vav protein during B lymphocyte signaling.J. Immunol. 159,3278-3287. Satoh, T., Nakafuku, M., Miyajima, A., and Kaziro, Y. (1991). Involvement of ras p21 protein in signal-transduction pathways from interleukin 2, interleukin 3, and granulocyte/ macrophage colony-stimulating factor, but not from interleukin 4. Proc. Nutl. Acud. Sci. U.S.A. 88, 3314-3318. Saxton, T. M., van Oostveen, I., Bowtell, D., Aebersold, R., and Gold, M. R. (1994). B cell antigen receptor cross-linkinginduces phosphorylation of the p21" oncoprotein activators SHC and mSOSl as well as assembly of complexes containing SHC, CRB-2, mSOS1, and a 145-kDa tyrosine-phosphorylated protein. J. Immunol. 153, 623-636. Scharenberg, A. M., and Kinet, J.-P. (1996). The emerging field of receptor-mediated inhibitory signaling: SHP or SHIP? CelZ 87, 961-964. Schlaepfer, D. D., Hanks, S. K., Hunter, T., and Van der Geer, P. (1994). Integrin-mediated signal transduction linked to Ras pathway via GRB2 binding to focal adhesion kinase. Nature 372, 786-791. Schmid Antomarchi, H., Benkirane, M., Breittmayer, V., Husson, H., Ticchioni, M., Devaux, C., and Rossi, B. (1996). HIV induces activation of phosphatidylinositol 4-kinase and mitogen-activated protein b a s e by interacting with T cell CD4 surface molecules. Eur. J. Immunol. 26, 717-720. Schneider, H., Cai, Y. C., Cefai, D., Raab, M., and Rudd, C. E. (1995a). Mechanisms of CD28 signahg. Res. Immunol. 146, 149-154. Schneider, H., Cai, Y. C., Prasad, K. V., Shoelson, S. E., and Rudd, C. E. (1995b). T cell antigen CD28 binds to the GRB-USOS complex, regulators of p21". Eur. J. Immunol. 25, 1044-1050. Schwartz, R. H. (1997). Clonal anergy. Curt-. Opifi. Immunol. 9, 351-357.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

95

Schwartz,M. A., Schder, M. D., and Ginsberg, M. H. (1995).Integrins: Emergingparadigms of signal transduction. Annu. Rev. Cell Deu. Biol. 11,549-599. Sen, J., KapeIler, R., Fragoso, R., Sen, R., Zon, L. I., and Burakoff, S. J. (1996). Intrathymic signals in thymocytes are mediated by p38 mitogen-activated protein kinase. I. Immunol. 156, 4535-4538. Settleman, J., Albright, C. F., Foster, L. C., and Weinberg, R. A. (1992a). Association between GTPase activators for Rho and Ras families. Nature 359, 153-154. Settleman, J., Narasimhan, V., Foster, L. C., and Weinberg, R. A. (1992b). MoIecular cloning of cDNAs encoding the GAP-associated protein p190: Implications for a signaling pathway from ras to the nucleus. Cell 69, 539-549. Sharp, L. L., Schwarz, D. A,, Bott, C. M., Marshall, C. J., and Hedrick, S. M. (1997). The influence of the MAPK pathway on T cell lineage commitment. Immunity 7 , 609-618. She, H.-Y., Rockow, S., Tang, J., Nishimura, R., Skolnik, E. Y., Chen, M., Margolis, B., and Li, W. (1997). Wiskott-Aldricb syndrome protein is associated with the adadpetr protein Grb2 and the epidermal growth factor receptor in living cells. Mol. Bwl. Cell 8, 1709-1721. Shearer, G. M., and Clerici, M. (1992). How human immunodeficiency virus ravages the immune system. C u m Opin. Immunol. 4, 463-465. Siegel, J. N., Klausner, R. D., Rapp, U. R., and Samelson, L. E. (1990). T cell antigen receptor engagement stimulates c-raf phosphorylation and induces c-raf-associated kinase activity via a protein kinase C-dependent pathway. J. Biol. Chem. 265, 18472-18480. Siegel, J. N., June, C. H., Yamada, H., Rapp, U. R., and Samelson, L. E. (1993). Rapid activation of C-Raf-1 after stimulation of the T-cell receptor or the muscarinic receptor type 1 in resting T cells. I. Immunol. 151,4116-4127. Sieh, M., Batzer, A., Schlessinger, J., and Weiss, A. (1994). GRBB and phospholipase Cgamma 1associatewith a 36- to 38-kilodaltonphosphotyrosine protein after T-cell receptor stimulation. Mol. Cell. Biol. 14, 4435-4442. Sinclair, N. R., and Panoskaltsis, A. (1989). B cell regulation through Fc receptor-mediated signals. Contrib. Microbiol. Immunol. 11, 96-123. Sirinian, M. I., Marchetti, A., Di Rocco, G., Starace, G., Jucker, R., and Nasi, S. (1993). Ras oncogene transformation of human B Iymphoblasts is associated with lymphocyte activation and with a block of differentiation. Oncogene 8, 157-163. Skolnik, E. Y.,Batzer, A,, Li, N., Lee, C. H., Lowenstein, E., Mohammadi, M., Margolis, B., and Schlessinger, J. (1993a). The function of CRB2 in linking the insulin receptor to Ras signaling pathways. Science 260, 1953-1955. Skolnik, E. Y., Lee, C. H., Batzer, A., Vicentini, L. M., Zhou, M., Daly, R., Myers, M. J., Jr., Backer, J. M., Ullrich, A,, White, M. F., and Schlessinger, J. (1993b). The SHWSH3 domain-containing protein GRBB interacts with tyrosine-phosphorylated IRSl and Shc: Implications for insulin control of Ras signding. EMBO J. 12, 1929-1936. Sleckman, B. P., Peterson, A,, Jones, W. K., Foran, J. A,, Greenstein, J. L., Seed, B., and Burakoff, S. J. (1987). Expression and function of CD4 in a murine T-cell hybridoma. Nature 328, 351-353. Sloan-Lancaster, J., Shaw, A. S., Rothbard, J. B., and Allen, P. M. (1994). Partial T cell signaling: Altered phospho-{ and lack of Zap70 recruitment in APL-induced T cell anergy. Cell 79,913-922. Srnit, L., de-Vries-Smits, A. L., Bos, J. L., and Borst, J. (1994). B cell antigen receptor stimulation induces formation of a Shc-Grb2 complex containing multiple tyrosine-phosphorylated proteins. I. Biol. Chem. 269, 20209-20212. Smit, L., van der Horst, G., and Borst, J. (1996). Formation of ShdGrb2- and Crk adaptor complexes containing tyrosine phosphorylated Cbl upon stimulation of the B-cell antigen receptor. Oncogene 13,381-389.

96

AMNON ALTMAN A N D MARCEL DECKERT

Smith, B. L., Krushelnycky, B. W., Mochly Rosen, D., and Berg, P. (1996). The HIV nef protein associates with protein kinase C theta. J. Biol. Chem. 271, 16753-16757. Smith, J., Thompson, N., Thompson, J., Armstrong, J,, Hayes, B., Crofts, A,, Squire, J., Teahan, C., Upton, L., and Solari, R. (1997). Rat basophilic leukemia (RBL) cells overexpressing Rab3a have a reversible block in antigen-stimulated exocytosis. Biochem. J. 323, 321-328. Song, J. S., Gomez, J., Stancato, L. F., and Rivera, J. (1996). Association of a p95 Vavcontaining signaling complex with the FcepsilonRI gamma chain in the RBL-2H3 mast cell line. Evidence for a constitutive in vivo association of Vav with Grb2, Raf-1, and ERK2 in an active complex. I. Biol. Chem. 271,26962-26970. Sozeri, O., Vollmer, K., Liyanage, M., Frith, D., Kour, G., Mark, G. E. D., and Stabel, S. (1992). Activation of the c-Raf protein kinase by protein kinase C phosphorylation. Oncogene 7, 2259-2262. Spiegel, S., Foster, D., and Kolesnick, R. (1996). Signal transduction through lipid second messengers. Curr. +in. Cell Biol. 8, 159-167. Springer, T. (1990). Adhesion receptors of the immune system. Nature 346, 425-434. Springer, T. A. (1994).Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 76, 301-314. Stokoe, D., and McCormick, F. (1997). Activation of c-Raf-1 by Ras and Src through different mechanisms: activation in vivo and in vitro. EMBO J. 16, 2384-2396. Stowers, L., Yelon, D., Berg, L. J., and Chant, J. (1995). Regulation of the polarization of T cells toward antigen-presenting cells by Ras-related GTPase CDC42. Proc. Natl. Acad. Sci. U.S.A. 92,5027-5031. Su, B., and Karin, M. (1996). Mitogen-activated protein kinase cascades and regulation of gene expression. Curr. Opin. Immunol. 8, 402-411. Su, B., Jacinto, E., Hibi, M., Kallunki, T.,Karin, M., and Ben Neriah, Y. (1994). JNK is involved in signal integration during costimulation of T lymphocytes. CeZZ 77,727-736. Sundstedt, A., Sigvardsson, M., Leanderson, T., Hedund, G., Kalland, T., and Dohlsten, M. (1996).In vivo anergized CD4'T cells express perturbed AP-1 and NF-KBtranscription factors. Proc. Natl. Acad. Sci. U.S.A. 93, 979-984. Sutherland, C. L., Heath, A. W., Pelech, S. L., Young, P. R., and Gold, M. R. (1996). Differential activation of the ERK, JNK, and p38 mitogen-activated protein b a s e s by CD40 and the B cell antigen receptor. J. Immunol. 157, 3381-3390. Swan, K. A., Alberola Ila, J.. Gross, J. A., Appleby, M. W., Forbush, K. A., Thomas, J. F., and Perlmutter, R. M . (1995). Involvement of p2lras distinguishes positive and negative selection in thymocytes. EMBO 1. 14, 276-285. Swat, W., Fujikawa, K., Ganiatsas, S., Yang, D., Xavier, R. J., Harris, N. L., Davidson, L., Ferrini, R., Davis, R. J., Labow, M. A., Flavell, R. A., Zon, L. I., and At, F. W. (1998). SEKl/MKK4 is required for maintenance of normal peripheral lymphoid compartment but not for lymphocyte development. Immunity 8, 625-634. Swat, W., Shinkai, Y., Cheng, H. L., Davidson, L., and At, F. W. (1996). Activated Ras signals differentiation and expansion of CD4'8' thymocytes. Proc. Natl. Acad. Sci. U.S.A. 93,4683-4687. Symons, M. (1995). Rho family GTPases: The cytoskeleton and beyond. Trends Biochem. Sci. 21, 178-181. Symons, M., Derry, J. M. J., Karlak, B., Jiang, S., Lemahieu, V., McCormick, F., Francke, U., and Abo, A. (1996). Wiskott-Aldrich Syndrome protein, a novel effector for the GTPase CDC42Hs, is implicated in actin polymerization. Cell 84, 723-734. Takaki, S., Watts, J. D., Forbush, K. A., Nguyen, N. T., Hayashi, J., Alberola Ila, J., Aebersold, R., and Perlmutter, R. M. (1997). Characterization of Lnk. An adaptor protein expressed in lymphocytes.J. Biol. Chem. 272, 14562-14570.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

97

Tamma, S. M., Chirmule, N., McCloskey, T. W., Oyaizu, N., Kalyanaraman, V. S., and Pahwa, S. (1997a). Signals transduced through the CD4 molecule interfere with TCW CD3-mediated ras activation leading to T cell anergy/apoptosis. Clin. lmmunol. lmmunopathol. 85, 195-201. Tamma, S. M., Chirmule, N., Yagura, H., Oyaizu, N., Kalyanaraman, V., and Pahwa, S. [ 199%). CD4 cross-linking (CD4XL) induces RAS activation and tumor necrosis factoralpha secretion in CD4+ T cells. Blood 90, 1588-1593. Tapon, N., and Hall, A. (1997). Rho, Rac and Cdc42 GTPases regulate the organization of the actin cytoskeleton. Curr. Opin. Cell Biol. 9, 86-92. Tarakhovsky, A., Turner, M., Schaal, S., Mee, P. J., Duddy, L. P., Rajewsky, K., and Tybulewicz, L. J. (1995). Defective antigen receptor-mediated proliferation of B and T cells in the absence of Vav. Nature 374, 467-470. Taylor, S. J., and Shalloway, D. (1996). Cell cycle-dependent activation of Ras. Cum. Biol. 6, 1621-1627. Taylor, N., Jahn, T., Smith, S., Lamkin, T., Uribe, L., Liu, Y., Durden, D. L., and Weinberg, K. (1997). Differential activation of the tyrosine kinases ZAP-70 and Syk after FcyRI stimulation. Blood 89, 388-396. Taylor Fishwick, D. A,, and Siegel, J. N. (1995). Raf-1 provides a dominant but not exclusive signal for the induction of CD69 expression on T cells. Eur. 1.Immunol. 25,3215-3221. Teramoto, H., Coso, 0. A., Miyata, H., Igishi, T., Mild, T., and Gutkind, J. S. (1996). Signaling from the small GTP-binding proteins Racl and Cdc42 to the c-Jun N-terminal kinasekress-activated protein kinase pathway. A role for mixed lineage kinase 3/proteintyrosine kinase 1, a novel member of the mixed lineage kinase family. J. Bid. Chem. 271, 27225-27228. Teramoto, H., Salem, P., Robbins, K. C., Bustelo, X. R., and Gutkind, J. S. (1997). Tyrosine phosphorylation of the vav proto-oncogene product links FcsRI to the Racl-JNK pathway. 1.Biol. Chem. 272, 10751-10755. Thompson, C. B., and Allison, J. P. (1997). The emerging role of CTLA-4 as an immune attenuator. Immunity 7 , 445-450. Thuillier, L., Hivroz, C., Fagard, R., Andreoh, C., and Mangeat, P. (1994). Ligation of CD4 surface antigen induces rapid tyrosine phosphorylation of the cytoskeletal protein ezrin. Cell lmmunol. 156, 322-331. Timmerman, L. A., Healy, J. I., Ho, S. N., Chen, L., Goodnow, C. C., and Crabtree, G. R. (1997). Redundant expression but selective utilization of nuclear factor of activated T cells family members. 1.lmmunol. 159, 2735-2740. Tominaga, T., Sugie, K., Hirata, M., Morii, N., Fukata, J., Uchida, A,, Imura, H., and Narumiya, S. (1993). Inhibition of PMA-induced, LFA-l-dependent lymphocyte aggregation by ADP ribosylation of the small molecular weight GTP binding protein, Rho. J Cell Biol. 120, 1529-1537. Tordai, A., Franklin, R. A., Patel, H., Gardner, A. M., Johnson, G. L., and Gelfand, E. W. (1994). Cross-linking of surface IgM stimulates the Ras/Raf-UMEWMAPK cascade in human B lymphocytes. 1.Biol. Chem. 269, 7538-7543. Toyoshima, G., Moriguchi, T., and Nishida, E. (1997). Fas induces cytoplasmic apoptotic responses and activation of the MKK7-JNWSAPKand MKK6-p38 pathways independent of CPP32-like proteases. 1.Cell B i d . 139, 1005-1015. Treisman, R. (1994).Ternary complex factors: Growth factor regulated transcriptional activators. Curr. Opin. Genet. Dev. 4, 96-101. Treisman, R. (1996). Regulation of transcription by MAP kinase cascades. Cum. @in. Cell Biol. 8, 205-215.

98

AMNON ALTMAN AND MARCEL DECKERT

Tridandapani, S., Chacko, G. W., Van Brocklyn, J. R., and Coggeshall, K. M. (1997a). Negative signalinginB cells causes reduced Ras activityby reducing Shc-Grb2interactions. J. lmmunol. 158, 1125-1132. Tridandapani, S., Kelley, T., Cooney, D., Pradhan, M., and Coggeshall, K. M. (1997b). Negative signaling in B cells: SHIP Grbs Shc. Immunol. Today 18,424-427. Truitt, K. E., Hicks, C. M., and Imboden, J. B. (1994). Stimulation of CD28 triggers an association between CD28 and phosphatidylinositol3-kinase in Jurkat T cells. J. Exp. Med. 179, 1071-1076. Truneh, A., Albert, F., Golstein, P., and Schmitt-Verhulst, A.-M. (1985). Early steps of lymphocyte activation bypassed by synergy between calcium ionophores and phorbol esters. Nature 313, 318-320. Tsukita, S., and Yonemura, S. (1997).ERM (ezrin/radixin/moesin)family: From cytoskeleton to signal transduction. CUT. +in, CeZE Blol. 9, 70-75. Tuosto, L., Michel, F., and Acuto, 0. (1996).p95”” associates with tyrosine-phosphorylated SLP-76 in antigen-stimulated T cells. J. Exp. Med. 184, 1161-1166. Turner, H., and Cantrell, D. A. (1997). Distinct Ras effector pathways are involved in FcsRl regulation of the transcriptional activity of Elk-1 and NFAT in mast cells. J. Exp. Med. 185,43-53. Turner, B., Rapp, U., App, H., Greene, M., Dobashi, K., and Reed, J. (1991). Interleukin 2 induces tyrosine phosphoylation and activation of p72-74 Raf-1 kinase in a T-cell line. Proc. Natl. Acad. Sci. U.S.A. 88, 1227-1231. Turner, B. C., Tanks, N. K., Rapp, U. R., and Reed, J. C. (1993). Interleukin 2 regulates Raf-1 kinase activity through a tyrosine phosphorylation-dependent mechanism in a Tcell line. Proc. Natl. Acad. Sci. U.S.A. 90, 5544-5548. Turner, H., Reif, K., Rivera, J., and Cantrell, D. A. (1995).Regulation of the adapter molecule Grb2 by the FcsRl in the mast cell line RBL-2H3. J. Biol. Chem. 270, 9500-9506. Turner, M., Mee, P. J., Walters, A. E., Quinn, M. E., Mellor, A. L., Zamoyska, R., and Tybulewicz, V. L. J. (1997).A role for the Rho-family GTP exchange factor Vav in positive and negative selection of thymocytes. Immunity 7,451-460. Ueno, H., Sasaki, K., Miyagawa, K., Honda, H., Mitani, K., Yazaki, Y., and Hirai, H. (1997). Antisense repression of proto-oncogene c-Cbl enhances activation of the JAK-STAT pathway but not the ras pathway in epidermal growth factor receptor signaling. J. Biol. Chem. 272, 8739-8743. Valiante, N. M., Phillips, J. H., Lanier, L. L., and Parham, P. (1996). Killer cell inhibitory receptor recognition of human leukocyteantigen (HLA)class I blocks formationof a pp36/ PLC-y signaling complex in human natural killer (NK) cells.]. Elrp. Med. 184,2243-2250. Valitutti, S., Dessing, M., Aktories, K., Gallati, H., and Lanzavecchia, A. (1995a).Sustained signaling leading to T cell activation results from prolonged T cell receptor occupancy. Role of T cell actin cytoskeleton.J. Exp. Med. 181, 577-584. VaIitutti, S., Miiller, S., Cella, M., Padovan, E., and Lanzavecchia, A. (1995b).Serial engagement of many T-cell receptors by a few peptide-MHC complexes. Nature 375,148-151. van Aelst, L., and D’Souza-Schorey, C. (1997).Rho GTPases and signaling networks. Genes Den 11,2295-2322. Van Seventer, G. A., Shimizu,Y., and Shaw, S. (1991).Roles of multiple accessory molecules in T-cell activation. Cum @in. lmmunol. 3, 294-303. Veillette, A,, Bookman, M. A., Horak, E. M., Samelson, L. E., and Bolen, J. B. (1989). Signal transduction through the CD4 receptor involves the activation of the internal membrane tyrosine protein kinase ~ ~ 5 6 ’Nature ” ~ . 338, 257-259. Van Aelst. L., Joneson,T., and Bai Sagi, D. (1996).Identification of a novel Rael-interacting protein involved in membrane r u f h g . Embo J. 15, 3778-3786.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

99

Venkataraman, L., Francis, D. A., Wang, Z., Liu, J., Rothstein, T. L., and sen, R. (1994). Cyclosporin A-sensitive induction of NF-AT in murine B cells. Immunity 1, 189-196. Verschueren, H., De Baetselier, P., De Braekeleer, J., Dewit, J., Aktories, K., and Just, I. (1997). ADP-ribosylation of Rho-proteins with botulinum C3 exoenzyme inhibits invasion and shape changes of T-lymphoma cells. Eur. J. Cell Biol. 73, 182-187. Villalba-Gonzales, M., and Altman, A. (1998). Submitted for publication. Viola, A., Salio, M., Tuosto, L., Linkert, S., Acuto, O., and Lanzavecchia,A. (1997).Quantitative contribution of CD4 and CD8 to T cell antigen receptor serial triggering. J. Exp. Med. 186, 1775-1779. Vita, F., Soranzo, M. R., Borelli, V., Bertoncin, P., and Zabucchi, G. (1996). Subcellular localization of the small GTPase Rab5a in resting and stimulated human neutrophils. Exp. Cell Res. 227,367-373. Vojtek, A. B., and Cooper, J. A. (1995). Rho family members: Activators of MAP kinase cascades. Cell 82, 527-529. Wakelam, M. J., Martin, A., Hodgldn, M. N., Brown, F., Pettitt, T. R., Cross, M. J., De Takats, P. G., and Reynolds, J. L. (1997). Role and regulation of phospholipase D activity in normal and cancer cells. Adu. Enzyme Regul. 37, 29-34. Wange, R. L., and Samelson, L. E. (1996). Complex complexes: Signaling at the TCR. Immunity 5, 197-205. Ward, S. G. (1996). CD28: A signalling perspective. Biochem. J. 318, 361-377. Wardenburg, J. B., Fu, C., Jackman, J. K., Flotow, H., Wilkinson, S. E., Williams, D. H., Johnson, R., Kong, G., Chan, A. C., and Findell, P. R. (1996). Phosphorylation of SLP76 by the ZAP-70 protein-tyrosine kinase is required for T-cell receptor function. J. Biol. Chem. 271, 19641-19644. Watanabe, G., Salto, Y., Madaule, P., Ishizaki, T., Fujisawa, K., Morii, N., Mukai, H., Ono, Y., Kakizuka, A., and Narumiya, S. (1996). Protein kinase N (PKN) and PKN-related protein rhophilin as targets of small GTPases Rho. Science 271, 645-648. Watanabe, N., Madaule, P., Reid, T., Ishizaki, T., Watanabe, G., Kakizuka, A,, Saito, Y., Nakao, K., Jockusch, B. M., and Narumiya, S. (1997). p140mDh,a mammalian homolog of Drosaphila diaphanous, is a target protein for Rho small GTPase and is a ligand for profilin. EMBO J. 16, 3044-3056. Weiss, A., and Littman, D. R. (1994). Signal transduction by lymphocyte antigen receptors. Cell 76, 236-274. Weiss, D. L., Hural, J., Tara, D., Timmerman, L. A., Henkel, G., and Brown, M. A. (1996). Nuclear factor of activated T cells is associated with a mast cell interleukin 4 transcription complex. Mol. Cell. Biol. 16, 228-235. Weng, W. K., Jarvis, L., and LeBien, T. W. (1994). Signaling through CD19 activates Vadmitogen-activated protein kinase pathway and induces formation of a CD19Navl phosphatidylinositol 3-kinase complex in human B cell precursors. J. Biol. Chem. 269, 32514-32521. Werge, T. M., Baldari, C. T., and Telford, J. L. (1994). Intracellular single chain Fv antibody inhibits Ras activity in T-cell antigen receptor stimulated Jurkat cells. FEBS Lett. 351,393-396. Werlen, G., Jacinto, E., and Karin, M. (1998). Calcineurin preferentially synergizes with PKC-0 to activate JNK and IL-2 promoter in T lymphocytes. EMBOJ. 117,3101-3111. Westwick, J. K., Lambert, Q. T., Clark, G . J., Symons, M., Van Aelst, L., Pestell, R. G., and Der, C. J. (1997). Rac regulation of transformation, gene expression, and actin organization by multiple, PAK-independent pathways. Mol. Cell. B i d . 17, 1324-35. White, M. A,, Nicolette, C., Minden, A,, Polverino, A., Van, A. L., Karin, M., and Wigler, M. H. (1995). Multiple Ras functions can contribute to mammalian cell transformation. Cell 80, 533-541.

100

AMNON ALTMAN AND MARCEL DECKERT

Whitehurst, C. E., and Geppert, T. D. (1996). MEKl and the extracellular signal-regulated kinases are required for the stimulation of the IL-2 gene transcription in T cells. /. Immunol. 156,1020-1029. Williams, D. H., Woodrow, M., Cantrell, D. A., and Murray, E. J. (1995). Protein kinase C is not a downstream effector of p21W in activated T cells. Eur. 1.Immunol. 25,42-47. Williams, S., Couture, C., Gilman, J., Jascur, T., Deckert, M., Altman, A., and Mustelin, T. (1997). Reconstitution of T cell antigen receptor-induced Erk2 kinase activation in Lck-negative JCaMl cells by Syk. Eur. 1.Biochem. 245,84-90. Wilson, D. J., Fortner, K. A., Lynch, D. H., Mattingly, R. R., Macara, I. G., Posada, J. A., and Budd, R. C. (1996). JNK, but not MAPK, activation is associated with Fas-mediated apoptosis in human T cells. Eur. /. Immunol. 26, 989-994. Woodrow, M., Clipstone, N. A., and Cantrell, D. (1993a). p21W and calcineurin synergize to regulate the nuclear factor of activated T cells. I. Exp. Med. 178, 1517-1522. Woodrow, M. A., Rayter, S., Downward, J., and Cantrell, D. A. (199313).p21" function is important for T cell antigen receptor and protein kinase C regulation of nuclear factor of activated T cells. 1.Immunol. 150, 3853-3861. Worthen, G. S., Avdi, N., Buhl, A. M., Suzuki, N., and Johnson, G. L. (1994). FMLP activates Ras and Raf in human neutrophils: Potential role in activation of MAP kinase. J Clin. Invest. 94, 815-823. Wotton, D., Ways, D. K., Parker, P. J., and Owen, M. J. (1993). Activity of both Raf and Ras is necessary for activation of transcription of the human T cell receptor p gene by protein kinase C, Ras plays multiple roles. 1.Biol. Chem. 268, 17975-17982. Wu, J., Katzav, S., and Weiss, A. (1995). A functional T-cell receptor signaling pathway is required for p95"" activity. Mol. Cell. Biol. 15, 4337-4346. Wu, J., Motto, D. G., Koretzky, G. A., and Weiss, A. (1996). Vav and SLP-76 interact and functionally cooperate in IL-2 gene activation. Immunity 4, 593-602. Xu, Y., Casey, G., and Mills, G. B. (1995). Effects of lysophospholipids on signaling in the human Jurkat T cell line. 1. Cell Physiol. 163, 441-450. Yamada, K. M., and Geiger, B. (1997). Molecular interactions in cell adhesion complexes. Cum. Opin. Cell Biol. 9, 76-85. Yamanashi, Y., and Baltimore, D. (1997). Identification of the Abl- and rasGAP-associated 62 kDa protein as a docking protein, Dok. Cell 88, 205-211. Yaseen, N. R., Maizel, A. L., Wang, F., and Sharma, S. (1993). Comparative analysis of NFAT (nuclear factor of activated T cells) complex in human T and B lymphocytes. 1.Biol. Chern. 268, 14285-14293. Yaseen, N. R., Park, J., Kerppola, T., Curran, T., and Sharma, S. (1994). A central role for Fos in human B- and T-cell NFAT (nuclear factor of activated T cells): An acidic region is required for in vitro assembly. Mol. Cell. Biol. 14, 6886-6895. Ye, Z.-S., and Baltimore, D. (1994). Binding of Vav to Grb2 through dimerization of Src homology 3 domains. Proc. Nutl. Acad. Sci. U.S.A.91, 12629-12633. Yin, L., Schwartzberg, P., Scharton Kersten, T. M., Staudt, L., and Lenardo, M. (1997). Immune responses in mice deficient in Ly-GDI, a lymphoid-specific regulator of Rho GTPases. Mol. Immunol. 34, 481-491. Yokoyama, W. M. (1995). Natural killer cell receptors. Cum. Opin. Immunol. 7,110-120. Yoon, C. H., Lee, J., Jongeward, G. D., and Sternberg, P. W. (1995). Similarity of sZi-1, a regulator of vulval development in C. elegans, to the mammalian proto-oncogene c-cbl. Science 269, 1102-1105. Zachary, I., and Rozengurt, E. (1992). Focal adhesion kinase (p125FAK):A point of convergence in the action of neuropeptides, integrins, and oncogenes. Cell 71, 891-894.

GTPases IN IMMUNE RECOGNITION RECEPTOR SIGNALING

101

Zhang, X., Settlemen, J., Kyriakis, J. M., Takeuchi-Suzuki, E., Elledge, J., Marshall, M. S., Bruder, J. T., Rapp, U. R., and Avruch, J. (1993). Normal and oncogenic p21" proteins bind to the amino-terminal regulatory domain of c-Raf-1. Nature 364, 308-313. Zhang, R., Alt, F. W., Davidson, L., Orkin, S. H., and Swat, W. (1995). Defective signalling through the T- and B-cell antigen receptors in lymphoid cells Iacldng the uau protooncogene. Nature 374,470-473. Zhang, S., Han, J.. Sells, M. A., Chernoff, J., Knaus, U. G., Ulevitch, R. J., and Bokoch, G. M. (1995). Rho family GTPases regulate p38 mitogen-activated protein b a s e through the downstream mediator Pakl. J. Biol. Chem. 270, 23934-23936. Zhang, C., Baumgartner, R. A., Yamada, K., and Beaven, M. A. (1997). Mitogen-activated protein (MAP) kinase regulates production of tumor necrosis factor-alpha and release of arachidonic acid in mast cells. Indications of communication between p38 and p42 MAP kinases. J. Biol. Chem. 272, 13397-13402. Zhang, W., Sloan-Lancaster, J., Kitchen, J., Trible, R. P., and Samelson, L. E. (1998). LAT: The ZAP-70 tyrosine b a s e substrate that links T cell receptor to cellular activation. CeZZ 92, 83-92. Zheng, L., Sjolander, A., Eckerdal, J., and Anderson, T. (1996).Antibody-induced engagement of & integrins on adherent human neutrophils triggers activation of p21" through tyrosine phosphorylation of the protooncogene product Vav. Proc. Natl. Acad. Sci. U.S.A. 93,8431-8436. Zmuidzinas, A., Fischer, K. D., Lira, S. A,, Forrester, S., Bryant, S., Bernstein, A., and Barbacid, M. (1995). The vau proto-oncogene is required early in embrygenesis but not for hematopoietic development in vitro. EMBO J. 14, 1-11. This article was accepted for publication on July 10, 1998.

This Page Intentionally Left Blank

ADVANCES IN IMMUNOLOGY, VOL. 72

Function of the CD3 Subunits of the Pre-TCR and TCR Complexes during T Cell Development BERNARD MAUSSEN, LAURENCE ARDOUIN, SHIH-YAO UN, ANNE GILLET, AND MARIE MAUSSEN Cmhs d'lmmunobgie INSERM-CNRS de Marseille-fuminy, 13288 Marseilki Cedax 9, Fmnce

1. Introduction

The specific recognition of antigen by T cells and its ensuing transduction into intracellular signals are accomplished by a multisubunit transmembrane complex known as the T cell antigen receptor (TCR)-CD3 complex. On the basis of the structure of the TCR antigen-binding subunits found in the TCR-CD3 complexes, mature T cells can be divided into two subsets. In adult mice, most T cells express TCR heterodimers consisting of ct and @ chains, whereas a minor population expresses an alternative TCR isoform consisting of y and S chains. The TCR a, /?, y, and 6 chains each comprise an amino-terminal, clonally variable (V) region and a carboxy-terminalconstant (C)region. Peptide loops homologous to immunoglobulin ( Ig) complementarity-determining regions (CDRs) protrude at the membrane-distal end of TCR V domains, where they constitute the binding site for antigens. Transport of the TCR heterodimers to the cell surface is dependent on their prior assembly with the invariant CD3 subunits (CD3-y, CD3-6, C D ~ - ECD3-6, , and CD3-7). As shown in Fig. 1, the CD3-y, -6, and -E subunits are expressed as noncovalently associated C D 3 - 6 ~and CD3-6.5 pairs, and the CD3-land CD3-7 polypeptides can combine to form multiple disulfide-linked homodimers and heterodimers. The stoichiometry of the TCR-ap (or TCR-yS), C D S -~ ECD3-6&, , CD3-5,, and C D 3 4 7 pairs present within a given complex is not known. Aside from their role in TCR cell surface expression, the various CD3 subunits are also responsible for coupling the antigen-binding TCR-ap or -76 heterodimers to intracellular signaling pathways. In contrast to the situation observed with numerous growth factor receptors (such as those for insulin or platelet-derived growth factor), none of the CD3 subunits possesses a cytoplasmic domain endowed with recognizable enzymatic activity. However, each CD3 subunit contains one or several copies of a conserved sequence that is exposed to the cytoplasm and referred to as an immunoreceptor tyrosine-based activation motif (ITAM). ITAMs are also found in the transducing subunits of the antigen receptor of B lymphocytes (BCR), the receptors for the Fc domain 103

Copynght 0 1999 by Academic Press All nghts of reproduchon m any form resewed 0065-277Wil $30 00

104

BERNARD MALISSEN et al.

a

b

TCR

P

TCR

TCR

P

a

FIG.1. Putative subunit composition of the pre-TCR (a) and TCR (b) complexes that are sequentially expressed on developing T cells. The pre-TCR-a chain (pTa) and TCRfi polypeptides are expressed within the pre-TCR-CD3 complex as &sulfide-linked heterodirners and possess short cytoplasmic tails (4 to 30 residues). The pTa transmembrane regions contains polar residues (D, R, and K in the single-letter amino acid code), two of

TCR COMPLEX CD3 SUBUNIT FUNCTION

105

of IgE ( FcERI) and IgG ( FcyRIIIA), and the activating natural killer (NK) cell receptor. The analysis of TCR-CD3 complexes devoid of functional CD3-5 subunits has indicated that wild-type TCR-CD3 complexes are composed of at least two parallel signal-transducing modules made of the CD3-y~/6c and CD3& subunits, respectively (Wegener et al., 1992). These results, together with studies of chimeric receptors composed of inert transmembrane and extracellular domains and of CD3 cytoplasmic domains, have led to the view that each of the CD3 ITAMs can potentially act as an autonomous transducer once expressed in the context of a whole TCR complex (Irving and Weiss, 1991; Romeo and Seed, 1991; Letourneur and Klausner, 1992). The modular architecture of the TCR-CD3 complex (Fig. 1) favors the occurrence of combinatorial isoforms consisting of distinct polypeptide pairs. For instance, in thymus-independent intestinal intraepithelial T lymphocytes, homodimers consisting of the y chain of the FcERI can be incorporated into TCR-CD3 complexes in lieu of CD3-5 homodimers (Liu et al., 1993; Malissen et al., 1993). Protein tyrosine kinases (PTKs) belonging to the Src family (e.g., Lck, Fyn) phosphorylate conserved tyrosine residues present within the ITAMs. Once phosphorylated, the ITAMs act as high-affinity docking sites for the SH2 (src-homology 2) domains found in certain intracellular adaptor and effector molecules, among which are PTKs belonging to the SykEAP-70 family (e.g.,Syk, ZAP-70). The relocalization and/or clustering of SywzAP70-family PTKs into the receptor complex promote their activation and contribute to the successful progression of the activation program. Recruitment and activation of protein tyrosine phosphatases (PTPs) are likely to occur at later timepoints and to be responsible for the termination of signal transduction. Therefore, the ligand-activated switch operated by the TCR-CD3 complex appears composed, minimally, of a ligand-binding unit which (R and K) can be rigorously aligned with the arginine and lysine found in the transmembrane segment of the TCR-(w and TCR-6 chains. A VpreT domain may exist but has not yet been found. The immunoreceptor tyrosine-based activation motifs ( ITAMs) found in each of the CD3 subunits are shown as cylinders containing tyrosine-based docking motifs (YLYL). The cytoplasmic tails of the CD3-&, CD3--y, and CD3-S subunits each contain a single ITAM. The CD3-5 polypeptide displays three concatenated copies of the ITAM. The noncovalent nature of the interactions keeping these various polypeptide pairs together at the cell surface permits, as documented in at least two instances (Ono et al., 1995; Kishimotoet at., 1995),the rapid turnover of a single pair of components independentIy of the rest of the complex. Putative N-linked carbohydrate sites are indicated by black dots. The circles depict sequence segments that either have been shown to fold as a C or V immunoglobulin domain or are predicted to do so (S-S, disulfide bond). As discussed in Section 111,A,2, the c D 3 - 6 ~pair constitutes a dispensable component of the pre-TCR.

106

BERNARD MALISSEN et al.

(the TCR-a/3 or -yS heterodimers), a set of SH2-docking sites (the CD3 ITAMs), a few members of the Src and SywzAP-70 families of PTKs, and at least one PTP. Gene targeting experiments aimed at understanding the function of the TCR complexes and of their proximal effectors are the focus of this review. The current approaches used for generating mice that carry intended mutations involve the introduction of null mutations into embryonic stem (ES) cells, from which homozygous mutant mice can be derived. Because the null mutations are carried in the germ line, they manifest their effects throughout ontogeny and sometimes result in embryonic lethality or prevent the development of a given cell lineage. As discussed below, the constitutive inactivation of some of the genes coding for the TCR-CD3 complex has resulted in an unexpected block of early T cell development and revealed the existence of a novel TCR isoform that is specifically expressed during early T cell development and denoted as the pre-TCR. Because these constitutive mutations affect the processes that govern T cell development, their possible late impact on mature peripheral T cells cannot be studied. [Note that this limitation might be solved by the use of conditional gene inactivation techniques (Kuhn et al., 1995).]Therefore, the experimental results discussed in this review mostly bear on the role played by the pre-TCR and TCR complexes in the developmental control points through which T cell precursors have to pass to differentiate successfully. II. Mouse cup T Cell Development

Most a/3 T cells develop in the thymus. The thymic stromal microenvironment comprising epithelial cells, mesenchymal cells, macrophages, and dendritic cells provides developing T cells with essential extracellular matrix components, cell surface ligands, and secreted cytokines. In adult mice, thymopoiesis is dependent on a low but continuous import of bone marrowderived precursor cells. Intrathymic T cell development proceeds through discrete stages defined on the basis of both the configuration of TCR gene loci and the expression of surface markers (Fig. 2). The most immature cells identified in the adult thymus are present in small numbers (representing 0.2% of all thymocytes) and express low levels of CD4. During the initial step of maturation, these “CD4’” precursors” stop expressing CD4 to become “triple-negative” (TN) (CD4-CDS-CD3-) cells. The differentiation of TN cells is marked by the transient expression of the IL-2 receptor a chain (CD25) and the gradual disappearance of both CD44 and CD117 (c-kit). Late TN cells can progress to the “double-positive” (DP) CD4’CDS’ stage via intermediates that express either CD8 (for the major-

TCR COMPLEX CD3 SUBUNIT FUNCTION

107

FIG.2. Intrathymic developmental stages followed by most T cells belonging to the a@ T cell lineage. In adult mice, bone marrow contains hematopoietic stem cells (HSC), which can give rise to all lymphoid populations through common lymphoid progenitors (CLP). The thymus is organized into subcapsular, cortical, and medullary areas. T cell precursors entering the thymus move in a centripetal fashion from the subcapsular region to the medullary area via the cortex. The two phases during which thymocytes undergo cell division are highlighted by curved arrows. TCR gene rearrangements occur during stages characterized by a low incidence of dividing cells. The bottom part of the figure [adapted from DiSanto and Rodewald (1998)] depicts the range of action of the antiapoptotic signals emanating from the expression of c-kit, IL-7R, pre-TCR, and TCR. The CD44""" CD25' and DP stages constitute crucial time windows during which the IL-7IUpre-TCR and preTCR/TCR pairs work in relays (see Section V). In the absence of either IL-7R or c-kit, a few early T cells still manage to live long enough to undergo TCR-P rearrangement and assemble functional pre-TCR. However, as exemplified in mice deficient in pre-TCR (see Section 111),cup TCR (see Section IV), yc + c-kit (Rodewald et al., 1997), or yc + pTa (DiSanto and Rodewald, 1998), any break in this coordinated sequence aborts development. TN, Triple-negative cells; ISP, immature single-positive cells; DP, double-positive cells; SP, single-positive cells.

ity of mouse strains) or CD4 in the absence of mature type cr/3 TCR complex and are therefore called "immature single-positive'' ( ISP) cells. A small percentage of the DP cells mature further into CD4+ CD8- or

108

BERNARD MALISSEN et al.

CD4-CD8+ ("mature SP") cells that correspond to the end products of the intrathymic ap T cell differentiation pathway and gradually exit from the thymus to reach peripheral lymphoid organs. It takes 11 to 15 days for the earliest CD4'"" precursor cells to develop into DP cells. This process is associated with two phases of cell expansion (Fig. 2). The first starts at the CD44TD25' T N stage and stops during or just after the onset of down-regulation of expression of CD44. Cell proliferation resumes concurrent with loss of expression of CD25 and continues up to the early DP stage, at which time it declines again (Shortman et al., 1990). During intrathymic differentiation, the genes encoding the TCR variable region are assembled by site-specific DNA recombination reactions. These cell-autonomous reactions, termed V( D)J rearrangements, result in the random recombination of V and J gene segments in TCR-a and TCR-7 chain, and of V, D, and J gene segments in TCR-P and TCR-6 genes. During V( D)J recombination, the accessible coding gene segment ends are generally subjected to various degrees of base deletion, addition, or both, before ligation. As a consequence, V(D)Jjoining reactions may result either in productive rearrangements that maintain an open reading frame throughout the gene, or in an out-of-frame nonfunctional gene. TCR-P gene rearrangements start around the transition to the CD44-""CD25+ TN stage, whereas the first TCR-a rearrangements are measurable close to the transition to the DP stage (Fig. 2). As outlined in Fig. 2, the CD4'"" precursors can also produce NK cells and thymic dendritic cells (DC). In contrast, cells belonging to the next developmental stage (CD44TD25') cannot generate NK cells but still give rise to thymic DCs. Commitment to the T cell lineage occurs at the next TN stage (CD44-"lWCD25'), coincident with the onset of TCR-P, TCR-y, and TCR-6 gene rearrangements. Finally, the irreversible decision to become an a/3 rather than a y8 T cell may not take place until both TCR-a loci have rearranged and consequently excised the TCR-6 loci. Developing T cells that do not rearrange their TCR genes, rearrange them nonproductively, or express TCR-aP combinations with inappropriate specificities are arrested at discrete developmental control points. Molecular sensors have evolved to couple the transition through these control points to the prior attainment of productive TCR gene rearrangements and to the specificity of the resulting TCR-a/3 heterodmers. One of these sensors, known as the pre-TCR complex, controls the transition from the TN to the DP stage and ensures that only cells with a productive TCR-P gene rearrangement undergo this transition (a step referred to as TCR-P selection), whereas another sensor is made of aP TCR and controls the transition from the DP to the SP stage (a step referred to as TCR-aP

TCR COMPLEX CD3 SUBUNIT FUNCTION

109

selection) (Jameson and Bevan, 1998). Several classes of mutations (Fig. 3 ) have helped to resolve the structure and function of these two sensors. 111. The Pre-TCR Sensor

A. COMPONENTS OF THE PRE-TCRSENSOR 1. TCR-/3

A host of experiments suggest that the TCR-P chain plays a unique role in early T cell development. For instance, mutations in genes encoding the lymphoid-specificcomponents required for effecting the double-strand breaks (DSB) associated with the initiation of V( D)J recombination [that is, the recombination-activating genes RAG-1 and RAG-2 (Mombaerts et al., 1992; Shinkai et al., 1992)l or in some general DSB repair factors [for example, the catalybc and Ku80 subunits of DNA-dependent protein kinase (DNA-PK) (Danska et al., 1996; Xu et al., 1996; Nussenzweig et al., 1996)] impair the completion of the V( D)J recombination reactions and prevent T cell development beyond the CD44-"""CD25+ stage. Considering that TCR-P gene rearrangements precede rearrangements at the TCR-a locus and are essentially completed prior to the transition to the CD44-"CD25TN stage, the early T cell development blockade observed in RAG- and DNA-PK-deficient mice suggested that the TCR-/3 chain plays a unique role in the transition beyond the CD44-"""CD25' stage, irrespective of its later function as a component of the a/3 TCR. Studies aiming at complementing DNA-PK- and RAG-deficient mice with productively rearranged TCR-a or TCR-P transgenes have revealed that the expression of TCRP chain (but not TCR-a chain) is sufficient to relieve the developmental blockade observed in these mutant mice and to restore maturation of their thymocytes up to the DP stage (Mombaerts et al., 1992; Shinkaiet al., 1993). (Note that further maturation into SP cells requires the complementation of RAG-deficient mice with matched pairs of TCR-a and TCR-/3 transgenes.) Consistent with the above observations, selective mutations preventing TCR-P gene rearrangements (Mombaerts et al., 1992; Bones et al., 1996; Bouvier et al., 1996) or proper intracellular distribution of TCR-P polypeptides (O'Shea et al., 1997) alIowed only limited development of DP cells. In marked contrast, null mutations of the TCR-a locus do not affect the progression to the DP stage and become manifest only at the DP to SP transition (Mombaerts et al., 1992; Philpott et al., 1992). Collectively, these data suggest that the achievement of productive TCR-/3 rearrangements constitutes a rate-limiting factor in the progression beyond the CD44-' '""CD25+ stage. Identification of a substitute for TCR-a, denoted as the pre-TCR-a (pTa) chain and expressed at the CD44-"OWCD25+stage (see

110

BERNARD MALISSEN et al.

EXPANSION

t

SEEDING

OAlVEN EXPANSION

1.

TCR B SELECTION

MATURATION EXPORT

TCRaBSELECTION

TCR COMPLEX CD3 SUBUNIT FUNCTION

111

below), has led Saint-Ruf et al. (1994) to infer that newly formed TCR-@ chains must associate with pTa to form a pre-TCR sensor that then triggers the selective proliferation of TCR@+TN cells and enables their differentiation into DP cells. 2. p T a Using an anti-TCR-/3 monoclonal antibody and a sensitive immunofluorescence assay with a detection limit of about 400 molecules/cell, L. Bruno and colleagues (personal communication) have been able to detect preTCR expression on the surface of -20% CD44-"CD25+ and of -60% CD44-""""CD25-TN cells. Due to the minute amounts of pre-TCR complexes that can be detected at the surface of TN thymocytes, most of the biochemical evidence supporting the existence of a pre-TCR sensor have been gained from studies of a transformed immature T cell line derived from a severe combined immunodeficient (scid) strain of mice deprived of the DNA-PK catalpc subunit. Following stable transfection with a productively rearranged TCR-/3 chain gene, these transformed T cells do express high levels of TCR-/3 polypeptides at their surface. The reason for such unique competence is unknown. In the absence of TCR-a chain, not expressed in this scid-derived cell line, the TCR-/3 chain was found covalently associated with the pTa subunit, a 33-kDa type I transmembrane glycoprotein containing a single extracellular Ig-like domain and a 30residue cytoplasmic tail (Fig. 1).This polypeptide is encoded by a nonrearranging gene that is expressed in TN cells and switched off in late DP thymocytes (Groettrup et al., 1993; Saint-Ruf et at., 1994; Bruno et at., 1995). The extracellular domain of pTa lacks a covalently associated Iglike V domain. By analogy with the pre-BCR sensor, the pre-TCR may contain an as yet unidentified VpreT component capable of pairing with the hydrophobic surface of the V/3 domain (Bentley et al., 1995).Although there is only 12% identity between the pTa and TCR-a C domains ( Fehling et al., 1995a; Del Port0 et al., 1995),the TCR-a C domain residues involved in polar interactions with the TCR-/3 C domain are all conserved in the pTa C domain. Therefore, it is likely that the mode of association of the ~~

____

~

FIG. 3. Extent of a/3 T cell development in the thymus of mice deficient for a few selected genes. The top portion of the figure [adapted from Shortman et al. (1990)] is intended to depict the development of a cohort of precursor cells after entering a wildtype adult thymus. Following thymus colonization (seeding), the progression of the cohort of precursors cells has been assumed to occur in a synchronous mode, although this is unlikely to happen in a real thymus. Also shown is the position of the two major developmental checkpoints (TCR-fi selection and TCR-aP selection) encountered by developing cr/3 T cells. In the lower portion, for each mouse mutant, the course of development is depicted by a line that narrows when development is impaired.

112

BERNARD MALISSEN et d.

pTa and TCR-P C domains is similar to that observed in the TCR CaCP module (J.-H. Wang et al., 1998). A few CD4’CDB- T cells expressing TCR-/3 at their surface can be found in the periphery of TCR-a-deficient mice (Mombaerts et al., 1994b; Viney et al., 1994). Such “p-only” T cells appear to use an isoform of pTa, referred to as pTab, in place of TCR-a (Barber et al., 1998).pTab results from alternative splicing of the pTa gene, lacks the Ig-like extracellular domain, but retains the cysteine residue that permits dimerization with TCR-6. pTab is coexpressed in the thymus witb the previously characterized “full-length” form of pTa, referred to as pTaa, whereas peripheral “P-only” T cells express pTab alone. The role of pTa during early T cell development was directly tested by generation of mice lacking a functional pTa gene (Fehling et al., 1995b; Xu et al., 1996). In contrast to the absolute developmental arrest observed in RAG-deficient mice, a few pTa-deficient thymocytes do progress to the DP and SP stages (in numbers that do not exceed 10% of that found in wild-type thymuses). As discussed below, this leakiness may be due in part to the few constitutive TCR-a rearrangements that may occasionally occur within the CD44-”OWCD25’TN cells before pre-TCR assembly. Expression of a transgenic pTa chain lacking its cytoplasmic portion can overcome the developmental defects associated with pTa deficiency (Fehling et al. 1997). These results argue against an essential signaling function of the pTa cytoplasmic segment and suggest that the CD3 subunits associated with the pTaTCR-P heterodimers (see below) probably account for most of the transducing capacity of the pre-TCR sensor (Punt et al., 1991; Groettrup et al., 1992; Van Oers et al., 1995). Consistent with this view, the 127 amino acids that constitute the cytoplasmic segment of the human pTa chain do not show any identity with the 30 amino acid residues that compose the mouse pTa cytoplasmic segment. 3. CD3-E Studies of mice with a mutation of the CD3-E gene that removes the exon coding for the extracellular domain, and referred to as CD3-eA5,have formally shown that some CD3 components are essential for the assembly or function of the pre-TCR sensor (Malissen et al., 1995). By abolishing the expression of intact CD3-e polypeptides and adventitiously reducing the transcription rate of the neighboring CD3-.)I and CD3-6 genes, the CD3-sA5mutation totally blocked the progression beyond the CD44-’ loWCD25+ stage. The TN subsets developing in CD3-eASA5 mice prior to the arrest point showed no gross distortion in their size and content of dividing cells (Tanaka et al., 1996).C D ~ - Ethymocytes ~ ~ ’ ~ ~are thus arrested at the very same developmental control point as thymocytes deficient in

TCR COMPLEX CD3 SUBUNIT FUNCTION

113

RAG. A second strain of CD3-&-deficient mice, termed CD~-E”/~’, has been generated by replacing the promoter and first two exons of CD3-& with a neomycin resistance cassette. T cell development in C D ~ - E ~ ’mice ”~ is blocked at the same stage as in C D ~ - E ”mice ~ ’ ~(N. ~ Wang et al., 1998). As previously documented for CD~-E’”’~mice, the neomycin cassette left within the targeted CD3-Egene caused a severe inhibition of the expression of the CD3-y and CD3-6 genes via a silencing mechanism yet to be determined. In contrast, the insertion of the same neomycin cassette within the contiguous CD3-y or CD3-6 genes was without effect on the transcriptional activity of CD3-S and CD3-.9 in the case of CD3-y-’- mice (Haks et al., 1998), and of CD3-y and CD3-E in the case of CD3-S-’- mice (Dave et al., 1997). Thus, neither the C D ~ - E ” nor ’ ~ ~the CD3-eAP’AP strain corresponds to a bona fide CD3-& deficiency. As a consequence, they should be more appropriately referred to as mutants of the whole CD3y 6 module ~ (Fig. 3). Considering that the C D ~ - E ” ”thymocytes ~ contain readily detectable Ievels of TCR-/3, pTa, and CD3-5 transcripts, it is likely that the simultaneous lack of CD3-ye and CD3-8.5 dimers prevented the remaining pTaTCR-/3 and CD3& dimers from assembling into functional pre-TCR-CD3 subcomplexes. Consistent with this view, thymocytes from mice selectively deficient in both CD3-6 and CD3-y subunits are blocked at the same stage (CD44-””CD25’) as C D ~ - E ’ ”thymocytes ~~ (B. Wang et at., 1998a). A genuine CD3-&-deficientstrain has been developed using a neomycin resistance cassette flanked by directly repeated loxP sites (DeJarnette et al., 1998). In a first step, most of exon 5 and all of exon 6 of CD3-E were replaced with the neomycin cassette and a strain of mice was derived from the corresponding ES cells. Homozygous mutant mice look phenotypically similar to the C D ~ - E ’ ~mice ’ ~ ~and likewise display both reduced amounts of CD3-y transcripts and nearly undetectable CD3-S transcripts. In a second step, in vivo deletion of the neornycin cassette was achieved by crossing the primary mutant mice with a strain expressing the Cre recombinise in germinal cells. Deletion of the neomycin cassette restored both CD3-y and CD3-S expression. However, the lack of CD3-E per se still prevented the progression beyond the CD44-”””CD25’ stage. These results formally demonstrate that the presence of CD3-E does not control the expression of CD3-y and CD3-6 genes. Moreover, when considered together with the effects of the CD3-cA5,CD3-eAp,and CD3-y + CD3-S mutations, they indicate that the CD3-ye and/or CD3-6.s heterodimers are mandatory for pre-TCR assembly or signal transduction. 4. CD3-6 Biochemical studies have suggested that the CD3-S chain is not at all or only loosely associated with the pre-TCR complex (Jacobs et al., 1994;

114

BERNARD MALISSEN et al,

Berger et al., 1997). Selective disruption of the CD3-Sgene has no measurable effect on the developmental steps controlled by the pre-TCR, and CDSSdeficient mice present normal numbers of DP cells (Dave et al., 1997).CD3-6, therefore, appears dispensable for the assembly and function of the pre-TCR complex. Whether the evolutionarily related CD3-y subunit is capable of being incorporated into the pre-TCR in lieu of the missing CD3-6 subunit has yet to be determined. Considering that most CD3-&deficient thymocytes do not progress beyond the DP stage, CD36 might become mandatory only at the DP to SP transition, as part of a (TCR-a CD3-6.5) building block in charge of substituting for pTa. 5. CD3-y . Progression beyond the CD44-""CD25' TN stage is severely impaired in mice selectively lacking expression of CD3-y (Haks et al., 1998). These findings demonstrate that the CD3-y chain plays an important role in preTCR assembly or function, and suggest that the related CD3-6 chain (expressed in CD3-y-I- thymus) cannot efficiently compensate for the loss of CD3-y. In contrast to the situation observed in both CD3-eA51A5 and CD3-7-I- CD3-6-I- mice, a few CD3-y-deficient thymocytes are capable of developing beyond the CD44-"CD25+ stage and of giving rise to DP and SP cells (in numbers that do not exceed 5% of those found in wildtype thymuses). These few DP and SP cells express severely reduced levels of CD3-e and TCR-a@ at their surface. Moreover, most of the DP cells emerging in CD3-y-'- mice still express CD25. Residual CD25 expression, originally observed on CD3-(/q-deficient DP cells, may reflect the fact that the transition to the DP stage occurred in these mutant mice without concomitant cell proliferation owing to the inefficient signals provided by the remaining CD3 subunits (Crompton et al., 1994). Considering that mice deficient in both CD3-y and CD3-8 lack both DP and SP cells, it remains possible that the few DP and SP cells that develop in CD3-ydeficient mice express distorted TCR-CD3 complexes in which CD3-6e heterodimers have been inefficiently incorporated in place of CD3-ye.

+

6. c D 3 - u ~ The CD3-5 and CD3-77 polypeptides result from alternative splicing of a single gene denoted CD3-5/77. As a consequence, the Sand polypeptides are identical for 122 amino acids and then display distinct carboxy-terminal ends. Disruption of the CD3-[/77 gene incompletely blocked the DN to DP transition (Liu et al., 1993; Love et al., 1993; Malissen et aE., 1993). Within a given CD3-5/77-/- inbred line, the number of DP cells varies considerably from animal to animal. For instance, CD3-4'77-I- thymuses can contain from 2- to 30-fold less DP cells than those from age-matched

TCR COMPLEX CD3 SUBUNIT FUNCTION

115

wild-type littermates. Substantial interindividual variations in the number of DP cells were also noticed in TCR-P-I- and Lck+ thymuses (Mombaerts et al., 1992; van Oers et al., 1996; Passoni et al., 1997; Kang et al., 1998a). The source of such variations has yet to be determined. Most DP cells found in CD3-&j-'- mice express cytoplasmic TCR-P chains and thus went through TCR-P-selection. However, they can be distinguished from bona fide wild-type DP cells because they show a profound reduction in the surface levels of TCR-CD3 complexes, they contain limited TCR-a rearrangement, they have reduced sensitivity to dexamethasone-induced apoptosis, and some still express CD25 (Crompton et al., 1994; Levelt et al., 1995b; Tanaka et al., 1995). This split pattern of phenotypic changes triggered by the pre-TCR in the absence of c D 3 - y ~suggests that the CD3-5lq module increases the levels of expression of the pre-TCR and concomitantly strengthens pre-TCR signals to allow them to reach the activation thresholds plausibly required by some demanding cellular responses associated with progression beyond the CD44-"""CD25+ stage (Ardouin et al., 1998).

7. FceRZy Another possibility accounting for the leakiness of the c D 3 - y ~null mutation is that in CD3-5/q-'- TN cells the pre-TCR complex may use FceRIy chains in lieu of c D 3 - y chains. ~ This possibility is consistent with the expression of FceRIy at early stages of T cell development (Rodewald et al., 1992,1993; Heiken et al., 1996) and would not constitute an unprecedented situation, because thymus-independent intestinal intraepithelial T cells bearing a0 or y6 TCRs coexpress CD3-5/77 and FceRIy chains and thereby display only moderately reduced levels of TCR at their surface in the absence of CD3-5/9 chains (Liu et al., 1993; Malissen et al., 1993; Guy-Grand et al., 1994; Koyasu et al., 1994; Heiken et al., 1996; Shores et al., 1998). Disruption of the FceRIy chain gene is without functional effect on the development of aP and y6 T cells (Takai et al., 1994; Heiken et al., 1996), and examination of thymocytes from mice deficient in both CD3-5/q and FccRIy chains revealed a phenotype essentially identical to that of mice that lack CD3-5,q alone (Shores et al., 1998). Therefore, FceRIy was not responsible for enabling the development of some T cells in the absence of CD3-5/q, and pre-TCR subcomplexes containing neither CD3-Yq nor FceRIy dimers can still transduce signals that suffice to drive the maturation of a few T cell precursors. A disulfide-bonded homodimer, DAP-WKARAP, possessing a cytoplasmic domain with a single ITAM has been found associated with activating NK cell receptors (Olcese et al., 1997; Lanier et al., 1998a,b; Smith et al., 1998). Beyond the structural homology noticed between DAP- WKARAP and FceRIy polypeptides,

116

BERNARD MALISSEN et d

the evolutionary relationship existing between DAP-l2/KARAP and other ITAM-containing receptor subunits has been highlighted by the analysis of the exon-intron organization of the DAP-l2/KARAP gene (E. Vivier, personal communication). The observation that DAP-12IKARAP is expressed in at least some T cells should prompt the analysis of its plausible role in T cell development and function. OF CD3 PAIRS TO PRE-TCRASSEMBLY OR B. UNEVEN CONTRIBUTION FUNCTION Based on genetic studies, the CD3 pairs identified first as mandatory components of a0 TCR complexes appear to contribute unevenly to the pre-TCR sensor (Fig. 3). The relative contribution of these various pairs to the pre-TCR can be ranked as follows: CD3-ye + CD3-Se > CD3Y E > CD3-&, whereas both CD3-Se and FceRIy, appear dispensable for inducing TCR-P selection. The salient contribution played by the CD3ye and CD3& dimers may result from their role in pre-TCR assembly or function. TCR-P selection was restored after reconstitution of CD3-4777deficient mice with transgenes encoding mutant CD3-{chains that retained the sequence necessary for TCR-CD3 assembly but were deprived of signaling capacity (Shores et al., 1994, 1997; L. Ardouin, unpublished results). Along the same line, expression of a transgenic CD3-e chain devoid of transducing capacity restored the developmental defects associated with the CD3-eA5mutation (L. Ardouin, unpublished results). Therefore, the contribution of CD3-e and CD3-5 to pre-TCR assembly appears to be at least as important as their contribution to pre-TCR signaling. It could even be that the ITAMs found in CD3-e or C D 3 4 are not endowed with any exclusive signaling function during TCR-P selection. The asymmetrical contribution of the CD3-y and CD3-S subunits to TCR-P selection may result from their distinct signaling or scaffolding properties. Consistent with the latter explanation, on artificial cross-linking with anti-CD3-e monoclonal antibodies, CD3-ye and CD3-SE heterodimers were found equally capable of inducing the maturation into DP cells RAG-2-’- and CD3-y-’- thymuses, of the TN cells found in CD3-S-’respectively (Haks et al., 1998; Dave et al., 1998a). Therefore, the CD3ITAM does not appear to mediate a specific signaling function during the TN to DP transition. Thus, the preponderant contribution played by the CD3-ye pair during TCR-P selection is likely to relate to its unique mode of assembly with the pre-TCR C domain module. The association between the a/3 TCR and CD3-6.9 depends primarily on residues located in their respective transmembrane segments, whereas that of the CD3Y E and a@ TCR heterodimers is mainly controlled by their ectodomains (Manolioset al., 1994).Based on the crystal structure of a complete Cup TCR

+

TCR COMPLEX CD3 SUBUNIT FUNCTION

117

ectodomain and on competition assays between monoclonal antibodies directed against the TCR C p domain and CD3-e Ig-like domain, it has been suggested that a cavity made of C p and C a residues specifically accommodates the extracellular domain of the CD3-e subunit that is paired with CD3-y (Ghendler et al., 1998). Therefore, if the CD3-ye pair associates with the pTaTCR-/3 dimer in a manner similar to that suggested for its association with ap TCR, the resulting (pTaTCR-PICD3-ye) core may constitute a mandatory assembly intermediate onto which CD3-C2 and, dispensably, CD3-6e subsequently nucleate. According to this hypothesis, the lack of CD3-ye should largely prevent the assembly of functional preTCR subcomplexes made of pTaTCR-/3 and CD3-& pairs, whereas, in the absence of CD3-5, pre-TCR subcomplexes made of pTaTCR-/3 and CD3-ye may still assemble and activate, albeit inefficiently, TCR-P selection. Therefore, in the context of a wild-type pre-TCR sensor, it is likely that the CD3-& homodimers merely increase the stability of a core made of pTaTCR-P and CD3-ye dimers, and coincidently contribute, by way of additional and functionally redundant ITAMs (Shinkai et al., 1995; Shores et al., 1997;van Oers et al., 1998),to strengthen the signals emanating from the ITAMs provided by the CD3-ye dimers. OF THE PRE-TCRSENSOR C. DOWNSTREAM EFFECTORS Analyses of mice deficient in cytoplasmic PTKs of the Src and Syk families have revealed their critical and redundant role in early T cell development. Lck-’- mice display a pronounced thymic atrophy associated with a dramatic reduction in the number of DP cells and an almost complete absence of SP cells (Molina et al., 1992). The deletion of Fyn had no measurable effect on T cell development (Appleby et at., 1992). However, combined disruption of Fyn and Lck totally prevented development beyond the CD44-”OWCD25+stage (Groves et al., 1996; van Oers et al., 1996). Therefore, although Lck constitutes the primary Src-family PTK responsible for regulating the progression beyond the CD44-”OWCD25+ stage, the above data indicate that Fyn may also contribute inefficiently to TCR-P selection when Lck is absent. Three additional observations support the pivotal role played by Lck during TCR-P selection. First, overexpression of a constitutively active form of Lck (termed Lck Y505F) was capable of restoring the progression of RAG-’- thymocytes to the DP stage and increasing the number of DP cells to within the range (or slightly higher) of that of wild type (Mombaerts et al., 1994a). [In contrast, expression of a constitutively active form of Fyn (termed Fyn Y258F) was unable of restoring the ability of RAG-’- TN cells to mature into DP cells (Groves et al., 1996). The functional differences existing between Lck and Fyn, and between their corresponding gain-of-function mutants, are likely to

118

’

BERNARD MALISSEN et al

be due to the fact that Lck and Fyn display distinct intracellular distribution and/or possess different intracellular substrates.] Interestingly, the unique signaling properties of Lck Y505F were independent of its ability to interact with the CD4 and CD8 coreceptors (Mombaerts et al., 1994a). Second, thymocytes overexpressing a transgene coding for a dominant negative (catalytically inactive) form of Lck (termed Lck K273R) are arrested at the very same early stage as RAG-/- thymocytes (Anderson et aZ., 1993). In contrast to RAG-deficient mice, however, and as expected from the fact that Lck acts downstream of TCR-@, the negative effect of the Lck K273F transgene cannot be reversed by the coexpression of a productively rearranged TCR-@transgene (Anderson et al., 1993; see also Wallace et al., 1995). (In Lck K273R mice, a@ T cell development is blocked at an earlier stage than in Lck-deficient mice. This discrepancy may be accounted for by the propensity of the transgenic Lck K273R PTK to compete not only with Lck, but also with Fyn.)Third, injection of anti-CD3-E antibodies into Lck+ X RAG-l-’- double-deficient mice results in the production of DP cells, although with a CD25+ phenotype and in numbers that do not exceed 15% of those obtained in RAG-1-’- mice injected with antiCD3 antibodies (Levelt et al., 199513). Mice deficient in Syk showed no major abnormalities in a@ T cell development (Cheng et al., 1995; Turner et al., 1995), whereas mice deficient in ZAP-70 had salient developmental defects that become manifest only at the DP to SP transition (Negishi et al., 1995; Wiest et al., 1997). Mice lacking both Syk and ZAP-70 are completely blocked at the transition from the CD44-”CD25+ stage (Cheng et al., 1997). Therefore, these results suggest that Syk and ZAP-70 are endowed with redundant functions during TCR-@selection. Thymocytes deficient in SLP-76, an adapter protein known to be phosphorylated rapidly by ZAP-70 and/or Syk after TCR (and most likely pre-TCR) engagement, do not progress beyond the CD44-/ “““CD25’ stage (Clements et al., 1998; Pivniouk et al., 1998). Thus, unlike the situation observed for the LcWFyn and ZAP-7O/Syk PTK pairs, there appears to be no redundancy at the level of SLP-76 during TCR-@selection. The generation of DP thymocytes was apparently unaffected in transgenic mice overexpressing either a dominant negative p21” protein [termed Ha-ras N17 (Swan et al., 1995)], or a catalytically inactive form of a MAP kinase [referred to as MEK-1 A97 (Alberola-Ila et al., 1995)l. This suggests that either the pre-TCR complex is not coupled to the ras raf + MEK + ERK signaling cassette, or that the signals initiated during TCR-@selection transit via ras and MEK-1 but need higher expression levels of these dominant negative forms to be impaired. Consistent with the latter view, expression of an activated form of ras (termed Haras””) into RAG-l-’- thymocytes was capable of restoring their ability to

TCR COMPLEX CD3 SUBUNIT FUNCTION

119

mature into DP thymocytes and to expand their number to wild-type levels (Swat et al., 1996). A few complexes devoid of TCR chains and consisting of calnexin and of either CD3-7.5 or C D 3 - 6 ~pairs are expressed at the surface of TN thymocytes (Wiest et al., 1995). Their cross-linking with antiCD3.5 monoclonal antibodies is sufficient to induce the faithful maturation of RAG-/- thymocytes into DP cells (Jacobs et al., 1994; Shinkai and Alt, 1994). Considering that the overexpression of some CD3-13 transgenes blocks thymocyte development at the CD44TD25- TN stage, such calnexin-bound subcomplexes have been hypothesized to be part of a putative pro-TCR complex involved in controlling the transition beyond the CD44TD25- stage (Wang et al., 1995). It is unlikely, however, that such partial complexes have a normal signaling function, because mice lacking most CD3 subunits (CD3-cA5IA5 + CD3-Wq-I- genotype) or with a disruption of Lck + Fyn, ZAP-70 + Syk, or SLP-76 produce immature T cells that can reach the CD44-""CD25+ stage and faithfully initiate TCR-/3 rearrangement (Groves et al., 1996; van Oers et al., 1996; Cheng et al., 1997; Ardouin et al., 1998; Clements et al., 1998; Pivniouk et al., 1998). [Note that the initiation of TCR-P gene rearrangement is likely to be differentially controlled in the thymus and in the intestine because the absence of the C D 3 - 7 8 ~module prevented both Dp-JP and VP-DPJP rearrangements in the precursors of thymus-independent intestinal intraepithelial T cells (Page et al., 1998).] Thus, in TN thymocytes, the CD3 subunits, the Lck + Fyn and ZAP-70 + Syk PTK pairs, as well as the SLP-76 adapter protein become mandatory only when the pre-TCR is expected to operate. Accordingly, it is tempting to arrange some of the mutations collated in Fig. 3 in the following linear (but oversimplistic) cascade: V(D)J recombination machinery + TCR-P polypeptides + assembly of pTaTCR-P-CD3 complexes + involvement of Lck and Fyn + activation of ZAP-7O/Syk + involvement of SLP-76 + involvement of ras + phenotypic changes associated with transition beyond the CD44-' 10wCD25+ stage (see Table I). These interactions have not yet been validated by the establishment of interaction suppressors or enhancers. In mice, achievement of these goals is currently laborious to the point of infeasibility. However, in their absence, we cannot rule out the possibility that the effectors Lck/Fyn and ZAP-7O/Syk and the adapter SLP-76 contribute to TCR-P selection via receptors that are distinct from the pre-TCR. D. THEPRE-TCRSENSOR:A LIGAND-INDEPENDENT SIGNALING DEVICE? The presence of a VP domain within the TCR-j3 polypeptide and the expression of class I and class I1 molecules encoded by the major histocom-

120

BERNARD MALISSEN et a1

TABLE I PHENOTYPIC CHANGES ASSOCLATEDWITH TCR-P SELECTION Effect Proliferation Temporary shut-off of RAG-1 and RAG-2 transcription CD4"" CD8"" CD25"" CD2"" c3390n (transient1 IMT-~O" (transient) PD-lup (tTaISient) CD27"" plow

Ref. Shortman et al. (1990) Wilson et al. (1994)

Mombaerts et al. (1992) Shinkai and Alt (1994) Shinkai and Alt (1994) Levelt et al. (1995b) Kishi et al. (1998) Nishimura et al. (1996) Gravestein et al. (1996) Lehuen et al. (1992); A. Lehuen, personal communication faso" Ogasawara et al. (1995) bcl-2"" Petrie et al. (1995) bcl-x,"" Ma et al. (1995) CD5"" Groves et al. (1996); Turner et al. (1997) BST-UBP-P Ishihara et al. (1996), Vicari et al. (1996) Egr-1"P Miyazaki (1997) TSA-1 (S~a-2)~"" MacNeil et al. (1993) Transcriptional activation of the TCR-a locus Wilson et al. (1994), Wilson and (TEA""), initiation of V a + Ja rearrangements, MacDonald (1993, Levelt et and concomitant deletion of the TCR-8 locus al. (1995a) Up-regulation of the transcription of the productively Malissen et al. (1995), B. Wang rearranged VDJC, units et al. (1998a,b) The TCR-/3 locus becomes inacessible to V(D)J Senoo and Shinkai (1998), recombinase, ensuring that the TCR-/3 alleles Chattopadhyay et al. (1998) remaining unrearranged are not the object of adventitious rearrangements in the setting of the RAG' DP cells Transcriptional repression of TCR-./ genes Kang et al. (1998b) pTa"' Bruno et al. (1995) The survival signals triggered by the pre-TCR relieve DiSanto and Rodewald (1998) the dependence on those emanating from the IL-7R Sensitivity to calcium-induced apoptosis Andjelic et al. (1993) Loss of responsiveness to calcium ionophore and Chen and Rothenberg (1993), phorbol ester, block in the inducibility of cytokine Zlotnik and Moore (1995), genes Sen et al. (1994), ZunigaPflticker et al. (1993)

TCR COMPLEX CD3 SUBUNIT FUNCTION

121

patibility complex (MHC) are dispensable for proper pre-TCR function ( Jacobs et al., 1994).Thus, in contrast to TCR-aP selection, which depends on the specificity of the TCR-a0 heterodimer (see below), TCR-P selection appears to occur independently of the specificity of the TCR VP domain. Moreover, pTaTCR-P heterodimers lacking most of their extracellular domains but retaining both the membrane-proximal cysteine residues required for heterodimerization and the transmembrane segments needed for assembly with some CD3 subunits are fully capable of relieving the developmental block observed in RAG-deficient mice and restoring a DP cell compartment identical to that induced by a wild-type pre-TCR sensor (Irvinget al., 1998).Collectively, these findings suggest that the recognition of a ligand by the pTaTCR-P heterodimer is dispensable for pre-TCR function. However, despite the fact that a Vpre-T domain has not yet been found associated with the pre-TCR (Fig. l),the above findings do not formally dismiss the possibility that the pre-TCR also serves as a folding template that prescreens VP domains for their ability to pair with Va domains at the DP stage (see, for instance, Kline et al., 1998). In mature T cells, after a TCR-CD3 complex and a CD4/CD8 coreceptor have engaged the same peptide/MHC complex, the Lck molecule bound to the CD4/CD8 cytoplasmic segment phosphorylates the juxtaposed CD3-ITAMs and allows each of them to recruit a ZAP-70 molecule. Next, the kinase activity of the recruited ZAP-70 molecules increases as a result of both phosphorylation by Lck and transautophosphorylation and permits the phosphorylation of key substrates (Peterson et al., 1998). Therefore, following ligand-induced phosphorylation, the CD3 subunits participate in T cell activation by recruiting ZAP-70 molecules and redistributing them close to their substrates. Triggering of the high-affinity receptor for the Fc portion of IgE (FcERI) follows an analogous mode and takes place when the FcERI-bound IgE molecules are clustered by soluble, multivalent antigens. FcsRI clustering leads to the phosphorylation of the F c ~ R 1 - pand -y chain ITAMS by the Src-family PTK Lyn, and to the subsequent recruitment of the Syk PTK. Using this more tractable receptor, it has been possible to demonstrate that when the relative Lynl F ~ E Rratio I is low, ligand-induced clustering is required for Lyn-mediated ITAM phosphorylation. In contrast, at a high Lyn/FcsRI ratio, the rate of ITAM phosphorylation is faster than its rate of dephosphorylation, which results in the accumulation of tyrosine-phosphorylated receptor subunits even in the absence of ligand-induced clustering (Scharenberg et al. 1995; Vonakis et al. 1997). Changing the intracellular balance of PTKs and PTPs may thus convert a ligand-activated switch into a constitutively active "switch." Accordingly, if CD44-""CD25+ DN cells express a high PTW PTP ratio [as hinted in Norment et al. (1997)],they may have the propensity

122

BERNARD MALISSEN et ol.

to phosphorylate constitutively the CD3 ITAMs and the ZAP-7O/Syk PTKs associated with the pre-TCR. Therefore, provided that a minimal set of CD3 ITAMs assembles with a pTaTCR-fl heterodimer, and as a direct consequence reaches a post-Golgi cell compartment (O'Shea et al., 1997), the ZAP-7O/Syk PTKs would become both constitutively activated and colocalized with their substrates and thereafter capable of triggering TCRp selection in a ligand-independent manner. According to this hypothesis, TCR-P selection should thus constitute an efficient cell-autonomous phenomenon, not constrained by the need for an extracellular ligand or by a limiting number of selecting stromal cell niches as documented for TCRa/3 selection (see Section V). E. MATURATION PATHWAYS INDEPENDENT OF pTaTCR-P EXPRESSION The pre-TCR/CD3 complex appears to constitute a molecular sensor operating at the CD44-"OWCD25+ TN stage and coupling further maturation to the prior achievement of productive TCR-/3 gene rearrangements. The complex gene programs mobilized by the pre-TCR are likely to permit the selective survival, expansion, and differentiation of TCR-@positive CD44-""CD25' cells (Table I ) . At this stage of development, the pTa and CD3 components of the pre-TCR are already available and it is the TCR-P polypeptides that constitute the rate-limiting factor in the assembly of the pre-TCR-CD3 complex. Accordingly, mice deficient in TCR-P polypeptide should resemble CD3-cA5lA5 mice and be ranked as "nonleaky" pre-TCR checkpoint mutants. However, the presence of small numbers of both CD44-""CD25- TN and DP cells in TCR-PP mutant mice (Mombaerts et al., 1992; Godfrey et al., 1994; Passoni et al., 1997) apparently contradicts the above view because it shows that the TCR-P chain is sufficient, but not necessary, for the progression of development beyond the CD44-"""CD25' stage (Kisielow and Boehmer, 1995). Note that the DP cells found in TCR-P-'- mice constitute bona fide DP cells on the basis of the coexpression of CD4 and CD8aj3, and the fact that they have deleted the TCR-6 locus on TCR-a gene rearrangement and contain fulllength TCR-a transcripts (Mombaerts et al., 1992; Mertsching et al., 1997; Livak et al., 1997). Disruption of the gene coding for the pTa subunit should also have prevented the assembly of pre-TCR complex and halted the progression beyond the CD44-"owCD25tTN stage. Nevertheless, pTa? mice were found to contain small numbers of DP cells that expressed normal levels of TCR-ap dimers on the cell surface and could undergo positive selection to become mature T cells (Fehling et d.,1995b). Moreover, despite being unable to assemble both pTaTCR-/3 and TCR-a TCR-/3 complexes, pTa-'- + TCRa-'- mice still contained some small numbers of DP cells (Buer et al., 1997). An interpretation of these data

TCR COMPLEX CD3 SUBUNIT FUNCTION

123

would be that the pre-TCR complex may constitute a leaky molecular sensor, capable of being occasionally fired in the absence of pTaTCR-P dimers. However, the finding that thymocytes present in RAG-/- and CD3-emfi mutant mice do not progress beyond the CD44-"w CD25+TN stage (Mombaerts et al., 1992; Shinkai and Alt, 1992; Godfrey et al., 1994; Malissen et al., 1995) suggests, rather, the existence of maturation pathway(s) independent of pTaTCR-P expression and capable of inefficiently promoting the transition through the CD44-"""CD25+ to CD44-' '"CD25- control point. It is worth emphasizing that TCR-P-I-, pTa-l-, mutant and pTa-l- + TCR-a-'- mice differ from RAG-'- and CD3-eA5IA5 mice in that they contain y6 T cells. The latter may provide, directly or indirectly (i.e., via interaction with the thymic stroma), transacting factors that promote the survival of a few CD44-"CD25+ cells and permit their differentiation to the DP stage. Consistent with this latter possibility, thymocytes from scid mice, the development of which is normally blocked at the CD44-"OWCD25+stage, can be induced to become DP cells by "trans" signals emanating from adoptively transfered TCR-aP+ (Shores et al., 1990) or TCR-y6+ (Lynch and Shevach, 1993) thymocytes. By ablating the development of y6 T cells, the introduction of a TCR-Gnu11 allele in the TCR-P-I- or pTa? background permitted to assess whether y6 T cells are responsible for the generation of the few DP cells found in TCRp- and pTa-deficient thymuses. In line with an implication of y6 T cells, the development of DP cells is more rigorously blocked in TCR-P-'- + TCR-6-l- mice than in mice deficient in TCR-j3 alone (Mombaerts et al., 1992; Godfrey et al., 1994). However, pTa-'- + T C R W thymuses display numbers of DP cells that are almost identical to those found in pTa-'thymuses, even though yS T cells are no longer present in the former (Buer et al., 1997). Intracytoplasmic staining with an anti-TCR-P antibody showed that most of the pTa? + TCR-S-'- DP cells express a TCR-/3 chain, whereas in pT0r-l- + TCR-a? mice only 15%of DP cells contain TCR-j3 polypeptides. Therefore, in pTa-'- + TCR-a? mice, the transition to the DP cells occurred irrespective of the TCR-P gene status, whereas in pTa-'- + T C R X - mice, DP cells were exclusively generated via a mechanism selecting for TCR-@positive precursors. As proposed by Buer et al. (1997), such a mechanism probably relies on the few constitutive TCR-a gene rearrangements that may occur in the CD44-"OWCD25+ TN cells prior to pre-TCR assembly. In the absence of pTa chain, those prematurely expressed TCR-a chains are likely to assemble into CUPTCR complexes capable of triggering development beyond the CD44-"OWCD25+ stage, Taken together, these data suggest that in the absence of a pTaTCRP heterodimer the transition to the DP stage can be rescued in part by cell-autonomous signals emanating from prematurely expressed TCR-aP

124

BERNARD MALISSEN et al.

complexes or via trans-acting factors emanating from neighboring y6 T cells. This last possibility is mainly deduced from the comparative analysis of T cell development in TCR-P-I- and TCR-P-I- + T C R X - mice, and on the observation that the DP cells found in pTa-l- + TCR-a? mice do not contain TCR-6 chains (Buer et al., 1997). Nevertheless, considering that DP cell development is normally accompanied by both the deletion of the TCR-6 locus and the transcriptional repression of TCR-.)I genes (Table I), the above observations are also consistent with an alternative model in which TCR y6 heterodimers act in cis (Leu,in a cell-autonomous fashion) to trigger the development of DP cells. According to this model, the down-regulation of TCR-y genes and/or deletion of the TCR-6 genes that take place during the transition to the DP stage should blunt the expression of such a y6 TCR, and thereby account for the strongly reduced numbers of DP cells that do arise in the absence of pTaTCR-P. Consistent with this model of “TCR-yS selection” to the DP stage, it has been observed that the DP thymocytes from TCR-P-I- mice exhibit a proportion of inframe TCR-S and TCR-y genes as high as do TCR-yS+ thymocytes (Livak et al., 1997; Passoni et al., 1997; Kang et al., 1998a).Furthermore, although earlier studies of TCR-P-’- + TCR-6-I- mice have emphasized the absence of DP cells (see above), Kang and colleagues (1998b) have showed that very small numbers of DP cells, approximating 0.3 to 1.2% of total thymocytes, can still be detected in these doubly deficient mice, and provided genetic evidence that TCR-P-’TCR-6-/- mice do assemble pTaTCRy-CD3 complexes capable of stimulating the development of DP cells. As suggested above for y6 TCR, the capacity of those pTaTCR-y complexes to stimulate DP cell development in cis is likely to be blunted by the transcriptional repression of TCR-y genes that is normally associated with the transition to the DP cell stage. Therefore, these data rule out the existence of trans-acting “differentiation” factors emanating from neighboring yS T cells and indicate that four TCR isoforms, made of pTaTCR-P, TCR-aTCR-P, TCR-yTCR-6, or pTaTCR-y heterodimers, can induce DP cell development in cis, however, with markedly distinct efficiencies.Along that line, it should be noted that the possibility that developing yS or aP T cells can induce neighboring scid thymocytes to differentiate into DP cells (see above) appears to be limited to the scid background, because when nornial and RAG-1-’- stem cells were allowed to codifferentiate, no development of DP cells occurred from the RAG-1-’- progenitors (Kang et al., 1998a).

+

F. ROLEOF THE PRE-TCRAND TCR-76 COMPLEXES IN aP/yS LINEAGE COMMITMENT y6 T cell development proceeds normally in mice deficient in TCR-P, pTa, TCR-a, or CD3-6 genes (Mombaerts et al., 1992; Philpott et al.,

TCR COMPLEX CD3 SUBUNIT FUNCTION

125

1992; Fehling et al., 1995b; Dave et al., 1997). In contrast, the TCR-6, CD3-8, CD3-y, and CD3-( subunits are mandatory for proper y6 T cell development (Itohara et al., 1993; Malissen et al., 1993, 1995; Haks et al., 1998; Liu et al., 1993). Thus, the y6 TCR resembles the pre-TCR in that its function requires CD3-&,CD3-7, and CD3-5 but tolerates the lack of CD3-6. In the a/3 T cell lineage, various combinations of cell surface markers (e.g., CD44CD25, CDUCD8) have permitted measurement of the stage of differentiation reached by developing T cells even in mice deprived of pre-TCR or aP TCR complexes. In contrast, the genetic dissection of y6 T cell development has been hampered by the paucity of cell surface markers allowing the staging of developing y8 T cells and mainly relied on the monitoring of the expression of the y6TCR. Therefore, it is still unclear to what extent y6 TCR expression is needed to trigger the gene program leading to y6 T cell maturation (Leclercq et al., 1993; Tatsumi et al., 1993; Groves et al., 1996; Lalli et al., 1996). However, the observation that none of the CD3 components is required for the completion of TCR-y and TCR-6 gene rearrangements suggests that the latter are probably not subjected to a two-checkpoint model analogous to that affecting TCR-a and TCR-P gene rearrangements (Ardouin et at., 1998) (see also Section V). The mechanisms that underlie the commitment to the y 6 and a/3T cell lineages and establish the corresponding lineage-associated gene expression patterns are still largely unknown. Two main models have been proposed to account for the relationship existing between TCR gene rearrangements and a/3 versus y6 T cell lineage commitment (reviewed in Robey and Fowlkes, 1998). In both models, the TCR-P, TCR-y, and TCR-6 gene rearrangements are postulated to occur concurrently witllin the CD44-’ ’“CD25’ TN cells. In the “instructive” model (Fig. 4B), cell fate is primarily determined by which TCR isotype bipotent CD44-””CD25+ precursors assemble first. According to that model, expression of tlie TCR-ya-CD3 complex should specifically turn on a gene cassette (symbolized by a black arrow in Fig. 4A) dedicated to y6 T cell lineage development, whereas expression of the pTaTCR-P-CD3 complex should trigger a distinct gene cassette (open arrow, Fig. 4B) capable of carrying out aP T cell differentiation. It should be noted that there are presently no data suggesting that the pre-TCR and y6 TCR might be coupled to distinct signaling cassettes. For instance, none of the CD3 components appears uniquely expressed in either of these two TCR isoforms, and the large intracytoplasmic segment of pTa is clearly dispensable for a/3 T cell commitment (Fehling et al., 1997).The observations that DP T cells can be generated through alternate TCR isoforms devoid of the pTa subunit (TCR-76 or TCR-aP; see Section II1,F) and, conversely, that aP TCR can replace y6 TCR to promote the

BERNARD MALISSEN et d.

126

A rearrangement

commitment

1

8

fate

B

I

Lineage Commitment

I

1 Uncommitted precursors

FIG.4. Two models of commitment to the ab and y6 T cell lineages. (A) Commitment to the ap and y6 lineages within the thymus occurs prior to and independently of TCR gene rearrangements. The symbol R corresponds to pTaTCR-P, TCR-yG, or TCR-abcontaining receptors. However, only those cells expressing TCR isoforms that match the predetermined cell fate will be capable of fully differentiating (see text for details). (B) The TCR isotype expressed in immature T cells is the primary determinant of intrathymic lineage commitment.

development of yS lineage cells (Bruno et al., 1996; Fritsch et al., 1998) are also difficult to reconcile with the “instructive” model. The second main model, referred to as the “stochastic” model, proposes that lineage commitment occurs prior to the onset of TCR gene rearrangement, and via unknown mechanisms (Fig. 4A). In this model, rearrangements of the TCR-P, TCR-7, and TCR-6 genes are postulated to happen in each precursor cell irrespective of its lineage commitment status. However, considering that on a@ T cell lineage commitment, transcriptional silencing events affect TCR-.)I genes (Kang et al., 199810) and that the TCR-S locus

TCR COMPLEX CD3 SUBUNIT FUNCTION

127

is deleted on both chromosomes (Malissen et al., 1992), only a cell that makes a TCR isoform that matches its predetermined fate will be allowed to differentiate fully. For instance, ap lineage-committed cells that fail to express TCR-P but express inappropriately TCR-yS heterodimers might still become DP cells but fail to differentiate further because of their inability to assemble a relaying TCR-a0 receptor. G. THEPRE-TCRIs ESSENTIAL FOR ALLELICEXCLUSION AT THE TCR-P Locus Because T lymphocytes are diploid cells, the V( D)J recombination process could, in principle, generate T cell clones expressing two productively rearranged TCR alleles and therefore more than one TCR-cw/P or TCRy/S chain combination. In the mouse, the expression of a productively rearranged TCR-/3 chain transgene has been shown to prevent complete V(D)J rearrangement of endogenous TCR-P genes (Uematsu et al., 1988), and this has led to the assumption that CD44-''OWCD25+T cell precursors have developed feedback inhibition mechanisms to ensure that most mature T cell clones express one, and only one, TCR-p chain. These events are referred to as the establishment of allelic exclusion at the TCR-P locus. Considering that the expression of a productively rearranged TCR-P transgene inhibits most endogenous V/3 to DpJP rearrangements, it has been suggested that the TCR-P chain, and by extension the pre-TCR-CD3 complex, play a pivotal role in the enforcement of allelic exclusion at the TCR-/3 locus. Therefore, disruption of the gene coding for the pTa subunit should have prevented assembly of a functional pre-TCR complex and affected the establishment of allelic exclusion at the TCR-P locus. However, in pTa-'- thymocytes, expression of a transgene coding for a functional TCR-/3 chain was found to inhibit endogenous VP to DPJP rearrangements to almost the same extent as in a pTcu+'+ background (Xu et al., 1996; Krotkova et al., 1997). Assuming that no other gene products can fully compensate for the loss of pTa (e.g., the products of prematurely expressed TCR-a genes; see below), these data are inconsistent with the suggestion that the pre-TCR-CD3 complex is required for signaling allelic exclusion at the TCR-/3 locus. Using CD3-y Se- and CD3-l/v-deficient mice harboring a productively rearranged TCR-P transgene, Ardouin and colleagues (1998) have shown that the C D 3 - 7 8 ~and CD3-47~modules are each essential for the establishment of allelic exclusion at the TCR-@locus. Their mandatory contribution to the activation of this negative feedback loop probably relates to the role they play in the assembly and function of the pre-TCR. The discrepancy that exists between the pTa- and CD3-deficient mice with regard to the establishment of allelic exclusion at the TCR-P locus can be explained by

128

BERNARD MALISSEN et a2

the presence within the CD25' TN cells of low constitutive levels of V a + J a recombination. In pTa? mice, and only in pTa-'- mice, the resulting TCR-a chains are likely to contribute to the premature assembly of TCR-a/3 complexes capable of signaling maturation as well as allelic exclusion via their associated CD3 subunits (Buer et al., 1997). However, if Va-Ja rearrangements do occur in CD25+ TN cells before TCR-/3 selection, it is at a frequency at least 100-fold lower than that observed in DP cells (Tanaka et al., 1995). Thus, premature TCR-a chain expression can account only for part of the effects observed with the transgenic TCR/3 chain in the absence of pTa. The capacity of the transgenic TCR-/3 used by Krotkova et al., (1997) to signal allelic exclusion independently of pTa may relate to its abnormal expression in a phosphatidylinositol-linkedform at the surface of CD25+ TN cells. Such a nonphysiological pattern of expression probably triggers artificial signals that mimic those normally emanating from the pre-TCR, a possibility strengthened by the fact that expression of this particular TCR-/3 transgene in a pTa-/- + RAG-2-/background (i'e.,in the absence of any endogeneous TCR chain) promotes the generation of significant numbers of DP cells (Krotkova et al., 1997). Therefore, the occurrence of TCR-P allelic exclusion in the absence of the pTa chain is likely to result from the combination of inappropriate expression of the transgenic TCR-/3 chains and premature TCR-a chain expression. To avoid any possible artifacts linked to the expression of a TCR-/3transgene, TCR-/3 gene rearrangements have been directly assessed by single-cell polymerase chain reaction in the small CD44-"OWCD25+cells sorted from pTa-'- and pTa+/+mice (Aifantis et al., 1997). This analysis showed that the pTa chain is indeed mandatory for allelic exclusion at the TCR-/3 locus, because in CD25+ cells isolated from pTa? mice V/3 + DPJP rearrangement appears to proceed despite the presence of an already productively rearranged TCR-/3 gene. Together, with the analysis of CD3deficient mice, these data clearly exclude a model in which TCR-/3 chains can signal TCR-/3 allelic exclusion in the absence of pTa and of the CD3 components thought to be part of the pre-TCR-CD3 sensor. Therefore, the pre-TCR-CD3 complex is essential for the establishment of delic exclusion at the TCR-/3 locus. IV. The cup TCR Sensor

Signaling through the pre-TCR triggers the induction of a high rate of TCR-a gene rearrangements among DP cells (Table I). On productive TCR-a gene rearrangements and substitution of pTa by TCR-a, TCRa@+D P cells are rescued from programmed cell death and induced to differentiate into SP cells only if their a/3 TCRs bind with low affinity to

TCR COMPLEX CD3 SUBUNIT FUNCTION

129

self-peptide-MHC complexes expressed on thymic stromal cells. Such a phenotypic shift, often denoted as TCR-aP selection, is ablated by a number of genetic defects (Fig. 3) affecting (1)the assembly and proper cellular display of peptide-MHC complexes [from mutations in µglobulin (B2m) (Zijlstra et al., 1990; Koller et al., 1990), MHC class I heavy chain (Tanchot et al., 1997), subunits of the TAP peptide transporter (Van Kaer et al., 1992), MHC class I1 molecules (Cosgrove et al., 1991; Grunsby et al., 1991),transcription factors regulating MHC class I1 chain genes (Mach et al., 1996), and H-2M (Fung-Leung et al., 1996; Martin et al., 1996; Miyazaki et al., 1996)l; (2) the synthesis of TCR-a chains [from mutations in the TCR-a locus (Mombaerts et al., 1992; Philpott et al., 1992) and TCR-a gene enhancer (Sleckman et al., 1997)l; (3)the transducing CD3 subunits and intracellular effectors associated with the aj3 TCR (below); and (4) the coreceptors CD4 and CD8 (Schilham et al., 1993). Again, it is tempting to arrange the above mutations in the following linear cascade: TCR-a gene rearrangements + TCR-a polypeptides + assembly of a/3 TCR-CD3 complexes 4 interaction of the clonally distributed a/3 TCR heterodimers with self-peptide-MHC ligands + concomitant involvement of the CDNCD8 coreceptors and of Lck + ZAP-70 activation + phenotypic changes associated with the transition beyond the DP stage. As shown in Fig. 3, each of the CD3-c/r] (Liu et al., 1993; Love et al., 1993; Malissen et al., 1993), CD3-y (Haks et al., 1998), Lck (Molina et al., 1992),Vav (Zhanget al., 1995;Turner et al., 1997),or CD45 (Byth et al., 1995) gene defects corresponds primarily to a “leaky” mutation of the preTCR sensor. Because ofthis characteristic, it has been possible to score their secondary effect on the ap TCR sensor and confirm that the pre-TCR and TCR share many functional components. Consistent with these structural homologies, some of the phenotypic changes triggered by the pre-TCR and TCR sensors are identical (e.g., the transient induction of the CD69 and IMT-1 cell surface molecules; Table I). However, in sharp contrast to preTCR engagement, a/3 TCR firing leads to cell differentiation in the mere absence of cell proliferation. Interestingly, CD3-6 is required for TCR-a/3 selection but is dispensable for the development of functional y6 T cells (Dave et al., 1997). Thus, the pre-TCR and yt3 TCR contrast notably with the ap TCR in that they appear to be able to tolerate loss of CD3-6. This may reflect that TCR-P selection (and most likely TCR-yt3selection) requires a lower signaling threshold than TCR-a0 selection (Berger et al., 1997; Irving et al., 1998; see also Lalli et al., 1996). Expression of a transgenic CD3-6 chain lacking most of its intracytoplasmic segment restored the developmental defects observed in CD3-6-’- mice (Dave et al., 1998b). This indicates that the CD3-6 ITAM does not mediate a specific signaling function during TCR-ab selection and suggests that the requirement for

130

BERNARD MALISSEN et d.

CD3-6 during TCR-aP selection is rather due to its structural role in the assembly of TCR-aP-CD3 complexes. The few DP cells present in Lck-’- thymus differ from wild-type DP cells because they express residual levels of CD25 (Levelt et al., 1995b), five- to sixfold lower levels of CD5 (Groves et al. 1996), and intermediate to high levels of aP TCR complexes at their surface (Molina et al., 1992). In the absence of Lck, however, these TCR complexes or their CDdCD8 coreceptors are probably unable to trigger the signals required for TCRaP selection. In humans deficient in ZAP-70, CD4+, but not CD8+, SP cells are found in the thymic medulla and periphery (Arpaia et al., 1994; Chan et al., 1994; Elder et al., 1994; Gelfand et al., 1995). (In the absence of ZAP-70, however, these peripheral CD4+ SP cells are unable to respond to TCR stimulation.) In contrast, mice deficient in ZAP-70 or expressing a catalytxally inactive ZAP-70 develop a normal complement of a/3 TCR+ DP thymocytes, but neither CD4+ nor CD8+ SP T cells (Negishi et al., 1995; Wiest et al., 1997). Therefore, in ZAP-70-l- mice, Syk alone is sufficientto induce the transition to the DP stage but is unable to contribute to the DP to SP transition. During mouse T cell development, it is likely that Syk is turned off at an earlier timepoint than in the human and is thus prevented from participating in TCR-aP selection (Gong et al., 1997). When coexpressed in a single T cell, the CD3-[and CD3-77 polypeptides can combine to form disulfide-linked homo- and heterodimers, each capable of associating with subcomplexes made of the TCR-aP, CD3-ye, and CD3& polypeptide pairs. TCR complexes only contain a minor fraction of CD3-577 or CD3-7777 dimers as compared with CD3-55 homodimers. Studies performed with T cell hybridomas have suggested that the few CD3-77 containing dimers are responsible for coupling TCR engagement to a “cell death signaling cassette” (Mercep et al., 1989).Accordingly, mice specifically deprived of CD3-77 polypeptides, via disruption of the exon utilized exclusively in CD3-77 chain mRNA, should have manifested substantial alterations in T cell homeostasis. However, little difference was observed between wild-type and CD3-cl77- mice with regard to surface expression of the TCR complex, cellularity,or subset composition in thymus and peripheral lymphoid organs (Koyasu et al., 1994; Ohno et aZ., 1994). Furthermore, the absence of CD3-77 subunits was found to alter neither positive nor negative intrathymic selection events. Thus, in contrast to earlier suggestions based on in vitro data, the CD3-77 polypeptide does not appear endowed with unique signaling capacities during the development and function of T cells. Conversely, targeting of the exon utilized exclusively in CD3-5 chain mRNA prevented the expression of CD3-5 polypeptides without affecting the levels of CD3-V polypeptides (Ohno et al., 1993). Consistent with in vitro studies indicating that the CD3-77 chain is less efficient than the CD3-5 chain in enabling the transport to the cell

TCR COMPLEX CD3 SUBUNIT FUNCTION

131

surface of TCR subcomplexes made of the TCR-a/3, CD3-7.5, and CD3SE pairs, the CD3-q polypeptides present in the CD3-4-l~’mice permitted the expression of only minute amounts of TCR at the surface of DP and SP thymocytes. Nevertheless, these levels (one-fifth that seen in wild-type littermates) were sufficient to allow the generation of DP and SP cells in numbers that do not exceed 10% of those found in wild-type thymus. Crossing the CD3-r/q+ mice with male antigen (H-Y)-specificTCR transgenic mice further revealed that the decrease in TCR signaling caused by the CD3-5 deficiency resulted in a shift from negative to positive selection in male TCRm/CD3-c/qf mice and a failure of positive selection in female TCRHY/CD3-C/v+mice (Yamazaki et al., 1997). V. Raison d’6tre of the Pre-TCR

None of the CD3 components is required for the onset and completion of TCR-y and TCR-S chain gene rearrangements (Ardouin et al., 1998). These results suggest that TCR-y and TCR-6 gene rearrangements are probably not subjected to stepwise epigenetic controls analogous to those that monitor the outcome of TCR-/3 and TCR-a gene rearrangements and rely on the sequential expression of the pre-TCR and TCR sensors (Fig. 5). In the a/3 T cell lineage, the raison dstre of the pre-TCR may be that a/3T cells undergo a second step of selection known as TCR-aO selection, during which most DP cells (-97%) crash either by default of selection or by negative selection (Fig. 6). Thereby, by triggering the selective expansion and maturation of only those T cell precursors expressing a TCR-/3 chain, the pre-TCR is likely to counterbalance in part the tremendous “crash factors” associated with TCR-crP selection and thereby contribute to maximize the diversity of the mature a/3 T cell repertoire (Fig. 7). As outlined in Fig. 8, if TCR-a and TCR-/3 gene rearrangements were to happen concomitantly (as expected for TCR-y and TCR-S gene rearrangements), and if their outcome was to be sensed in a single step, the yield in SP T cells will be almost twice lower and the load in TCR-negative DP cells 1.5 times higher than in a situation in which TCR-/3 and TCR-a rearrangements occur sequentially and with a causal relationship (compare Figs. 7 and 8). Whether the different TCR gene rearrangement strategies followed by a/3 and y6 T cell precursors reflect that the development of yS T cell precursors occurs irrespective of TCR-yS specificity or only requires a “low-grade’’engagement of their TCR, akin to TCR-/3 selection, remains to be determined. VI. Are the Roles of the Pre-TCR and TCR Complexes Limited to Ensure Cell Survival?

The interconnection of the signaling pathways triggered by antigen receptors (pre-TCRs, TCRs), cytokine receptors [e.g., the receptor for

132

BERNARD MALISSEN et al

Phase of intense prolieration,

serves to generate a pool of cells large enough for the constitution of a rejwtoire of diverse TCR fl chains.

1

I

Phase of intense proliferation, serves to generate a pool of TCR p'DP cells large enough for the constitution of a repeaoire of diverse TCR a chains.

I

TCR p rearrangements

-

11 CEECKPOMT 1

Developing T cells do not progress beyond this point unless they have assembled a functional pre-TCR complex. This first quality control does not depend on the Vp specificity and ensures that all the DP cells that are produced express a TCR p chain.

I

TCR a rearrangements 11-11

A CHECKPOEW 2

Developing T cells do not progress beyond this point unless they have assembled a TCR afi complex capable of binding with low affinity to self-peptides presented by MHC molecules. This second quality control increases the efficiency of the peripheral T-cell repertoire by avoiding its overload with "useless" T cells.

FIG.5. A two-checkpoint model of a/3 T cell development. Two CD3-operated checkpoints, corresponding to the sequential assembly of the pre-TCR (checkpoint 1)and TCR (checkpoint 2) sensors, couple the survival of developing T cells to the achievement of successful TCR gene rearrangement. Prior to the onset of TCR gene rearrangement, redundant signals emanating from c-kit and the IL-7 receptor are necessary for the survival, expansion, and differentiation of early T cells.

interleukin-7 (IL-7R)], and growth factor receptors (e.g., c-kit) has made it difficult to decipher their respective roles in the survival, proliferation, and differentiation events that govern T cell development (see, for instance, Galandrini et al., 1997). Whether a given receptor is necessary for inducing differentiation beyond a given stage, or for allowing developing cells to survive and respond to intrinsic (cell-autonomous) or extrinsic (intercellular) differentiation cues, cannot be generally sorted out until the signaling cassettes operated by this receptor are worked out (Corcoran et al., 1996; Wiest et al., 1997). However, hints of these challenging issues have been obtained by enforcing the expression of the antiapoptotic proteins Bcl-2

TCR COMPLEX CD3 SUBUNIT FUNCTION

133

FIG.6. The pre-TCR and TCR complexes that are sequentially expressed on developing a/3 T cells sense the outcome of TCR V(D)J rearrangements. Considering that V -+ (D)J joining events are random with respect to the reading frame, from a cohort of 18 TN thymocytes (each of them being symbolized by a square), 6 (1/3) might be expected to produce a functional V/3 gene as a result of their first attempt of rearrangement (see Malissen et nl., 1992) and thus assemble pre-TCR complexes capable of triggering their transition to the DP stage. As soon as assembled, the pre-TCR activates a negative feedback loop that will close the accessibility of the second TCR-/3 allele to V(D)J recombinase, thereby rendering the TCR-/3 locus inaccessible to further V( D)J recombination and restricting the

134

BERNARD MALISSEN et al.

and Bcl-xl in the lymphocytes that develop in mice deficient in RAG, IL7R, or the common cytokine y chain (yc).(The yc subunit is shared by the receptors for IL-2, IL-4, IL-7, IL-9, and IL-15, and its deletion drastically reduces the number of thymocytes without affecting the relative represen-

6 selected cells to the expression of a single productively rearranged TCR-/3 allele. Of the remaining 12 TN cells with a nonproductively rearranged TCR-P allele, an additional third might be expected to produce a functional V/3 gene on rearrangement of the second TCR/3 allele. Thus, 8 out of the 18 original cells would end up with two out-of-frame V@ genes (-/- TN cells) and so would not be capable of assembling a pre-TCR and of progressing beyond the TN stage. In the absence of secondary rearrangements, these failed cup T cells would either die or attempt to become y6 T cells (see Section 111,E).The 10 pre-TCR+ TN cells (denoted as +) will experience TCR-/3 selection and give rise via seven to nine consecutive cell divisions to -2560 DP cells capable of initiating rearrangements at the TCR-a chain locus [for simplicity, only a fraction (2560 cells X 1/32 = 80 cells) of the resulting DP cells have been represented]. Among the 80 DP thymocytes that are depicted, 4 of 9 would not be capable of assembling a TCR-aP heterodimer due to nonproductive TCR-a chain gene rearrangements (Set 1). These failed a/3T cells (denoted as -DP cells) will not be rescued from programmed cell death. Of 9 DP cells (denoted as +DP cells), 5 are expected to undergo V a gene rearrangements that maintain an open reading frame and to subsequently express TCR-a@heterodimers at their surface (Sets 2, 3, and 4). As many as 1 in 3 a/3’ DP cells might express TCR-af3 heterodimers capable of engaging in productive interactions with self-peptide/MHC complexes (Merkenschlager et al., 1997). Among the ap DP cells capable of productive interactions, those that express TCRs that react too strongly toward self-peptide/MHC complexes will be eliminated (Set 3), whereas the few that express TCR capable of interacting with low affinity with self-peptide/MHC complexes will be rescued from programmed cell death and allowed to mature into SP cells (Set 4).The remaining a@’ DP cells (Set 2) express TCR-aP heterodimers that cannot engage in productive interactions with self-peptide/MHC complexes due (1)to the presence of polymorphic MHC residues that prevent proper TCR docking, or (2) to “mismatched CDR-3cr/CDR-3/3 combinations that are not capable of interacting optimally with the peptide/MHC surface (see Mazza et al., 1998). Set 2 may also include DP cells expressing TCR-a/? heterodimers that are potentially selectable in the context of the inherited MHC alleles but have not entered the positive selection process because (1)they have failed to reach the limited number of stromal cell niches capable of supporting positive selection (Merkenschlager et al., 1997), (2) they have been confronted with a repertoire of MHCbound self-peptides that was not capable of positively selecting them (Sant’Angelo et al., 1997), or (3) the CD4 or CD8 coreceptor they retained at their surface was not matched with the MHC class specificity of their TCR. As a result of all these discrete processes, DP cells belonging to Sets 1, 2, and 3 die in situ (shaded areas), whereas those belonging to Set 4 survive and become mature SP cells. The role of positive selection is likely to increase the efficiency of the peripheral T cell repertoire by avoiding its overload with “useless” T cells. The number of cell divisions occurring at the TN to DP transition and the percentage of cells attributed to each of the four individualized D P sets were estimated based on Zerrahn et al. (1997), Merkenschlager et al. (1997), Malissen et al. (1992), Shortman et al. (1990), PBnit et al. (1995), and Ignatowicz et al. (1996). Note that in s h a v contrast to TCR/3 selection, TCR-a/3 selection occurs without cellular expansion (Shortman et al., 1990).

TCR COMPLEX CD3 SUBUNIT FUNCTION

135

FIG.7 . Conditioning the onset of TCR-(w gene rearrangements to the prior occurrence of productive TCR-/3 gene rearrangements is likely to maximize the generation of a diverse repertoire of peripheral cup T cells. This two-checkpoint model of (w/3 T cell development is to be compared with the single-checkpoint model of a/3 T cell development shown in Fig. 8. Considering that CD44-””CD25’ TN cells appear to comply with the “regulated” model of allelic exclusion (Malissen et nl., 1992; Aifantis et ul., 1997), from a cohort of 18 CD44-”OWCD25+ cells, 10 will be capable of expressing TCR-/3 chains and experiencing TCR-/3 selection. Due to combinatorial and junctional diversities, each of the 10 TCR-/3+ cells is expected to contain a distinct V/3 gene (denoted b l to b10). Among the -2560 DP cells resulting from TCR-P selection, approximately 5 out of 9 will assemble TCR-(w/3 heterodiiners at their surface. Importantly, because TCR-/3 selection includes a phase of

136

BERNARD MALISSEN d al.

tation of the CDdCD8 subsets.) Considering that the constitutive expression of a Bcl-2 transgene partially corrects the hypocellularity observed in mice deficient in IL-7R or yc (Kondo et al., 1997; Maraskovsky et al., 1997; Akashi et al., 1997), it has been inferred that the IL-7R operates in part via Bcl-2 to promote survival of TN T cells prior to TCR-/3 selection (von Freeden-Jeffiy et al., 1997; but see DiSanto and Rodewald, 1998). On productive TCR-/3 gene rearrangements, the resulting pre-TCR complexes are likely to induce Bcl-xl-dependent signals that relay the decaying Bcl-2 survivals signals and thus prolong the life-span of DP cells up to the time of TCR-aP selection (Petrie et al., 1993). Then, the few DP T cells capable of being positively selected according to the specificity of their a/3 TCR reactivate their Bcl-2 gene and regain resistance to apoptosis on entering the SP stage. Therefore, when considered together with the phenotype observed in mice deficient in various members of the Bcl-2 family (Knudson and Korsemeyer, 1997; Ma et al., 1995; Motoyama et al., 1995), the sequential expression of the IL-7R, pre-TCR, and a/3 TCR is likely to provide partially overlapping antiapoptotic signals to developing thymocytes (Fig. 2, bottom part). A scenario similar to that outlined for a/3 T cells is likely to govern B cell development. For instance, in transgenic mice constitutively expressing Bcl-2 or Bcl-xl genes in the B lineage, pro-B cells accumulate that have not productively rearranged their heavy chain genes. Being defective in pre-BCR assembly, cells with such a genotype are normally targeted for cell death in physiological situations. Importantly, in no case has the constitutive expression of Bcl-2 or Bcl-xl been found capable of relieving the developmental block observed in RAG-deficient B cells. Therefore, causing early B cells to survive is not in itself sufficient for differentiation, and development beyond the pro-B cell stage requires the triggering of both survival and differentiation signals via the pre-BCR (Tarlinton et al., 1997; Fang et al., 1996; Young et al., 1997). In contrast, analysis of RAG-deficient mice in

cell proliferation, a given TCR VP gene will end up being assorted with many distinct TCR V a genes [denoted a1 to a44 in the small fraction (1132) of DP cells that is depicted]. Taking into account the “crash factors” associated with TCR-a/3 selection (see Fig. 6), -96 out of the -1408 unique TCR-crIP combinations that have been assembled will be positively selected. Three additional points should be made. First, followingthis two-checkpoint model of a@ T cell development, the stromal cell niches capable of supporting TCR-aP selection will be swamped by TCR-P+TCR-a!- cells (denoted as “cold competitors) that represent -44% of the pool of newly generated DP cells. Second, the figures indicated in this model have been rounded and are only suggested to illustrate the essence of the argument. Third, the fact that TCR-a! chain gene rearrangements do not obey a “regulated’ model of delic exclusion (Malissen et aE., 1992) does not affect the above calculations.

TCR COMPLEX CD3 SUBUNIT FUNCTION

137

FIG.8. A virtual single-checkpoint model of ap T cell development. This figure shows the expected outcome of TCR-a6 selection in the case that TCR-a and TCR-P genes will have rearranged concomitantly and the attainment of productive TCR-a and TCR-/3 gene rearrangements will have been checked in one step. (A) The simultaneous rearrangements of the TCR-a and TCR-/3 loci are postulated to occur only once developing T cells have undergone a phase of cytokine-driven expansion and progressed to the DP stage. Using the V gene reading-frame errors and TCR-aP selection “crash” factors discussed in Figs. 6 and 7 , -64 out of the -800 TCR-dD combinations that have been assembled will be selected and the “cold’ competitors’ Ioad will represent 69% of the DP cell pool. (B) In the case the TCR-a and TCR-fl gene rearrangements would have simultaneously occurred at the CD44-””’”CD25+TN stage, the diversity of the (YPT cell repertoire would have been dramatically reduced due to the small size of the CD44-””CD25+ compartment.

138

BERNARD MALISSEN et al.

FIG.8-Continued

which Bcl-2 or Bcl-xl was constitutively expressed in the T cell lineage showed the presence of a small number of DP cells (Linette et al., 1994; Chao and Korsmeyer, 1997). These last studies are not entirely congruent (see, for instance, Maraskovsky et al., 1997), but nevertheless, suggest that the function of the pre-TCR may not be directly necessary for inducing differentiation toward the a/3 T cell lineage, but rather may be limited to trigger signals that promote survival and/or proliferation of TCRP', CD44-' '"CD25' cells and enable them to utilize an intrinsic differentiation program or respond to differentiation signals triggered by pathways that function independently of the pre-TCR. VII. The Limits of Genetic Analysis: Redundancy and Adaptive Response to Certain Mutations

Genetic redundancy has often been invoked as an explanation when a null mutation in a gene results in no detectable phenotypic abnormalities

TCR COMPLEX CD3 SUBUNIT FUNCTION

139

or less severe abnormalities than expected. In some cases, functional redundancy constitutes a sound explanation (Thomas, 1993), but an alternative possibility is that the gene performs a nonredundant function that can be unraveled only under challenging experimental or environmental conditions (Yasunaga et al., 1995). Adaptive mechanisms can also be set in motion to compensate for a missing gene product and to buffer the severity of the mutation. Such compensatory mechanisms occur in multigene families, the members of which display overlapping expression patterns [for example, Src-family kinases (Lowell and Soriano, 1996) and Bcl-2-family members (Knudson and Korsemeyer, 1997)l.Because of the epigenetic mechanisms that shape the repertoire of TCRs expressed on mature T cells, lymphoid development is particularly prone to compensatory mechanisms. For instance, the denTCR-CD3 complexes expressed by DP thymosity and output of the cytes contribute to setting the threshold for TCR-a/3 selection. However, it is possible to reset TCR selection windows artificially by engineering mutations that either reduce ( T u n e r et al., 1997; Shores et al., 1997; Lin el al., 1997) or increase (Tarakowskyet al., 1995) TCR-transmitted signals. When these mutations were bred into mice capable of generating a complete repertoire of aUp TCRs, T cell development was unabated, and selection of a@TCRs with lower or higher affinity for self-MHC was thought to compensate for those TCR-CD3 complexes with a boosted or impeded signaling capacity, respectively. Consistent with this view, the effect of most of these mutations was revealed when they were bred into mice in which ap TCR variability was neutralized by expressing matched pairs of TCR-a and TCR-P transgenes that had been calibrated in a normal thymus. Along the same line, it should be noted that despite the almost complete absence of SP thymocytes noted in CD3-5/q-’- and CD3-6-l- mice, the spleen and lymph nodes of those mutant mice contained SP T cells that express at their surface low (CD3-6-’-) to undetectable (CD3-5/7-’-) levels of partial TCR-CD3 complexes. The peripheral SP cells found in CD3J/q+ mice accumulate over time in a process dependent on the expression of MHC molecules (Simpson et al., 1995). It is therefore likely that in CD3-~7q-I- DP thymocytes, some partial TCR-CD3 complexes can be expressed at the cell surface in amounts not detectable by flow cytometry, but sufficient to trigger TCR-ap selection. Consistent with this possibility, peripheral CD3-47q-l- SP T cells do contain afi TCRs that readily react against self-MHC products once artificially expressed at densities comparable to those found on wild-type T cells and together with a complete constelhion of CD3 subunits (Lin et al., 1997). It is likely that once expressed in a wild-type DP T cells, these very TCRs would readily trigger negative selection by overtly reacting with self-peptide/MHC complexes.

140

BERNARD MALISSEN d al.

The above observations are reminiscent of those obtained in mice that lack &m (Glas et al., 1994; Cook et al., 1995; Ljunggren et al., 1995) and express MHC class I molecules at levels considerably lower than normal. In these mice, even the low levels of conformed class I heavy chains that can reach the surface appeared capable of selecting CD8+ T cells, but in numbers lower than in wild-type mice. The few CD8+ T cells selected in &m+ mice were capable of mounting strong allospecific responses after in vivo priming. However, in marked contrast to alloreactive CD8+ T cells generated in &m-positive littermates, the alloreactive T cells induced in a &rn-’- context readily cross-reacted toward syngeneic cells derived from &m+’+.As suggested by Glas et al., (1994), the most likely interpretation of these data is that in Pzm-’- animals, negative selection does not eliminate T cells capable of recognizing &.m-positive cells expressing self-peptide/ MHC complexes at physiological levels. Consequently, by readjusting T cell selection windows, mutations in the CD3-LJr) and Pzm genes resulted in two phenotypes that display striking symmetrical relationships. Finally, it may be worth emphasizing that some individual signaling components associated with antigen receptors are required for the sequential activation of both positive and negative regulatory loops. Consequently, disruption of the corresponding genes often leads to compound phenotypes, which are in part determined by the relative efficiency of the mechanisms capable of compensating the missing elements during the activation of the positive and negative loops. For instance, B cells deficient in the Src-family PTK Lyn display an enhanced proliferative response to antiIgM stimulation, in apparent contradiction to the finding that the earliest BCR signaling events they exhibit were delayed and slightly decreased (Chan et al., 1997,1998; Nishizumi et al., 1998). Such contrasting observations can be reconciled only by postulating that Lyn plays both a positive and a redundant role during early BCR firing, and, at later timepoint, a nonredundant role in the initiation of the mechanisms that negatively regulate BCR signaling. VIII. Conclusions

The analysis of the in vivo consequences of constitutive gene inactivation has clear limitations. For instance, when a gene defect results in embryonic lethality or prevents the development of a given lineage, any late potential impacts will be missed. This problem might be, at least in part, solved by the use of conditional gene inactivation techniques (see, for instance, Kuhn et al., 1995). Other limitations stem from the fact that gene inactivation studies are usually restricted to known genes and that the choice of the gene to be inactivated often depends on preconceived ideas (but see

TCR COMPLEX CD3 SUBUNIT FUNCTION

141

Zambrowicz et al., 1998). In spite of the above limitations, the analysis of engineered mouse mutants has constituted a powerful approach to identifylng the molecular sensors that couple the survival of developing lymphocytes to the achievement of successful gene rearrangements at the TCR loci. Besides identifylng cytokines, growth factors, and transcription factors involved in lymphocyte development, genetic analysis also makes it possible to organize most of these protagonists into gene networks that control critical events in the life of developing lymphocytes. Nevertheless much remains to be learned about the sequence of events that facilitate the survival, expansion, and differentiation of developing T lymphocytes.

ACKNOWLEDGMENTS Some of the studies described in this paper were supported by institutional grants from INSERM and CNRS, and by specific grants from the Commission of the European Communities, Association pour la Recherche sur le Cancer, and Ligue Nationale contre le Cancer. We thank Pierre Golstein, PhiIippe Massol, Arek Miazek, Lee Leserman, and AnneMarie Schmitt-Verhulst for comments on the manuscript; Ludovica Bruno, Jim DiSanto, Dietmar Kappes, Paul Love, David Raulet, Cox Terhorst, and Eric Vivier for sharing unpublished data with us; Corinne B6ziers-La-Fosse for graphic art; and Noelle Guglietta and Veronique Pr6au for typing the manuscript.

REFERENCES Aifantis, I., Buer, J., von Boehmer, H., and Azogui, 0. (1997). Immunity 7, 601-607. Akashi, K., Kondo, M., von Freeden-Jeffry, U., Murray, R., and Weissman, I. L. (1997). Cell 89,1033-1041. Alberola-Ila, J., Forbush K. A,, Seger, R., Krebs, E. G., and Perlmutter, R. M . (1995). Nature 373, 620-623. Anderson, S. J., Levin, S. D., and Perlmutter, R. M. (1993). Nature 365,552-554. Andjelic, S., Jain, N., and Nikolic-Zugic (1993).J. Exp. Med. 178, 1745-1751. Appleby, M. W., Gross, J. A., Cooke, M. P., Levin, S. D., Qian, X., and Perlmutter, R. M. (1992). Cell 70, 751-763. Ardouin, L., Ismaili, J., Malissen, B., and Malissen, M. (1998).J. Exp. Med. 187, 105-116. Arpaia, E., Shahar, M., Dadi, H., Cohen, A,, and Roifman, C. M. (1994). Cell 76,947-958. Barber, D. F., Passoni, L., Wen, L., Geng, L., and Hayday, A. C. (1998). J. Immunol. 161,ll-16. Bentley, G. A,, Boulot, G., Kajalainen, K., and Mariuzza, R. A. (1995). Science 267,19841987. Berger, M. A,, D a d , V., Rhodes, M. R., Bosma, G. C., Bosma M. J., Kappes, D. J., and Wiest, D. L. (1997).]. Exp. Med. 186, 1461-1467. Bones, J.-C., Demengeot, J., Davidson, L., and Alt, F. W. (1996). Proc. Nutl. Acad. Sci. U.S.A. 93, 7871-7876. Bouvier, G . ,Watrin, F., Naspetti, M., Verthuy, C., Naquet, P., and Ferrier, P. (1996).Proc. Natl. Acad. Sci. U.S.A. 93, 7877-7881. Bruno, L., Rocha, B., Rolink, A,, Von Boehmer, H., and Rodewald, H. R. (1995). Eur. J. Zmmunol. 25, 1877-1882. Bruno, L., Fehling, H. J., and von Boehmer, H. (1996). Zmmunity 5, 343-352.

142

BERNARD MALISSEN et al.

Buer, J., Aifantis, I., DiSanto, J. P., Fehling, H. J., and van Boehmer, H. (1997).J. Exp. Med. 185, 1541-1547. Byth, K. F., Conroy, L. A., Howlett, S., Smith, A. J. H., May, J., Alexander, D. R., and Holmes, N. (1995).J. Exp. Med. 183, 1707-1718. Chan, A. C., Kadelcek, T. A,, Elder, M. E., Filipovich, A. H., Kuo, W.-L., Iwashima, M., Parslow, T. G., and Weiss, A. (1994). Science 264, 1599-1602. Chan, V. W., Meng, F., Soriano, P., DeFranco, A. L., and Lowell, C. A. (1997).Immunity 7, 69-81. Chan, V. W. F., Lowell, C. A,, and DeFranco, A. L. (1998). Cum Biol. 8, 545-553. Chao, D. T., and Korsmeyer, S. J. (1997). Int. Immunol. 9, 1375-1384. Chattopadhyay. S., Whitehurst, C. E., Schwenk, F., and Chen J. (1998). J. Immunol. 160, 1256-1267. Chen, D., and Rothenberg, E. V. (1993). Mol. Cell. Biol. 13, 228-237. Cheng, A. M., Rowley, B., Pao, W., Hayday, A., Bolen, J. B., and Pawson, T. (1995). Nature 378, 303-306. Cheng, A. M., Negishi, I., Anderson, S. J., Cha, A. C., Bolen, J., Loh, D. Y., and Pawson, T. (1997). Proc. Nutl. Acud. Sci. U.S.A. 94, 9797-9801. Clements, J. L., Yang, B., Ross-Barta, S. E., Eliason, S. L., Hrstka, R. F., Wikamson, R. A., and Koretzky, G. A. (1998). Science 281, 416-419. Cook, J. R., Solheim, J. C., Connolly, J. M., and Hansen, T. H. (1995). J. Immunol. 154, 47-57. Corcoran, A. E., Smart, F. M., Cowling, R. J., Crampton, T., Owen, M. J., andvenkitaraman, A. R. (1996). EMBO 1. 15, 1924-1932. Cosgrove, D., Gray, D., Dierich, A., Kaufman, J., Lemeur, M., Benoist, C., and Mathis, D. (1991). Cell 66, 1051-1066. Crompton, T., Moore, M., MacDonald, H. R., and Malissen, B. (1994). Eur. J. Immunol. 25, 1903-1907. Danska, J. S., Holland, D. P., Mariathasan, S., Williams, K. M., and Guidos, C. J. (1996). Mol. Cell. Biol. 16, 5507-5517. Dave, P. D., Cao, Z., Browne, C., Alarcon, B., Fernandez-Miguel, G., Lafaille, J., de la Hera, A., Tonegawa, S. and Kappes, D. J. (1997). EMBO J. 16, 1360-1370. Dave, P. D., Keefe, R., Berger, M. A., Drbal, K., Punt, J. A., Wiest, D. L., Alarcon, B., and Kappes, D. J. (1998a). Int. Immunol. 10, 1481-1490. Dave, P. D., Sdancewics, M., and Kappes, D. J. (1998b).J. Exp. Med. (in press) Dejamette, J. B., Sommers, C. L., Huang, K., Woodside, K. J., Emmons, R., Katz, K., Shores, E. W., and Love, P. E. (1998). Proc. Nutl. Acud. Sci. U.S.A. 95, 14909-14914. Del Porto, P., Bruno, L., Matt&, M.-G., Van Boehmer, H., and Saint-Ruf, C. (1995).Proc. Nutl. Acud. Sci. U.S.A. 92, 12105-12109. DiSanto, J. P., and Rodewald, H.-R. (1998). Cum @in. Immunol. 10, 196-207. Elder, M. E., Lin, D., Clever, J., Chan, A. C., Hope, T. J., Weiss, A., and Parslow, T. G. (1994). Science 264, 1596-1599. Fang, W., Mueller, D. L., Pennell, C. A,, Rivard, J. J., Li, Y.-S., Hardy, R. R., Schlissel, M. S., and Behrens, T. W. (1996). Immunity 4, 291-299. Fehling, H. J., Laplace, C., MattBi, M.-G., Saint-Ruf, C., and Von Boehmer, H. (1995a). lmmunogenetics 42,275-281. Fehling, H. J., Krotkova, A., Saint-Ruf, C., and Van Boehmer, H. (1995b).Nature 375, 795-798. Fehling, H. J., Iritani, B. M., Krotkova, A,, Forbush, K. A., Laplace, C., Perlmutter, R. M., and Van Boehmer, H. (1997). Immunity 6, 703-714.

TCR COMPLEX CD3 S U B U N I T F U N C T I O N

143

Fritsch, M., Andersson, A., Peterson, A., and Ivars, F. (1998).Eur.]. Inmiunol. 28,828-837. Fling-Leung, W.-P., Surh, C. D., Liljedahl, M., Pang, J., Leturcq, D., Peterson, P. A,, Webb, S. R., Karlsson, L. (1996). Science 271, 1278-1281. Galandrini, R., Henning, S. W., and Cantrell, D. A. (1997). Inmwnity 7, 1133-174. G e h n d , E. W., Weinberg, K., Mazer, B. D., Kadecek. T. A,, and Weiss, A. (1995).]. Exp. Men. 182, 1057-1066. Ghendler,Y., Sniolyar,A,, Chang, H.-C., and Reinherz, E. L. (1998).]. Exp. Men. 187,15291536. Glas, R., Ohlen, C., Hijglund, P., and Kirre, K. (1994).J . Exp. Med. 179, 661-672. Godfrey, D. I., Kennedy, J., Mombaerts, P., Tonegawa, S.,and Zlotnik, A. (1994).J.Zmmnunol. 152, 4783-4792. Gong, Q., White, L., Johnson, R., White, M., Negishi, I., Thomas, M., and Chan, A. C. (1997). Zmn~unity7, 369-377. Gravestein, L. A., van Ewijk, W., Ossendorp, F., and Borst, J. (1996).J. Exp. Med. 184, 675-685. Groves, T., Smiley, P., Cooke, M. P., Forbush, K., Perlmutter, R. M., and Guidos, C. J. (1996). Zrninunity 5, 417-428. Groettrup, M., Baron, A,, Griffiths, G., Palacios, R., and Von Boehmer, H. (1992). EMBO J. 11,2735-2746. Groettrup, M., Ungewiss, K., Azogui, O., Palacios, R., Owen, M. J., Hayday, A. C., and Von Boehmer, H. (1993). Cell 75, 283-294. Grunsby, M. J., Johnson, R. S., Paparannou, V. E., and Glimcher, L. H. (1991). Science 253, 1417-1420. Guy-Grand, D., Rocha, B., Mintz, P., Malassis-Seris, M., Selz, F., Mdissen, B., and Vassdli, P. (1994).]. Exp. Med. 180, 673-679. Haks, M. C., Krimpenfort, P., Borst, J., and Kruisbeek, A. M. (1998). EMBO J. 17, 18711882. Heiken, H., Schulz, R.-J., Ravetch, J. V., Reinherz, E. L., and Koyasu, S. (1996). Eur. /. Iinmunol. 26, 1935-1943. Ignatowicz, L., Kappler, K., and Marrack, P. (1996). Cell 84, 521-529. Irving, B. A,, and Weiss, A. (1991). Cell 64, 891-901. Irving, B. A,, Alt, F. W., and Killeen, N . (1998). Science 280, 905-908. Ishihara, K., Kobune, Y., Okuyama, Y., Itoh, M., Lee, 0. K., Murdoka, O., and Hirano, T. (1996). Int. Imninunol. 8, 1395-1404. Itohara, S., Mombaerts, P., Lafaille, J., Iacomini, J., Nelson, A., Clarke, A. R., Hopper, M. L., Far,A,, and Tonegawa, D. (1993). Cell 72, 337-348. Jacobs, H., Vandeputte, D., Tolkamp, L., De Vries, E., Borst, J., and Berns, A. (1994).Eur. 1.Znimunol. 24, 934-939. Jameson, S. C., and Bevan, M. J. (1998). Cum. Opin. Imnmnunol. 10, 214-219. Kang, J., Coles, M., Cado, D., and Raulet, D. H. (1998a). bnnmnity 8, 427-438. Kang, J., Fehling, J. J., Laplace, C., Malissen, M., Cado, D., and Raulet, D. H. (1998b). Inmunity 8, 713-721. Kishi, H., Tong, J.-J., Nagdtd, T., and Muraguchi, A. (1998). Int. Immnunol. 10, 951-960. Kishimoto, H., Kubo, R. T., Yorifuji, H., Nakayarna, T., Asano, Y., and Tada, T. (1995). 1. Exp. Med. 182, 1997-2006. Kisielow, P., and Boehmer, H. V. (1995). Adu. Immunol. 58, 87-209. Kline, G. H.. Hartwell, L., Beck-Engeser, G. B., Keyna, U., Zaharrevitz, S., Klinman, N. R., a i d rick, H.-M. (1998).J. Immunol. 161, 1608-1618. Knudson, C. M., and Korsmeyer, S. J. (1997). Nature Genet. 16, 358-363. Koller, B. H., Marrack, P., Kappler, J. W., Smithies, 0. (1990). Science 248, 1277-1230.

144

BERNARD MALISSEN el a!

Kondo, M., Akashi, K., Domen, J., Sugamura, K., and Weissman, I. L. (1997). Immunity 7, 155-162. Koyasu, S., Hussey, R. E., Clayton, L. K., Lerner, A., Pedersen, R., Delany-Heiken, P., Chau, F., and Reinherz, E. L. (1994). EMBO J. 13, 784-797. Krotkova, A,. von Boehmer, H., and Fehling, H. J. (1997).J. Exp. Med. 186,767-775. Klihn, R., Schwenk, F., Aguet, M., and Rajewsky, F. (1995). Science 269, 1427-1429. Lalli, E., Sassone-Corsi, P., and Ceredig, R. (1996). EMBO J. 15, 528-537. Lanier, L. L., Corliss, B., Wu, J. Leong, C., and Philipps, J. H. (1998a).Nature 391,703-707. Lanier, L. L., Corliss, B., Wu, J., and Philipps, J. H. (1998b). Immunity 8, 693-701. Leclercq, G., Plum, J,, Nandi, D., De Smedt, M., and Allison, J. P. (1993).J. Exp. Med. 178,309-315. Lehuen, A,, Monteiro, R. C., and Kearney, J. F. (1992). Eur. J. Immunol. 22,2373-2380. Letourneur, F., and Klausner, R. D. (1992). Science 255, 79-82. Levelt, C. N., Wang, B., Ehrfeld, A,, Terhorst, and Eichmann, K. (1995a).Eur. J. Immunol. 25, 1257-1261. Levelt, C. N., Mombaerts, P., Wang, B., Kohler, H., Tonegawa, S., Eichmann, K., and Terhorst, C . (1995b). Immunity 3,215-222. Lin, S.-Y., Ardouin, L., Gillet, A,, Malissen, M., and Malissen, B. (1997). J. Exp. Med. 185, 707-715. Linette, G. P., Grusby, M. J., Hedrick, S. M., Hansen, T. H., Glimcher, L. H., and Korsmeyer, S. J. (1994). Immunity 1, 197-205. Liu, C. P., Ueda, R., She, J., Sancho, J., Wang, B., Weddell, G., Loring, J., Kurahara, C., Dudley, E. C., Hayday, A., et al., (1993). EMBO J. 12,4863-4875. Livak, F., Wilson, A., MacDonald, H. R., and Schatz, D. G. (1997). Eur. J. Immunol. 27, 2948-2958. Ljunggren, H.-G., Van Kaer, L., Sabatine, M. S., Auchincloss, Jr., H., Tonegawa, S., and Ploegh, H. L. (1995). Int. Immunol. 7, 975-984. Love, P. E., Shores, E. W., Johnson, M. D., Tremblay, M. L., Lee, E. J., Grinberg, A., Huang, S. P., Singer, A., and Westphal, H. (1993). Science 261,918-921. Lowell, C. A., and Soriano, P. (1996). Genes Deu. 10, 1845-1857. Lynch, F., and Shevach, E. M. (1993). Int. Immunol. 5, 991-995. Ma, A., Pena, J. C., Chang, B., Margosian, E., Davidson, L., A t , F. W., and Thompson, C. B. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 4763-4767. Mach, F., Steimle, V., Martinez-Soria, E., and Reith. W. (1996). Annu. Reu. Immunol. 14,301-331. MacNeil, I., Kennedy, J., Godfrey, D. I., Jenkins, N. A., Masciantonio, M., Mineo, C., Gilbert, D. J., Copeland, N. G., Boyd, R. L., and Zlotnik, A. (1993).]. Immunol. 151,6913. Malissen, M., Trucy, J., Jouvin-Marche, E., Cazenave, P. A,, Scollay, R., and Malissen, B. (1992). Immunol. Today 13,315-322. Malissen, M., Gillet, A., Rocha, B., Trucy, J., Vivier, E., Boyer, C., Kontgen, G., Brun, N. K., Mazza, G., Spanopoulou, E., et al. (1993). EMBO J. 12,4347-4355. Malissen, M., Gillet, A., Ardouin, L., Bouvier, G., Trucy, J., Fenier, P., Vivier, E., and Malissen, B. (1995). EMBO J, 14,4641-4653. Manolios, N., Kemp, O., and Li, Z. G. (1994). Eur. J. Immunol. 24, 84-92. Maraskovsky, E., O’Reilly, L. A,, Teepe, M., Corcoran, L. M., Peschon, J. J., and Strasser, A. (1997). Cell 89, 1011-1019. Martin, W. D., Hicks, G. G., Mendiratta, S. K., Leva, H. I., Rdey, H. E., and Van Kaer, L. (1996). Cell 84, 543-550. Mazza, G., Housset, D., Piras, C., Grbgoire, C., Lin, S.-Y., FonteciUa-Camps, J. C., and Malissen, B. (1998). Immunol. Rev. 163, 187-196.

TCR COMPLEX CD3 SUBUNIT FUNCTION

145

Mercep, M., Weissman, A. M., Franck, S. J., Klausner, R. D., and Ashwell, J. D. (1989). Science 246, 1162-1165. Merkenschlager, M., Graf, D., Lovatt, M., Bomhardt, U., Zamoyska, R., and Fisher, A. G. (1997).J. Exp. Med. 186, 1149-1158. Mertsching, E., Wilson, A,, MacDonald, H. R., and Ceredig, R. (1997). Eur. J. Zmmunol. 27, 389-396. Miyazaki, T. (1997).J. Exp. Med. 186, 877-885. Miyazaki, T., Wolf, P., Tourne, S., Waltzinger, C., Dierich, A,, Barois, N., Ploegh, H., Bonoist, C., and Mathis, D. (1996). Cell 84, 531-541. Molina, T. J., Kishihara, K., Siderovski, D. P., Van Ewijk, W., Narendran, A,, Timms, E., Wakeham, A., Paige, C. J., Hartmann, K. U., Veillette, A,, Davidson, D., and Mak, T. W. (1992). Nature 357, 161-164. Mombaerts, P., Clarke, A. R., Rudnicki, M. A., Iacomini, J., Itohara, S., Lafaille, J. J., Wang, L., Ichikawa, Y., Jaenisch, R., Hooper, M. L., and Tonegawa, S. (1992). Nature 360, 225-231. Mombaerts, P., Anderson, S. J., Perlmutter, R. M., Mak, T. W., and Tonegawa, S. (1994a). Zmmunity 1, 261-267. Mombaerts, P., Mizoguchi, E., Ljunggren, H.-G., Iacomini, J., Ishikawa, H., Wang, L., Grusby, M. J., Glimcher, L. H., Winn, H. J., Bhan, A. K., and Tonegawa, S. (1994b). Int. Zmmunol. 6, 1061-1070. Motoyama, N., Wang, F., Roth, K. A,, Sawa, H., Nakayama, K.-I., Nakayama, K., Negishi, I., Senju, S., Zhang, Q., Fujii, S., and Loh, D. Y. (1995). Science 267, 1506-1510. Negishi, I., Motoyama, N., Nakayama, K.-I., Nakayama, K., Senju, S., Hatakeyama, S., Zhang, O., Chan, A. C., and Loh, D. Y. (1995). Nature 376,435-438. Nishimura, H., Agata, Y., Kawasaki, A,, Sato, M., Imamura, S., Minato, N., Yagita, H., Nakano, T., and Honjo, T. (1996). Znt. Zmmunol. 8, 773-780. Nishizumi, H., Horikawa, K., Mlinaric-Rascan, I., and Yamamoto, T. (1998).J. Exp. Med. 187, 1343-1348. Norment, A. M., Forbush, K. A,, Nguyen, N., Malissen, M., and Perlmutter, R. M. (1997). J. Exp. Med. 185, 121-130. Nussennveig, A,, Chen, C., da Costa Soares, V., Sanchez, M., Sokol, K., Nussennveig, M. C., and Li, G. C. (1996). Nature 382, 551-555. Ogasawara, J., Suda, J., and Nagata, S. (1995).J. Exp. Med. 181, 485-491. Ohno, H., Aoe, T., Tab, S., Kitamura, D., Ishida, Y., Rajewski, K., and Saito, T. (1993). E M B O J . 12,4357-4366. Ohno, H., Goto, S., Taki, S., Shirasawa, T., Nakano, H., Miyatake, S., Aoe, T., Ishika, Y., Maeda, H., Shirai, T., Rajewsky, K., and Saito T. (1994). E M B O ] . 13, 1157-1165. Olcese, L., Cambia@, L., Semenzato, G., Bottino, C., Moretta, A., and Vivier, E. (1997). J. Zmmunol. 158, 5083-5086. Ono, S., Ohno, H., and Saito, T. (1995). Zmmunity 2, 639-644. O’Shea, C. C., Thornell, A. P., Rosewell, I. R., Hayes, B., and Owen, M. J. (1997).Immunity 7,591-599. Page, S. T., Bogatzki, L. Y., Hamerman, J. A,, Malissen, M., Perlmutter, R. M., and Pullen, A. M. (1998). Proc. Natl. Acad. Sci. U.S.A. 95, 9459-9464. Passoni, L., Hoffman, E. S., Kim, S., Crompton, T., Pao, W., Dong, M.-Q., Owen, M. J., and Hayday A. C. (1997). Immunity 7, 83-95. PBnit, C., Lucas, B., and Vasseur, F. (1995).J. Zmmunol. 154, 5103-5113. Peterson, E. J., Clements, J. L., Fang, N., and Koretzky, G. A. (1998).Cum. @in. Immunol. 10,337-344. Petrie, H. T., Livak, F., Burtrum, D., and Mazel, S. (1995).J. Exp. Med. 182, 121-127.

146

BERNARD MALISSEN ef al.

Petrie, H. T., Livak, F., Schatz, D. G., Strasser, A,, Crispe, I. N., and Shortman, K. (1993). 1.Exp. Med. 178, 615-622. Philpott, K. L., Viney, J. L., Kay, G., Rastan, S., Gardiner, E. M., Chae, S., Hayday, A. C., and Owen, M. J. (1992). Science 256, 1448-1452. Pivniouk, V., Tsitsikov, E., Swinton, P., Rathbun, G., Alt, F. W., and Geha, R. S. (1998). Cell 94,229-238. Punt, J. A,, Kubo, R. T., Saito, T., Finkel, T. H., Kathiresan, S., Blank, K. J., and Hashimoto, Y. (1991).]. Exp. Med. 174, 775-783. Robey, E., and Fowlkes, B. J. (1998). Cum. +in. lmmunol. 10, 181-187. Rodewald, H.-R., Moingeon, P., Lucich, J, L., Dosiou, C., Lopez, P., and Reinherz, €3. L. (1992). Cell 69, 139-150. Rodewald, H.-R., Awad, K., Moingeon, P., D’Adamio, L., Rabinowitz, D., Shinkai, Y., Alt, F. W., and Reinherz, E. L. (1993).J.Exp. Med. 177, 1079-1092. Rodewald, H.-R., Ogawa, M., H d e r , C., Waskow, C., and DiSanto, J. P. (1997). Immunity 6, 265-272. Romeo, C., and Seed, B. (1991). Cell 64, 1037-1046. Saint-Ruf, C., Ungewiss, K., Groettrum, M., Bruno, L., Fehling, H. J., and Von Boehmer, H. (1994). Science 266, 1208-1212. Sant’Angelo, D. B., Waterbury, P. G., Cohen, B. E., Martin, W. D., Kaer, L. V., Hayday, A. C., and Janeway, Jr., C. A. (1997). Immunity 7, 517-524. Scharenberg, A. M., En. S., Cuenod, B., Yamamura, H., and Kinet, J.-P. (1995). EMBO 1. 14,3385-3394. Schilham, M. W., Fung-Leung, W.-P., Rahemtulla, A,, Kuendig, T., Zhang, L., Potter, J.. Millert, R. G., Hengartner, H., and Mak, T. W. (1993). Eur. J. Immunol. 23,1299-1304. Sen, J., Shinkai, Y., Alt, W., Sen, R., and Burakoff, S. J. (1994).]. Exp. Med. 180,2331-2327. Senoo, M., and Shinkai,Y. (1998). Int. Immunol. 10, 553-560. Shinkai, Y., and Alt, F. W. (1994). Int. lmmunol. 6, 995-1001. Shinkai, Y., Koyasu, S., Nakayama, K. I., Murphy, K. M., Loh, D. Y., Reinherz, E. L., and Alt, F. W. (1993). Science 259, 822-825. Shinkai, Y., Ma, A., Cheng, H. L., and Alt, F. W. (1995). Immunity 2,401-411. Shinkai, Y., Rathbun, G., Lam, K. P., Oltz, E. M., Stewart, V., Mendelsohn, M., Charron, J., Datta, M., Young, F., Stall, A. M., et al. (1992). Cell 68, 855-867. Shores, E. W., Sharrow, S. O., Uppenkamp, I., and Singer, A. (1990). Eur. J. lmmunol. 20,69-77. Shores, E. W., Huang, K., Tran, T., Lee, E., Grinberg, A., and Love, P. E. (1994). Science 266, 1047-1050. Shores, E. W., Tran, T., Grinberg, A., Sommers, C. L., Shen, H., and Love, P. E. (1997). J. Exp. Med. 185, 893-900. Shores, E. W., Ono, M., Kawabe, T., Sommers, C. L., Tran, T., Lui, K., Udey, M. C., Ravetch, J., and Love, P. E. (1998).]. Exp. Med. 187, 1093-1101. Shortman, K., Egerton, M., Spangrude, G., and Scollay, R. (1990).Semin. lmmunol. 2,3-12. Simpson, S., Holliinder, G., She, J., Levelt, C., Huang, M., and Terhorst, C. (1995). Int. lmmunol. 7, 287-293. Sleckman, B. P., Bardon, C. G., Ferrini, R., Davidson, L., and Alt, F. W. (1997). lmmunity 7, 505-515. Smith, K. M., Wu, J., Bakker, A. B. H., Philipps, J. H., and Lanier, L. L. (1998).]. lmmunol. 161, 7-10. Swan, K. A,, Alberola-IIa, J., Gross, J. A., Appleby, M. W., Forbush, K. A., Thomas, J. F., and Perlmutter, R. M. (1995). EMBO 1. 14,276-285.

TCR COMPLEX CD3 SUBUNIT FUNCTION

147

Swat, W., Shinkai, Y., Cheng, H.-L., Davidson, L., and Alt, F. W. (1996). Proc. Natl. Acad. Sci. V.S.A. 93,4683-4687. Takai, T., Li, M., Sylvestre, D., Clynes, R., and Ravetch, J. V. (1994). Cell 76, 519-529. Tanaka, Y., Ardouin, L., Gillet, A., Lin, S.-Y., Magnan, A,, Malissen, B., and Mdissen, M. (1995). lmmunol. Reu. 148, 171-199. Tanchot, C., Lemonnier, F. A,, Perarnau, B., Freitas, A. A., Rocha, B. (1997). Science 276, 2057-2062. Tarakhovsky, A,, Turner, M., Schaal, S., Mee, P. J., Duddy, L. P., Rajewsky, K., and Tybulewicz, L. J. (1995). Nature 374,467-470. Tarlinton, D. M., Corcoran, L. M., and Strasser, A. (1997). Int. Immunol. 9, 1481-1494. Tatsumi, Y., Pena, J. C., Matis, L., Deluca, D., and Bluestone, J. A. (1993).J. Immunol. 151,3030-3041. Thomas, J. H. (1993). Trends Genet. 9, 395-398. Turner, M., Mee, P. J., Costello, P. S., Williams, O., Price, A. A,, Duddy, L. P., Furlong, M. T.,Geahlen, R. L., and Tybulewicz, L. J. (1995). Nature 378, 298-302. Turner, M., Mee, P. J., Walters, A. E., Quinn, M. E., Mellor, A. L., Zamoyska, R., and Tybulewicz, L. J. (1997). Immunity 7, 451-460. Uematsu, Y., Ryser, S., Dembic, Z., Borgulya, P., Krimpenfort, P., Berns, A,, von Bohemer, H., and Steinmetz, M. (1988). Cell 52, 831-841. V,m Kaer, L., Ashton-Rickardt, P. G., Ploegh, H. L., and Tonegawa, S. (1992).Cell 71,12051214. van Oers, N. S. C., Van Boehmer, H., and Weiss, A. (1995).J. Exp. Med. 182, 1585-1590. van Oers, N. S. C., Lowin-Kropf,B., Finlay, D., Connolly, K., and Weiss, A. (1996).Immunity 5,429-436. van Oers, N. S. C., Love, P. E., Shores, E. W., and Weiss, A. (1998).J. lmmunol. 160, 163-170. Vicari, A. P., Bean, AZ. G., and Zlotnik, A. (1996). Int. Immunol. 8, 183-191. Viney, J, L., Dianda, L., Roberts, S. J.. Wen, L., Mallick, C. A., Hayday, A. C., and Owen, M. J . (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 11948-11952. Vonaks, B. M., Chen, H., Haleem-Smith, A,, and Metzger, H. (1997). J. BioZ. Chem. 272,24072-24080. von Freeden-Jeffry, U., Solvason, N., Howard, M., and Murray, R. (1997). Immunity 7, 147-154. Wallace, V. A,, Kawai, K., Levelt, C. N., Kishihara, K., Molina, T., Timms, E., Pircher, H., Penninger, J., Ohashi, P. S., Eichmann. K., and Mak, T.W. (1995). Eur. J. lmmunol. 25, 1312-1318. Wang, B., Levelt, C., Salio, M., Zheng, D., Sancho, J., Liu, C.-P., She, J.. Huang, M., Higgins, K., Sunshine, M.-J., et al. (1995). Int. 1. Immunol. 7, 435-448. Wang, B., Wang, N., Salio, M., Sharpe, A,, Allen, D., She, J., and Terhorst, C. (1998a). J. Exp. Med. 188, 1375-1380. n’ang, B., Wang, N., Whitehurst, C. E., She, J., Chen, J., and Terhorst, C. (1998b). J. Immunol. (submitted) Wang, J.-H., Lirn, K., Smolyar, A,, Teng, M.-K., Liu, J.-H., Tse, A. G. D., Liu, J., Hussey, R. E., Chishti, Y., Thomson, C. T., Sweet, R. M., Nathenson, S. G., Chang, H.-C., Sacchettini, J. C., and Reinherz, E. L. (1998). EMBO J. 17, 10-26. Wang, N., Wang, B., Salio, M., Allen, D., She, J., and Terhorst, C. (1998). Int. Immunol. (in press) Wegener, A. M. K., Letourneur, F., Hoeveler, Z. A., Brocker, T.,Luton, F., and Malissen, B. (1992). Cell 68, 83-95.

148

BERNARD MALISSEN et al.

Wiest, D. L., Burgess, W. H., McKean, D., Kearse, K. P., and Singer, A. (1995). EMBO 1. 14, 3425-3433. Wiest, D. L., Ashe, J. P., Howcroft, T. K., Lee, H.-M., Kemper, D. M., Negishi, I., Singer, D. S., Singer, A., and Abe, R. (1997). Immunity 6, 663-671. Wilson, A,, and MacDonald, R. (1995). Int. Immunol. 7, 1659-1664. Wilson, A,, Held, W., and MacDonald, H. R. (1994).1.Exp. Med. 179, 1355-1360. Xu, Y., Davidson, L., Alt, F. W., Baltimore, D. (1996).Proc. Natl. Acad. Sci. U.S.A.93,21692173. Yamazaki, T., Arase, H., Ono, S., Ohno, H., Watanabe, H., and Saito, T. (1997).]. Immunol 158, 1634-1640. Yasunaga, M.,Adachi, S . , Itoh, N., and Nishikawa, %-I.(1995).Semin. Immunol. 7,185-196. Young, F., Mizoguchi, E., Bhan, A. K., and Alt, F. W. (1997). Immunity 6,23-33. Zambrowicz, B. P., Friedrich, G. A., Buxton, E. C., Lilleberg, S. L., Person, C., and Sands, A. T. (1998). Nature 392, 608-611. Zerrahn, J,, Held, W., and Raulet, D. H. (1997). Cell 88, 627-636. Zhang, R., Alt, F. W., Davidson, L., Orkin, S. H., and Swat,W. (1995).Nature 374,470-473. Zhu, C . , Bogues, M. A., Lim,D.-S., Hasty, P., and Roth, D. B. (1996). Cell 86, 379-389. Zijlstra, M., Bix, M., Simister, N. E., Loring, J. M., Raulet, D. H., and Jaenisch, R. (1990). Nature 344, 742-746. Zlotnik, A., and Moore, T. A. (1995). C u m +in. Immunol. 7, 206-213. Zbiiiga-Plucker,J. C., Schwartz, L., andLenardo, M. J. (1993).J.Exp. Med. 178,1139-1149. This article was accepted for publication on August 28, 1998.

ADVANCES IN IMMUNOLOGY. VOL 72

Inhibitory Pathways Triggered by ITIM-Containing Receptors SllVlA B O U N D AND JEFFREY V. RAVETCH f i e Rockefeller Uniersify, New York, New Yo& 10021

1. introduction

Cell activation triggered by immune receptors, such as the B cell, T cell, or Fc receptor (BCR, TCR, or FcR), display the capacity to undergo fine modulation to adjust to different developmental and environmental conditions. The response to a particular stimulus represents the balance between stimulatory and inhibitory signals and its magnitude will determine the fate of the cell. Thus, the strength of BCR stimulation determines whether a particular cell will proliferate or enter an apoptotic pathway (Nossal, 1994; Rajewsky, 1996; Healy and Goodnow, 1998). Similarly, TCR stimulation can result in either positive or negative selection, determined by the signal intensity transduced through the receptor (Hogquist et al., 1994; Sebzda et al., 1994).Regulation of the immune response also requires the termination of activation signals when the response has met the immediate need. For example, Fc receptor coligation to the BCR by antibodyantigen complexes abrogates B cell proliferation and antibody secretion, terminating the activation response that sets the program in motion (Ravetch, 1994,1997; Daeron, 1997).An important element in this fine-tuning of cell signals is the newly discovered class of inhibitory receptors, which modulate cell responses by increasing thresholds for activation and terminating stimulatory signals. The characterization of several inhibitory receptors in recent years has permitted the identification of a family of receptors that reveal similar characteristics (Vivier and Daeron, 1997;Vely and Vivier, 1997). These receptors are inert when self-aggregated but are able to abolish cellular signals when coligated to stimulatory receptors. Their cytoplasmic domains contain one or more immune receptor tyrosinebased inhibitory motifs (ITIMs), defined by the six-amino acid sequence (ILV)xYxx(LV)(Vivier and Daeron, 1997; Burshtyn et al., 1997; Vely and Vivier, 1997). In a number of cases, ITIM sequences have been shown to be phosphoylated on receptor coligation to create a binding site for Srchomology 2 (SH2) domain-containing cytoplasmic factors that can transmit the inhibitory signal intracellularly (Muta et al., 1994; Fry et al., 1996; Olcese et al., 1996; Burshtyn et al., 1996; Mason et al., 1997; Nakamura et al., 1997; Kuroiwa et al., 1998; Adachi et al., 1998). The ITIM-bearing receptors commonly inhibit activating signals triggered by receptors that contain the immune receptor activation motif 149

Copyright 6 1989 by Academic Press. All rights of reproduction in any fonn reserved. 0065-2776/99 $30.00

150

SILVIA BOLLAND AND JEFFREY V. RAVETCH

(ITAM) in their cytoplasmic tails, such as the BCRs, TCRs, and FcRs. Cell activation mediated by ITAM-containing receptors involves the phosphorylation and activation of several tyrosine kinases and subsequent activation of phospholipase Cy and phosphatidylinositol 3-kinase (PLCy and PI3-K), together leading to the production of phosphoinositol messengers and a sustained increase in cytoplasmic Cazt (Bijsterbosch et al., 1985; Weiss and Littman, 1994;Alberola-Ilaet al., 1997) (Fig. 1A). It has become evident in the past few years that inhibitory receptors use two different strategies to terminate cell activation depending on the type of molecule that is recruited to the phosphorylated ITIM sequences (Gupta et al., 1997; Ono et al., 1997; Vely et al., 1997). Inactivation can be attained by protein dephosphorylation mediated by the tyrosine phosphatases SHP-1 and/or SHP-2, in which case the most proximal events in the activation cascade are abrogated so that Ca2+mobilization is completely abolished (Fig. 1B). A second mode of inhibitory signal utilizes the phosphoinositol phosphatase SHIP to hydrolyze phosphoinositol messengers (Fig. 1C). This type of inhibitory signal does not affect proximal events triggered by the activating receptor, such as the activation of kinases, receptor phosphorylation, or Ca2+release from intracellular stores, but it specifically impedes extracellular Ca2+influx and therefore blocks a sustained increase in cytoplasmic Ca". Analyses of the mechanism of action of FcyRII and KIR, the best studied inhibitory receptors to date, show that they utilize distinct and nonredundant pathways. Although the FcyRII signal is dependent on the phosphoinositol phosphatase SHIP and not the tyrosine phosphatase SHP-1, the KIR signal is dependent on SHP-1 and not SHIP (Gupta et al., 1997; Ono et al., 1997; Vely et al., 1997). Thus, there is fine specificity regarding which events are inhibited by each receptor so that the type of inhibitory receptor engaged in a particular situation will determine the kind of suppression that is achieved. The three molecules recruited by inhibitory receptors, SHP-1, SHP-2, and SHIP, are part of the general regulation of immune receptor activation. They have been found associated with antigen, Fc, growth factor, and cytokine receptors and their absence results in augmented cell activation and proliferation (Tsui et al., 1993; Pani et al., 1995, 1996; Cyster and Goodnow, 1995; Lioubin et al., 1996; Helgason et al., 1998). Thus, they not only function through recruitment to inhibitory receptors, but also via association with stimulatory receptors to set thresholds or to prevent unsolicited cell activation. II. Inhibitory Receptors and Activating Counterparts

The family of inhibitory receptors is expanding rapidly; at least 14 families of receptors have been found that contain one or more ITIM motifs in

INHIBITION BY ITIM-CONTAINING RECEPTORS

151

FIG. 1. Schematic representation of immune receptor signaling pathways. (A) Crosslinking of activating immune receptors leads to ITAM phosphorylation and recruitment of protein kinases via SH2 domain interactions. Membrane association of PIS-kinase results in the production of PIP3 and recruitment of PH domain-containing factors such as Btk and PLCy. Both Src kinases and Btk phosphorylate PLCy. Activated PLCy produces IP3, the release of intracellular Ca2+,and subsequent Ca2+influx from the extracellular medium. (B ) Coengagement of the KIR inhibitory receptor completely abolishes Ca" mobilization triggered by immune receptor cross-linking. KIR-phosphorylated ITIM sequences recruit the tyrosine phosphatase SHP-1, which dephosphorylates receptor ITAMs and associated kinases, abrogating all the proximal events in the activation cascade. (C) Coengagement of

152

SILVIA BOLLAND AND JEFFREY V. RAVETCH

FIG. l-Continued FcyRII inhibitory receptor abrogates Ca2+influx but does not impede Ca2+release from intracellular stores. FcyRII-phosphorylated ITIM recruits SHIP, which hydrolyzes PIP3 into PIP2, and impedes the recruitment to the membrane of PH domain-containing factors such as Btk and PLCy.

their intracellular domains, as listed in Table I. Inhibitory receptors have been identified in all types of hematopoietic cells and also in some nonhematopoietic cells. Structurally, they are single-chain receptors that belong to the immunoglobulin or the lectinlike superfamilies. A few of them are expressed at the surface as homodimers (p70 KIR, Ly-49, CD72) or heterodimers (CD94NKG2A). Details of the different receptors and their ligands have been discussed in previous publications (Orourke et al., 1997; Daeron, 1997; Lanier, 1997,1998a,b; Moretta et al., 1997;Vely and Vivier, 1997). An interesting point is that inhibitory receptors have an activating receptor counterpart that contains a highly homologous extracellular domain and a short cytoplasmic portion that lacks signaling capacity. The transmembrane domains of these activation receptors are characterized by the presence of a charged amino acid, a hallmark of receptors that associate with accessory subunits containing the activation ITAM motif (Vely and Vivier, 1997). At the present time three such accessory subunits have been identified: the FcR y chain, the CD3 chain, and DAP-12 (Kurosakiet al., 1991;Wirthmueller et al., 1992; Olcese et al., 1997; Smith et al., 1998; Lanier et al., 1998a,b).These molecules are structurally related, with a short extracellular sequence containing a cysteine residue that medi-

TABLE I INHIBITORY RECEPTORS Class Ig-SF

Receptof

Ligand

Activating partner

Expression

Ref.

(nl/h)FryRIIB

ITYSLL

IgG complex

FcyRIII

Myeloid and B cells

(h)KIR (p58/p70)

VTYAQL IVYELL VTYAQL VTYAEV VTYAQV VTYAQL ITYAAV IVYAQV VTYAQL ITYADL LTYADL VSYAIL IHYSEL VDYVTL VTYSTL IIYSEV

HLA-A, -B, -C

KAR

NK and T cells

Unhown HLA-G

PIR-A ILTl

Myeloid and B cells All immune cells

Aniigorena et al. (1992), Muta et al. (1994) Colonna and Samaridis (1995), D'Andrea et al. (1995) Kubagawa et al. (1997) Samaridis and Colonna (1997)

Unlolown

-

T and NK cells

Meyaard et al. (1997)

Unknown

-

Myeloid and NK cells

Castells et al. (1994)

Hematopoietic and nonhematopoietic cells B cells

Kharitonenkov et al. (1997)

(m)PIR-B (h)ILT2/3/4/5 (h)LAIR-l (m)gp49B1 (h)SIRPa (mn/h)CD22 (m/h)CD66a

Lectinlike

ITIM sequence

(m)CD5 (m/h)CTLA-4 (m)Ly49A/C/G2 (h)NKG2A/B (di)CD72

-b -I,

VTYSTV VIYSDL ITYAEL ITYADL ITYENV

Growth factor Sialic acid Unknown CD72 CD80, CD86 MHC I HLA-E, -G CD5

SIRPP -

CD66d -

Ly49D/H NKG2C -

Doody et al. (1995)

QZ. (1991)

Neutrophils and embryonic cells T and B cells T cells NK and T cells NK and T cells

Van de Velde et QZ. (1991) Brunet et al. (1987) Yokoyama and Seaman (1993) Chang et ~ 2 (1995) .

B cells

Nakayama et al. (1989)

* m, mouse; 11, human. 'These receptors have no canonical ITIM motif in their cytoplasmic domains, but have been sbown to be inhibitory.

Thompson et

154

SILVIA BOLLAND AND JEFFREY V. RAVETCH

ates the formation of homo- or heterodimers with related accessory subunits, a charged amino acid in the transmembrane domain, and an intracytoplasmic sequence containing one or more ITAM motifs. They function both in receptor assembly and signal transduction. Absence of the accessory subunit results in little or no surface expression of the ligand-binding subunit, whereas the ITAM motif is required for mediating cellular activation on receptor cross-linking. In those cells in which both activating and inhibitory receptors with homologous extracellular domains are present simultaneously,the relative expression level of the two receptors will determine the activation state of the cell. For example, the FcyRIIB/FcyRIII pair is present on mast cells, macrophages, and neutrophils. Both receptors bind immunoglobulin G (IgG) immune complexes with low affinity, so that cells that express low levels of the inhibitory receptor FcyRIIB, such as skin mast cells or alveolar macrophages, are very responsive to the presence of immune complexes, whereas cells that express high levels of FcyRIIB, such as bone marrow-derived mast cells or resident macrophages, are relatively unresponsive (Bonnerot and Daeron, 1994; Castells, 1994; Katz and Lobell, 1995; Daeron, 1997; R. Clynes and J. V. Ravetch, unpublished). In the same way, natural killer (NK) cells can express many combinations of activating or inhibiting receptors, some of which will recognize the same HLA allele (Biassoni et al., 1996). How inhibitory and activating major histocompatability complex (MHC) class I receptors interact to regulate the activity of NK cells is not clear at the moment. At the very least, the function of activating receptors should be to recruit kinases that phosphorylate ITIMs on the inhibitory receptor. The overall cell response in this case seems to be the inhibition of NK cytotoxicity as a result of the dominant signal coming from the inhibitory receptor. The PIR family of surface receptors represents another example of this dichotomy of activation and inhibitory receptors expressed on the same cells. They include the activation molecule PIR-A, which associates with the FcR y chain as its activatory subunit, and the homologous PIR-B inhibitory receptor (Maeda et al., 1998a; Kubagawa et al., 1997, 1998; Blery et al., 1998). The ligands for these molecules have not been identified, but the presence of both activation and inhibitory molecules with highly homologous extracellular domains suggests that they function in concert to set thresholds for lymphoid and myeloid cells. 111. FcflI-Mediated Inhibitory Signal

FcyRIIB is a low-affinity receptor for the Fc portion of immunoglobulin G, with IgG immune complexes as its natural ligands (Ravetch and Kinet, 1991). It is a widely expressed receptor, present on all hematopoietic

INHIBITION BY ITIM-CONTAINING RECEPTORS

155

lineages with the exception of red blood cells and NK cells. FcryRII was the first receptor known to abrogate cellular responses when coligated to an activating receptor. Its function was initially characterized in B cells, where it modulates antibody production, lymphokine release, and cell proliferation. It was known for a long time that immune complexes are strong inhibitors of humoral immune responses (Chan and Sinclair, 1971). Initial evidence for the involvement of FcyRII in this immune complexmediated regulation of B cell activationwas provided by Phillips and Parker (1983, 1984), who established that although F(ab')e fragments of anti-Ig antibodies (Abs) induced B cell proliferation, intact IgG Abs did not. This experiment implicated an FcR, later defined as FcyRIIB, the only IgGFc binding molecule on B cells, in the inhibitory effect of intact IgG. The ITIM sequence was first defined after the identification of a 13amino acid sequence (EANTITYSLLKH) in the cytoplasmic domain of FcyRII that is required for negative signaling (Amigorena et al., 1992). Muta et al. (1994) formally demonstrated that this ITIM sequence is sufficient for inhibitory signaling when it is expressed in an inert receptor context and that a tyrosine included in the sequence is essential for FcyRIImediated inhibitory signaling. On BCR coligation, the ITIM on the FcyRII cytoplasmic portion is phosphorylated on Tyr-309 by the BCR-activated 1p kinase (Wang et al., 1996; Malbec et al., 1998). This phosphorylation event creates a domain recognized by SH2-containing molecules and suggested that the mechanism of inhibition by FcyRIIB is mediated by the recruitment of such a molecule to the membrane. The molecular mechanism of the FcyRII inhibitory signal has been best studied in B cell lines. Antigen-mediated aggregation of the BCR promotes several distinct intracellular pathways, yet there is fine specificity as to which ones are subject to inhibition by FcyRII. BCR cross-linking induces the tyrosine phosphorylation of many proteins, including associated receptors CD19 and CD22, the BCR signaling chains Iga and I& the kinases Syk, lyn, PI3-K, and Btk, and other factors such as PLCy, Shc, Grb2, and Vav (Bijsterbosch et al., 1985; Tuveson et al., 1993; Weng et al., 1994; Takata et al., 1994, 1995; Zhang et al., 1995; Doody et al., 1995; Takata and Kurosaki, 1996; Smit et al., 1996; Yamanashi et al., 1997; Sugawara et al., 1997; Kurosaki, 1997). Activated PLCy generates inositol3-phosphate (IP3) and triggers Ca2+ release from intracellular stores. Depletion of Ca2+from intracellular stores stimulates Ca" influx from the extracellular medium by an unknown mechanism. Altogether, the sustained increase in cytoplasmic Ca2+results in the transcriptional activation that leads to proliferation and/or differentiation of the cells (Dolmetsch et al., 1997). Simultaneous coligation of FcyRII with the BCR leads to a dominant negative signal that inhibits some of these events. Although there is no

156

SILVIA BOLLAND AND JEFFREY V. RAVETCH

change in phosphorylation of the activated kinases and CaZt release from intracellular stores, there is a blockage at the Ca2+influx stage with consequent arrest of transcriptional activation and cell proliferation and/or differentiation (Choquet et al., 1993; Muta et al., 1994). Because FcyRII inhibition does not completely abrogate BCR-triggered Ca" mobilization, it might selectively inhibit transcription pathways that require high Ca2' levels for activation, For example, it has been reported that the amplitude and duration of Ca2' signals in B cells control differential activation of NF-KB, JNK1, and NFAT (Dolmetsch et al., 1997). Selective inhibition of transcription pathways could be a way to allow the necessary stimulation for B cell survival without inducing cell proliferation. In vitro, the FcyRII-phosphorylated ITIM can bind to three different SH2-containing molecules: the tyrosine phosphatases SHP-1 and SHP-2, and the phosphoinositol polyphosphate phosphatase SHIP (D'Ambrosio et al., 1995; Ono et al., 1996). The role of these three phosphatases in FcyRII-mediated signaling in uivo has been the subject of some controversy. SHP-1 has been proposed as the mediator of FcyRII inhibitory signaling via its direct effect on the phosphorylation of CD19, a BCRassociated coactivator (Hippen et al., 1997). CD19 dephosphorylation by SHP-1 has been proposed as a mechanism that would lead to a decrease in PI3-K recruitment and abrogation of cell activation (Tuveson et al., 1993). Several lines of evidence argue against this conjecture. First, bone marrow-derived mast cells from moth-eaten mice (melme),which are naturally deficient in SHP-1, still show FcyRII-mediated inhibition of FcRtriggered degranulation (Ono et al., 1996). Moreover, a B cell line derived from m e l m lymphocytes retains FcyRII-mediated inhibition of BCRtriggered activation. Experiments using this cell line established that SHP1 is not required for the FcyRII-evoked decrease in CD19 tyrosine phosphorylation (Nadler et al., 1997). Although it is still formally possible that the tyrosine phosphatase SHP-2, which binds in uitro to the phosphorylated ITIM, can substitute for SHP-1 in this role, this is unlikely. FcyRII signaling is functional in the chicken B cell line DT40, deficient for both SHP-1 and SHP-2 (S. Bolland and J. V. Ravetch, unpublished). Altogether, these genetic studies provide strong evidence against a prominent role for tyrosine phosphatases in the mechanism of FcyRII inhibition of Ca2' signals. Out of the three SH2-containing phosphatases capable of interaction with the phosphorylated ITIM of FcyRIIB, SHIP seems to be the molecule preferentially recruited to the cytoplasmic domain of FcyRII in uiuo, as demonstrated by immunoprecipitation following coligation to the BCR or to FcsRI (Ono et al., 1996, 1997; Fong et al., 1996). Genetic proof of the significance of this association came from a DT40 SHIP knockout cell line, which is impaired for FcyRILmediated inhibition of BCR-triggered Ca2+

INHIBITION BY ITIM-CONTAINING RECEPTORS

157

mobilization, clearly demonstrating that SHIP protein is necessary for this pathway (Ono et al., 1997). As will be summarized below, SHIP-deficient murine B cells display a similar phenotype, further confirming the biological relevance of the association of SHIP with FcyRIIB in mediating the inhibitory effect of this receptor (Liu et al., 1998). Characterization of mice with disruption of the FcyRIIB gene has helped to further elucidate the biological function of this inhibitory receptor (Takai et al., 1996).Developmental defects were not noted, nor were autoantibodies detected. Antibody responses are enhanced in these mice, which have serum antibody titers 3-10 times higher following immunization with both T cell-dependent and T cell-independent antigens. This response confirms FcyRII as a modulator of B cell antibody production. However, based on in vitro studies, the magnitude of this enhancement is not as substantial as was expected, a fact that can be explained in several ways. FcyRII is expressed in cells other than B cells, which are involved in modulating the antibody response. For example, follicular dendritic cells ( FDCs) express FcyRII as their sole FcR, functioning in the retention of antigen as immune complexes. This positive role in shaping the B cell repertoire through positive selection of high-affinity BCRs could therefore counterbalance the negative effect of FcyRII on B cells. Furthermore, FcyRII might modulate responses of naive B cells but not cells that establish the extent of antibody responses, such as memory B cells or plasma cells. The FcyRIIdeficient mice display a defect in the primary antibody response, with elevated serum titers of IgM as well as IgG, supporting the hypothesis that this receptor has a pleiotropic role in regulating the antibody response. Obviously, the presence of additional regulatory pathways still active in the FcyRII-deficient B cells cannot be dismissed. In addition to the inhibitory function on B cell activation, FcyRII can also inhibit activation signals in many other hematopoietic cells where it is expressed. It has been shown to inhibit FcR-triggered mast cell degranulation when coligated to FcyRIII or FceRI (Daeron et al., 1995a,b).Accordingly, mice deficient in FcyRII are hyperresponsive for peripheral cutaneous anaphylaxis (PCA) to IgG complexes (Takai et al., 1996). These mice also seem to exhibit enhanced responses in various models of inflammation, such as immune complex-induced alveolitis, collagen-induced arthritis, and systemic anaphylaxis, confirming that FcyRII functions to set thresholds for immune complex stimulation in systems in which FcyRIII is the primary effector (Clyneset al., 1998; Ujike et aZ., 1999;Yuasa et al., 1999).Moreover, FcyRIIB-deficient mice display an enhanced sensitivity to IgE-mediated systemic anaphylaxis, indicating an unexpected interaction between FcERI and FcyRIIB in setting thresholds for IgE-triggered mast cell activation. This enhanced IgE-mediated response may result from the ability of IgE

158

SILVIA BOLLAND AND JEFFREY V. RAVETCH

to interact with FcyRII with low affinity, or it may result from competition between FceRI and FcyRII for SHIP, thereby altering the threshold for stimulation. As described below, the SHIP-deficient mouse displays a similar enhancement of IgE signaling (Helgason et al., 1998). IV. Mechanism of Inhibition by SHIP

The SH2-containing inositol 5-polyphosphate phosphatase SHIP has been recognized as a general regulator of immune receptor activation, in addition to its role in mediating FcyRII inhibitory signaling. SHIP is widely expressed in all hematopoietic lineages and functions to set thresholds in a variety of systems such as growth factor-induced cell proliferation and antigen receptor or FcR stimulation (Damen et al., 1996; Lioubin et al., 1996; Kavanaugh et al., 1996;Ono et al., 1996). It is a 145-kDa protein with multiple domains that can potentially signal through several intracellular pathways and provide diverse protein interactions. It contains an aminoterminal SH2 domain that binds to phosphotyrosine receptors and mediates recruitment to the membrane. This domain has been reported to interact with the phosphorylated ITIM from FcyRII, as well as the phosphorylated ITAMs from FcERI p chain and TCR-J chain (Crowley et al., 1996; Osborne et al., 1996; Kimura et al., 1997a). In addition, SHIP contains two tyrosine residues that, in the phosphorylated form, can bind phosphotyrosine-binding (PTB) domain-containing molecules, such as the adapter molecule Shc or the tyrosine phosphatase SHP-2 (Liu et d., 1994, 1997; Kavanaugh et al., 1995; Lamkin et al., 1997; Pradhan and Coggeshall, 1997). Its catalytic domain contains inositol polyphosphate 5-phosphatase motifs; this enzymatic activity has been observed in immunoprecipitates of SHIP alone, or SHIP associated with Shc or FcyRII (Jefferson and Majerus, 1995; Damen et al., 1996; Lioubin et aZ.,1996; Ono et al., 1996). Finally, the proline-rich carboxy-terminaldomain of SHIP could potentially interact with any of the reported SH3 domain-containing proteins. SHIP has been found to be tyrosine phosphorylated and associated with Shc following activation of receptors for numerous cytokines including erythropoietin (Epo), c-Kit, interleukin 3 (IL-3), IL-2, granulocyte/macrophage colony-stimulating factor (GM-CSF), and M-CSF (Liu et al., 1994, 1996; Damen et al., 1996; Hunter and Avalos, 1998). SHIP is also phosphorylated following cross-linking of antigen receptors in B and T cells or following activation of FceRI in mast cells (Chacko et ab., 1996; Osborne et al., 1996; Pradhan and Coggeshall, 1997). SHIP phosphorylation can occur as a consequence of its recruitment nearby an activating receptor. It is also possible that SHIP is constitutively associated with some receptors, so that upon receptor cross-linking SHIP becomes phosphorylated, associatedwith

INHIBITION BY ITIM-CONTAINING RECEPTORS

159

Shc, and at the same time dissociated from the receptor. In this context, SHIP would function as a gatekeeper to avoid unintentional activation. So far, there is no clear evidence to show that phosphorylation affects SHIP activity, or its recruitment to the membrane. Nevertheless, SHIP has been proposed to be a down-modulator of cell activation because its overexpression causes inhibition of cell growth (Lioubin et al., 1996). More detailed investigations have arisen from the analysis of the SHIPmediated inhibitory signal triggered by FcyRII coligation. In this case, SHIP protein is sufficient to abrogate the Ca2+influx triggered by BCR cross-linking, as evidenced by a chimeric receptor that substitutes the FcyRII cytoplasmic portion for SHIP and is competent for inhibitory signaling (On0 et al., 1997). Mutation of the inositol phosphatase motif in this construct abolishes the inhibitory function, implying that the enzymatic activity of SHIP is necessary for inhibition of Ca" influ. SHIP catalyzes the hydrolysis of the 5'-phosphate of two specific substrates: inositol1,3,4,5tetrakisphosphate (IP4) and phosphatidylinositol3,4,5-trisphosphate (PIP3 or PIns3,4,5P3) (Damen et al., 1996; Lioubin et al., 1996). IP4 is a cytosolic inositol phosphate that has been shown to activate certain plasma membrane Ca" channels (Luckhoff and Clapham, 1992). PIP3 is a membranebound phosphoinositol that has been observed to appear after stimulation of virtually every receptor type (Traynor-Kaplan et al., 1988; Auger et al., 1989; Coughlin et al., 1989; Backer et al., 1992; Carpenter and Cantley, 1996; Toker and Cantley, 1997). Thus, both the hydrolysis of IP4 and of PIP3 by SHIP could have an inhibitory effect on receptor-triggered Ca" mobilization. PIP3 has already been observed to be involved in immune receptor-triggered activation signals, because it is originated from PIns4,5P2 by PI3-K, an enzyme recruited and activated by ITAM crosslinking (Tuveson et al., 1993; Pleiman et al., 1994; Gold and Aebersold, 1994; Ward et al., 1996; Aagaard-Tillery and Jelinek, 1996). PIP3 has been proposed to interact with proteins that contain a plecstrin-homology (PH) domain to promote their association with the plasma membrane and place them in proximity to their substrates (Lemmon et al., 1996). Examples of such PH domain-containing factors that have been shown to bind PIP3 in vitro are PKC, PDK1, Grpl, PLCy, and the Btk/Itk/Tec kinases (Nakanishi et al., 1993; Toker et al., 1994; Kojima et al., 1997; Rameh et al., 1997). SHIP, by reducing the levels of PIP3, can prevent the recruitment to the membrane of these molecules, and in this way can abrogate cellactivating signals. The necessity of PH domain-mediated recruitment to the membrane has been extensively analyzed in the case of Bruton's tyrosine kinase (Btk). Spontaneous mutations in this protein are responsible for X-linked agammaglobulinemia in humans and X-linked immunodeficiency in mice (Smith

160

SILVIA BOLLAND AND JEFFREY V. RAVETCH

et al., 1994; de Weers et al., 1994), in both cases demonstrating that Btk is necessary for B cell development and activation. Some of these spontaneous mutations alter the PH domain of Btk such that it is no longer able to bind PIP3. These mutations generate cells that are compromised in mediating BCR-triggered cellular activation (Salimet al., 1996; Hyvonen and Saraste, 1997). Conversely, mutations in the Btk PH domain that lead to increased membrane association display enhanced Ca2+mobilization on BCR cross-linking (Li et al., 1995). When overexpressed, Btk enhances the sustained increase in cytoplasmic Ca2+following BCR cross-linking, affecting mostly the Ca2+influx step (Bolland et al., 1998; Fluckiger et al., 1998). This enhancement in Ca2+mobilization is even larger when Btk is expressed as a membrane chimera, probably because of the proximity to its substrate. Therefore, PIPS-dependent membrane recruitment of Btk is essential to maintaining the sustained Ca" responses required for B cell activation. SHIP enzymatic activity, by reducing the PIP3 levels, can impair Btk recruitment to the membrane and the sustained Ca2+response. In fact, the inhibitory effect of FcyRII or SHIP can be suppressed by expression of Btk as a membrane-associated chimera, and pathways that result in decreased PIP3 levels reduce Btk membrane association and accordingly reduce Ca" mobilization. Meanwhile, cells deficient in SHIP show a higher level of Btk associated with the membrane and a concomitant enhancement of Ca2' mobilization following BCR cross-linking (Bolland et al., 1998). Although Btk is primarily expressed in B and mast cells, additional tyrosine kinases with significant homology to Btk are expressed in both hematopoietic and nonhematopoietic cells. Examples include Itk, with expression restricted to T, NK, and myeloid cells, and the kinase Tec, present in myeloid cells and nonimmune tissues (Siliciano et al., 1992). Itk has been observed to be recruited to the membrane through PH domain interactions following TCR activation (August et al., 1997). Because SHIP is widely expressed in myeloid and lymphoid cells, it is conceivable that its inhibitoly function regulates membrane association of Btk homologs in the same manner as observed for Btk in B cells. In addition to attenuating the recruitment of Tec kinases, SHIP could also have an effect on PLCy PHmediated membrane recruitment and activation,thereby reducing IP3 production. Membrane-associated Btk could affect Ca" influx directly by acting on the Ca2+channel, or indirectly through phosphorylation of an intermediate. Several proteins have been reported to associate with Btk: PLCy, G protein /3/y subunits, protein kinase C, and two novel proteins of unknown function, BAP-135 and Sab (Langhans-Rajasekaranet al., 1995; Yang and Desiderio, 1997; Yao et al., 1997; Matsushita et al., 1998). Some of these factors could be substrates of the Btk kinase activity: tyrosine phosphorylation of PLCy2,

INHIBITION BY ITIM-CONTAINING RECEPTORS

161

for example, has been shown to be partly mediated by Btk. In DT40 B cells deficient in Btk, PLCy phosphorylation is less intense following BCR cross-linking.Conversely, PLCy phosphorylation is enhanced when cotransfected with Btk and PI3-K (Scharenberget al., 1998). Btk-mediated stimulation of PLCy activity should result in increased IP3 levels, complete Ca2+depletion of intracellular stores, and concomitant enhancement of Ca" influx. Consistent with this hypothesis, IP3 production triggered by BCR cross-linking in DT40 cells is completely dependent on Btk (Takata and Kurosaki, 1996). Remarkably, FcyRII-mediated inhibitory signaling does not completely abrogate PLCy or Btk activity, because Ca" release from intracellular stores seems unchanged. This suggests that SHIP activity can prevent Btk and/or PLCy recruitment to the membrane but does not completely abrogate their activity. V. Inhibition by KIR Receptors

Experiments on the rejection of autologous tumors in mice provided the first indication that NK cells eliminate cells that have lost expression of MHC class I molecules (Karre et al., 1986). Subsequent experiments showed that NK cells recognize MHC class I molecules with allele specificity and clonal diversity (Moretta et al., 1990). This finding predicted the existence of multiple NK cell receptors that recognize polymorphic MHC determinants and transmit an inhibitory signal to prevent NK cell-mediated killing. Since then, several families of inhibitory receptors with specificity for MHC class I molecules have been identified, all of them containing ITIM sequences in their cytoplasmic tails. Human NK cells express p58 and p70 KIRs, which contain two and three Ig domains in their extracellular portion, respectively (Moretta et al., 1993; D'Andrea et al., 1995; Colonna and Samaridis, 1995). These cells also express the lectinlike family of receptors NKG2, including the inhibitor receptor NKGSA, which forms heterodimers with CD94 (Chang et al., 1995; Brooks et al., 1997; Carretero et al., 1997). The murine form of NK inhibitory receptor is the lectinlike Ly49 family (Karlhofer et al., 1992; Yokoyama and Seaman, 1993). Each of the NK cell inhibitory receptors abrogates cytotoxicity when coligated with ITAM-dependent receptors such as FcyRIII or any of the activating isoforms of p50 (KAR), Ly49, and NKG2 (Fry et al., 1996; Houchins et al., 1997). Signals triggered by these receptors are very similar to the antigen receptor-mediated activation of B or T cells. They engage PLCy to produce IP3, which stimulates Ca2+mobilization from intracellular stores and subsequent extracellular Ca" influx (Azzoni et al., 1992; Ting et al., 1992; Kaufman et al., 1995). Simultaneous cross-linking of the inhibitory receptors results in the tyrosine phosphorylation of the

162

SILVIA BOLLAND AND JEFFREY V. RAVETCH

cytoplasmic tail of KIRs (Campbell et at., 1996; Fry et at., 1996; Olcese et al., 1996; Burshtyn et al., 1996). Human KIRs contain two ITIM sequences in their cytoplasmic domain. Each of these ITIMs, in its phosphorylated form, can bind in vitro to SHP-1 and SHP-2, but not to SHIP (Vely et al., 1997). It follows that SHIP has additional sequence requirements other than the ITIM motif WxYxxL. It also seems that in most cases SHP1 prefers the sequence VxYxxL. The N-terminal ITIM (VTYAQL) binds to both of the SHP-1 SH2 domains, and it does it more efficiently than the C-terminal ITIM (IWELL). The N-terminal ITIM has been found to be sufficient for the inhibitory signal in deletion studies (Fry et al., 1996). Out of the two SH2 domains of SHP-1, the N-proximal one not only acts as a recruitment unit, but also as a regulator of SHP-1 phosphatase activity (Pei et al., 1994, 1996). Indeed, phosphorylated ITIMs have been reported to activate the SHP-1 tyrosine phosphatase function (D’Ambrosio et al., 1995). It remains to be determined if both ITIMs in KIR simultaneously bind the two SHP-1 SH2 domains in physiological conditions, or whether SHP-1 and SHP-2 can conjointly bind the KIR cytoplasmic tail. Evidence for a definitive role of SHP-1 in KIR-mediated inhibitory signaling came from experiments showing that overexpression of a dominant negative SHP-1 mutant in NK cell clones prevents MHC class Imediated inhibition of NK cell lysis (Burshtyn et al., 1996). This point was genetically tested by the analysis of DT40 cells deficient in SHP-1. A chimeric receptor containing the cytoplasmic portion of human KIR (p58) was shown to inhibit BCR-triggered activation in DT40 wild-type and SHIP-’- cells, but not in SHP-l-’- cells (On0 et al., 1997). In these experiments, a chimeric receptor containing SHP-1 in the cytoplasmic portion was found to be sufficient to deliver an inhibitory signal and dependent on the phosphatase catalytic domain. So far, the direct substrates of SHP1 activity have not yet been definitively identified. Studies of target cells protected from NK lysis by expression of KIRs have detected a reduction in CD3 chain and PLCy phosphorylation and the absence of PIP2 hydrolysis, all of which abrogate Ca2+mobilization completely (Kaufman et al., 1995; Binstadt et al., 1996; Blery et at., 1997). Other factors that have been found associated with SHP-1 are ZAP-70, Lck, and Src kinases (Plas et al., 1996; Raab and Rudd, 1996; Somani et al., 1997). It is likely that SHP-1 acts to dephosphorylate, in a nonspecific manner, several of the factors involved in NK activation signaling. Whether the signal is completely abrogated will depend on the time required to recruit and activate SHP-1, which in some cases will allow an initial spike of activation. Many inhibitory receptors other than KIRs seem to recruit SHP-1 and/ or SHP-2 for signaling. Phosphorylated ITIMs from gp49B1, CD22, CD66, CD72, Ly-49, ILT-2, ILT-3, LAIR-1, NKG2-A, and PIR-B have been

INHIBITION BY ITIM-CONTAINING RECEPTORS

163

found associated with SHP-1 and SHP-2 (Doody et al., 1995; Mason et al., 1997; Nakamura et al., 1997; Blery et al., 1998; Ledrean et al., 1998; Adachi et al., 1998; Carretero et al., 1998; Maeda et al., 199813).Phosphorylated ITIMs from gp49B1 and PIR-B also bind SHIP, although their inhibitory function seems unaffected in the DT40 SHIP-’- cells (Kuroiwa et al., 1998; Maeda et al., 199813). VI. Lessons from SHIP, SHP-1,and SHP-2 Knockout Mice

The naturally occumng mutation of the SHP-1 gene is the primary gene defect in moth-eaten (me)mice (Tsui et al., 1993; Shultz et al., 1993). These mice suffer from severe combined immunodeficiency in association with an autoimmune syndrome, and usually die of pulmonary complications within the first weeks of life. The me phenotype is characterized by extensive neutrophilic infiltration of dermal tissues and accumulation of macrophages and granulocytes in the lung, with markedly reduced lymphocyte populations in the bone marrow. The development of systemic autoimmunity in these mice can be explained as a consequence of an abnormal expansion of the B-1 subset that results in the production of autoantibodies. In addition to these obvious abnormalities, in vitru experiments have shown that bone marrow macrophages from me mice grow independently of CSF1, and that B and T cells are hyperresponsive to antigen cross-linking. In general, this phenotype is consistent with the absence of a major regulator of immune cell signaling that controls the growth and development of a large variety of hematopoietic cells. Because SHP-1 has been observed associated with several cytokine and antigen receptors (Yi et al., 1993b; Klingmuller et al., 1995; Pani et al., 1995, 1996; Konkozlowski et al., 1996; Paulson et al., 1996; Lopez et al., 1997; Yu et al., 1998; Kozlowski et al., 1998), it is likely that the pleiotropic phenotype in me mice results from multiple primary cell defects. The phenotype of the SHIP-deficient mouse is remarkably similar to the moth-eaten phenotype. Absence of SHIP results in a myeloproliferative-like syndrome, with profound splenomegaly and massive myeloid cell accumulations in the lungs that lead to death at approximately 10 weeks of age (Helgason et al., 1998). In addition, peripheral blood from SHIP-’mice contains an increased number of circulating monmytes and mature neutrophils, with decreased lymphocyte counts. In vitro, bone marrow progenitors exhibit an enhanced sensitivity to GM-CSF and IL-3, and mast cells from SHIP-’- mice are hyperresponsive to Fc receptor activation. Also, SHIP-deficient DT40 B cells are hyperresponsive to BCR stimulation, confirming SHIP as a general regulator of B cell activation (Bolland et al., 1998). The lymphopenia observed in the SHIP knockout mouse could

164

SILVIA BOLLAND AND JEFFREY V. RAVETCH

then be explained because of excessive signaling through the BCR that negatively selects SHIP-/- cells. Accordingly, SHIP-deficient B cells obtained by RAG-’- blastocyst complementation are hyperresponsive to BCR activation and show abnormal differentiation (Liu et al., 1998). The similarities between SHIP- and SHP-l-deficient mice suggest that these two molecules regulate equivalent activating signals in the same cell types. Both are widely expressed in the immune system, are recruited by SH2 domain interactions, and have been found associated with the same stimulatory receptors, such as IL-3, c-Kit, CSF, FceRI, TCR, BCR, or erythopoietin receptor (Yi and Ihle, 1993;Yi et al., 1993,1995; Klingmuller et al., 1995; Pani et al., 1995; Lorenz et al., 1996; Kimura et al., 1997b),This raises the question of whether SHIP and SHP-1 function independently, in a redundant manner, or with complementary functions. The fact that deficiencies in either of them result in a marked phenotype argues against redundant functions. Most likely, SHIP and SHP-1 act simultaneously to prevent unintentional cell activation,but neither of them alone at physiological levels is sufficient to provide full repression. Coengagement of a particular inhibitory receptor might specifically recruit SHIP or SHP-1, so that the local concentration of that factor is increased and full inhibition of specific pathways is achieved. Despite the similar phenotypes in deficient mice, SHIP and SHP-1 modulate activation responses in significantlydifferent ways, suggesting that in a normal physiological context, the type of inhibitory signal will be important. Because the perturbations in these molecules result in pleiotropic effects, conditional deficiencies will need to be generated and analyzed to determine the consequence of each pathway in modulating activation signals. The phenotype of SHIP and SHP-1 knockout mice is reminiscent of the phenotype of lyn-deficient mice (Table 11).These mice have decreased numbers of mature peripheral B cells, greatly elevated serum IgM and IgA, and production of autoantibodies (Hibbs et al., 1995). I n uitro, Iyndeficient splenic B cells are hyperresponsive to BCR stimulation (Nishizumi et al., 1995; Wang et al., 1996; Chan et al., 1997). This pattern suggests a role for 1yn in the negative regulation of B cell activation. This function could be performed by 1yn by phosphorylating inhibitory receptors such as CD22 or FcyRII (Malbec et al., 1998; Smith et al., 1998; Nishizumi et al., 1998; Cornall et al., 1998), although the severe phenotype observed in the absence of lyn relative to the CD22 or FcyRII deficiencies indicates that other inhibitory signals must be dependent on lyn function. The tyrosine phosphatase SHP-2 shares domain structure and considerable overall sequence identity (55%) with SHP-1, but its function in inhibitory signaling is not as clear as for SHP-1. Although SHP-1 expression is mostly limited to hematopoietic cells, SHP-2 is ubiquitously expressed. Early studies indicated that SHP-2, as well as its Drosophila homolog

165

INHIBITION BY ITIM-CONTAINING RECEPTORS

TABLE I1 KNOCKOUTMICE Viability Cellularity Lung (monocytesl neutrophils) Spleen Peripheral B cells Monocyteslneutrophils Peritoneal (B1 cells) Immune function Autoantibodies Serum IgM In vitro stimulation Anti-IgM IgE CSF/ILS FcyRII

1y”-/-

SHIP-’(12 weeks)

SHIP-1-’(2 weeks)

(24 weeks)

FcyRIIP (normal)

t t

t

n.d:

n.d.

T

t

Normal

.1

t

Normal n.d. Normal

.1

t

n.d.

n.d. n.d. n.d.

t t

t

Normal

t

t

n.d.

t t

t 1

t t

t t 1

t

Normal

n.d.

n.d.

1

.1

md.. not determined.

Corkscrew (CSW), played a positive role in growth factor signaling (Tang et al., 1995). A targeted deletion of the SH2 domain of SHP-2 leads to an embryonic lethality at midgestation in homozygous mutant mice, and reduced hematopoietic activity in differentiated SHP-2-’- ES cells (Qu et al., 1997). These results suggest that SHP-2 function is not restricted to immune cells, and that it is essential for signaling in tissue development. In this context, SHP-2 could positively regulate kinases by dephosphorylating inhibitory tyrosine phosphorylation sites, as has been shown for SHP-1 and Src kinases (Somani et al., 1997). An inhibitory function of SHP-2 could also explain embryonic lethality, if an excessive signal from growth factor receptors due to the absence of the inhibitory molecule SHP-2 is detrimental for cell development. Recent studies on the inhibitory receptors SIRPa, CTLA-4, and PIR-B have shown that their inhibitory signals are at least partially dependent on SHP-2 function (Marengere et al., 1996; Kharitonenkov et al., 1997; Maeda et al., 199813). These results point to an inhibitory role for SHP-2, consistent with its association, together with SHP-1, with most of the ITIM sequences in immune receptors. VII. Conclusions

Inhibitory receptors were unknown until quite recently; since their discovery the complexities of their actions have begun to be appreciated. It

166

SILVIA BOLLAND AND JEFFREY V. RAVETCH

is now understood that inhibitory receptors share ligand specificity with activating counterparts and work in concert to determine threshold levels for cellular responses. The rapidly expanding family of inhibitory receptors raises the intriguing question as to their function in regulating the immune response. To date, well-defined ligands and clear physiological functions have been elucidated for the Fc receptor system and the NK inhibitory system. With these two molecules as prototypes of the family, significant differences have been observed in their mechanism of conferring inhibitory signals. The outcome of these pathways is either to abrogate all calcium responses or to modulate calcium responses by blocking influx and thereby impeding sustained calcium responses. At present, the physiological significance of this fine regulation of calcium is not apparent. Future studies will need to determine the functional consequences of engaging these discrete pathways as well as the ligands involved.

ACKNOWLEDGMENTS The authors wish to acknowledge the support of the National Institutes of Health and the S.L.E. Foundation of New York. The contribution of Raphael Clynes and Toshi Takai, whose unpublished results have been discussed in this review, is greatly appreciated. We are grateful to the members of the laboratory for helpful discussions and critical comments.

REFERENCES Aagaard-Tillery, K. M., and Jelinek, D. F. (1996). Phosphatidylinositol 3-kinase activation in normal human B lymphocytes. 1.lmmunol. 156,4543-4554. Adachi, T., Flaswinkel, H., Yakura, H., Reth, M., and Tsubata, T. (1998).The B cell surface protein CD72 recruits the tyrosine phosphatase SHP-1 upon tyrosine phosphorylation. I. Immunol. 160,4662-4665. Alberola-Ila, J., Takaki, S., Kemer, J. D., and Perlmutter, R. M. (1997). Differential signaling by lymphocyte antigen receptors I . Annu. Rev. lmmunol. 15, 125-154. Amigorena, S., Bonnerot, C., Drake, J. R., Choquet, D., Hunziker, W., Guillet, J. G., Webster, P., Sautes, C., Mellman, I., and Fridman, W. H. (1992). Cytoplasmic domain heterogeneity and functions of IgG Fc receptors in B lymphocytes. Science 256, 18081812. Auger, K. R., Serunian, L. A., Soltoff, S. P., Libby, P., and Cantley, L. C. (1989). PDGFdependent tyrosine phosphorylation stimulates production of novel polyphosphoinositides in intact cells. Cell 57, 167-175. August, A,, Sadra, A., Dupont, B., and Hanafusa, H. (1997). Src-induced activation of inducible T cell kinase (ITK) requires phosphatidylinositol 3-kinase activity and the Pleckstrin homology domain of inducible T cell kinase. Proc. Natl. Acad. Sci. U.S.A. 94, 11227-11232. Azzoni, L., Kamoun, M., Salcedo, T. W., Kanakaraj, P., and Perussia, B. (1992). Stimulation of Fc gamma RIIIA results in phospholipase C-gamma 1 tyrosine phosphorylation and p561ck activation. J. Exp. Med. 176, 1745-1750. Backer, J. M., Myers, M. J., Shoelson, S. E., Chin, D. J., Sun, X. J., Miralpeix, M., Hu, P., Margolis, B., Skolnik, E. Y., Schlessinger, J., et al. (1992). Phosphatidylinositol3’-kinase is activated by association with IRS-1 during insulin stimulation. EMBOJ. 11,3469-3479.

INHIBITION BY ITIM-CONTAINING RECEPTORS

167

Biassoni, R., Cantoni, C., Falco, M., Verdiani, S., Bottino, C., Vitale, M., Conte, R., Poggi, A., Moretta, A., and Moretta, L. (1996). The human leukocyte antigen (HLA)-C-specific “activatory”or “inhibitory” natural killer cell receptors display highly homologous extracellular domains but differ in their transmembrane and intracytoplasmic portions. J . Exp. Med. 183,645-650. Bijsterbosch, M. K., Meade, C. J., Turner, G. A,, and Klaus, G. G. (1985). B lymphocyte receptors and polyphosphoinositide degradation. Cell 41, 999-1006. Binstadt, B. A., Brumbaugh, K. M., Dick, C. J., Scharenberg, A. M., Williams, B. L., Colonna, M., Lanier, L. L., Kinet, J. P., Abraham, R. T., and Leibson, P. J. (1996). Sequential involvement of Lck and SHP-1 with MHC-recognizing receptors on NK cells inhibits FcR-initiated tyrosine kinase activation. Immunity 5, 629-638. Blery, M., Delon, J., Trautmann, A., Cambiaggi, A., Olcese, L., Biassoni, R., Moretta, L., Chavrier, P., Moretta, A., Daeron, M., and Vivier, E. (1997). Reconstituted killer cell inhibitory receptors for major histocompatibility complex Class I molecules control mast cell activation induced via immunoreceptor tyrosine-based activation motifs. J. Biol. Chem. 272,8989-8996. Blery, M., Kubagawa, H., Chen, C. C., Vely, F., Cooper, M. D., and Vivier, E. (1998). The paired Ig-like receptor PIR-B is an inhibitory receptor that recruits the protein-tyrosine phosphatase SHP-1. Proc. Nut. Acad. Sci. U.S.A. 95, 2446-2451. Bolland, S., Pearse, R. N., Kurosaki, T., and Ravetch, J. V. (1998). SHIP modulates immune receptor responses by regulating membrane association of Btk. Immunity 8, 509-516. Bonnerot, C., and Daeron, M. (1994). Biological activities of murine low-affinityFc receptors for IgG. lnimunomethods 4, 41-47. Brooks, A. G., Posch, P. E., Scorzelh, C. J., Borrego, F., and Coligan, J. E. (1997). NKG2A complexed with CD94 defines a novel inhibitory natural killer cell receptor. J. Exp. Med. 185, 795-800. Brunet, J. F., Denizot, F., Luciani, M. F., Row-Dosseto, M., Suzan, M., Mattei, M. G., and Golstein, P. (1987). A new member of the immunoglobulin superfamily-CTLA-4. Nature 328, 267-270. Burshtyn, D. N., Scharenberg, A. M., Wagtmann, N., Rajagopalan, S., Berrada, K., Yi, T., Kinet, J. P., and Long, E. 0. (1996). Recruitment of tyrosine phosphatase HCP by the killer cell inhibitor receptor. Immunity 4, 77-85. Burshtyn, D. N., Yang, W., Yi, T., and Long, E. 0. (1997). A novel phosphotyrosine motif with a critical amino acid at position -2 for the SH2 domain-mediated activation of the tyrosine phosphatase SHP-1. J. Bid. Chem. 272, 13066-13072. Campbell, K. S., Dessing, M., Lopezbotet, M., Cella, M., and Colonna, M. (1996).Tyrosine phosphorylation of a human killer inhibitory receptor recruits protein tyrosine phosphatase 1C. J. Exp. Med. 184, 93-100. Carpenter, C. L., and Cantley, L. C. (1996). Phosphoinositide kinases. Curr. Opin. Cell Bid. 8, 153-158. Carretero, M., Cantoni, C., Bellon, T., Bottino, C., Biassoni, R., Rodriguez, A., Perez-Villar, J. J., Moretta, L., Moretta, A,, and Lopez-Botet, M. (1997). The CD94 and NKG2-A Ctype lectins covalently assemble to form a natural killer cell inhibitory receptor for HLA class I molecules. Eur. J. Immunol. 27, 563-567. Carretero, M., Palmieri, G., Llano, M., TulLio, V., Santoni, A,, Geraghty, D. E., and Lopezbotet, M. (1998). Specific engagement of the CD94lNKG2A killer inhibitory receptor by the HLA-E Class IB molecule induces SHP-1 phosphatase recruitment to tyrosinephosphorylated NKG-A-Evidence for receptor function in heterologous tranfectants. Eur. J. Immunol. 28, 1280-1291.

168

SILVIA BOLLAND AND JEFFREY V. RAVETCH

Castells, M. C. (1994). Surface makers for mast cell subtypes: Low &nity IgG receptors and gp49 family. Allergie Immunol. 26, 127-131. Castells, M. C., Wu, X.,Arm, J. P., Austen, K. F., and Katz, H. R. (1994). Cloning of the gp49B gene of the immunoglobulin superfamily and demonstration that one of its two products is an early-expressed mast cell surface protein originally described as gp49. 1.Biol. Chem. 269, 8393-8401. Chacko, G. W., Tridandapani, S., Damen, J. E., Liu, L., Krystd, G., and Coggeshall, K. M. (1996). Negative signaling in B lymphocytes induces tyrosine phosphorylation of the 145-kDa inositol polyphosphate 5-phosphatase, SHIP. I. Immunol. 157,2234-2238. Chan, P. L., and Sinclair, N. R. S. C. (1971). Regulation of the immune response: V. An analysis of the function of the Fc portion of antibody in suppression of an immune response with respect to interaction with components of the lymphoid system. Immunology 21, 967-987. Chan, V. W. F., Meng, F. Y.,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 down-regulation. Immunity 7, 69-81. Chang, C., Rodriguez, A,, Carretero, M., Lopez-Botet, M., Phillips, J. H., and Lanier, L. L. (1995).Molecular characterization of human CD94: A type I1 membrane glycoprotein related to the C-type lectin superfamily. Eur. J. Immunol. 25, 2433-2437. Choquet, D., Partiseti, M., Amigorena, S., Bonnerot, C., Fridman, W. H., and Kom, H. (1993). Cross-linking of IgG receptors inhibits membrane immunoglobulin-stimulated calcium influx in B lymphocytes. J. Cell Biol. 121, 355-363. Clynes, R., Maizes, J., Guinamard, R., and Ravetch, J. V. (1998). Modulation of immune complex induced inflammation by the coordinate expression of activation and inhibitory receptors. J. Exp. Med., in press. Colonna, M., and Samaridis, J. (1995). Cloning of immunoglobulin-superfamily members associated with HLA-C and HLA-B recognition by human natural killer cells. Science 268,405-408. Cornall, R. J., Cyster, J. G., Hibbs, M. L., Dunn, A. R. 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. Coughlin, S. R., Escobedo, J. A,, and Williams, L. T. (1989). Role of phosphatidylinositol lanase in PDGF receptor signal transduction. Science 243, 1191-1194. Crowley, M. T., Harmer, S. L., and DeFranco, A. L. (1996). Activation-induced association of a 145-kDa tyrosine-phosphorylated protein with Shc and Syk in B lymphocytes and macrophages. J. Biol. Chem. 271, 1145-1152. Cyster, J. G., and Goodnow, C. C. (1995). Protein tyrosine phosphatase 1C negatively regulates antigen receptor signaling in B lymphocytes and determines thresholds for negative selection. Immunity 2, 13-24. Daeron, M. (1997). Fc receptor biology. Annu. Rev. Immunol. 15, 203-234. Daeron, M., Latour, S., Malbec, O., Espinosa, E., Pina, P., Pasmans, S., and Fridman, W. H. (1995a). The same tyrosine-based inhibition motif, in the intracytoplasmic domain of Fc gamma RIIB, regulates negatively BCR-, TCR-, and FcR-dependent cell activation. Immunity 3, 635-646. Daeron, M., Malbec, 0.. Latour, S., Arock, M., and Fridman, W. H. (1995b). Regulation of high-affinity IgE receptor-mediated mast cell activation by murine low-affinity IgG receptors. 1.Clin. Invest. 95, 577-585. D’Ambrosio, D., Hippen, K. L., Minskoff, S. A., Mellman, I., Pani, G., Siminovitch, K. A,, and Cambier, J. C. (1995). Recruitment and activation of PTPlC in negative regulation of antigen receptor signaling by Fc gamma RIIB1. Science 268, 293-297.

INHIBITION BY ITIM-CONTAINING RECEPTORS

169

Damen, J. E., Liu, L., Rosten, P., Humphries, R. K., Jefferson, A. B., Majerus, P. W., and Krystal, G. (1996). The 145-kDa protein induced to associate with Shc by multiple cytokines is an inositol tetraphosphate and phosphatidylinositol 3,4,5-trisphosphate 5phosphatase. Proc. Natl. Acad. Sci. U.S.A. 93, 1689-1693. D'Andrea, A., Chang, C., Franz-Bacon, K., McClanahan, T., Phillips, J. H., and Lanier, L. L. (1995). Molecular cloning of NKB1. A natural killer cell receptor for HLA-B allotypes.J. lmmunol. 155, 2306-2310. 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 alteredin immunodeficient xid mice. Hurnan Mol. Genet. 3, 161-166. Dolmetsch, R. E., Lewis, R. S., Goodnow, C. C., and Healy, J. I. (1997). Differential activation of transcription factors induced by Ca2+response amplitude and duration. Nature 386, 855-858. Doody, G. M., Justement, L. B., Delibrias, C. C., Matthews, R. J., Lin, J., Thomas, M. L., and Fearon, D. T. (1995). A role in B cell activation for CD22 and the protein tyrosine phosphatase SHP. Science 269, 242-244. Fluckiger, A. C., Li, Z., Kato, R. M., Wahl, M. I., Ochs, H. D., Longnecker, R., Kinet, J. P., Witte, 0. N., Scharenberg, A. M., and Rawlings, D. J. (1998). Btk/Tec kinases regulate sustained increases in intracellular Ca" following B-cell receptor activation. EMBO 1. 17, 1973-1985. Fong, D. C., Malbec, O., Arock, M., Cambier, J. C., Fridman, W. H., and Daeron, M. (1996). Selective in uivo recruitment of the phosphatidylinositol phosphatase SHIP by phosphorylated Fc-gamma-RIIB during negative regulation of IgE-dependent mouse mast cell activation. lmmunol. Lett. 54, 83-91. Fry, A. M., Lanier, L. L., and Weiss, A. (1996). Phosphotyrosines in the killer cell inhibitory receptor motif of NKBl are required for negative signaling and for association with protein tyrosine phosphatase 1C. J. Exp. Med. 184, 295-300. Gold, M. R., and Aebersold, R. (1994). Both phosphatidylinositol3-kinase and phosphatidylinositol4-kinase products are increased by antigen receptor signalingin B cells.]. lmmunol. 152,42-50. Gupta, N., Scharenberg, A. M., Burshtyn, D. N., Wagtmann, N., Lioubin, M. N., Rohrschneider, L. R., Kinet, J. P., and Long, E. 0. (1997). Negative signaling pathways of the killer cell inhibitory receptor and Fc-gamma-RIIB 1require distinct phosphatases. 1.Exp. Med. 186,473-478. Healy, J. I., and Goodnow, C. C. (1998). Positive versus negative signaling by lymphocyte antigen receptors. Annu. Rev. lmmunol. 16, 645-670. Helgason, C. D., Damen, J. E., Rosten, P., Grewal, R., Sorensen, P., Chappel, S. M., Borowski, A., Jirik, F., Krystal, G., and Humphries, R. K. (1998). Targeted disruption of SHIP leads to hemopoietic perturbations, lung pathology, and a shortened life span. Genes Deu. 12, 1610-1620. Hibbs, M. L., Tarlinton, D. M., Armes, J., Grail, D., Hodgson, G., Maglitto, R., Stacker, S. A., and Dunn, A. R. R. (1995). Multiple defects in the immune system of lyn-deficient mice, culminating in autoimmune disease. Cell 83, 301-311. Hippen, K. L., Buhl, A. M., Dambrosio, D., Nakamura, K., Persin, C., and Cambier, J. C. (1997). Fc-gamma-RIIB 1 inhibition of BCR-mediated phosphoinositide gydrolysis and Ca2+mobilization is integrated by CD19 dephosphorylation. Immunity 7 , 49-58. 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.

170

SILVIA BOLLAND AND JEFFREY V. RAVETCH

Houchins, J. P., Lanier, L. L., Niemi, E. C., Phillips, J. H., and Ryan, J. C. (1997). Natural killer cell cytolytic activity is inhibited by NKG2-A and activated by NKG2-C.J. Immunol. 158,3603-3609. Hunter, M. G., and Avalos, B. R. (1998). Phosphatidylinositol3'-kinaseand SH2-containing inositol phosphatase (SHIP) are recruited by distinct positive and negative growthregulatory domains in the granulocyte colony-stimulating factor receptor. J. Immunol. 160,4979-4987. Hyvonen, M., and Saraste, M. (1997). Structure of the PH domain and Btk motif from Bruton's tyrosine kinase-Molecular explanations for X-linked agammaglobulinaemia. EMBO J. 16,3396-3404. Jefferson, A. B., and Majerus, P. W. (1995). Properties of type I1 inositol polyphosphate 5-phosphatase. J. Biol. Chem. 270,9370-9377. Karlhofer, F. M., Ribaudo, R. K., and Yokoyama, W. M. (1992). MHC class I dloantigen specificity of Ly-49+ IL-%activated natural killer cells. Nature 358, 66-70. Karre, K., Ljunggren, H. G., Piontek, G., and Kiessling, R. (1986). Selective rejection of H-2-deficient lymphoma variants suggests alternative immune defence strategy. Nature 319, 675-678. Katz, H. R., and Lobell, R. B. (1995). Expression and function of Fc gamma R in mouse mast cells. Int. Arch. Allergy Immunol. 107, 76-78. Kaufman, D. S., Schoon, R. A., Robertson, M. J., and Leibson, P. J. (1995). Inhibition of selective signaling events in natural killer cells recognizing major histocompatibility complex class I. Proc. Nat. Acad. Sci. U.S.A. 92,6484-6488. Kavanaugh,W. M., Turck, C. W., and Williams, L. T. (1995).PTB domain binding to signaling proteins through a sequence motif containing phosphotyrosine. Science 268,1177-1 179. Kavanaugh, W. M., Pot, D. A,, Chin, S. M., Deuterreinhard, M., Jefferson, A. B., Noms, F. A, Masiarz, F. R., Cousens, L. S., Majerus, P. W., and Williams, L. T. (1996). Multiple forms of an inositol polyphosphatase 5-phosphatase form signaling complexes with Shc and Grb2. C u m Biol. 6, 438-445. Kharitonenkov, A., Chen, Z., Sures, I., Wang, H., Schilling, J., and Ullrich, A. (1997). A family of proteins that inhibit signalling through tyrosine kinase receptors. Nature 386, 181-186. Kimura, T., Sakamoto, H., Appella, E., and Siraganian, R. P. (1997a). The negative signaling molecule SH2 domain-containing inositol- polyphosphate 5-phosphatase (SHIP) binds to the tyrosine- phosphorylated beta subunit of the high affinity IgE receptor. J . Biol. Chem. 272,13991 -13996. Kimura, T., Zhang, J., Sagawa, K., Sakaguchi, K., Appella, E., and Siraganian, R. P. (1997b). Syk-independent tyrosine phosphorylation and association of the protein tyrosine phosphatases SHP-1 and SHP-2 with the high affinity IgE receptor. J. Immunol. 159,4426-4434. Klingmuller, U., Lorenz, U., Cantley, L. C., Neel, B. G., and Lodish, H. F. (1995). Specific recruitment of SH-PTP1 to the erythropoietin receptor causes inactivation of JAK2 and termination of proliferative signals. Cell 80, 729-738. Kojima, T., Fukuda, M., Watanabe, Y., Hamazato, F., and Mikoshiba, K. (1997). Characterization of the pleckstrin homology domain of Btk as an inositol polyphosphate andphosphoinositide binding domain. Bwchem. Biophys. Res. Commun. 236, 333-339. Konkozlowski, M., Pani, G., Pawson, T., and Siminovitch, K. A. (1996).The tyrosinephosphatase PTPlC associates with vav, grb2, and msosl in hematopoietic cells. J. Biol. Chem. 271,3856-3862. Kozlowski, M., Larose, L., Lee, F., Le,D. M., Rottapel, R., and Siminovitch, K. A. (1998). SHP-1 binds and negatively modulates the c-Kit receptor by interaction with tyrosine 569 in the c-Kit juxtamembrane domain. Mol. Cell. Biol. 18,2089-2099.

INHIBITION BY ITIM-CONTAINING RECEPTORS

171

Kubagawa, H., Burrows, P. D., and Cooper, M. D. (1997).A novel pair of immunoglobulinlike receptors expressed by B cells and rnyeloid cells [see comments]. Proc. Nutl. Acnd. Sci. U.S.A. 94, 5261-5266. Kubagawa, H., Chen, C. C., Ho, L. H., Gartland, L., Shimada, T., Mashburn, C., Ravetch, J. V., and Cooper, M. D. (1998). Biochemical nature and cellular distribution of the Paired Immunoglobulin-Likereceptors, PIR-A and PIR-B. J. Exp. Med., in press. Kuroiwa, A., Yamashita, Y., Inui, M., Yuasa, T., Ono, M., Nagabukuro, A., Matsuda, Y., and Takai, T. (1998). Association of tyrosine phosphatases SHP-1 and SHP-2, inositol5phosphatase SHIP with gp49B1, and chromosomal assignment of the gene.]. Biol. Chem. 273, 1070-1074. Kurosaki, T. (1997). Molecular mechanisms in B cell antigen receptor signaling. Cum. Opin. Immunol. 9,309-318. Kurosaki, T., Gander, I., and Ravetch, J. V. (1991).A subunit common to an IgG Fc receptor and the T-cell receptor mediates assembly through different interactions. Proc. Natl. Acad. Sci. U.S.A. 88, 3837-3841. Lamkin, T. D., Walk, S. F., Liu, L., Damen, J. E., Krystal, G., and Ravichandran, K. S. (1997). Shc interaction with Src homology 2 domain containing inositol phosphatase (SHIP) in uiuo requires the Shc-phosphotyrosine binding domain and two specific phosphotyrosines on SHIP. J. Biol. Chem. 272, 10396-10401. Langhans-Rajasekaran, S. A,, Wan, Y., and Huang, X. Y. (1995). Activation of Tsk and Btk tyrosine kinases by G protein beta gamma subunits. Proc. Nut]. Acud. Sci. U.S.A. 92,8601-8605. Lanier, L. L. (1997). Natural killer ceIl receptors and MHC class I interactions. Cum. Opin. Inmwnol. 9, 126-131. Lanier, L. L. (1998a). Follow the leader: NK cell receptors for classical and nonclassical MHC class I. Cell 92, 705-707. Lanier, L. L. (1998b). NK cell receptors. Annu. Reu. Immunol. 16, 359-393. Lanier, L. L., Corliss, B. C., Wu, J., Leong, C., and Phillips, J. H. (1998a). Immunoreceptor DAPl2 bearing a tyrosine-based activation motif is involved in activating NK cells. Nature 391, 703-707. Lanier, L. L., Corliss, B. C., Wu, J., and Phillips, J. H. (1998b). Association of DAP12 with activating CD9mKG2C NK cell receptors. Immunity 8, 693-701. Ledrean, E., Vely, F., Olcese, L., Cambiaggi, A,, Guia, S., Krystal, G., Gervois, Moretta, A., Jotereau, F., and Vivier, E. (1998). Inhibition of antigen-induced T cell response and antibody-induced NK cell cytotoxicity by NKG2A: Association of NKG2A with SHP-1 and SHP-2 protein-tyrosine phosphatases. Eur. 1.Immunol. 28, 264-276. Lemmon, M. A., Ferguson, K. M., and Schlessinger, J. (1996). PH domains: Diverse sequences with a common fold recruit signaling molecules to the cell surface. Cell 85, 621-624. Li, T., Tsukada, S., Satterthwaite, A,, Havlik, M. H., Park, H., Takatsu, K., and Witte, 0. N. (1995). Activation of Bmton’s tyrosine kinase (BTK) by a point mutation in its pleckstrin homology (PH) domain. Immunity 2,451-460. Lioubin, M. N., Algate, P. A,, Tsai, S., Carlberg, K., Aebersold, A., Rohrschneider, and LR. (1996). pl50Ship, a signal transduction molecule with inositol polyphosphate-5phosphatase activity. Genes Deu. 10, 1084-1095. Liu, L., Damen, J. E., Cutler, R. L., and Krystal, G. (1994). Multiple cytokines stimulate the binding of a common 145-kilodalton protein to Shc at the Grb2 recognition site of Shc. Mol. Cell. Biol. 14, 6926-6935. Liu, L., Jefferson, A. B., Zhang, X., Norris, F. A,, Majerus, P. W., and Krystal, G. (1996). A novel phosphatidylinositol-3,4,5-trisphosphate 5-phosphatase associates with the interleukin-3 receptor. /. Biol. Chem. 271, 29729-29733.

172

SILVIA BOLLAND AND JEFFREY V. RAVETCH

Liu, L., Damen, J. E., Ware, M. D., and Krystal, G. (1997). Interleukin-3 induces the association of the inositol 5-phosphatase SHIP with SHP2. I. Biol. Chem. 272, 10998-

11001. Liu, Q., Oliveira-Dos-Santos, A. J., Mariathasan, S., Bouchard, D., Jones, J., Sarao, R., Kozieradzki, I., Ohashi, P. S., Penninger, J. M., and Dumont, D. J. (1998). The inositol polyphosphate 5-phosphatase Ship is a crucial negative regulator of BCR signaling. J. Exp. Med. 188, 1333-1342. Lopez, F., Esteve, J. P., Buscail, L., Delesque, N., Saint-Laurent, N., Theveniau, Nahmias, C., Vaysse, N., and Susini, C. (1997). The tyrosine phosphatase SHP-1 associates with the sst2 somatostatin receptor and is an essential component of sst2-mediated inhibitory growth signaling. J. Biol. Chem. 272, 24448-24454. Lorenz, U., Bergemann, A. D., Steinberg, H. N., Flanagan, J. G., Li, X. T., G&, S. J., and Nee], B. G. (1996).Genetic analysisreveals cell type-specificregulation of receptor tyrosine kinase c-kit by the protein tyrosine phosphatase SHP1. J. Exp. Med. 184, 1111-1126. Luckhoff, A., and Clapham, D. E. (1992). Inositol 1,3,4,5-tetrakisphosphateactivates an endothelial Ca(2+)-permeable channel. Nature 355, 356-358. Maeda, A., Kurosaki, M., and Kurosaki, T. (1998a). Paired immunoglobulin-like receptor (P1R)-A is involved in activating mast cells through its association with Fc receptor y chain. J . Exp. Med. 188,991-995. Maeda, A,, Kurosaki, M., Ono, M., Takai, T., and Kurosaki, T. (199813). Requirement of SH2-containingprotein tyrosine phosphatases SHP-1 and SHP-2 for paired immunoglobulin-like receptor B (PIR-B)-mediated inhibitory signal. J. Exp. Med. 187, 1355-1360. Malbec, O., Fong, D. C., Turner, M., Tybulewicz, V. L., Cambier, J. C., Fridman, W. H., and Daeron, M. (1998). Fc epsilon receptor I-associated lyn-dependent phosphorylation of Fc gamma receptor IIB during negative regulation of mast cell activation.J. Immunol. 160, 1647-1658. Marengere, L. E., Waterhouse, P., Duncan, G. S., Mittrucker, H. W., Feng, G. S., Mak, and TW. (1996). Regulation of T cell receptor signaling by tyrosine phosphatase SW association with CTLA-4. Science 272, 1170-1173. Mason, L. H., Gosselin, P., Anderson, S. K., Fogler, W. E., Ortaldo, J. R., and McVicar, D. W. (1997). Differential tyrosine phosphorylation of inhibitory versus activating Ly-49 receptor proteins and their recruitment of SHP-1 phosphatase. J. Immunol. 159,41874196. Matsushita, M., Yamadori, T., Kato, S., Takemoto, Y., Inazawa, J., Baba, Y., Hashimoto, S., Sekine, S., Arai, S., Kunikata, T., Kurimoto, M., Kishimoto, T., and Tsukada, S. (1998). Identification and characterization of a novel SH3-domain binding protein, Sab, which preferentially associates with Bruton's tyrosine kinase (BtK). Biochem. Biophys. Res. Commun. 245,337-343. Meyaard, L., Adema, G. J., Chang, C., Woollatt, E., Sutherland, G. R., Lanier, L. L., and Phillips, J. H. (1997). LAIR-1, a novel inhibitory receptor expressed on human mononuclear leukocytes. Immunity 7 , 283-290. Moretta, A., Bottino, C., Pende, D., Tripodi, G., Tambussi, G., Vide, O., Orengo, A., Barbaresi, M., Merli, A., and Ciccone, E. (1990). Identification of four subsets of human CD3-CD16' natural killer (NK) cells by the expression of clonally distributed functional surface molecules: Correlation between subset assignment of NK clones and ability to mediate specific alloantigen recognition. J. Exp. Med. 172, 1589-1598. Moretta, A., Vitde, M., Bottino, C., Orengo, A. M., Morelli, L., Augugliaro, R., Barbaresi, M., Ciccone, E., and Moretta, L. (1993). P58 molecules as putative receptors for major histocompatibility complex (MHC) class I molecules in human natural killer (NK) cells.

INHIBITION BY ITIM-CONTAINING RECEPTORS

173

Anti-p58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities.J . Erp. Med. 178, 597-604. Moretta, A., Biassoni, R., Bottino, C., Pende, D., Vitale, M., Poggi, A., Mingari, M. C., and Moretta, L. (1997). Major histocompatibility complex class I-specific receptors on human natural killer and T lymphocytes. ImmunoE. Rev. 155, 105-117. Muta, T., Kurosaki, T., Misulovh, Z., Sanchez, M., Nussennveig, M. C., and Ravetch, J. V. (1994). A 13-amino-acid motif in the cytoplasmic domain of Fc gamma RIIB modulates B-cell receptor signalling. Nature 368, 70-73. Nadler, M. J. S., Chen, B. B., Anderson, J. S.,Wortis, H. H., and Neel, B. G. (1997). Proteintyrosine phosphatase SHP-1 is dispensable for Fc-gamma-RIIB-mediated inhibition of B cell antigen receptor activation. J. Biol. Chem. 272, 20038-20043. Nakamura, M. C., Niemi, E. C., Fisher, M. J., Shultz, L. D., Seaman, W. E., and Ryan, J. C. (1997). Mouse Ly-49A interrupts early signalingevents in natural killer cell cytotoxicity and functionally associates with the SHP-I tyrosine phosphatase. J. Erp. Med. 185, 673-684. Nakanishi, H., Brewer, K. A,, and Exton, J. H. (1993). Activation of the zeta isozyme of protein kinase C by phosphatidyliriosito13,4,5-trisphosphate.1.Biol. Chem. 268, 13-16. Nakayama, E., Von, H. I., and Parnes, J. R. (1989).Sequence of the Lyb-2 B-cell differentiation antigen defines a gene superfamily of receptors with inverted membrane orientation. Proc. Natl. Acad. Sci. U.S.A. 86, 1352-1356. 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. Nishizumi, H., Horikawa, I., Mlinaric-Rascan,I., and Yamamoto, T. (1998).A double-edged kinase lyn: A positive and negative regulator for antigen receptor-mediated signals. J. Erp. Med. 187, 1343-1348. Nossd, G. J. (1994). Negative selection of lymphocytes. Cell 76, 229-239. Olcese, L., Lang, P., Vely, F., Cambiaggi, A,, Marguet, D., Blery, M., Hippen, KL, Biassoni, R., Moretta, A., Moretto, L., Cambier, J. C., and Vivier, E. (1996). Human and mouse killer-cell inhibitory receptors recruit PTPlC and PTPlD protein tyrosine phosphatases. J. Immuml. 156,4531-4534. Olcese, L., Cambiaggi, A,, Semenzato, G., Bottino, C., Moretta, A,, and Vivier, E. (1997). Human killer cell activatory receptors for MHC Class I molecules are included in a multimeric complex expressed by natural killer cells. J. Immunol. 158, 5083-5086. Ono, M., Bolland, S., Tempst, P., and Ravetch, J. V. (1996). RoIe ofthe inositol phosphatase SHIP in negative regulation of the immune system by the receptor Fc-gamma-RIIB. Nature 383, 263-266. Ono, M., Okada, H., Bolland, S., Yanagi, S., Kurosaki, T., and Ravetch, J. V. (1997). Deletion of SHIP or SHP-1 reveals two distinct pathways for inhibitory signaling. Cell 90,293-301. Orourke, L., Tooze, R., and Fearon, D. T. (1997). Co-receptors of B lymphocytes. Curr. Opin. Immunol. 9, 324-329. Osborne, M. A,, Zenner, G., Lubinus, M., Zhang, X., Songyang, Z., Cantley, L. C., Majerus, P., Bum, P., and Kochan, J. P. (1996).The inositol5'-phosphatase SHIP binds to immunoreceptor signaling motifs and responds to high affinity IgE receptor aggregation. J. Biol. Chem. 271,29271-29278. Pani, G., Kozlowski, M., Cambier, J. C., Mills, G. B., and Siminovitch, K. A. (1995). Identification of the tyrosine phosphatase PTPlC as a B cell antigen receptor-associated protein involved in the regulation of B cell signaling. J. Exp. Med. 181, 2077-2084. Pani, G., Fischer, K. D., Mlinaricrascan, I., and Siminovitch, K. A. (1996). Signaling capacity of the T cell antigen receptor is negatively regulated by the PTPlC tyrosine phosphatase. I. Exp. Med. 184, 839-852.

174

SILVIA BOLLAND AND JEFFREY V. RAVETCH

Paulson, R. F., Vesely, S., Siminovitch, K. A., and Bernstein, A. (1996). Signalling by the wkit receptor tyrosine kinase is negatively regulated in vivo by the protein tyrosine phosphatase SHP1. Nature Genet. 13,309-315. Pei, D., Lorenz, U., Klingmuller, U., Neel, B. G., and Walsh, C. T. (1994). Intramolecular regulation of protein tyrosine phosphatase SH- PTP1: A new function for Src homology 2 domains. Biochemistry 33, 15483-15493. Pei, D., Wang, J., and Walsh, C. T. (1996). Differential functions of the two Src homology 2 domains in protein tyrosine phosphatase SH-FTP1. Proc. Nutl. Acud. Sci. U.S.A. 93, 1141-1145. Phillips, N. E., and Parker, D. C. (1983). Fc-dependent inhibition of mouse B cell activation by whole anti-mu antibodies. J. lmmunol. 130, 602-606. Phillips, N. E., and Parker, D. C. (1984). Cross-linkingof B lymphocyte Fc gamma receptors and membrane immunoglobulin inhibits anti-immunoglobulin-induced blastogenesis. J. lmmunol. 132,627-632. Plas, D. R., Johnson, R., Pingel, J. T., Matthews, R. J., Dalton, M., Roy, G., Chan, A. C., and Thomas, M. L. (1996). Direct regulation of ZAP-70 by SHP-1 in T cell antigen receptor signaling. Science 272, 1173-1176. 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. Pradhan, M., and Coggeshall, K. M. (1997). Activation-induced bi-dentate interaction of SHIP and Shc in B lymphocytes C. J. Cell. Biochem. 67,32-42. Qu, C. K., Shi, Z. Q., Shen, R., Tsai, F. Y., Orkin, S. H., and Feng, G. S. (1997). A deletion mutation in the SH2-N domain of Shp-2 severely suppresses hematopoietic cell development. Mol. Cell. Biol. 17, 5499-5507. Raab, M., and Rudd, C. E. (1996). Hematopoietic cell phosphatase (HCP) regulates p561ck phosphorylation and ZAP-70 binding to T cell receptor zeta chain. Biochem. Biophys. Res. Commun. 222,50-57. Rajewsky, K. (1996). Clonal selection and learning in the antibody system. Nature 381, 751-758. Rameh, L. E., Arvidsson Ak, Carraway, K. L. 3., Couvillon, A. D., Rathbun, G., Crompton, A., VanRenterghem, B., Czech, M. P., Ravichandran, K. S., Burakoff, S. J., Wang, D. S., Chen, C. S., and Cantley, L. C. (1997). A comparative analysis of the phosphoinositide binding specificity of pleckstrin homology domains. J. Biol. Chem. 272, 22059-22066. Ravetch, J. V. (1994). Fc receptors: Rubor redw. Cell 78, 553-560. Ravetch, J. V. (1997). Fc receptors. Cuw. Opin. lmmunol. 9, 121-125. Ravetch, J. V., and Kinet, J. P. (1991). Fc receptors. Annu. Reu. lmmunol. 9,457-492. Salim, K., Bottomley, M. J., Querfurth, E., Zvelebil, M. J., Gout, I., Scaife, R., Margolis, R. L., Gigg, R., Smith, C. I., Driscoll, P. C., Waterfield, M. D., and Panayotou, G. (1996). Distinct specificity in the recognition of phosphoinositides by the pleckstrin homology domains of dynamin and Bruton’s tyrosine kinase. EMBO J. 15, 6241-6250. Samaridis, J., and Colonna, M. (1997). Cloning of novel immunoglobulin superfamily receptors expressed on human myeloid and lymphoid cells: Structural evidence for new stimulatory and inhibitory pathways. Eur. J. lmmunol. 27, 660-665. Scharenberg, A. M., El-Hillal, O., Fruman, D. A., Beitz, L. O., Li, Z., Lin, S., Gout, Candey, L. C., Rawlings, D. J., and Kinet, J. P. (1998). Phosphatidylinositol-3,4,5-trisphosphate (PtdIns-3,4,5-P3)/Tec base-dependent calcium signaling pathway: A target for SHIPmediated inhibitory signals. EMBO J. 17, 1961-1972. Sebzda, E., Wallace, V. A., Mayer, J., Yeung, R. S., Mak, T. W., and Ohashi, P. S. (1994). Positive and negative thymocyte selection induced by different concentrations of a single peptide. Science 263, 1615-1618.

INHIBITION BY ITIM-CONTAINING RECEPTORS

175

Shultz, L. D., Schweitzer, P. A., Rajan, T. V., Yi, T., Ihle, J. N., Matthews, R. J., Thomas, M. L., and Beier, D. R. (1993). Mutations at the murine motheaten locus are within the heinatopoietic cell protein-tyrosine phosphatase (Hcph) gene. Cell 73, 1445-1454. Sihciano,J. D., Morrow, T. A., and Desiderio, S. (1992). Itk, a T-cell-specifictyrosine kinase gene inducible by interlleukin 2. Proc. Natl. Acnd. Sci. U.S.A. 89, 11194-11198. Smit, L., van der Horst, G., and Borst, J. (1996). Sos, Vav, and C3G participate in B cell receptor-induced signaling pathways and differentially associate with Shc-Grb2, Crk, and Crk-L adaptors. J. Biol. Chem. 271, 8564-8569. Smith, C . I., Islam, K. B., Vorechovsky, I., Olerup, 0..Wallin, E., Rabbani, H., Baskin, B., and Hammarstrom, L. (1994).X-Linked agammaglobulineniiaand other immunoglobulin deficiencies. Irnmunol. Reo. 138, 159-183. Smith, K. C., Tarlinton, D. M., Doody, G. M., Hibbs, M. L., and Fearon, D. T. (1998). Inhibition of the B cell by CD22-A requirement for lyn. 1.Exp. Med. 187, 807-811. Smith, K. M., Wu, J., Bakker, A. H., Phillips, J. H., and Lanier, L. L. (1998). Ly-49H associate with mouse DAPl2 and form activating receptors. J. Zmmunol. 161, 7-10. Somani, A. K., Bignon, J. S., Mills, G. B., Siminovitch, K. A,, and Branch, D. R. (1997). Src kinase activity is regulated by the SHP-1 protein-tyrosine phosphatase. J Biol. Chem. 272,21113-21119. Sugawara, H., Kurosaki, M., Takata, M., and Kurosaki, T. (1997). Genetic evidence for involvement of type 1, type 2, and type 3 inositol 1,4,5-trisphosphate receptors in signal transduction through the B-cell antigen receptor. EMBO J 16, 3078-3088. Takai, T., Ono, M., Hihda, M., Ohmori, H., and Ravetch, J. V. (1996).Augmented humoral and anaphylactic responses in Fc gamma RII-deficient mice. Nature 379, 346-349. Takata, M., and Kurosaki, T. (1996). A role for Brutons tyrosine kinase in B cell antigen recepor-mediated activation of phospholipase C-gamma-2.1. Exp. Med. 184, 31-40. Takata, M., Sabe, H., Hata, A., Inazu, T., Homma, Y., Nukada, T., Tamamura, H., and Kurosaki, T. (1994). Tyrosine kinases Lyn and Syk regulate B cell receptor-coupled Ca” mobilization through distinct pathways. EMBO J. 13, 1341-1349. Takata, M., Homma, Y., and Kurosaki, T. (1995). Requirement of phospholipase C-gamma2 activation in surface immunoglobulin M-induced B cell apoptosis. 1. Exp. Med. 182, 907-914. Tang, T. L., Freeman, R. M., Jr., OReilIy, A. M., Neel, B. G., and Sokol, S. Y. (1995). The SHZ-containing protein-tyrosine phosphatase SH-PTP2 is required upstream of MAP kinase for early Xenopus development. Cell 80, 473-483. Thompson, J. A,, Grunert, F., and Zimmermann, W. (1991). Carcinoembryonic antigen gene family: Molecular biology and clinical perspectives. 1.Clin. Lab. And. 5, 344-366. Ting, A. T., Karnitz, L. M., Schoon, R. A,, Abraham, R. T., and Leibson, P. J. (1992). Fc gamma receptor activation induces the tyrosine phosphorylation of both phospholipase C (PLC)-gamma 1 and PLC-gamma 2 in natural killer cells.]. Exp. Med. 176,1751-1755. Toker, A., and Cantley, L. C. (1997). Signalling through the lipid products of phosphoinositide-3-OH kinase. Nature 387, 673-676. Toker, A,, Meyer, M., Reddy, K. K., Falck, J. R., Aneja, R., Aneja, S., Parra, A., Bums, D. J., Ballas, L. M., and Cantley, L. C. (1994). Activation of protein kinase C family members by the novel polyphosphoinositides PtdIns-3,4-P2 and PtdIns-3,4,5-P3, J. B i d . Chem. 269, 32358-32367. Traynor-Kaplan, A. E., Harris, A. E., Thompson, B. L., Taylor, P., and SMar, L. A. (1988). An inositol tetrakisphosphate-containing phospholipid in activated neutrophils. Nature 334, 353-356. Tsui, H. W., Siminovitch, K. A., de Souza, L., and Tsui, F. W. (1993). Motheaten and viable motheaten mice have mutations in the haematopoietic cell phosphatase gene. Nature Genet. 4, 124-129.

176

SILVIA BOLLAND AND JEFFREY V. RAVETCH

Tuveson, D. A., Carter, R. H., Soltoff, S. P., and Fearon, D. T. (1993). CD19 of B cells as a surrogate kinase insert region to bind phosphatidylinositol 3-base. Science 260, 986-989. Ujike, A,, Ishikawa, Y., Ono, M., Yuasa, T., Yoshino, T., Fukumoto, M., Ravetch, J. V., and Takai, T. (1999). Modulation of IgE-mediated systemic anaphylaxis by low affinity Fc receptors for IgG. (in preparation) Van de Velde, H., van Hoegen, I., Luo, W., Parnes, J. R., and Thielemans, K. (1991).The B-cell surface protein CD72/Lyb-2 is the ligand for CD5. Nature 351, 662-665. Vely, F., and Vivier, E. (1997). Conservation of structural features reveals the existence of a large family of inhibitorycell surfacereceptors and noninhibitory/activatorycounterparts. 1.lmmunol. 159, 2075-2077. Vely, F., Olivero, S., Olcese, L., Moretta, A,, Damen, J. E., Liu, L., Krystal, G., Cambier, J. C., Daeron, M., and Vivier, E. (1997). Differential association of phosphatases with hematopoietic co-receptors bearing immunoreceptor tyrosine-based inhibition motifs. Eur. J. lmmunol. 27,1994-2000. Vivier, E., and Daeron, M. (1997).Immunoreceptortyrosine-basedinhibition motifs. lmmunol. Today 18,286-291. Wang, J. Y., Koizumi, T., and Watanabe, T. (1996). Altered antigen receptor signaling and impaired Fas-mediated apoptosis of B cells in lyn-deficient mice. 1. Exp. Med. 184,831-838. Ward, S. G., June, C. H., and Olive, D. (1996). PI 3-kinase: A pivotal pathway in T-cell activation? Immunol. Today 17, 187-197. Weiss, A., and Littman, D. R. (1994).Signal transduction by lymphocyte antigen receptors. Cell 76, 263-274. Weng, W. K., Jarvis, L., and LeBien, T. W. (1994). Signaling through CD19 activates Vadmitogen-activated protein kinase pathway and induces formation of a CD19Navl phosphatidylinositol 3-kinase complex in human B cell precursors. J. Bid. Chem. 269, 32514-32521. Wirthmueller, U., Kurosaki, T., Murakami, M. S., and Ravetch,J. V. (1992).Signal transduction by Fc gamma RIII (CD16) is mediated through the gamma chain. 1. Exp. Med. 175, 1381-1390. Yamanashi, Y., Fukuda, T., Nishizumi, H., Inazu, T., Higashi, K., Kitamura, D., Ishida, T., Yamamura, H., Watanabe, T., and Yamamoto, T. (1997). Role of tyrosine phosphorylation of HS1 in B cell antigen receptor-mediated apoptosis. J. Exp. Med. 185, 1387-1392. Yang, W., and Desiderio, S. (1997). BAP-135, a target for Bruton’s tyrosine kinase in response to B cell receptor engagement. Proc. Natl. Acad. Sci. U.S.A. 94, 604-609. Yao, L., Suzuki, H., Ozawa, K., Deng, J., Lehel, C., Fukamachi, H., Anderson, W. B., Kawakami, Y., and Kawakami, T. (1997). Interactions between protein kinase C and pleckstrin homology domains. Inhibition by phosphatidylinositol 4,5-bisphosphate and phorbol 12-myristate 13-acetate.J. Biol. Chem. 272, 13033-13039. Yi, T., and Ihle, J. N. (1993). Association of hematopoietic cell phosphate with c-Kit after stimulation with c-Kit ligand. Mol. Cell. Biol. 13, 3350-3358. Yi, T., Mui, A. L., Krystal, G., and Ihle, J. N. (1993). Hematopoietic cell phosphatase associates with the interleukin-3 (IL-3) receptor beta chain and down-regulates IL-3induced tyrosine phosphorylation and mitogenesis. Mol. Cell. Biol. 13, 7577-7586. Yi, T., Zhang, J., Miura, O., and Ihle, J. N. (1995).Hematopoietic cellphosphatase associates with erythropoietin (Epo) receptor after Epo-induced receptor tyrosine phosphorylation: Identification of potential binding sites. Blood 85, 87-95. Yokoyama, W. M., and Seaman, W. E. (1993). The Ly-49 and NKR-P1 gene families encoding lectin-like receptors on natural killer cells: The NK gene complex. Annu. Reu. lmmunol. 11, 613-635.

INHIBITION BY ITIM-CONTAINING RECEPTORS

177

Yu, Z., Su, L., Hoglinger, O., Jaramillo, M. L., Banville, D., and Shen, S . H. (1998). SHP1 associates with both platelet-derived growth factor receptor and the p85 subunit of phosphatidylinositol 3-kinase. 1.Biol. Chem. 273, 3687-3694. Yuasa, T., Kubo, S . , Yoshino, T., Ujike, A., Matsumura, K., Ono, M., Ravetch, J. V., and Takai,T. (1999). Deletion of FcyRllB renders H-2bmice susceptible to collagen-induced arthritis. /. Exp. Med., in press. Zhang, R., Alt, F. W., Davidson, L., Orkin, S. H., and Swat, W. (1995). Defective signalling through the T- and B-cell antigen receptors in lymphoid cells lacking the vav protooncogene. Nature 374,470-473. This article was accepted for publication on August 28, 1998.

This Page Intentionally Left Blank

ADVANCES IN IMMUNOLOGY, VOL. 72

ATM in Lymphoid Development and Tumorigenesis YANG XU w h e n t of B i o k , UniVeniiy of Calihio, San Diego, Lo Mla, California92093-0322

1. lymphoid Defects and Tumorigenesis in Ataxia-Telangiectasia Patients

Immune defects in ataxia-telangiectasia (A-T)patients are characterized

by thymic hypoplasia or absence of thymus, reduced circulating T cells, and, in some patients, a selective deficiency of immunoglobulin isotypes, including IgA, IgE, IgG2, and IgG4 (Peterson and Good, 1968; Strober et al., 1968). White blood cells derived from A-T patients exhibit defects in cytokine production, decreased blast transformation, and proliferation in response to in vitro antigenic stimulation, including with common microbial antigens (Beatty et al., 1986; PaganelIi et aE., 1984; Tamura et at., 1980). In addition, A-T patients are defective in their cell-mediated immune responses to human cytomegalovirus infection (Levis et al., 1979; Tamura et al., 1980).Together, these findings indicate that A-T patients are defective in both humoral and cellular immune responses. About 40% of A-T patients suffer from increased tumorigenesis, mostly of the lymphoid system. A 250-fold or 750-fold increase in the risk of lymphomas was caculated for Caucasian or African-American A-T patients, respectively, when compared with the normal population (Morrell et al., 1986). Among the lymphomas in A-T patients, there appears to be a 4- to 5-fold increase in the frequency of T cell tumors compared to B cell tumors. Lymphoid tumors in A-T patients usually harbor the characteristic chromosomal translocations also found in the their circulating lymphocytes (Kojis et al., 1991). These chromosomal translocations typically involve the immunoglobulin gene loci in chromosome bands 2p12, 14q32, and 22qll and the T cell receptor (TCR) gene loci in chromosome bands 7p14,7q35, 14q11.2, and 14q32, suggesting that these chromosomal translocations may be due to illegitimate joining during V(D)J recombination. A similar spectrum of chromosomal translocation has also been identified in Burkitt’s Iymphoma and adult T cell leukemia, suggesting that these chromosomal rearrangements may promote T cell tumorigenesis (Taylor et at., 1996). Consistent with this notion, molecular analysis of these translocations indicates that the translocation involves an oncogene (Hecht and Hecht, 1987; Narducci et al., 1995; Stem et al., 1989). In addition, T cell clones with these chromosomal translocations usually expand into T cell leukemias in older A-T patients (Hecht et al., 1973; Oxford et al., 1975). 179

Copyright Q 1999 by Academic Press. AU rights of reproduction in any form reserved. 0065-2776/99$30.00

180

YANG XU

II. The ATM Gene

A gene consistently mutated in A-T patients, denoted ATM, has been identified through positional cloning (Savitsky et al., 1995). The ATM gene encodes a large kinase that belongs to a family of kinases containing a highly conserved C-terminal kinase domain related to the phosphatidylinositol3kinase domain (Savitsky et al., 1995). Members of this kinase family have been shown to function in DNA repair and cell cycle checkpoint control following DNA damage (Zakian, 1995). As such, several members of this family, including yeast proteins Rad3, Mecl, and Tell, are involved in DNA repair, genomic stability, meiotic recombination, and maintenance of telomere length (Greenwell et al., 1995; Jayaraman et al., 1997; Morrow et al., 1995; Paulovich and Hartwell, 1995).Other members of this family, such as the yeast proteins Torl, Tor2, and their mammalian counterpart FRAP, control cell cycle progression from G1 to S phase (Brown et al., 1995). A direct link between this kinase family and DNA double-stranded break repair was implicated by the findings that the DNA-dependent protein kinase catalpc subunit (DNA-PKcs) is directly involved in DNA double-stranded break repair, including V(D)J recombination in lymphocytes (Hartley et al., 1995; Blunt et al., 1995). Mice defective in DNAPKcs lack any mature lymphocytes and develop severe immunodeficiency (Hartley et al., 1995). Studies of the A-T cellular defects indicate that ATM might be involved in controlling cell cycle checkpoints and possibly other cellular functions in response to DNA strand break damage. Following ionizing radiation (IR), normal cells typically arrest their cell cycle at multiple checkpoints at the Gl/S border, in the S phase, and at the G$M border, thus ensuring the repair of DNA damage and genome integrity. However, all three cell cycle checkpoints in response to IR are defective in cells derived from A-T patients (Jayaraman et al., 1997). Although the basis for the S phase and G$M checkpoint defects of A-T cells in response to IR is not clear, the defective GI cell cycle arrest following IR in A-T cells is Iikely due to impaired p53 responses (Canman et al., 1994; Kastan et al., 1992; Khanna and Lavin, 1993). Cells derived from A-T patients are hypersensitive to DNA-damaging agents such as ionizing radiation and restriction enzymes that introduce double-stranded DNA breaks (Levis et al., 1979; Metcalfe et al., 1996). Several studies have failed to identify gross defects in the kinetics of repair of DNA single- and double-stranded breaks in A-T cells (Taylor et al., 1996; Lehmann, 1982). However, additional studies showed that although A-T cells can rejoin double-stranded breaks at the same rate as normal cells, the percentage of DNA double-stranded breaks that did not rejoin was five to six times higher in A-T cells compared to the normal

ATM IN LYMPHOID DEVELOPMENT AND TUMORIGENESIS

181

cells (Cornforth and Bedford, 1985). In addition, A-T cells exhibit a significant increase in the frequency of induced aberrations, such as chromatid breaks, gaps, and rearrangement, following the induction of DNA doublestranded breaks (Cohen and Levy, 1989). Therefore, A-T cells may have subtle defects in repairing DNA double-stranded breaks. In this context, V(D)J recombination represents a repair process of DNA double-stranded breaks. It has been proposed that defects in V(D)J recombination might account for the immune defects in A-T (Peterson and Funkhouser, 1989). 111. Dissecting the lymphoid Defects and Tumorigenesis in A-T Mouse Models

Mouse models have become invaluable tools to elucidate the molecular basis of defects in human diseases. In an attempt to create a mouse model to study the pleiotropic defects in A-T patients, several groups independently disrupted different regions of the Atm gene in mice through homologous recombination (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996). All Atm mutant mice express a number of A-T phenotypes, including growth retardation, abolished germ cell development, immune defects, and a high incidence of thymic lymphomas (Barlow et al., 1996; Elson et al., 1996; Xu et al., 1996). In addition, primary cells derived from the Atrn-l- mice display cell cycle checkpoint defects characteristic of A-T, including hypersensitivity to y-irradiation and defective cell cycle checkpoint control following y-irradiation (Barlow et al., 1996; Baskaran et al., 1997; Xu and Baltimore, 1996). Similar to cells derived from A-T patients, primary fibroblasts derived from Atm-/- mice also exhibit A-Trelated proliferative defects, including slower proliferation, defective cell cycle GJS progression, and premature senescence (Barlow et al., 1996; Elson et al., 1996; Xu and Baltimore, 1996). Therefore, the Atm-/- mice represent a valid mouse model to further study this spectrum of A-T defects. The A-T-related immune defects in Atm-/- mice include thymus hypoplasia, defective lymphoid differentiation, and impaired T-dependent immune responses. Compared to the normal mice, Atm-/- mice display an approximately threefold reduction of total thymocyte number. In addition, the number of CD4' or CD8' single-positive (SP) mature thymocytes, particularly that of the CD4+ ones, is more dramatically reduced in the Atm-/- mice, with an absolute number about 10% that of normal controls (Barlow et al., 1996; Elson et al., 1996; Xu and Baltimore, 1996). Accordingly, the number of T cells in the peripheral lymphoid organs of Atm-/- mice is also greatly reduced. However, although the absolute number of thymocytes is reduced in Atm-/- mice, the percentage of CD4-CD8- thymocytes representing the early precursor T lineage cells

182

YANG XU

is increased. Thus, the actual number of CD4-CD8- thymocytes is similar between the Atm-’- mice and normal controls, and the reduction of total thymocyte number in Atrn-l- mice is not due to fewer T lineage precursor cells in these mice (Xu et al., 1996). Similar to the T cell developmental defects, the number of B220’IgM- pre-B cells is also decreased in the bone marrow of Atm-’- mice when compared to that of normal mice (Xu et al., 1996). The reduction of the pre-B cells is not due to fewer B lineage precursor cells, because the number of B220TD43’ pro-B cells, which are the precursor cells differentiating into pre-B cells, is similar between the Atm-’- and normal control mice (Xu et al., 1996). In addition, the number of mature B cells remains similar between the normal and Am-/mice (Barlow et al., 1996; Elson et al., 1996; Xu and Baltimore, 1996). A-T-related immunodeficiency was also evident in Atm-’- mice. By challenging the Atm-‘- mice with T-dependent and T-independent antigens, Atm-’- mice were found defective in their T-dependent immune responses but normal in T-independent antibody responses. This suggests that Atm-’- B cells are functionallynormal whereas the T-help cell function and/or T-B cell interactions are defective in the mutant mice (Xu et al., 1996). However, studies of several Epstein-Barr virus-transformed human A-T B cell lines showed that cross-linking of the B cell receptor failed to induce mitogenic responses and Ca” mobilization from internal stores, suggesting that signaling through the B cell antigen receptor is impaired in the transformed A-T B cells (Khanna et al., 1997). In addition, although some A-T patients lack certain immunoglobulin isotypes, the serum levels of the various immunoglobulin isotypes are grossly normal in Atm-’mice. Therefore, it appears that B cell functions are better preserved in Atm-’- mice than in A-T patients. However, this conclusion should be confirmed by direct analysis of the signaling pathways in primary B cells derived from Atm-’- mice as well as from A-T patients. RAG-2 is required for V(D)J recombination, and disruption of the RAG-2 gene in mice leads to blockage of thymocyte development at the CD4-CD8- stage with a thymic cellularity of only about 1%of normal mice (Shinkai et al., 1992). Because introduction of a transgenic TCR-/3 chain into RAG-2-l- mice promotes thymocyte expansion, yielding a 100fold increase in thymus cellularity, it has been thought that a productively rearranged TCR-/3 chain is required for thymocyte expansion (Mombaerts et al., 1992; Shinkai et al., 1993). In addition, a productively rearranged TCR-a chain is required for thymocyte transition from the double-positive to the single-positive stage (von Boehmer, 1994). Therefore, the impaired T cell development in Atm-’- mice, represented by a 3-fold reduction of total thymocyte number and an average 10-fold reduction in mature thymocyte number, could be due to two possibilities:

ATM IN LYMPHOID DEVELOPMENT AND TUMORIGENESIS

183

(1)defects in thymocyte proliferation and (2) impaired V( D)J recombination processes leading to a lower frequency of productive V(D)J recombination. If the T cell developmental defects in Atm-’- mice are due to impaired V(D)J recombination, a productively rearranged transgenic TCR-aP chain should be able to rescue these defects in TCR-aP+Atm-’- mice. Indeed, initial analysis indicates that a productively rearranged TCR-aP chain could rescue the defective T cell differentiation in Atm-/- mice, suggesting that ATM, although not essential, is involved in the V(D)J recombination (Y. Xu, unpublished data). The V( D)J recombination process involves the introduction of DNA double-stranded breaks by RAG-1/2 at the recombination signal sequences flanking the V(D)J gene segments (Schatz, 1997). Subsequently, the recombination signaling sequences are precisely joined to form signal joints and the coding ends are usually imprecisely joined to form coding joints sometimes with the addition of N regions (Alt et al., 1987). ATM might play two possible roles during V(D)J recombination. Based on its structural similarity to DNA-PKcs, which is essential in the joining processes of the DNA double-stranded break repair, ATM might perform a similar function in the joining processes of the V(D)J gene segments (Blunt et al., 1995; Hartley et al., 1995). However, in contrast to the severe combined immune deficiency mice whose defect in DNA-PKcs leads to the absence of any IgM+ B cells, the B cells in Atm-’- mice produce functional antibody, suggesting that Atm might not be important in the joining processes of the DNA double-strand breaks during V(D)J recombination. In addition, although lymphocytes defective in DNA-PKcs fail to join correctly the coding ends of the V, D, and J gene segments (Mdynn et at., 1988), analysis of the V( D)J recombination activity using plasmid substrates in A-T human cells indicates that the signal and coding joint formation during V(D)J recombination is normal in A-T cells (Hsieh et al., 1993). Nevertheless, these data do not rule out the possibility that ATM and DNA-PKcs might have other redundant functions in signaling cellular responses to DNA damage, especially in light of the report that similar to ATM, DNA-PK also activates p53 responses following DNA strand-break damage (Woo et al., 1998). While there have been no data supporting a direct role of ATM in the repairing processes during the V(D)J recombination, it has been well established that ATM is required for the cell cycle checkpoint control at the GJS border, S phase, and G$M checkpoints following DNA strandbreak damage (Baskaran et al., 1997; Meyn, 1995; Shiloh, 1995; Xu and Baltimore, 1996). Because V(D)J recombination involves the repair of DNA double-stranded breaks, it is possible that ATM functions in the cell cycle regulation of V(D )J recombination. Based on the findings that RAG-

184

YANG XU

U2-induced DNA double-stranded breaks accumulate only in the Go/GI phase of the cell cycle (Lin and Desiderio, 1995; Schlissel et al., 1993), it has been suggested that V(D)J recombination is restricted to the Go/G1 phase in order to preserve the chromosomal integrity of lymphocytes by ensuring that DNA double-stranded breaks introduced during V(D)J recombination are not present during S phase (Lin and Desiderio, 1995). Consistent with this notion, DNA double-stranded breaks introduced by RAG-l/RAG-2 during V(D)J recombination could also activate ATMdependent signaling pathways leading to cell cycle GI arrest until V(D)J recombination is completed (Fig. 1).Therefore, in the absence of ATM activity, lymphocytes might enter the S phase before V(D)J recombination is completed. This could lead to a lower frequency of productive V(D)J recombination and increased chromosomal translocation involving immunoglobulin and TCR loci, because the V(D)J recombination-induced DNA double-stranded breaks may not be efficiently resolved in S phase due to reduced DNA end-joining activity, leading to increased potential of these broken ends to be misjoined to randomly generated DNA broken ends (Giaccia et al., 1985; Li et al., 1995; Paulovich et al., 1997). All published reports show that Atm-’- mice develop thymic lymphomas by the age of 4 months (Barlow et al., 1996; Elson et al., 1996; Xu and Baltimore, 1996). The lymphoma is composed of highly proliferating GO/Gl lymphocytes undergoing V(D)J recombination

n

/rDNAmpair7 \ Without Lymphocytes enter

Reduced frequencyof productive V(D)J recombination

completion of V@)J

Increased frequency of chrnmoeomal translocation involving Ig or TCR loci

FIG.1. A speculative role of ATM in V(D)J recombination. The introduction of DNA double-stranded breaks by FiAG-1/2 activates ATM through an unhown mechanism. The activated ATM signals the cell cycle checkpoint to arrest lymphocytes at GI phase until the RAG-l/2induced DNA double-stranded breaks are repaired. In the absence of a functional ATM, lymphocytes undergoing V(D)J recombination might progress through cell cycle S and G2/M phases before the RAG-1/2-induced DNA double-stranded breaks are repaired. Two outcomes can be predicted: a less efficient resolution of the DNA broken ends of the coding strands, leading to a lower frequency of productive V(D)J joints, and an elevated possibility of misjoining of RAG-1/2-induced DNA broken ends with random DNA broken ends.

ATM I N LYMPHOID DEVELOPMENT AND TUMORIGENESIS

185

lymphoblasts that were metastatic in one case (Barlow et al., 1996), but noninfiltrating in another (Xu et al., 1996). Nevertheless, in all cases, the tumors eventually kill the animals by filling up the chest cavity, suppressing the heart and lung. Analysis of the tumor cells with spectral karyotyping indicated multiple and recurrent chromosomal translocations (Barlow et al., 1996). In addition, chromosomal abnormalities involving the T cell receptor p locus have also been identified in the monoclonal tumor cells derived from Atm-’- mice (Xu et aE., 1996). Therefore, similar to the lymphoid tumorigenesis in A-T patients, genetic instability appears to play a role in the early onset of the lymphoid tumors in the Atrn-l- mice. As predicted by the potential role of ATM in V(D)J recombination (Fig. l),it is possible that the abnormal V(D)J recombination in ATM-/lymphocytes might promote frequent chromosomal translocations involving the TCR loci or immunoglobulin loci, leading to a high incidence of lymphomas by deregulating the expression of protooncogenes (Kojis et al., 1991; Xu et al., 1996). In support of this hypothesis, the onset of thymic lymphomas appears to be significantly delayed or suppressed in Atm-’RAG-2-I- mice that are completely deficient in V( D)J recombination (Y. Xu, unpublished observation). The suppression of tumor onset cannot be due solely to the greatly reduced thymocyte number in RAG-2-l- mice because the p53-’-RAG-2-’- mice develop thymic lymphomas with the same kinetics and at the same frequency as the p53-I- mice. This also suggests that the underlying mechanism of the high incidence of thymic lymphomas in Atm-’- and p53-I- mice may be distinct (Liao et al., 1998; Nacht and Jacks, 1998). Consistent with this notion, studies of the interactions between ATM and p53 in tumor suppression in the Atm-’-p53-’- mice showed that these double-mutant mice develop lymphomas significantly earlier than the single-mutant mice, indicating that Atm and p53 cooperate in tumor suppression (Westphal et al., 1997; Xu et al., 1998). While it is known that ATM is mainly responsible for signaling the p53 response following DNA strand-break damage, it is important to note that ATM and p53 also appear to have distinct functions in cellular responses to DNA damages. In this context, ATM signals the p53 responses only to DNA strand-break damage but is dispensable in the p53 responses to other genotoxic and cellular stresses, such as DNA damage induced by ultraviolet irradiation (Khanna and Lavin, 1993;Xu and Baltimore, 1996). In addition, following DNA strand-break damage, ATM also signals p53-independent pathways, including the ones leading to cell cycle S phase checkpoint control (Baskaran et al., 1997; Painter and Young, 1980). IV. Future Perspectives

Work summarized here has documented applications of the A-T mouse model, the Atm-/- mice, to elucidate the basis of lymphoid defects and

186

YANG XU

tumorigenesis in A-T. These studies have provided genetic evidence suggesting that ATM plays a role in the regulation of V(D)Jrecombination, and abnormality in this process could contribute to the genetic instability and tumorigenesis in ATM-’- lymphocytes. Thus, elucidation of the exact mechanism of such potential functions of ATM will be crucial to understand A-T lymphoid tumorigenesis. Genetic studies have so far failed to support the idea that ATM and DNA-PK have similar functions in V(D)J recombination, but the potentially redundant functions of these two related tumor suppressors in signaling cellular responses to DNA damage should be considered in both a genetic and biochemical context. Finally, identification of additional physiological substrates of the ATM kinase should help to delineate the role of ATM in V(D)Jrecombination as well as in general DNA double-stranded break repair. REFEERENCES Alt, F. W., Blackwell, T. K., and Yancopoulos, G. D. (1987). Development of the primary antibody repertoire. Science 238, 1079-1087. Barlow, C., Hirotsune, S., Paylor, R., Liyanage, M., Eckhaus, M., Collins, F., Shiloh, Y., Crawley, J. N., Ried, T., Tagle, D., and Wynshaw-Boris, A. (1996). Atm-deficient mice: A paradigm of ataxia telangiectasia. Cell 86, 159-171. Bashan, R., Wood, L. D., Whitaker, L. L., Canman, C. E., Morgan, S. E., Xu, Y., Barlow, C., Baltimore, D., Wynshaw-Boris, A,, Kastan, M. B., and Wang, J. Y. (1997). Ataxia telangiectasia mutant protein activates c-Abl tyrosine kinase in response to ionizing radiation. Nature 387,516-9. Beatty, D. W., Arens, L. J., and Nelson, M. M. (1986).Ataxia-telangiectasia.X,14 translocation, progressive deterioration of lymphocyte numbers and function, and abnormal in vitro immunoglobulin production. S. Af. Med. J. 69, 115-118. Blunt, T., Finnie, N. J., Taccioli, G. E., Smith, G. C., Demengeot, J., Gottlieb, T. M., Mizuta, R., Varghese, A. J., Alt, F. W., Jeggo, P. A., et al. (1995). Defective DNAdependent protein kinase activity is linked to V(D)J recombination and DNA repair defects associated with the murine scid mutation. Cell 80, 813-823. Brown, E. J., Bed, P. A., Keith, C. T., Chen, J., Shin, T. B., and Schreiber, S. L. (1995). Control of p70 s6 kinase by kinase activity of FRAP in viva Nature 377, 441-446. Canman, C. E., Chen, C. Y.,Lee, M. H., and Kastan, M. B. (1994). DNA damage responses: p53 induction, cell cycle perturbations, and apoptosis. CSH Symnp. Quant. Biol. 59, 277-286. Cohen, M. M., and Levy, H. P. (1989). Chromosome instability syndromes. Adv. Human Genet. 18,43-149. Cornforth, M. N., and Bedford, J. S. (1985). On the nature of a defect in cells from individuals with ataxia-telangiectasia. Science 227, 1589-1591. Elson, A,, Wang, Y., Daugherty, C. J., Morton, C. C., Zhou, F., Campos-Torres, J., and Leder, P. (1996).Pleiotropic defects in ataxia-telangiectasiaprotein-deficient mice. Proc. Natl. Acad. Sci. U.S.A. 93, 13084-13089. Giaccia, A., Weinstein, R., Hu, J., and Stamato, T. D. (1985). Cell cycle-dependent repair of double-strand DNA breaks in a gamma-ray-sensitive Chinese hamster cell. Somut. Cell. Mot. (2nd.11, 485-491.

ATM IN LYMPHOID DEVELOPMENT AND TUMORICENESIS

187

Greenwell, P. W., Kronmal, S. L., Porter, S. E., Gassenhuber, J., Obermaier, B., and Petes, T. D. (1995). TEL1, a gene involved in controlling telomere length in S. cereoisiae, is homologous to the human ataxia telangiectasia gene. Cell 82, 823-829. Hartley, K. O., Cell, D., Smith, G. C., Zhang, H., Divecha, N., Connelly, M. A,, Admon, A., Lees-Miller, S. P., Anderson, C. W., and Jackson, S. P. (1995). DNA-dependent protein kinase catalytx subunit: A relative of phosphatidylinositol3-kinase and the ataxia telangiectasia gene product. Cell 82, 849-856. Hecht, F., and Hecht, B. K. (1987). Chromosome changes connect immunodeficiency and cancer in ataxia-telangiectasia.Am. J. Pediatr. Henuztol. Oncol. 9, 185-188. Hecht, F., McCaw, B. K., and Koler, R. D. (1973). Ataxia-telangiectasia-Clonal growth of translocation lymphocytes. N . Engl. J. Med. 289, 286-291. Hsieh, C. L., Arlett, C. F., and Lieber, M. R. (1993). V(D)J recombination in ataxia telangiectasia, Bloom’s syndrome, and a DNA ligase I-associated immunodeficiency disorder. J. Biol. Chem. 268,20105-20109. Jayaraman, L., Murthy, K. G., Zhu, C., Curran, T., Xanthoudakis, S., and Prives, C. (1997). Identification of redoxhepair protein Ref-1 as a potent activator of p53. Genes Deu. 11,558-570. Kastan, M. B., Zhan, Q., el-Deiry, W. S., Carrier, F., Jacks, T., Walsh, W. V., Plunkett, B. S., Vogelstein, B., and Fornace, A. J., Jr. (1992). A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia.Cell 71,587-597. Khanna, K. K., and Lavin, M. F. (1993). Ionizing radiation and UV induction of p53 protein by different pathways in ataxia-telangiectasiacells. Oncogene 8, 3307-3312. Khanna, K. K., Yan, J., Watters, D., Hobson, K., Beamish, H., Spring, K., Shiloh, Y., Gatti, R. A., and Lavin, M. F. (1997). Defective signaling through the B cell antigen receptor in Epstein-Barr virus-transformed ataxia-telangiectasiacel1s.J. Biol. Chem. 272,9489-9495. Kojis, T. L., Gatti, R. A,, and Sparkes, R. S. (1991).The cytogenetics of ataxia telangiectasia. Cancer Genet. Cytogenet. 56, 143-156. Lehmann, A. R. (1982). “Ataxia-Telangiectasia: A Cellular and Molecular Link between Cancer, Neuropathology and Immune Deficiency” (B. A. Bridges and D. G. Hamden, eds.), pp. 83-102. Wiley, New York. Levis, W. R., Dattner, A. M., and Shaw, J. S. (1979). Selective defects in T cell function in ataxia-telangiectasia.Clin. Exp. Immunol. 37, 44-49. Li, Z., Otevrel, T., Gao, Y., Cheng, H. L., Seed, B., Stamato, T. D., Taccioli, G. E., and Alt, F. W. (1995). The xRCC4 gene encodes a novel protein involved in DNA doublestrand break repair and V(D)J recombination. CelZ 83, 1079-1089. Liao, M. J., Zhang, X. X., Hill, R., Gao, J., Qumsiyeh, M. B., Nichols, W., and Van Dyke, T. (1998). No requirement for V(D)J recombination in p53-deficient thymic lymphoma. Mol. Cell. Biol. 18, 3495-3501. Lin, W. C., and Desiderio, S. (1995). V(D)J recombination and the cell cycle. Zmmunol. Today 16, 279-289. Malynn, B. A,, Blackwell, T. K., Fulop, G . M., Rathbun, G. A,, Furley, A. J., Ferrier, P., Heinke, L. B., Phillips, R. A,, Yancopoulos, G. D., and Alt, F. W. (1988). The scid defect affects the final step of the immunoglobulin VDJ recombinase mechanism. Cell 54,453-460. Metcalfe, J. A,, Parkhill, J., Campbell, L., Stacey, M., Bigs, P., Byrd, P. J., and Taylor, A. M. (1996). Accelerated telomere shortening in ataxia telangiectasia. Nature Genet. 13,350-353. Meyn, M. S. (1995). Ataxia-telangiectasia and cellular responses to DNA damage. Cancer Res. 55, 5991-6001.

188

YANG XU

Mombaerts, P., Clarke, A. R., Rudnicki, M. A., Iacomini, J., Itohara, S., Lafaille, J. J., Wang, L., Ichikawa, Y., Jaenisch, R., Hooper, M. L., et al. (1992). Mutations in T-cell antigen receptor genes alpha and beta block thyrnocyte development at different stages. Nature 360,225-331. Morrell, D., Cromartie, E., and Swift, M. (1986). Mortality and cancer incidence in 263 patients with ataxia-telangiectasia.1.Natl. Cancer lnst. 77, 89-92. Morrow, D. M., Tagle, D. A., Shiloh, Y.,Collins, F. S., and Hieter, P. (1995). TEL1, an S . cerevisiae homolog of the human gene mutated in ataxia telangiectasia, is functionally related to the yeast checkpoint gene MEC1. Cell 82, 831-840. Nacht, M., and Jacks, T. (1998). V(D)J recombination is not required for the development of lymphoma in p53-deficient mice. Cell Growth B f e r . 9, 131-138. Narducci, M. G., Virgilio, L., Isobe, M., Stoppacciaro, A., Elli, R., Fiorilli, M., Carbonari, M., Antonelli, A,, Chessa, L., Croce, C. M., et al. (1995). TCLl oncogene activation in preleukemic T cells from a case of ataxia-telangiectasia. Blood 86, 2358-2364. Oxford, J. M., Harnden, D. G., Parrington, J. M., and Delhanty, J. D. (1975). Specific chromosome aberrations in ataxia telangiectasia. J. Med. Genet. 12, 251-262. Paganelli, R., Capobianchi, M. R., Matricardi, P. M., Cioe, L., Seminara, R., Dianzani, F., and Aiuti, F. (1984).Defective interferon-gamma production in ataxia-telangiectasia. Clin. lmmunol. lmmunopathol. 32, 387-391. Painter, R. B., and Young, B. R. (1980). Radiosensitivity in ataxia-telangiectasia:A new explanation. Proc. Natl. Acad. Sci. U.S.A. 77, 7315-7317. Paulovich, A. G., and Hartwell, L. H. (1995). A checkpoint regulates the rate of progression through S phase in S. cerevisiae in response to DNA damage. Cell 82, 841-847. Paulovich, A. G., Toczyski, D. P., and Hartwell, L. H. (1997). When checkpoints fail. Cell 88,315-321. Peterson, R. D., and Funkhouser, J. D. (1989). Speculations on ataxia-telangiectasia:Defective regulation of the immunoglobulin gene superfamily. lmmunol. Today 10,313-314. Peterson, R. D. A,, and Good, R. A. (1968). Ataxia-telangiectasia. In “Birth DefectsImmunologic Deficiency Diseases in Man” (D. Bergsma and R. A. Good), pp. 370-377. National Foundation, March of Dimes, New York, New York. Savitsky, T. K., Bar-Shira, A., Gilad, S., Rotman, G., Ziv, Y., Vanagaite, L., Tagle, D. A., Smith, S., Uziel, T., Sfez, S., et ul. (1995). A single ataxia telangiectasia gene with a product similar to PI-3 kinase. Science 268, 1749-1753. Schatz, D. G. (1997). V(D)J recombination moves in uitro. Semin. Immunol. 9, 149-159. Schlissel, M., Constantinescu, A., Morrow, T., Baxter, M., and Peng, A. (1993). Doublestrand signal sequence breaks in V(D)J recombination are blunt, 5’-phosphorylated, RAGdependent, and cell cycle regulated. Genes Den 7, 2520-2532. Shiloh, Y. (1995). Ataxia-telangiectasia: Closer to unraveling the mystery. Eur. J. Human Genet. 3, 116-138. Shinkai, Y., Rathhun, G., Lam, K. P., Oltz, E. M., Stewart, V., Mendelsohn, M., Charron, J., Datta, M., Young, F., Stall, A. M., et al. (1992). RAG-2-deficient mice lack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell 68, 855-867. Shinkai, Y., Koyasu, S., Nakayama, K., Murphy, K. M., Loh, D. Y., Reinherz, E. L., and Alt, F. W. (1993).Restoration of T cell development in RAG-2-deficient mice by functional TCR transgenes. Science 259,822-825. Stem, M. H., Theodorou, I., Aurias, A., Maier-Redelsperger, M., Debre, M., Debre, P., and Griscelli, C. (1989). T-cell nonmalignant clonal proliferation in ataxia telangiectasia: A cytological, immunological, and molecular characterization. Blood 73, 1285-1290. Strober, W., Wochner, R. D., Barlow, M. H., McFarlin, D. E. and Waldmann, T. A. (1968). Immunoglobulin metabolism in ataxia telangiectasia. 1.Clin. Znuest. 47, 1905-1915.

ATM IN LYMPHOID DEVELOPMENT AND TUMORIGENESIS

189

Tamura, T., Chiba, S., Abo, W., Chiba, Y., and Nakao, T. (1980). Cytomegalovirus-specific lymphocyte transformations in subjects of different ages with primary immunodeficiency. Infect. Immun. 28, 49-53. Taylor, A. M., Metcalfe, J. A., Thick, J., and Mak, Y. F. (1996). Leukemia and lymphoma in ataxia telangiectasia. Blood 87, 423-438. von Boehmer, H. (1994). Positive selection of lymphocytes. CeZt 76, 219-228. Westphal, C. H., Rowan, S., Schmaltz, C., Elson, A,, Fisher, D. E., and Leder, P. (1997). atm and p53 cooperate in apoptosis and suppression of tumorigenesis, but not in resistance to acute radiation toxicity. Nature Genet. 16, 397-401. Woo, R. A., McLure, K. G., Lees-Miller, S. P., Rancourt, D. E., and Lee, P. W. (1998). DNA-dependent protein ldnase acts upstream of p53 in response to DNA damage. Nature 394, 700-704. Xu, Y., and Baltimore, D. (1996). Dual roles of ATM in the cellular response to radiation and in cell growth control. Genes Deu. 10, 2401-2410. Xu, Y., Yang, E. M., Brugarolas, J., Jacks, T., and Baltimore, D. (1998). Involvement of p53 and p21 in cellular defects and tumorigenesis in atm--/- mice. Mol. Cell. Biol. 18,4385-4390. Xu, Y., Ashley, T., Brainerd, E. E., Bronson, R. T., Meyn, M. S . , and Baltimore, D. (1996). Targeted disruption of ATM leads to growth retardation, chromosomal fragmentation during meiosis, immune defects, and thymic lymphoma. Genes Deu. 10, 2411-2422. Zakian, V. A. (1995). ATM-related genes: What do they tell us about functions of the human gene? Cell 82,685-687. This articIe was accepted for publication on September 9, 1998.

This Page Intentionally Left Blank

ADVANCES IN IMMUNOLOGY, VOL. 72

Comparison of Intact Antibody Structures and the Implications for Effector Function USA J. HARRIS, STEVEN 8. LARSON, AND ALEXANDER McPHERSON Deportment of Malecubr Bio/ogy and Biochemishy, Universily of California, /wine, Calibrnia 92697

1. Inkduction

Immune systems of vertebrates rely on antibody molecules, structurally dynamic glycoproteins, for coupling foreign body recognition with elimination processes. An intact immunoglobulin G (IgG) molecule is composed of four polypeptide chains, two identical light (L) chains and two identical heavy (H) chains, with variable (V) and constant (C) regions, which form two Fab segments and one Fc segment for a total of 12 domains. The domains associate in pairs (VL:VH, CL:CHi,CHz:CHz, and c H 3 : c H 3 ) , producing six globular units within the three segments. Hypervariable loops (L l , L2, L3, H1, H2, and H3) make up the antigen-binding site on the variable domain pair ( VL:VH)of each Fab, whereas effector sites are located primarily on the Fc (for further review, see Davies and Chacko, 1993). Inherent flexibility of the immunoglobulin yields a spectrum of molecular conformations, making crystallization problematic. Consequently, progress in intact IgG structural elucidation has been painfully slow, spanning the past 30 years. Eight intact IgG crystals have so far been reported: Dob (Terry et al., 1968), Mcg (Edmundson et al., 1970), Kol (Palm and Colman, 1974), Zie (Ely et al., 1978), Mab231 (Larson et al., 1991; Harris et al., 1995), Mab4B7 (Stura et al., 1994), Mab24-404.1, and Mab61.1.3 (Harris et al., 1995). Five of these yielded structures, or partial structures; Zie, Mab4B7, and Mab24-404.1 are the exceptions. Other classes of intact immunoglobulin, i.e., IgA, IgD, IgE, and IgM, have not been crystallized, only IgG. As opposed to entire Ig, fragments of antibodies, such as Fabs, Fc, and Fvs, are more readily crystallized, and from their analyses the /3-barrel topologies of individual antibody domains have been defined at high resolution. Fragment studies have, in addition, provided insight into mechanisms of antigen binding by the complementarity-determining regions (CDRs), and of some effector protein interactions with the Fc (for reviews, see Colman, 1988; Sheriff, 1993; Wilson and Stanfield, 1994; Braden and Poljak, 1995; Edmundson et al., 1995; Davies and Cohen, 1996; Padlan, 1996). Structural information on the intact molecule is currently confined to Dob, Mcg, Kol, Mab231, and Mab61.1.3. In every case, the IgG was visualized in its free state, not associated with either antigen or effector. 191

Copyright 0 1909 by Academic Press. All rights of reproduction ~ 1 1any form reserved. 0065-2776/9(1$30.00

192

LISA J. HARRIS et al

Comparison of the five intact antibodies whose structures are known delineates to some extent the dynamic range of IgG, and it serves to clarify some features of antigen clearance through effector function processes. This is of relevance in deducing mechanisms by which binding of antigenbearing targets is allied with the activities of effectors such as C l q and Fcy receptors. Our intent here is to summarize the structures of the intact antibodies available, both visually and geometrically (Table I), and to describe the dispositions of biologically relevant sites on the molecules. II. Hinge-Deleted Dob and Mcg

Dob, Mcg, and Kol are human myeloma proteins obtained from patients. Both Dob and Mcg are of the IgG1 subclass but contain “genetic” hinge deletions (Table 11). A mere seven amino acids of the CH2gene product (structural lower hinge residues in normal antibodies) serve as a hinge bypass peptide. Thus these IgGs, without upper and core hinge regions, have no inter-heavy-chain disulfide bonds. Light chains instead are linked by disulfide bridges between their terminal cysteines. This further restricts the molecule’s flexibility, particularly that of the Fabs. The deleted hinge also causes the Fc to be drawn tightly against the Fab segments. As a consequence, Dob and Mcg are conformationally restricted and their structures exhibit compact, twofold symmetrical T shapes (Figs. 1 and 2, see color plate) (Silverton et al., 1977; Guddat et al., 1993). Although the natural antigen is unknown for both Dob and Mcg, the latter is known to bind a variety of synthetic peptides on solid supports (Edmundson et al., 1995), thus normal antigen-binding properties are not believed altered. Effector functions are, however, obstructed (see below). While Dob and Mcg are unable to perform in the same capacity as antibodies possessing hinges, it is still an open question as to whether they play some meaningful role in the immune system, About 1%of a normal human IgG pool consists of hinge-deleted antibodies. These could represent genetic recombination errors (Steiner and Lopes, 1979), but this is still very speculative. In Fig. 2, a model of Mcg, based on a 3.2 A resolution crystal structure determination (Guddat et al., 1993), is presented. The two halves of the molecule are related by an exact (crystallographic) twofold axis. Lack of hinge polypeptides, and the presence of an L-L disulfide bond joining light chain termini, not only impose a compact T shape on the molecule, but they also result in noncanonical domain pairing for VL:VH, CJ.:CH~, C$:CH~,and cH3:CH3.Consequently, Fabs, as well as the Fc of Mcg, do not have the characteristic quaternary structures that are found for other antibody fragments, or for Fab and Fc segments within intact antibodies

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

193

containing normal hinges. Furthermore, the Fabs are inclined slightly toward the Fc. Thus the angle between Fabs is obtuse at 185", and the hinge angle between each Fab and the Fc is acute at 87.5". Intact, hinge-deleted Dob is a lower resolution structure for which only alpha carbon coordinates are available. It was originally modeled at 6 A by fitting Fab and Fc segments, determined from fragment crystals by X-ray diffraction, into the electron density for the Dob crystals (Silverton et al., 1977). Like Mcg, this model has a crystallographic dyad relating halves, an angle between Fabs of 189, and hinge angles of 87.5" (Fig. 1). Domains were later fitted to 4 A resolution X-ray data (Sarma and Laudin, 1982) but, as yet, coordinates from that investigation are not available. 111. Partial Struchrre of Kol

Similar to Dob and Mcg, the immunoglobulin Kol is a human IgG1,but possesses normal hinge polypeptides (Table 11). Crystals ofintact Kol yielded an electron density map that revealed only approximately two-thirds of the molecule (Marquart et al., 1980). The Fc segment was crystallographically disordered. The two Fab segments were defined, however, in the Fourier maps, including also the upper and core hinge regions. Statistical disorder appeared with the lower hinge and propagated to the Fc, which, probably as a rigid body, occupied at least several orientations in the crystal lattice. This suggested that, for antibodies with normal hinges, the disposition of the Fc is, in general, not fixed with regard to the Fab segments, but is free to assume a variety of positions, perhaps even an unlimited number. Though the structure of intact antibody Zie has not been determined, evidence indicates that the Fc in its crystal is also mobile (Ely et al., 1978), as for Kol. Zie, of human subclass IgG2,was crystallized in its intact form, but F(ab'), fragments made from Zie were crystallized in addition. Crystals of the two were essentially isomorphous, with mean differences in corresponding intensities nearly within experimental error. This implied that the Fc did not contribute significantly to the diffraction pattern of the intact IgG. The observation that both Kol and Zie had multiple conformations of Fc in their crystals proved a discouraging reminder of how the flexibility of immunoglobulins could hinder crystallographic analyses. For Kol, as for Dob and Mcg, an exact, crystallographic twofold axis related the Fab segments (Fig. 3, see color plate). Although the Fc portion of the molecule was not visible, the angle between long axes of the Fabs, defined as the line connecting the centers of masses of VL:VHand C,:C,i, was 132", lending a Y shape to the molecule (Fig. 3A). Figure 3B shows that the Fab long axes do not intersect but are offset by 11A. The corresponding translation offset of Fab axes is 9 A for Mcg, 10 A for

TABLE I COMPARISON OF INTACT ANTIBODY STRUCTURES

E

Antibody Characteristic

Dob

Mcg

Kol

Mab231

Mab61.1.3

2IG2 Human myeloma IgGu Unknow

lIGT Mouse hybridoma IgG,, Canine lymphoma 23

lIGY Mouse hybridoma

IgGK

Yes Cryoglobulin-like 4% PEG 3350, pH 8, 18°C 2.8 A Triclinic, P1

No None 12% PEG 3350, pH 5, 23°C 3.2 A Monoclinic, P21

____~ ~

PDB code Source Isotype Antigen Number of residues in hingeb Number of inter-heavychain disulfides in core hinge Activates complement Special solubility characteristic Crystal conditions Resolution of structure Space group

Not deposited” Human myeloma IgcI, Unknown 7 0

No Clyoglobulin 0.1 M Tris, pH 8, 4°C 4.0A Monoclinic, C2

lMCO Human myeloma I@lA UlllalOwJl

~~

7 0

22

No Euglobulin Deionized water, 20°C 3.2 A Orthorhombic, c222,

unlcnown Cryoglobulin 1.5 MAS, pH 5

2

3.0 A Trigond, P3,21

Phenobarbital 17 3

Asymmetric unit Overall symmetly Angle between Fabs Translation offset between Fab long axes Fab elbow angles Hinge angles Angle between Fab dyad and Fc dyad Translation offset of Fab dyad and Fc dyad Translation between Fabs along Fab dyad Distance between H3s Distance from CH3sto H3 of Fabl and Fab2 Overall conformation of structure

4i Ah Exact dyad 185" 9A

4i Ah Exact dyad 132" 11 A

1 Ab None 172" 10 A

6.k

174" da'

0"

118" 87.5" 0"

n/a

159" and 143" 66" and 113" 128'

155" 78" and 123" 107"

OA

OA

da

26

A

OA

O A

OA

OA

0.9 '4

9.4

156 A 99 A

120 A 95 A

n/a

151 A 131 hi and 119 '4

118 A 108 A and 126 A

Symmetric T

Symmetric T

Symmetric Y

Asymmetric, distorted T

Asymmetric, distorted Y

Y2 Ab

Exact dyad 185" OA 147" 87.5"

142

A

Dob alpha carbon coordinates, at 6 A resolution, were kindly provided by David R. Davies for this analysis For Dab and Mcg these residues are considered a hinge "bypass;" see Table I1 also. 'd a , not applicable. a

1 Ab None 115"

196

LISA J. HARRIS et d.

TABLE I1 HINGESEQUENCES FOR INTACT IgG STRUCTURES“ Antibody

Upper

Dob Mcg Kol Mab231 Mab61.1.3

- -- _ - - - - - - -- - - - - - - -

Core

__--_ -_-_-

EPKSCDKTHT CPPCP EPRGPT I KP CPPCKCP VPRDCG CKPCICT hing-Genetic -Structural hinge-

Lower APELLGG APELLGG APE LLGG APNLLGG VPEV

Heavy chain sequences are compared from amino acid 226 to 250 [in Kabat numbering (Kabat et al., 1991)] corresponding to the hinge polypeptides. AU Cys in the hinge core form inter-heavy-chain disulfide bonds. For Kol and Mab61.1.3, the Cys of the upper hinge supplies the heavy chain’s contribution to the heavy-light chain disulfide bridge. For Mab231, Cys128 of the CH1domain engages in a disulfide with the light chain terminus Cys-214. The structural lower hinge residues are genetically part of the CHz gene product.

Mab231, and 6 A for Mab61.1.3 (Table I). The Dob structure, at 6 A resolution, was modeled with no offset. This, however, preceded the higher resolution structures when an offset was unanticipated. In Kol, the normal human IgGl hinge is such that the terminal cysteine of each light chain bonds with a cysteine in the upper hinge of the molecule. Two inter-H-chain disulfides form the hinge core, producing a polyproline helix through the assembly (Fig. 3A). The packing of the Kol molecule in the unit cell utilized hypervariable loops (CDRs) interacting with the upper and core hinge regions of crystallographically related molecules, thereby stabilizing those sections of the hinge. Thus packing contacts were likely responsible for the clarity with which those portions of the hinge polypeptides appeared in electron density maps.

N.Mab231 Obtaining a well-ordered crystal is to arrange conformationallyidentical molecules in a periodic, three-dimensional array. Even with a conformationally variable molecule, however, this can be achieved if the crystal selects only a single conformer for incorporation. In such cases, X-ray crystallography may yield a single “still image” of this select conformation if all segments are immobilized in the crystal through favorable packing interactions. Mab231 and Mab61.1.3 (murine hybridoma Mabs) have normal hinge polypeptides representative of their respective subclasses, IgGz,

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

197

and IgGI. In both cases the three-dimensional structures of the entire molecules were deduced by X-ray crystallography (Harris et al., 1992,1997, 1998). They exhibit strikingly different conformations, both adopted as a consequence of specific, stabilizing crystal lattice interactions. These X-ray crystallographic images provide but two states, among a likely multitude, that are possible for IgG molecules with normal hinge polypeptides. Mab231, specific for canine lymphoma cells, exhibits a distorted T shape (Fig. 4, see color plate). The angle between Fab long axes is 172", and the Fc is obliquely disposed with hinge angles of 66" and 113", while making an angle of 128"with respect to the plane of the Fabs. The structure contains no global symmetry operators as were found for Dob, Mcg, and the ordered portion of Kol. Though the structure appears overall asymmetric, local pseudotwofold axes (dyads) do pertain within or between segments of the IgG. One dyad relates only constant domain pairs of the Fabs. CL:CHl of Fabl is related to CL:CHlof Fab2 by a rotation of 179.4". A second, independent, pseudodyad relates heavy chains composing the Fc. Fab elements have been shown, by earlier crystallographic studies of Ig fragments, to be flexible at their switch peptides, which connect variable and constant domains. The Fab "elbow" is defined as the angle bemeen the pseudotwofold axes relating VL to VH,and CL to CHI. Elbow angles have been observed from 127-227" (for a review, see Wilson and Stanfield, 1994). In Mab231, one Fab assumes an elbow angle of 159" while that of the second is less obtuse at 143". As a consequence of the 16" difference in elbow angles, VL:VH pairs are not related by the local dyad, which relates Fabs by their CL:CHl pairs. For convenience, we will refer to the local dyads within Mab231 as the "Fab dyad' and the "Fc dyad," though both are only approximate. As noted above, the angle between the Fab dyad and the Fc dyad is 128". Moreover, the dyads do not intersect one another, but are offset by 26 A (Fig. 4).This translation further emphasizes the independent nature of the Fc with respect to the Fabs. The canine lymphoma antibody packed in the crystallographic unit cell such that hypervariable loops of each Fab contacted Fc segments of latticerelated molecules, specifically at the switch peptide junctions between CH2and CH3 domains. Contacts between CDRs and the Fcs of adjacent molecules stabilized both Fabs as well as the Fc segment. This was not the case for crystals of antibody Kol, and is probably why its Fc segment was not visible in electron density maps. The Mab231 hinge polypeptides, on the other hand, did not benefit from packing contacts, and although their paths were ultimately determined from difference Fourier maps, they were seen to exhibit high thermal factors indicative of some statistical disorder (Harris et al., 1997).

198

LISA J. HARRIS et QZ.

V. Mab61.1.3

A model of the intact monoclonal antibody against phenobarbital, Mab61.1.3, is shown in Fig. 5 (see color plate) (Harris et al., 1998). Like the IgGz, for canine lymphoma, this IgGl conformation exhibits no overall symmetry, though it too possesses local dyads relating segments. Hinge angles, elbow angles, the angle between Fabs, and the angle between dyads are different, thus providing another "still image" of an IgG. This antibody displays a distorted Y shape with an angle between Fab long axes of 115". The Fc is oriented at 107" with respect to the plane of the Fabs. Heavy chains of the Fc segment are approximately related by a local twofold axis, similar to Mab231 and, indeed, to all Fc fragment structures. As for other structures of Fc segments, the twofold axis tends to degenerate with respect to the CH2domains, and for Mab61.1.3 the rotation relating CH2domains is about 175". IgG, Fab segments both have elbow angles of 155" and they are related by a rotation of 179.7'. There is, however, a relative translation of the Fabs along the rotation axis of 9 A (Fig. 5A), thus it is more properly described as a screw axis. For Dob, Mcg, and Kol, exact crystallographic symmetry precludes any possibility for such a translation between Fabs. For Mab231, which crystallized in a triclinic unit cell lacking any symmetry operators, the translation is 0.9 A, which is but marginally significant. The structure of Mab61.1.3 thus identifies yet another degree of freedom available to the IgG. Again, for convenience, we may refer to the axis relating Fab segments as the "Fab dyad," but with the provision of a 9 A translational component. In contrast to Mab231, the IgGl Fab dyad exactly intersects the dyad relating Fc heavy chains, that is, there is no translational offset. The 107' angle between the Fab and the Fc dyads and the differing hinge angles of 78" and 123" establish the independence of segments within the molecule. The angle between Fab segments of 115" is the sharpest angle observed for intact antibody structures and, indeed, may represent the limit for this subclass. Any angle much more acute would result in unacceptable steric interactions between murine IgGl Fabs. The distribution of molecules in crystals of Mab61.1.3 was such that the VH domain of every Fab was in contact with the VH domain of a nearby IgG. These interactions involved, specifically, the CDR regions. Intermolecular P ribbons were formed as a result of hydrogen bonding between the strand following hypervariable loop H2 (residues 55-59)' and the corresponding strand of the VH of an adjacent antibody in the unit cell. While Fabs engaged in VHlattice contacts, their V, domains contacted the Fc of another antibody. These specific interactions stabilized the three The residue numbering convention used herein is that of Kabat et al. (1991)

CH. 5, FIG.1. The 6 A resolution crystal stiucture of the hinge-deleted immunoglobulin Doh. Light chains are in gold, and heavy chains are two shades of orange. The crystdogrdphic twofold axis is shown by a solid white line. The long axes of the two Fdbs are indicated by white spheres. As seen here, the angle between Fab long axes is 18.5".The angles between each Fab long axis and the Fc dyad, tvrined hinge angles, are 87.5". The molecule exhibits a compact, syininetrical T shape.

CH.5 , FIG.2. Molecular structure of the hinge-deleted antibody Mcg at 3.2 A resolution. Light chains are beige, and heavy chains are colored purple and blue. The two oligosaccharide chains are indicated in red and orange. An interchain disulfide bond, which connects light chain termini, is shown in yellow. No inter-heavy-chain disulfides appear i n genetic hinge-deleted antibodies. The crystallographic dyad that relates halvcs of the IgG is displayed as a lime green line. Like Dob, Mcg assumes a compact, s y n nietrical T shape.

C H .5 , FIG.3. Partial structure of the intact imtnunoglohulin Kol based on 3 A resolution x-ray diffraction data. Fab segments and the upper and core hinge regions were visible in electron density maps, whereas the lower hinge region and Fc were not. Light chains are pink, and heavy chains of the two Fabs are red. (A) The two light-heavy chain &sulfide bonds of the upper hinge, and the two inter-heavy-chain disulfide bonds of the core hinge, are highlighted in yellow. The crystallographic dyad is indicated by a pale pink line. The angle between Fab long axes is 132", yielding a symmetric Y shape. ( B ) It is evident that Fab long axes do not intersect hut are offset by 11 A.

Cii. 5, FIG.4. The 2.8 A resolution structure of Mab231, an IgC& against canine lymphoma cells. Light chairis arr lime green; h e a y chains are orange and red. Fab long axes are white, and the local Fab dyad and Fc dyad are both in blue. (A) The molecule is viewed perpendicular to the pseudodyad relating C L : C H ~domain pairs of the Fab segments. The Fab long axes arc nearly colinear, with an angle between them of 172". Fah long axes do not intersect, but as seen in A are offset by 10 A. (B) The IgGe, is viewed approximately perpendicular to the pseudodyad relating Iieay chairis of the Fc. In both A and B it is apparent that the Fat) and Fc dyads do riot intersect

but are offset by a translation of 26 A. In B, hinge angles are clearly observed: the angle between one Fah long cuis and the Fc dyad is 66", and thr hinge angle hetween the other Fab and the Fc ib 113". ( C )The angle between the Fab and Fc dyads, 128", is most evident. The Fc segment assuined an orientation 128"with respect to the plane of the Fabs. The conformation of Mab231 is best described

a

CII.S, FIG.5. Crystallographic structure of Mab61.1.3. an IgGl specific for phenobarbital. based on 3.2 resolution x-ray diffraction data. Light chains are yellow, and heavy ch;uns are shown in cyan and blue. Rib long axes are white, and the local Fab dyad and Fc dyad are in red. (A and B ) The angle between Falls is seen as being a rather closed 115", with the Fc segment oriented rouglily perpendicular (107")to the plane of the Fabs. The two local pseudodyads for this Mab exactly intersect, in contrast to Mab231, which exhjbits a 26 offset. The angle between dyads is 107" lor Mabti1.1.3, in contrast to Mab231. where it is 128". In A. a relative translation o f 9 A tietween Fahs along their dyad can be secn, and in C, there is a translation offset of 6 A for their long axes. (C) The molecule is viewed along the pseudodyad relating Fall segnieiits, which hest illustrates the differing hinge angles. The hinge angle lietween one Fab long axis and the Fc dyad is 78". whereas that between the second Fah and the Fc is 123". The conformation of Mabti1.1.3 is best descrihed as a hstorted Y shape.

a

CH. 5, FIG.6. Comparison of conformations for the two intact monoclonal antibodies, Mab231 and Mab61.1.3. Intact Mab231 is displayed in red and Mab61.1.3 in yellow. (A) The C L : C H ~domains of one Fah of Mab61.1.3 are superimposed (as a unit) onto the corresponding domains of FabZ from Mab231. Variable domains do not superimpose but are shifted by 12", as a consequence of differing Fah elbow angles (Table I). (B and C) The cH3:cH3 domain pairs of the two Fc segments are superimposed. The conformations of the two antibodies are quite unique.

CH. 5, FIG.7 . Mapping of the distribution of IgC residues involved in key antigen and effector inolecule inter-

actions. Structures of (A) Mahfil.l.3, (B) Mab231, and ( C ) Mcg. I n all cases heavy chains are white, while light chains are blue, lime green, and purple, respectively. Antigen binding sites are indlcated by ball and stick at the extremes ofthe Fab arms. The four lower hinge residues (cyan)immediately preceding Pro-251 arcs involved in Fcy rcxrptor binding. Lower hinge amino acids have also been implicated in Clq binding specificity. Core Clq binding residues (orange), carried by all IgG isotypes, are Glu-337, Lys-3.39, and Lys-341 (or Arg-341 in ~ n n r i wIgG,). A hinge-proximill loop containing Pro-350 (magenta) is snggested to participate in both Clq and Fcy receptor inter3 bacterial proteins A and C: have coninion contact points shown in green: action. At the c : ~ Z - c ~ junction, residnes specific for protein A are in red and for protein 6, dark purplc. Protein G has a second interaction site on CHI domain (yellow). Rheiimatoid factor and rat neonatal Fc receptor hintling sites overlap the area where ~ Oligosaccharides (gray) are generally required for proper elfector proteins A and G contact C H ~ - C H(green). functioning but have not, as yet, been implicated in forming binding sites. (A) For Mab61.1.3, it should be noted that although lower hinge residues are colored cyan, human F q R l does not bind mouse IgC 1 because these residues lack the critical Leu-Leu-Gly-Cly-Pro-2.51binding sequence present in inousc IgCz, and human IgG] . ( C )For comparison purposes, effector sites are mapped on Mcg, but notct that hinge-deleted Mcg does not have nornial effector hinction capal)ilities like molecules with hinges.

CH. 6, FIG.3. Selective expression of the mucosal vascular addressin, MAdCAM-1, on postcapillary venules involved in lymphocyte homing to gastrointestinal sites. Immunohistologic staining reveals MAdCAM-1 on high endothelid venules in Peyer’s patches (PP) and on postcapillary venules in the small intestinal lamina propria (SI).In contrast, high endothelial venules in axillary lymph nodes (PLNs) are negative.

CH. 6 , FIG.5B. Stereo view of the crystal structure of an N-terminal two-domain fragment of human MAdCAM-I (68).The two most significant Ioops involved in a4PS binding, the GLDTScontaining CD loop on domain 1 and the negatively charged DE p ribbon, or “antenna,” on domain 2, are highlighted in yellow (with piirple stripes). The key D42 side-chain in domain 1 is shown as a stick-and-ball model with red oxygens: R70, important for maintenance of the structure of the CD and EF loops (the EF loop, with a short piece of helix in D1, is in blue), is shown with its oxygens in blue. The most critical residues in the domain 2 DE antenna, D156, and E151, E152, and E157, are also highlighted.

CII.7 , FIG.2. Morphology of dendritic cells. (A) Cluster of DCs obtained by culturing blood CD34' hematopoietic progenitors with GM-CSF anti TNF 1x40). (B) Giemsa staining of a cytocentrifuged DC clrister ( ~ 1 0 0 )(C) . During culture of CD34' hematopoietic progenitors, a fraction of DCs is strongly adherent to the vessel surface (Giemsa staining, ~ 4 0 )(D) . Interdigitating DCs within the T-cell-rich areas of tonsils (blue, CD3 staining of T cells; red, CD83 staining of DCs); a germinal center is identified on the upper right corner. ( E ) CDla DCinrm(red) infiltrate breast carcirioina tissue. (F) CD83 DCrllnt(red) surround breast carcinoma together with CD4' T cells (hlnc);the purple color of the DCs indicates that they also express CD4.

CH. 8. FIG. 2. Molecular surface representations of integrin-binding domains for ICAM-2, VCAM-I, and FN10. The critical acidic residues involved i n integiin binding are shaded red. Note that thr binding regions in VCXM-1 and FK10 protruck from the protein surface: in contrast, the corresponding binding region of ICAM-2 is on a flat plane. [Reprinted by perinission from Nnturp (Casasnovas ef nl., 1997).copyriglit 1997, Macmillan Magazines Ltd.]

This Page Intentionally Left Blank

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

199

segments of the IgG. Though the hinge polypeptides were free of any intermolecular packing contacts, stabilization of the Fabs and Fc significantly restricted the conformation of the hinge. The 9 A translation between Fabs stretched the hinge through its upper section to the first interheavy-chain disulfide. The core hinge was further stabilized by making an “intramolecular” contact with one glycosylated C,z loop and its attendant oligosaccharide. Because of this immobilization due to tension, electron density for the murine IgG, hinge was considerably more interpretable at an early stage of X-ray analysis than was the hinge region for the IgG2,. domain pair of the IgGl antiphenoIn Fig. 6A (see color plate), a CL:CH1 of the IgG2,, barbital antibody (Mab61.1.3) is superimposed on Fab2 CL:CHi canine lymphoma antibody (Mab231). A 12” shift in variable domains resulting from the different elbow angles, 155” for Mab61.1.3 and 143” for Fab2 of Mab231, is evident. Fabl of Mab231, on the other hand, has an elbow angle of 159”,which is closer to those of Mab61.1.3 (Table I). In Fig. 6A-C, it can be seen that the two IgG conformations are substantially different, yet both exhibit two pronounced local dyads. Because the dispositions of segments are unique, those of corresponding dyads are as well. These Mab structures provide images of two entire IgG molecules possessing normal hinge polypeptides (Figs. 4-6). The angle between the long ‘axes ofthe Mab61.1.3 Fabs is 115”,producing aY appearance, whereas that between Mab231 Fabs is 172”,yielding a T shape. In both cases, the Fc is obliquely oriented with respect to the Fabs and thus “distorts” the Y and T. In solution, antibodies transform within a spectrum of conformations. Taking the conformations of these two intact antibodies asolimits, a range of antigen combining-site distances of about 120 to 150 A would be expected (Table I). Fab segments can, however, adjust to accommodate many diverse antigen arrangements, for example, those presented on the surfaces of lymphoma cells. As evidenced by the phenobarbital antibody, the Fabs can also make translational adjustments to optiomizeantigen targeting. For Mab61.1.3 this Fab translation was only 9 A, but murine IgGl has a relatively short upper hinge. Antibodies with longer upper hinge polypeptides could likely undertake greater translational adjustments in collaboration with rotational and hingelike movements. VI. Biological Implications and Effector Functions

Antigen clearance mechanisms are either cellular, through binding of Fc to Fcy receptors FcRI, FcRII, and FcRIII on leukocytes, or humoral, dependent on activation of the complement cascade. Fc effector interactions which lead to elimination occur only after the antibody has established contact with antigen. For the case of complement, antibody Fc-Fc aggrega-

200

LISA J. HARRIS el al.

tion must also occur, following antigen contact, in order to activate C1 and subsequently the cascade. Details as to how effector functions are mediated are, in general, lacking. How antigen binding, given the flexible nature of the antibody molecule, triggers effector responses at the Fc is a particular mystery. Some possibilities have been reviewed by Metzger, who discusses (1)an allosteric model, (2) a distortive model, and (3) a purely associative or aggregation model (Metzger, 1978). Burton, on the other hand, describes (1)a disordered adapter model and (2) an ordered adapter model, both of which are distinct from constructs implying allosteric conformational changes upon IgG cornplexation with antigen (Burton, 1990a; Burton and Woof, 1992). Recombinations of different elements of these proposals are possible, whereas specific features of the models are difficult to assess by physical methods. The flexible nature of the antibody allows for the Fc to move out of the plane of the Fabs, while adjusting in a variety of ways to both target and effector. The intact monoclonal antibody structures, Mab231 and Mab61.1.3, discussed above, exemplify the possibilities (Figs. 4-6). Both exhibit dislocated Fcs, yielding arbitrary conformations that could permit Fc-Fc association. This would fulfill the minimum binding requirement for two of the six heads of C l q needed for complement activation, and conformations that allow clear access to Fcy receptor sites in the lower hinge regions. Although these few structures do not provide definitive answers to the salient questions regarding effector function, information does emerge by mapping, as in Fig. 7 (see color plate), the distribution of key residues involved in antigen and effector interactions. Our discussion of IgG effector functions will be primarily restricted to human IgG1,murine IgG1,murine IgG2,, and human hinge-deleted IgG1,which are the isotypes of the structures reviewed here (Table I). The Fcy receptor family has been characterized for human and mouse immunoglobulins and consists of FcyRI, FcyRII, and FcyRIII in each species. The critical IgG recognition sequence for the Fcy receptor has been identified as lower hinge residues 247-251, Leu-Leu-Gly-Gly-Pro, of the CHzgene product (EU 234-238). Particular weight has been assigned to the two leucines. This sequence has been implicated as a binding site, or overlapping binding site, for all three human Fcy receptors, and for mouse FcyRII as well (Duncanet al., 1988; Lundet al., 1991,1992; Sarmay et al., 1992; Morgan et al., 1995). Human FcyRI binds with high affinity to both monomeric human IgGl and murine IgGzabut fails to bind to rnurine IgGl (Duncan et al., 1988). Sequence 247-251 is identical for human IgGl and murine IgGza(Table 11), and the isotypes are virtually indistinguishable in terms of their affinities for human FcyRI. Murine IgG,, having a lower hinge that is three residues shorter, does not contain

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

20 1

the binding sequence (Table 11). Variation in lower hinge sequences for the IgG subclasses may dictate specificity of antibody interaction for mouse and human Fcy receptors (Lund et al., 1992). Dob and Mcg contain the crucial Leu-Leu-Gly-Gly-Pro lower hinge sequence, because they have only genetic hinge deletions (Table 11).FcyRI has some binding affinity for Mcg [J. M. Woof and D. R. Burton, as quoted in Burton (1990b)], but not for intact Dob (Klein et al., 1981). An engineered chimeric IgG3 having a genetic hinge deletion, but a disulfide bond inserted at the N terminus of the lower hinge, showed high activity for FcyRI-mediated phagocytosis. Hinge-deleted chimeric IgG3 without an H-H disulfide bridge yielded low to no activity (Brekke et al., 1993, 1996; Michaelsen et al., 1994). Thus the key role of the “genetic” hinge (upper and core structural hinge), required for high-efficiency FcyR signaling, appears to be maintenance of at least one inter-H-chain disulfide bond. This joins heavy chains just prior to the lower hinge residues. The genetic hinge acting as a long spacer, or endowing Fab-Fab flexibility, seems unnecessary for FcyR interaction (Michaelsen et al., 1994). Optimal association with human FcyRI suggests that lower hinge residues act in combination with a hinge-proximal CH2loop, sequence 344-350 (EU numbering 325-331), containing a particularly significant Pro-350 (Canfield and Morrison, 1991). The loop likely forms part of the binding platform for FcyRI. In the canine lymphoma antibody structure, the lower hinge polypeptide of one heavy chain does in fact make contact with CH2loop residues 344-350 on that same heavy chain. This confirms that lower hinge residues can be in intimate contact with the hinge-proximal loop containing Pro-350 (EU Pro-331) (see Fig. 7B). In Mcg, accessibility of Leu-Leu permits receptor binding (see above), but Pro-350 was noted to be sterically blocked by the bypass peptides (Edmundson et al., 1995). A locus essential for C l q binding has been identified as three specific polar residues of the CH2 domain (Duncan and Winter, 1988). All IgG isotypes contain this “core” C l q site (Fig. 7). In most IgGs, including human IgGl and mouse IgGPd,the residues are Glu-337, Lys-339, and Lys-341. For mouse IgGl residue 341 is an arginine. Because antibody subclasses show differing abilities to activate complement, specificity must reside elsewhere on the IgG. It has been assigned to the c H 2 gene product, implying residues in the structural lower hinge or CH2domain (Clackson and Winter, 1989; Tao et al., 1991, 1993; Greenwood et al., 1993; Brekke et al., 1994; Morgan et al., 1995). Lower hinge residues and the hingeproximal CH2loop that were found to be important for FcyR binding have additionally been implicated in C l q specificity (Morgan et al., 1995; Tao et al., 1993; Brekke et al., 1994). Again, within this CH2 loop, Pro-350 appears to be of particular importance. Oligosaccharides have also been

202

LISA J. HARRIS et al.

shown essential for activation (Nose and Wigzell, 1983), as are paired CH3domains (Utsumi et al., 1985). Human IgG, and mouse IgGz, are strong activators of complement, whereas mouse IgGl is poor, and genetic hinge-deleted human Dob and Mcg fail altogether. Fluorescent depolarization studies of normal IgG subclasses, and the examples of Dob and Mcg, encouraged the idea that hinge flexibility was correlated with an ability to activate complement (Oi et al., 1984; Dangl et al., 1988). A genetic hinge-deleted, chimeric IgG3, with Ala-Ala-Cys-Ala introduced between Ala-244 and Pro-245 of the lower hinge region, was shown, however, to have an even greater complementmediated lysis (CML) activity than wild-type IgG3. The corresponding hinge-deleted IgG3 without an inserted disulfide bond had essentially no CML activity (Brekke et al., 1993, 1996; Michaelsen et al., 1994). The genetic hinge (upper and core hinge) functioning as a flexible spacer was, therefore, shown unnecessary for complement activation. The only role of the “genetic” hinge for complement activation appears to be maintenance of an inter-H-chain disulfide bond( s) linking heavy chains at the N terminus of the lower hinge. This result corroborates a longtime observation that reduction of normal IgG, under conditions where only interchain disulfides are cleaved, causes loss of Fc-mediated effector functions (Dorrington, 1978). Because Mcg and Dob lack an inter-heavy-chain disulfide, hinge bypass segments (lower hinges) are not drawn together seven residues N terminal to the structural CH2 domain. These residues, instead, take a path that could obstruct access to residues involved in the binding of Clq, and/or their path may be such that CHzdomains do not maintain correct orientations for activation. Although the core Clq-binding motif of Mcg is exposed (Fig. 7 ) ,the hinge-proximal loop containing residues 344-350 is sterically obstructed by the bypass residues (Edmundson et al., 1995). The fact that Mcg does not have a canonical C H q C H e relationship lends some credence to the proposal that CH2 domains may not be oriented in an appropriate manner for C l q interaction. Removal or degradation of the immunoglobulin oligosaccharides has also been shown to alter significantly not only complement interactions, but most effector functions in general (Nose and Wigzell, 1983). IgGs lacking carbohydrates lose ability to activate complement, to bind Fc receptors, and to induce antibody-dependent cellular cytotoxicity (ADCC). Galactose deficiencies, in particular, have been associated with rheumatoid arthritis and other chronic diseases (Parekh et al., 1985; Tsuchiya et al., 1989; Furukawa and Kobata, 1991). Even though sugar residues have not been directly implicated in effector binding sites, there is abundant evidence that they play an essential role in the function of Fc. We now know from the structure of Mab61.1.3 that core and lower hinge polypeptides

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

203

can contact the oligosaccharide, as well as the CHzpolypeptide loop that N links the carbohydrate chain (Harris et al., 1998).This contact, although likely transient in solution, could be important for effector function (see Fig. 7A). The interfaces between CHzand CH3domains within the Fc are additional regions of physiological interest, because several biological macromolecules are known to recognize and bind to this area, or to include it in overlapping sites. Biochemical studies have indicated the interface to be important in regulating both transcytosis and catabolism (Ghetie and Ward, 1995). Crystallographic studies have identified the CH2-CH3junction as the site of interaction for rheumatoid factor (Corper et al., 1997), rat neonatal Fc receptor (Burmeister et al., 1994;Vaughn et al., 1997), and Staphylococcus atireus proteins A (Deisenhofer, 1981) and G (Sauer-Eriksson et al., 1995). Ig-binding proteins A and G are considered Fc receptors; however, both proteins utilize secondary binding sites on the Fab segments as well (Derrick and Wigley, 1992). These molecules, A and G, are found on the surfaces of bacteria and are relevant to bacterial virulence (Kastern et al., 1990; Raeder and Boyle, 1993). Protein A has been shown to bind hingedeleted Dob (Klein et d., 1981) and Mcg, even though the Fc interaction site on Mcg is enlarged due to its noncanonical c H z - c H 3 interface (Edmundson et al., 1995). In all crystallographic structures of Fc fragments, or Fc segments within intact antibodies, the first loop of the CH2 domain (residues 264-267) remains stationary with respect to the interface with CH3,while the remainder of the CHndomain may show appreciable movement. This loop appears to serve as a “pivot” allowing the CH2domain to undergo rigid body motion with respect to the CH2-CW3interface (Harris et al., 1997). While the interface is maintained, the CH”cH3 pivot could modulate effector functions by altering the spatial relationship of binding sites. Although segmental flexibility and spacer properties of the genetic hinge (upper and core hinge) were shown of little importance in complement activation and FcyR signaling, these features do affect antigen accessibility. The lengths of the upper hinge polypeptides dictate Fab-Fab flexibility and thus give differing “reach” capabilities, rotational and translational, during antigen complexation. IgGs are able to interact with epitopes of varying dispositions as a consequence of genetic hinge diversity. This diversity may be most important when the IgG is membrane bound on B cells (Brekke et al., 1995). In this case its reach would be almost entirely dependent on the genetic hinge sequence. VII. Concluding Remarks

The intersegmental flexibility illustrated by the intact antibody structures is well established, and was previously documented using a variety of

204

LISA J , HARRIS et nl

lower resolution methods, such as electron microscopy and fluorescence depolarization spectroscopy (Wrigley et al., 1983; R o u , 1984; Wade et al., 1989; Oi et al., 1984). These high-resolution crystallographic images do, however, allow a more careful analysis and exact definition of degrees of freedom (Table I and Figs. 1-6). Mutational studies pertinent to hinge sequence and effector interactions, taken in the structural context of the intact antibody models, suggest that some ideas regarding antibody flexibility should possibly be reevaluated. The genetic hinge, composed of the upper and core structural hinge residues (Table 11),is primarily responsible for providing Fab-Fab movements for antigen binding. It does not appear to play a major role in effector processes, except to provide at least one inter-heavy-chain disulfide bond, thereby tethering the chains, and thus the segments. Variability in genetic hinge sequences provides various reaching capabilities for the IgG subclasses, and this, perhaps, is its primary function. The flexible-spacer role of the genetic hinge is apparently unnecessary for effector processes, including complement activation, Fcy receptor signaling, and antibodydependent cellular cytotoxicity (see review, Brekke et al., 1995, 1996). Polypeptides C terminal to the genetic hinge are structurally regarded as the lower hinge (Table 11), but they are encoded by the CH2exon. These residues alone provide the flexibility necessary for effector functions. C l q and FcyR recognition sites on IgG are accessible even when the upper and core hinge regions are deleted, as long as a disulfide link is inserted, and properly formed, at the N terminus of the lower hinge. Residues essential for Clq and FcyR specificity and activation are included within the CH2gene product, and are composed of the lower hinge and CH2domain (Fig. 7 ) .The interface between c H z - c H 3 domains in the Fc is an additional locus for interaction with a variety of biological molecules, some of which are involved in effector mechanisms (Fig. 7 ) . A critical CH2loop at the interface serves as a structural “pivot” for rigid body motions of C,z domain(s) with respect to the CH3:CH3pair. One somewhat unexpected result to emerge from the crystallographic studies discussed above, consistent with fragment studies as well, is the independence of the two CHzs.These are the only domains in the antibody not paired through noncovalent interactions; all other domain pairings exhibit extensive contacts. The CH2s are glycosylated in a manner that sequesters the carbohydrates, unlike most glycoproteins. They are clearly the most mobile of the domains, but, at the same time, they carry a variety of effector sites. In some cases, CH2domains seem to function in concert with lower hinge residues, and in others cases in collaboration with cH3s. Disulfide bridging of heavy chains in the hinge core does impose

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

205

limits on CH2movements that are, perhaps, significant for proper orientation of CH2domains during effector activation. The intact antibodies with hinge polypeptides, Kol, Mab231, and Mab61.1.3, taken together, provide a dynamic range of motion for immunoglobulin G during antigen recognition and subsequent effector functions. Although the crystal structures represent “snapshots” or “still images” of IgG, in solution the antibodies can be considered as transforming between these representative conformations, as well as a myriad of others. Genetic hinge-deleted antibodies such as Dob and Mcg are clearly the most rigid IgG molecules, but these also, in solution, would be expected to have some flexibility, albeit severely reduced.

ACKNOWLEDGMENTS The authors are supported by grants from NASA and QED Bioscience. We thank the San Diego Supercomputer Facility for providing computational resources. We also extend special thanks to Aaron Greenwood for assistance with figures.

REFERENCES Braden, B. C., and Poljak, R. J. (1995).Structural features ofthe reactions between antibodies and protein antigens. Faseh J . 9, 9-16. Brekke, 0. H., Michaelsen, T. E., Sandin, R., and Sandlie, I. (1993). Activation of complement by an IgC molecule without a genetic hinge. Nature 363, 628-630. Brekke, 0. H., Michaelsen, T. E., Aase, A,, Sandin, R., and Sandlie, I. (1994). Human IgC isotype-specific amino acid residues affecting complement-mediated cell lysis and phagocytosis. Eur. J. lmmiinol. 24, 2542-2547. Brekke, 0. H., Michaelsen, T. E., and Sandlie, I. (1995). The structural requirements for complement activation by IgC: Does it hinge on the hinge? lmmtinol. Toclny 16,85-90. Brekke, 0. H., Michaelsen, T. E., Sandin, R., and Sandlie. I. (1996). Activation of complement by an IgC molecule without a genetic hinge [published erratum]. Nature 383, 103. Burmeister, W. P., Huber, A. H., and Bjorkman, P. J. (1994). Crystal structure of the complex of rat neonatal Fc receptor with Fc. Nature 372, 379-383. Burton, D. R. (1990a). Antibody: The flexible adaptor molecule. Trends Biochem. Sci. 15, 65-69. Burton, D. R. (1990b). The conformation of antibodies. In “Fc Receptors and the Action of Antibodies” (H. Metzger, ed.), pp. 31-54. American Society for Microbiology,Washington D.C. Burton, D. R., and Woof, J. M. (1992). Human antibody effector function. Adu. lmmunol. 51, 1-84. Canfield, S. M., and Morrison, S. L. (1991). The binding affinity of human IgG for its high affinity Fc receptor is determined by multiple amino acids in the CH2 domain and is moduIated by the hinge region. /. Exp. Med. 173, 1483-1491. Clackson, T., and Winter, G. (1989). Sticky feet-directed mutagenesis and its application to swapping antibody domains. Nucleic Acids Res. 17, 10163-10170. Colman, P. M. (1988). Structure of antibody-antigen complexes: Implications for immune recognition. Adu. lmmunol. 43,99-132. Coyer, A. L., Sohi, M. K., Bonagura, V. R., Steinitz, M., Jefferis, R., Feinstein, A., Beale, D., Taussig, M. I., and Sutton, B. J. (1997). Structure of human IgM rheumatoid factor

206

LISA J , HARRIS et al

Fab bound to its autoantigen IgG Fc reveals a novel topology of antibody-antigen interaction. Nature Struct. Biol. 4, 374-381. Dangl, J. L., Wensel, T. G., Morrison, S. L., Stryer, L., Henenberg, L. A., and Oi, V. T. (1988). Segmental flexibility and complement fixation of genetically engineered chimeric human, rabbit and mouse antibodies. EMBO J. 7, 1989-1994. Davies, D. R., and Chacko, S. (1993). Antibody structure. Acc. Chem. Res. 26, 421-427. Davies, D. R., and Cohen, G . H. (1996). Interactions of protein antigens with antibodies. Proc. Natl. Acad. Sci. U.S.A. 93, 7-12. Deisenhofer, J. (1981). Crystallographic refinement and atomic models of a human Fc fragment andsits complex with fragment B of protein A from Staphylococcus aureus at 2.9- and 2.8-A resolution. Biochemistry 20, 2361-2370. Derrick, J. P., and Wigley, D. B. (1992). Crystal structure of a streptococcal protein G domain bound to an Fab fragment. Nature 359,752-754. Dorrington, K. J. (1978).The structural basis for the functional versatility of immunoglobulin G1. Can. J . Biochem. 56, 1087-1101. Duncan, A. R., andwinter, G. (1988).The binding site for Clq on IgG. Nature 332,738-740. Duncan, A. R., Woof, J. M., Partridge, L. J., Burton, D. R., and Winter, G . (1988). Localization of the binding site for the human high-affinity Fc receptor on IgG. Nature 332, 563-564. Edmundson, A. B., Wood, M. K., Schiffer, M., Hardman, K. D., Ainsworth, C. F., Ely, K. R., and Deutsch, H. F. (1970). A crystallographic investigation of a human IgG immunoglobulin. J. Biol. Chem. 245,2763-2764. Edmundson, A. B., Guddat, L. W., Rosauer, R. A., hdersen, K. N., Shan, L., and Fan, Z.-C. (1995).Three-dimensional aspects of IgG structure and function. In “The Antibodies” (M. Zanetti and J. D. Capra, eds.), pp. 41-100. Hanvood Academic Publishers, New York, New York. Ely, K. R., Colman, P. M., Abola, E. E., Hess, A. C., Peabody, D. S., Parr, D. M., Connell, C.E., Laschinger, C. A., and Edmundson, A. B. (1978). Mobile Fc region in the Zie IgG2 cryoglobulin: Comparison of crystals of the F(ab‘), fragment and the intact immunoglobulin. Biochemistry 17, 820-823. Furukawa, K., and Kobata, A. (1991). IgG galactosylation-Its biological significance and pathology. Mol. lmmunol. 28, 1333-1340. Ghetie, V., and Ward, E. S. (1995). Genetic manipulation of antibodies: From variable domains to constant regions. In “The Antibodies” (M. Zanetti and J. D. Capra, eds.), pp. 169-211. Hanvood Academic Publishers, New York, New York. Greenwood, J., Clark, M., and Waldmann, H. (1993). Structural motifs involved in human IgG antibody effector functions. Eur. J. Zmmunol. 23, 1098-1104. Guddat, L. W., Herron, J. N., and Edmundson, A. B. (1993). Three-dimensional structure of a human immunoglobulin with a hinge deletion. Proc. Nutl. Acad. Sci. U.S.A. 90,42714275. Harris, L. J., Larson, S. B., Hasel, K. W., Day, J., Greenwood, A., and McPherson, A. (1992). The three-dimensional structure of an intact monoclonal antibody for canine lymphoma. Nature 360,369-372. Harris, L. J., Skaletsky, E., and McPherson, A. (1995). Crystallization of intact monoclonal antibodies. Proteins: Struct. Fund. Genet. 23, 285-289. Harris, L. J., Larson, S. B., Hasel, K. W., and McPherson, A. (1997). Refined structure of an intact IgG2a monoclonal antibody. Biochemistry 36, 1581-1597. Harris, L. J., Larson, S. B., Skaletsky, E., and McPherson, A. (1998). Comparison of the conformations of two intact monoclonal antibodies with hinges. Immunological Reviews 163,35-43.

INTACT ANTIBODY STRUCTURES AND EFFECTOR FUNCTION

207

Harris, L. J., Skaletsky, E., and McPherson, A. (1998). Crystallographic structure of an intact IgGl monoclonal antibody. /. Mol. Biol. 275, 861-872. Kabat, E. A., Wu, T. T., Perry, H. M., Gottesman, K. S., and Foeller, C. (1991). In “Sequences of Proteins of Immunological Interest,” 5th Ed. U.S. Public Health Service, NIH, Washington, D.C. Kastern, W., Holst, E., Nielsen, E., Sjobring, U., and Bjorck, L. (1990). Protein L, a bacterial immunoglobulin-binding protein is a possible virulence factor. Infect. Immun. 58, 1217-1222. Klein, M., Haeffner-Cavdon, N., Isenman, D. E., Rivat, C., Navia, M. A,, Davies, D. R., and Dorrington, K. J. (1981). Expression of biological effector functions by immunoglobulin G molecules lacking the hinge region. Proc. Natl. Acad. Sci. U.S.A.78, 524-528. Larson, S., Day, J., Greenwood, A., Skaletsky, E., and McPherson, A. (1991).Characterization of crystals of an intact monoclonal antibody for canine 1ymphoma.j. Mot. B i d . 222,17-19. Lund, J., Winter, G., Jones, P. T., Pound, J. D., Tanaka, T., Walker, M. R., Artymiuk, P. J., Arata, Y., Burton, D. R., Jefferis, R., and Woof, J. M. (1991). Human FcyRI and FcyRII interact with distinct but overlapping sites on human IgG. /. Immunol. 147, 2657-2662. Lund, J., Pound, J. D., Jones, P. T., Duncan, A. R., Bentley, T., Goodall, M., Levine, B. A., Jefferis, R., and Winter, G. (1992). Multiple binding sites on the CH2domain of IgG for mouse FcyR11. Mol. Immunol. 29,53-59. Marquart, M., Deisenhofer, J., Huber, R., and Palm, W. (1980). Crystallographic refinement and atomic models of the $tact immunoglobulin molecule Kol and its antigen-binding fragment at 3.0 A and 1.9 A resolution. 1.Mol. Biol. 141, 369-391. Metzger, H. (1978).The effect of antigen on antibodies: Recent studies. In “Contemporary Topics in Molecular Immunology” (R. A. Reisfeld and F. P. Inman, eds.), pp. 119-152. Plenum, New York. Michaelsen, T. E., Brekke, 0. H., Aase, A., Sandin, R. H., Bremnes, B., and Sandlie, I. (1994). One disulfide bond in front of the second heavy chain constant region is necessary and sufficient for effector functions of human IgG3 without a genetic hinge. Proc. Nutl. h a d . Sci. U.S.A. 91, 9243-9247. Morgan, A,, Jones, N. D., Nesbitt, A. M., Chaplin, L., Bodmer, M. W., and Emtage, J. S. (1995). The N-terminal end of the CH2domain of chimeric human IgGl anti-HLA-DR is necessary for Clq, FcyRI and FcyRIII binding. Immunology 86, 319-324. Nose, M., and Wigzell, H. (1983). Biological significance of carbohydrate chains on monoclonal antibodies. Proc. Natl. Acad. Sci. U.S.A. 80, 6632-6636. Oi, V. T., Vuong, T. M., Hardy, R., Reidler, J., Dangle, J., Herzenberg, L. A., and Stryer, L. (1984). Correlation between segmental flexibility and effector function of antibodies. Nature 307, 136-140. Padlan, E. A. (1996).X-Ray crystallographyof antibodies. In “Advances in Protein Chemistry, Antigen Binding Molecules: Antibodies and T-cell Receptors” (E. Haber, ed.), Vol. 49, pp. 57-133. Academic Press, New York. Palm, W., and Colman, P. M. (1974). Preliminary X-ray data from well-ordered crystals of a human immunoglobulin G molecule. J. Mol. Biol. 82, 587-588. Parekh, R. B., Dwek, R. A,, Sutton, B. J., Fernandes, D. L., Leung, A., Stanworth, D., Rademacher, T. W., Mizuochi, T., Taniguchi, T., Matsuta, K., Takeuchi, F., Nagano, Y., Miyamoto, T., and Kobata, A. (1985). Association of rheumatoid arthritis and primary osteoarthritis with changes in the glycosylation pattern of total serum IgG. Nature 316,452-457. Raeder, R., and Boyle, M. D. P. (1993). Association of type I1 immunoglobulin G-binding protein expression and survival of group A streptococci in human blood. Infect. Immun. 61,3696-3702.

208

LISA J. HARRIS

d

al.

Roux, K. H. (1984). Direct demonstration of multiple VH allotopes on rabbit Ig molecules: Allotope characteristics and Fab arm rotational flexibility revealed by immunoelectron microscopy. Eur. J . Immunol. 14,459-464. G. (1982). The three-dimensional structure of a human IgGl Sarma, R., and Laudin, immunoglobulin at 4 A resolution: A computer fit of various structural domains on the electron density map. J. Appl. Crystallogr. 15, 476-481. Sarmay, G., Lund, J., Rozsnayay, Z., Gergely, J., and Jefferis, R. (1992). Mapping and comparison of the interaction sites on the Fc region of IgG responsible for triggering antibody dependent cellular cytotoxicity (ADCC) through different types of human Fc receptors. Mol. lmmunol. 29, 633. Sauer-Eriksson, A. E., Kleywegt, G. J., Uhlen, M., and Jones, T. A. (1995). Crystal structure of the C2 fragment of streptococcal protein G in complex with the Fc domain of human IgG. Structure 3, 265-278. o ~222-227. s Sheriff, S. (1993). Antibody-protein compiexes. ~ m m u n o M e t ~ 3, Silverton, E. W., Navia, M. A,, and Davies, D. R. (1977). Three-dimensional structure of an intact human immunoglobulin. Pmc. Natl. Acad. Sci. U.S.A. 74,5140-5144. Steiner, L. A,, and Lopes, A. D. (1979). The crystallizable human myeloma protein Doh has a hinge-region deletion. Biochemistry 18, 4054-4067. Stura, E. A., Satterthwait, A. C., Calvo, J. C., Stefanko, R. S., Langeveld, J. P., and Kaslow, D. C. (1994). Crystallization of an intact monoclonal antibody (4B7) against Plasmodium falciparum malaria with peptides from the Pfs25 protein antigen. Acta Crystallogr. D50,556-562. Tao, M. H., Canfield, S. M., and Momson, S. L. (1991). The differential ability of human IgGl and IgG4 to activate complement is determined by the COOH-terminal sequence of the CH2domain. J. Exp. Med. 173, 1025-1028. Tao, M. H., Smith, R. I. F., and Morrison, S. L. (1993). Structural features of human immunoglobulin G that determines isotype-specific differences in complement activation. J. Exp. Med. 178,661-667. Teny, W. D., Matthews, B. W., and Davies, D. R. (1968). Crystallographic studies of a human immunoglobulin. Nature 220, 239-241. Tsuchiya, N., Endo, T., Matsuta, K., Yoshinoya, S., Aikawa, T., Kosuge, E., Takeuchi, F., Miyamoto, T., and Kobata, A. (1989). Effects of galactose depletion from oligosaccharide chains on immunological activities of human IgG. J. Rheumutol. 16, 285-290. Utsunii, S., Okada, M., Udaka, K., and Amano, T. (1985).Preparation and biologic characterization of fragments containing dimeric and monomeric C gamma 2 domain of rabbit IgG. Mol. lmmunol. 22, 811-819. Vaughn, D. E., Milburn, C. M., Penny, D. M., Martin, W. L., Johnson, J. L., and Bjorkman, P. J. (1997). Identification of critical IgG binding epitopes on the neonatal Fc receptor. J. Mol. B i d . 274, 597-607. Wade, R. H., Taveau, J. C., and Lamy, J. N. (1989).Concerning the axial rotational flexibility of the Fdb regions of immunoglobulin G. J. Mol. Biol. 206,349-356. Wilson, I. A,, and Stanfield, R. L. (1994). Antibody-antigen interactions: New structures and new conformational changes. Curt-. Opin. Struct. Biol. 4, 857-867. Wrigley, N. G., Brown, E. B., and Skehel, J. J. (1983). Electron microscopic evidence for the axial rotation and inter-domain flexibility of the Fab regions of immunoglobulin G. J. Mot. Biol. 169, 771-774.

a.

This article was accepted for publication on September 9, 1998.

ADVANCES IN IMMUNOLOGY, VOL 72

Lymphocyte Trafficking and Regional Immunity EUGENE C. BUTCHER,',' MARNA WILLIAMS,' KENNETH YOUNGMAN,' LUSUAH ROlT,' AND MICHAEL BRISKIN+ 'Lobomiory of immunology ond Vascular Biology, Deporimeni of Pathology, Stanfod University school of Medicine, Sfanlord, C a l h i o 94305, ond Center for Mokcular Biology and Medicine, Vetemns Affoirs Palo Alto Health Core System, Pal0 Alto, C a h h i a 94304; and bukosil8, hc., Cambidge, Mossachuselk 02 142

I. Introduction

Lymphocytes are migratory cells, trafficking from their sites of origin in the bone marrow and thymus, and homing to and recirculating through specialized lymphoid and extralymphoid tissues in the periphery ( 1-14). Like all leukocytes, lymphocytes develop with characteristic trafficking properties. Lymphocytic components of the innate immune system, such as natural killer cells, as well as subsets of y 8 T cells and specialized lymphoid dendritic populations, appear to be preprogrammed with particular tissue or inflammation tropisms during their development in primary lymphoid organs. For lymphocytes of the adaptive immune system, comprising the bulk of cup receptor-expressing T cells and of B cells in the adult animal, initial homing properties-a pronounced tropism for secondary lymphoid organs, including lymph nodes (LNs), Peyer's patches (PPs), and spleen-also seem to be determined developmentally, prior to or in association with emigration from the thymus or bone marrow (see, for example, Ref. 15).However, whereas most leukocytes pass through the blood only once in their migration to terminal sites of participation in inflammatory responses, these T and B cells recirculate continously from blood to lymph. Moreover, antigen-dependent stimulation appears to reprogram the trafficking properties of naive B and T cells in the periphery, inducing or selecting for memory and effector cells that not only home much more effectively than their naive precursors to extralymphoid sites of tissue inflammation, but that can also display striking selectivity for trafficking to the gastrointestinal tract, skin, or other specific tissues or organs ( 1 , 4-10, 12-14, 16-18) (reviewed in Refs. 19-21). For example, B and T immunoblasts and T memory cells induced in gastrointestinal lymphoid tissues traffic preferentially into the intestinal wall and/or gutassociated lymphoid organs (PPs, appendix, and mesenteric LNs; see below). Conversely, immunocytes generated in response to cutaneous immunization home better to inflamed skin (see Fig. 1).

' To whom correspondence should be addressed. 209

Copynght Q 1999 by Acddemic Precs

AU nghts of reproduction in any form rrservrd

0065-2776BY $30 W

210

EUGENE C. BUTCHER et al

Lymphocytes of the adaptive immune response p8 B cells -> seconda

+ &l

c

K

E

c m

Other cells of the immune system granulocytes and monocytes -> inflammation natural killer cells -> inflammation yS T cells -> epithelia, etc. CD5+ B cells -> serosal surfaces dendritic cells -> lymphoid, other various neonataVperinatal lymphocytes

J \ B and T effector & memory subsets

.$

a.4p7c

E

cn.-

5.-

skin

CNS gastrointestinal tract other -> extralymphoid inflammatory sites +/. brnphold tissues

FIG.1. Developmental and antigen-dependent regulation of cellular traffic in the immune system. Developmentally preprogrammed trafficldng properties control the selectivity of leukocyte recruitment from the blood, and direct the recirculation of naive B and T lymphocytes through Peyer’s patches, lymph nodes, and spleen. Lymphocytes of the adaptive immune system (including cub T cell receptor-expressing T cells and @+B cells) undergo reprogramming as a function of microenvironmental influences during antigenic stimulation, resulting in the tissue-selectiverecirculation and homing patterns of antigen-specificmemory and effector lymphocytes.

A central tenet of current thinking in the field is that this tissue-specific targeting of antigen-reactive populations must increase the efficiency of immune surveillance by circulating memory cells, as well as enhancing local immune responses. It may also orchestrate the systemic competition between immunocytes that underlies the homeostatic regulation of memory and immune reactivity (22),and may serve to help reduce inappropriate (autoimmune) reactivity of activated lymphocytes, for example, by decreasing opportunities for gut antigen-responsive lymphocytes to encounter self-determinants in extralymphoid tissues elsewhere in the body. From the perspective of regional physiology and pathology, selective trafficking may also provide a mechanism for segregating the specialized immune response modalities characteristic of intestinal versus systemic immune responses. Recent studies have not only identified many key molecules that help target lymphocyte trafficking, but have also begun to define the relationship between selectively recirculating lymphocytes and the cellular carriers of

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

21 1

immunologic memory for regionally presented antigens. This review outlines our present understanding of tissue-selective lymphocyte trafficking, focusing in particular on the gut and its associated lymphoid tissues. The sophisticated cellular, developmental, and molecular mechanisms involved are discussed, with emphasis on the importance of these mechanisms to understanding and manipulating regional immune responses in the gastrointestinal tract . II. Interaction of Blood Lymphocytes with Endothelium Involves Multiple Steps That Control the Specificity of Lymphocyte Recruitment

A. SPECIALIZED EDOTHELIAL CELLSDETERMINE SITESOF LYMPHOCYTE EXTRAVASATION The recruitment of circulating lymphocytes from the blood into intestinal tissues, as elsewhere, begins with interaction with the blood vascular endothelium, principally within specialized postcapillary venules. Lymphocytes can distinguish between and adhere differentially to venular endothelium in different sites in the body (18, 23, 24), an ability that helps determine their trafficking patterns in vivo (19-21, 25-27). In lymph nodes and in the mucosal lymphoid organs-the PPs and appendix-the venules involved are lined by metabolically active, plump or “high’ endothelium (26, 27). In the intestinal lamina propria venules supporting lymphocyte extravasation are less distinctive histologically, but here too the endothelial cells are highly specialized for their role in recruiting lymphocytes from the blood (see, for example, Refs. 28-30). MULTIPLE STEPS REQUIRED FOR RECRUITMENT OF CIRCULATING LYMPHOCYTES FROM THE BLOOD Blood-borne lymphocytes flow with remarkable speed in relation to their size, requiring specialized mechanisms for interaction with postcapillary venule endothelium. As first proposed as an extrapolation from studies of neutrophils (31, 32), these mechanisms embody a multistep process in which specialized adhesion and signaling molecules can participate to mediate each of a series of essential steps (Fig. 2 ) (22).In the first step, microvillous processes, extending from the lymphocyte surface and deploying constitutively active adhesion receptors, make contact with the specialized endothelium and initiate rolling of the lymphocyte along the vascular lumen. In some instances (notably the homing of naive lymphocytes to intestinal PPs; see below), additional rolling receptors appear to be required to slow rolling sufficiently for subsequent events to occur. Rolling delays the transit of lymphocytes dramatically, allowing “sampling” of the local

B.

212

EUGENE C. BUTCHER et al.

-4000pm/sec midline velocity

-4OudSec

contact through microvillous receptors

-

1-20 seconds

+

-

through G-protein linked receptors

minutes

dependent arrest (reversible over minutes)

4- djapedesis

FIG.2. The multistep model of lymphocyte/endothelialcell recognition and recruitment from the blood. The sequential requirement for at least four independently regulated receptor-ligand interactions allows combinatorial determination of the specificity of lymphocyte homing (31),implying that the specificity of the overall process can greatly exceed that of its component steps. The velocities of free-flowing (noninteracting) and rolling lymphocytes from in situ microscopic observations of lymphocytes in Peyer’s patch high endothelial venules are given (76); these values may differ in other vascular beds. (Adapted with permission from Ref. 22.)

microenvironment for activating factors that act primarily through serpentine receptors of the Gai protein-linked chemoattractant receptor subfamily. In this second step, these activating factors trigger rapid intracellular signaling in the leukocyte, leading to functional alterations and activation of preexisting cell surface integrins (and perhaps other activatable cell surface receptors), which then mediate firm arrest of the cell on the vessel wall, resistant to continuing blood shear forces (step 3). Recent studies indicate that chemoattractant cytokines (chemokines) presented by vascular endothelium can play a critical role as triggers of lymphocyte arrest (33,34).Triggered integrin-mediated arrest is eventually reversible so that, unless additional signals lead to transendothelial migration (step 4) and recruitment into the surrounding tissues, within several minutes the adherent cell will begin to roll again, passing downstream and returning to the blood. Thus each step in this multistep process represents a “yesho” decision point, so that the successful recruitment of lymphocytes from the blood requires engagement of receptor-ligand pairs at each step. Because several receptor-ligand pairs can be used interchangeably for each event, the multistep process provides a combinatorial mechanism for precise regulation of lymphocyte (and other leukocyte) subset recruitment as a function of the regional site, of the developmental stage of the leukocyte, of the tissue microenvironment, and of the nature of any ongoing pathologic inflammation.

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

213

111. Molecules Involved in Lymphocyte Interactions with Intestinal Endothelium

A. ADHESIONMOLECULES IMPLICATED I N LYMPHOCYTE HOMING TO INTESTINAL SITES Here we begin with discussion of the intestinal vascular addressin, MAdCAM-1, and of its lymphocyte receptor, the integrin a4P7. These molecules appear to function in a selective fashion in vivo for support of the immune system in intestinal and closely related mucosal sites. Although a4P7 and MAdCAM-1 participate in trafficking to other tissues as well, more than any other single adhesion receptorhgand pair they appear to be closely associated with, and indeed to dominate, the homing of specialized intestinal lymphocyte subsets to intestinal lymphoid tissues and to the gastrointestinal lamina propria (29, 35-38). 1. The Mucosal Vascular Addressin, MAdCAM-1 MAdCAM-1 acts as a key vascular “addressin,” or address code molecule for intestinal tissues (29).It is an immunoglobulin superfamily member, related to other vascular adhesion molecules, e.g., vascular cell adhesion molecule-1 (VCAM-l), and intercellular adhesion molecules (ICAM-1, ICAM-2, and ICAM-3) that function as ligands for leukocyte integrins (39, 40). MAdCAM-1 is selectively though not exclusively expressed by postcapillary venules, defining sites of lymphocyte extravasation into the intestinal lamina propria and the intestine-associated lymphoid tissues, especially the PPs and mesenteric LNs (Fig. 3, see color plate) (29, 41). It also defines vessels involved in lymphocyte trafficking into the inflamed pancreas (42, 43), to the lactating mammary gland (29, 44), and in the preneoplastic thymus in AKR mice (45).MAdCAM-1 expression can be up-regulated over constitutive levels in the intestinal lamina propria during active inflammation, for example, in the effected intestines of patients with ulcerative colitis or Crohn’s disease (41). Increased MAdCAM-1 expression on intestinal venules has also been reported in untreated celiac disease (46). MAdCAM-1 is also expressed by vascular zone endothelium in the developing mouse placenta at the stage of active trophoblast invasion, where it may participate with P-selectin in recruiting a unique population of monocyte-related mononuclear cells (47).It is also present at the lining of the marginal sinus in the spleen (48, 49), and is expressed by choroid plexus epithelium in the central nervous system (50),observations currently of undetermined significance. In contrast, although MAdCAM-1 can be up-regulated in vitro by proinflammatory cytokines (51),in the adult animal MAdCAM-1 is absent (or has been reported only at very low levels, of uncertain physiologic significance) on venules in most nongastrointestinal sites of inflammation. Low levels have been observed, for example, on

214

EUGENE C. BUTCHER et al

inflamed peripheral lymph node high endothelial venules (HEVs), in inflamed bronchial tissues, and in extraordinary circumstances in inflamed joints [in murine Lyme disease models (52)] and in the central nervous system [in chronic relapsing experimental allergic encephalomyelitis (5311. Other adhesion pathways dominate trafficking in these and other extragastrointestinal tissues, however, thus helping segregate intestinal from nonintestinal trafficking networks. In this context, it is worth emphasizing that intestinal trafficking cannot be equated with “mucosal” lymphocyte trafficking. For example, MAdCAM-1 and its lymphocyte receptor a4P7 appear to play little or no role in lymphocyte homing to pulmonary or bronchial sites of inflammation (54-56). Similarly, the oral mucosa and tonsils lack significant histologic expression of MAdCAM-1(41). These considerations suggest that grouping bronchial and oral mucosa with the intestinal mucosa into a “common mucosal immune system” may be an oversimplification of the immune physiology of these distinct organs. They also raise important questions about the mechanisms by which IgA plasma cell responses are disseminated among different mucosal tissues following local immunization (57). MAdCAM-1 may also contribute to lymphocyte homing to the inflamed genitourinary tract. Murine genital infection with Chlamydia trachomtis is associated with up-regulation of ICAM-1 and VCAM-1 (58, 59), and also MAdCAM-1 on subsets of involved genital tract venules, suggesting that lymphocyte recruitment in genital tract infection may be regulated in part by a4P7IMAdCAM-1 interactions (58).However, in a parallel study (L. Perry, H. Caldwell, and E. C. Butcher, unpublished) MAdCAM-1 expression was primarily limited to the involved fallopian tubes, where it was expressed with VCAM-1. VCAM-1 but not MAdCAM-1was prominent in the inflamed uterus. Thus, the observations in this one model suggest that there may be significant differences between the addressin expression and mechanisms of recruitment to fallopian tubes versus the uterine wall, Functional studies of in vivo lymphocyte recruitment will be required to resolve the relative contributions of MAdCAM-1 versus VCAM-1 (and other adhesion pathways) in recruitment into the various microenvironments of the inflamed female genital tract. Interestingly, in mice, MAdCAM-1 is much more broadly expressed in the fetus and in the perinatal period than in the adult. In the Dll-14 embryo, many endothelial cells in vessels throughout the body are MAdCAM-1’. By D14 most endothelial cells in the mouse intestine display MAdCAM-1 at high levels, prior to development of the peripheral lymphoid system. Interestingly, MAdCAM-1 is expressed highly by HEVs in developing LNs, as well, and it appears to be the dominant “addressin” in both peripheral and mucosal lymphoid tissues during the perinatal

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

215

period (26, 60). During the immediate postnatal period, MAdCAM-1 on HEVs in lymph nodes recruits a unique CD4+CD3- thymus-derived lymphocyhc stem cell that can give rise to natural killer cells, dendritic antigen-presenting cells, and follicular stromal cells of uncertain naturebut not to conventional T or B cells (61).Loss of MAdCAM-1by peripheral (nonmucosal) lymph node HEVs, with progressive up-regulation of the peripheral lymph node addressin (PNAd, which comprises L-selectin ligands) occurs gradually over 3-6 weeks after birth in the mouse, when the adult pattern of differential expression is achieved. This switch in vascular specificity parallels a dramatic change in the populations of lymphocytes recruited to lymph nodes (60), so that in the adult, naive alp+T cell receptor (TCR)-expressing T cells and pS Ig-expressing B cells are numerically dominant lymph node homing populations. This switch in lymph node homing subsets is but one illustration of the fact that unique populations of lymphocytes and distinctive patterns of lymphocpe trafficking characterize the developing immune system. In fact, the adult homing mechanisms for segregating and targeting effector responses of the adaptive immune system, which we emphasize in this review, develop in parallel with the development and peripheral export of a@ TCR-expressing T cells and pS immunoglobulin heavy chain-expressing B cells (62). Interestingly, however, the developmental regulation of vascular addressin expression does not depend on lymphocyte recirculation, because MAdCAM-1 is expressed normally in the gut and lymphoid tissues of T cell-deficient nu/ nu mice, and in lymphocyte-deficient severe combined immune-deficient (SCID) mice. Instead, studies of ectopically transplanted lymph nodes suggest that microenvironmental factors during the pre- and perinatal period may imprint on the developing lymphoid organ features that determine the patterns of MAdCAM-1 (vs. PNAd) expression in the adult (26, 63). In addition to its expression by intestinal venules, MAdCAM-1 is also displayed by follicular dendritic cells (FDCs) in the mucosal lymphoid organs (41, 64), especially within PPs and the appendix (Fig. 4) (64). MAdCAM-1 decorates interlacing dendritic cells throughout PPs and appendix B cell follicles, and is accentuated on FDCs of the germinal center light zone, and on “junctional” dendritic cells overlapping the B zone border into the outer T zone and subepithelial dome region, sites associated with microenvironmental homing decisions and antigen presentation. In contrast, MAdCAM-1 is rarely displayed by FDCs in primary (resting) peripheral lymph nodes (PLNs), and is largely confined to the germinal center after lymph node immunization. Functional studies confirm that FDC-associated MAdCAM-1plays an important and selective role in follicular binding of lymphocytes in PPs, when assayed in vitro on frozen tissue

216

EUGENE C. BUTCHER et d.

FIG. 4. MAdCAM-1 is displayed by follicular dendritic cells, especially in intestinal lymphoid tissues. (A) Immunohistologic staining of a mouse Peyer’s patch shows intense MAdCAM-1 expression on high endothelial venules (HEV) and on follicular dendritic cells in the germinal center, as well as reactivity of dendritic cells throughout the follicle and into the subepithelial dome region (SDR), the site of antigen recruitment by M cells from the gastrointestinal lumen. (B) Interlacing dendritic cells displaying MAdCAM-1 in the follicular corona (FDC) and SDR of a mouse Peyer’s patch. (C) Absence of MAdCAM-1 in primary B cell follicles in unimmunized mouse peripheral lymph nodes. The border of the primary follicle, defined by anti-IgD staining, is outlined. (Adapted from Ref. 64.)

sections (64). Display of MAdCAM-1 by junctional and coronal FDCs in mucosal lymphoid organs may selectively support the development, retention, and function of lymphocytes expressing the mucosal homing receptor a4P7, thus helping define and target intestinal memory and effector cells. It may also help influence the generation of intestinal homing properties (a4p7 expression) during the antigen-specific generation of memory and effector B and T cells in PPs, working in conjunction with other local microenvironmental influences (e.g., cytokines) to determine the homing properties of lymphocytes responding to antigens presented in intestinal lymphoid organs. Indeed, these observations suggest that regional differences in the lymphoid stromal cells may help control the specialization of intestinal versus nonintestinal immune responses. The domain structure of human and mouse MAdCAM-1 is illustrated in Fig. 5A (39, 40). The dominant splice variant of mouse MAdCAM-1 contains three Ig domains, the most membrane proximal of which (domain 3) is absent in human MAdCAM-1. Otherwise, the structure of the human and mouse mucosal addressin is similar, consisting of two N-terminal immunoglobulin (Ig) domains homologous to the other Ig family vascular EC ligands for leukocyte integrins (VCAM-1 and the ICAMs), and a proximal mucinlike sequence. Less abundant splice variants have been reported in both the mouse [lacking the mucin and third Ig domain (65)l and humans (66), although expression of these at the protein level has not yet been confirmed. Domain swapping and construction of chimeric soluble forms of MAdCAM-1 have shown that Ig domains 1 and 2 are both required (and

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

-

A GLDTSL-

/

217

GLDTSL

8 amlno acld

Mouse \

non-

homologous

eequences

IgAhomolcgy (mouse only)

-10 nm

COOH

FIG.5 . Structure of the mucosal vascular addressin, MAdCAM-1. (A) Schematic diagram of the domain structure of mouse and human MAdCAM-1. Human and mouse MAdCAM1 share strong homology in two N-terminal Ig domains, which together are required (and sufficient) for efficient binding of‘a4P7 on lymphocytes. They also share prominent mucin domains, including an &amino acid motif that is repeated several times in the human sequence. Repulsion between a negatively charged DE /3 loop “antenna” and the negatively charged mucin domain (and the cell surface glycocalyx) is postulated to help orient the integrin-binding face of the N-terminal Ig domain (including a critical GLDTSL in the CD loop) for presentation to a4p7 on lymphocytes (68; and see text). The mucin domain also serves as a stalk, extending the N-terminal Ig domains above the surrounding glycocalyx, and it can be modified by L-selectin-binding carbohydrates when expressed in HEVs. Mouse but not human MAdCAM-1 also has a proximal Ig domain homologous to the Fc-binding region of IgA,: this domain is missing in an alternatively spliced mouse variant (39).Approximate dimensions of Ig domains (4x 2.5 nm) and of extended mucin domains are predicted from electron microscopic studies of ICAM-1 and CD43 (79). (B) See color plate.

sufficient) for efficient a4P7 binding (67). A ribbon representation of the structure of the two Ig domain fragments is presented in Fig. 5B (see color plate) (68). Mutagenesis studies have identified a potentially important motif, defined by the sequence GLTDSL, that is essential for binding a4P7 (67, 69, 70). All Ig family integrin ligands contain similar motifs in their integrin-binding domains, characterized by an absolute requirement for a core acidic residue (D for a4 ligands MAdCAM-1 and VCAM-1, and E for the P2 integrin ligands, ICAM-1, -2, and -3). This sequence is thought to be functionally related to LDV-containing motifs associated with a4 integrin binding sites in fibronectin and in the a4 chain. Although the Nterminal domains of MAdCAM-1, VCAM-1, ICAM-1, and ICAM-2 are structurally similar in overall configuration [allcontaining an “I1 set” immunoglobulin fold (71)],the architecture and location of these conserved sequence motifs are quite different, potentially contributing to the selectivity of these vascular ligands for their respective integrin receptors. In the P2 integrin ligands ICAM-1 and -2, the essential acidic residue lies at the

218

EUGENE C . BUTCHER et al.

end of the C /3 strand in a flat recognition surface, lacking a well-defined CD loop. In contrast, in both MAdCAM-1 and VCAM-1, the critical acidic residue ( D in both receptors) lies in a protruding CD loop (68),illustrated for MAdCAM-1 as the yellow and purple ribbon in domain 1in Fig. 5B. MAdCAM-1 differs from VCAM-1, however, in the position of the critical aspartate residue in the CD loop: when VCAM-1 and MAdCAM-1 are superimposed, these D residues lie 8 A apart from one another. Moreover, the CD loop of MAdCAM-1 is shorter (5 amino acids in length) than that of VCAM-1 (7 amino acids) (68). These differences may be critical for affinity of interactions because exchange of the entire CD loops severely diminishes the binding activity of both receptors (69). Interestingly, however, exchange of the CD loop consensus sequence of VCAM-1, QIDSP, for the equivalent 5-amino acid GLDTS sequence in MAdCAM-1 results in a superadhesive VCAM-1 mutant, with enhanced binding to both a4P7 and a 4 p l (69). The corresponding chimera of MAdCAM-1, however, was diminished in its capacity to bind a4p7 (while retaining the characteristic inability of MAdCAM-1 to bind a4pl). Thus the CD loop of MAdCAM-1 may be a better a 4 binding site than that of VCAM-1. In keeping with this suggestion, the isolated LDTSL peptide of MAdCAM-1is a more potent inhibitor of both a4@7binding to MAdCAM1 and a 4 p l binding to VCAM-1 compared to the corresponding CD loop peptide of VCAM-1 (72, 73; M. Briskin, unpublished). Most importantly, the results indicate that the CD loop is not sufficient to confer the characteristic selectivity of MAdCAM-1 for a4p7 versus a 4 p l binding. This selectivity (i-e.,the inability of MAdCAM-1 to bind a4p1, which distinguishes it from VCAM-1) appears to reside in domain 1, however, as indicated by correlation of binding specificitywith the N-terminal Ig domain in domainswap chimeras. In addition to the differences in size of the CD loop, human MAdCAM1 contains a unique arginine residue (R70) that is buried within domain 1, surrounded by hydrophobic residues that fail to neutralize its charge (see Fig. 5B). This arginine is capable of donating three hydrogen bonds, two of which interact with D42 and T43 in the CD loop (68). These hydrogen bonds appear to constrain the backbone to a novel y-turn configuration around D42, which contrasts with a @-turnaround T37 and D40 in VCAM-1. Additionally, R70 also interacts with A66 in the EF loop, suggesting a role in maintaining the structure of both the E F and CD loops. Mutagenesis has shown that alanine substitution for R70 completely abolishes a407 binding, confirming the importance of this conserved residue in MAdCAM-1, and providing an additional structural distinction from VCAM-1 that may contribute to the specificity of integrin interactions (M. Briskin, unpublished observations).

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

219

Another unique feature of MAdCAM-1 is that its second Ig domain, like the N-terminal domain, belongs to the “I1 set” of‘ Ig domains (68). This contrasts with all other Ig family adhesion molecules (and with CD2 and CD4) whose second Ig domains lack a D strand and that are classified as I2 set domains (71).Domain 2 of MAdCAM-1 is 114 residues in length, which is extremely large for an I set domain. The most significant feature of this domain is a negatively charged @-ribbon loop of 11 amino acids between the D and E strands (see Fig. 5B). Seven of the 11 residues in this loop are aspartate or glutamate, and mutation of several of these residues results in inhibition of a4P7 binding. Importantly, this negatively charged “antenna” extends outward from domain 2, reaching into close proximity to the conserved GLDTS motif in domain 1, where it may potentially contribute to integrin binding or may help orient the integrinbindmg face of MAdCAM-1 in an optimal position for interaction. The coulombic field set up by the sialic acid-rich inucin region of MAdCAM1 (in addition to the negative charge of the cell surface) could repel the “DE antenna,” thus forcing it and the adjacent integrin-binding face of domain 1 to point outward, accessible for integrin interaction (see Fig. 5A) (68). Consistent with this model, the importance of the negatively charged antenna in domain 1presentation has been confirmed by swapping the MAdCAM-1 DE antenna and the corresponding domain 2 loop from VCAM-1: replacing the MAdCAM-1 antenna with the corresponding sequence from VCAM-1 results in a nonfunctional molecule. In the converse situation, the MAdCAM-1 antenna significantly alters the presentation of VCAM-1 to a4P7, rendering it more “MAdCAM-1-like”: a4P7 binding to this VCAM-1 mutant is inhibited by the monoclonal antibody (mAb) Act1,which normally blocks a4P7 binding to MAdCAM-1 but not to VCAM1 (74). In summary, the N-terminal Ig domain is responsible for the selectivity of MAdCAM-1 for a4@7, but it has not yet been possible to assign this selectivity to a more restricted sequence or structure within this Ig domain, suggesting that its interaction with a4P7 may be complex. In addition, normal and efficient presentation of MAdCAM-1 domain 1 to a4P7 requires domain 2, and in particular the negatively charged domain 2 DE antenna. In addition to a potential role in orientation of the distal a4P7 integrin binding domains, the membrane-proximal mucin region likely also serves to present the a4P7-binding Ig domains away from the endothelial surface, thus facilitating molecular interactions with lymphocyte homing receptors. The human MAdCAM-1 mucin domain, for example, is predicted to form a stalk extending -20 nm from the cell membrane, well beyond the surface glycocdyx (see Fig. 5A). In addition, when MAdCAM-1 is expressed by

220

EUGENE C. BUTCHER et ol.

high endothelial venules in mesenteric lymph nodes and Peyer’s patches, the 0-linked glycans of the mucin domain also present L-selectin-binding carbohydrates (75), allowing MAdCAM-1 in these lymphoid organs to support lymphocyte interactions through L-selectin as well as a4P7. In contrast, modification of MAdCAM-1 by L-selectin-binding carbohydrates does not appear to occur in venules in the normal mouse intestine. This facultative display of L-selectin-binding carbohydrates by MAdCAM-1 probably helps define the differential homing patterns of naive versus memory/effector lymphocytesto mucosal lymphoid organs versus the extralymphoid lamina propria (76) (see also below).

2. The Intestinal Homing Receptor, a4P7 a4P7 is a member of the heterodimeric integrin adhesion receptor family (77).Leukocyte integrins, including a4/37, are displayed in relatively inactive or low avidity states on most resting and blood-borne lymphocytes, but can be triggered to undergo dramatic functional activation in response to a variety of stimuli (32, 78, 79). In particular, chemoattractant receptors can trigger functional activation, assessed by lymphocyte adhesion to integrin ligands, within seconds (33,80).It is not known whether this functional activation involves changes in receptor affinity through conformational alterations, changes in receptor avidity through clustering, or enhanced mobility in the membrane, or a combination of these effects. Triggering of a4P7 and of LFA-1 (aLP2)by HEVs or lamina propriavenule-associated activating signals may play an important role in lymphocyte arrest (see below). Interestingly, however, a4 integrins display a significant (if low affinity/avidity)activity even in the absence of stimulation. Moreover, circulating lymphocytes display these a4 integrins concentrated on microvilli (8l)-for this reason, in some settings they can play an important role in activation-independent initial contact formation, or “tethering,” under shear (81, 82) [which depends on presentation of receptors on microvilli (83)]. In the absence of activation, a4 integrins can also support reversible lymphocyte rolling (81, 82). a4P7 binds MAdCAM-1, but also other ligands, including the CS1 peptide-containing splice variant of fibronectin, the a4 integrin chain, and VCAM-1 (84-86). These interactions can be differentially inhibited by mAbs against different epitopes of a4/37, suggesting involvement of unique as well as overlapping contact sites (87). These additional ligands may lymphocyte cell-cell and cell-matrix be important in gut-homing ~x4/37~’ interactions within lymphoid tissues and within the lamina propria; however, there is as yet no evidence to support a significant role for a4P7 adhesion through VCAM-1 (or fibronectin or a4) in the recruitment of lymphocytes from the blood. Indeed, at least on most circulating lympho-

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

221

cytes (including ~ r 4 0 7T~ cells, ' which tend to be a4pl''' but not negative), binding to VCAM-1 is dominated by a4P1, with a4P7-VCAM-1 interaction being difficult to demonstrate except under artificial experimental conditions (88, 89). Moreover, VCAM-1, although like MAdCAM-1 an Ig family member that can be expressed by endothelium, is clearly distinguished from MAdCAM-1 by its patterns of expression. Although abundant in the stroma of PPs and the gastrointestinal lamina propria, VCAM-1 is largely excluded from expression by mucosal endothelium-even during chronic inflammation in inflammatory bowel disease (90).Instead, VCAM1is induced on vessels involved in mononuclear cell recruitment in nonintestinal sites of inflammation, notably the skin, heart, and CNS. Thus, although it cannot be ruled out that activated a4P7 on, for example, immunoblasts may participate in VCAM-l-dependent extravasation in nonmucosal sites, current data suggest that, in the context of leukocyte homing from the blood, a4P7-MAdCAM-1 and a4p1-VCAM-1 are largely independent homing receptorhascular "addressin" interactions that play a major role in the differential trafficking of intestinal (a4@7+,a4@I") versus nonintestinal (a4P7- a4j37') memory and effector populations in uiuo. Fibronectin has been implicated in a@l-dependent lymphocyte adhesion to endothelium in vitro, but its ability to support physiologic a4p7dependent adhesion and its importance to in vivo recruitment remain to be explored (91). The ability of a4P7 (and a 4 p l ) to bind to the a 4 chain may be important for cell-cell interactions within tissues, but at least potentially may also be relevant to lymphocyte trafficking given that endothelial cells can also express a 4 integrins (85, 92). In addition to its expression by naive and intestinal memory lymphocytes (74, 93-95) (see below) a4@7is displayed by a variety of other leukocyte subtypes, including human eosinophils, basophils, and rat mast cells. In vitro data have shown that all of these classes of leukocytes are capable of interacting with MAdCAM-1 (via a4@7),interactions that may facilitate their recruitment into gastrointestinal tissues (96-98). Circulating monocytes express very little p7, but the p7 chain can be up-regulated on monocytoid cells and on monocytes by differentiating stimuli, and p7 expression can be dramatic on tissue macrophages (99) and on plasma cells (100).a4P7 likely plays distinct roles in the function and tissue-based interactions of these relatively sessile (nonrecirculating) leukocyte populations. The critical role of a4p7 in lymphocyte homing to PP and to intestinal lamina propria, first indicated based on its involvement in lymphocyte binding to PP-HEVs in vitro (35),was confirmed through in vivo homing studies using antie4 and anti-a4/37 mAbs: mAbs capable of blocking MAdCAM-1 binding in vitro also inhibited lymphocyte homing from the

222

EUGENE C. BUTCHER et al

blood into PPs in uiuo, and substantially reduced lymphocyte entry into the small and large intestinal lamina propria as well, paralleling the effects of anti-MAdCAM-1 mAb (36, 76, 101) (see below for further discussion). Subsequent studies of 07-deficient mice were entirely consistent with these early investigations,confirming the critical role of 0 7 in lymphocyte homing and in the function of intestinal PPs (38). Interestingly, Peyer’s patches were poorly developed and there was a dramatic reduction in plasma cells and T cells in the intestinal lamina propria, and of intraepithelial leukocytes (IELs) in the intestines of these 07-deficient mice as well. Although the three-dimensional structure of a407 has not been determined, it contains within its two chains regions of sequence homology with functionally important integrin subdomains whose structure has been determined or considered in other integrins. First, 0 7 (like all integrin P chains) contains an “I domainlike” structural homology unit, encompassing amino acid residues 150 to 322 in the 0 7 chain sequence. Interestingly, six out of seven function-blocking anti-P7 mAbs recognized epitopes within a single short region (residues 176-250) of this “I domain” homology unit (102). Furthermore, in functional assays with chimeric 67/01 integrins, this region was required for adhesion to MAdCAM-1, further supporting its involvement in ligand binding. This 0 7 I-like domain is related to the true I domains found in the a chains of 0 2 integrins (but not in a4). Importantly, in the case of cxM [the Mac-1 a chain, C D l l b (103)]and a L [LFA-1 cx chain, C D l l a (104)],these isolated I domains bind directly to 0 2 integrin ligands, and are thought to play a critical role in ligand specificity. The presence of I domains in the 0 but not a chains of a4 integrins may explain the importance of 0 7 versus 01 in determining MAdCAM-1 binding selectivity, whereas the I domain-containing Q chains in the 0 2 integrins are critical determinants of differential interaction with the ICAMs. Oxygenated residues from I domains (and likely from the I domainlike region of 0 7 as well) can coordinate binding of divalent metal cations (Mg2’). Indeed, the I-like domain of 0 7 contains a metal ion-dependent adhesion site (MIDAS) consensus sequence, DLSYS, at residues 159-163, in a location similar to those of the MIDAS motifs within the I domains of a L and a M . In the a M I-domain crystal structure with MIDAS-bound M$+, a glutamate side chain from a neighbor molecule in the crystal lattice completes the octahedral Mg2+ coordination complex. This glutamate may be acting as a ligand mimetic (105),mirroring the fact that all integrin ligands contain acidic resides as a key feature of their ligand-binding motifs. Taken together, therefore, the available structural information leads to the hypothesis that Mg2+may serve as a key coordinating element for a407-MAdCAM-1 interaction, by binding to and linking the MIDAS motif

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

223

in the 07 I-like domain to the conserved aspartate residue D42 in the protruding MAdCAM-1 domain 1 CD loop. The second common integrin element found in a407 is within the a 4 chain-a “0 propeller” domain. One model suggests that all integrin a subunits share this common structural motif (106),which was first observed in the crystal structure of the 0 chain of heterotrimeric G proteins. Each a chain contains seven homologous repeats that are predicted to form a propeller-like cyclic structure consisting of seven four-stranded 0 sheets. Mg2’ ions may bind to sites on the upper face of the propeller, in regions apparently involved in ligand binding, whereas Ca2+binding motifs on the lower face of the propeller may play a structural role, because Ca” modulates integrin affinity. Interestingly, the I domains are structurally homologous to the G protein a subunits, and in a M and a L the I domains are linked to and predicted to be located on top of the propeller domain, tethered by a hinge that should allow movement of domains relative to one another (106). It is intriguing to speculate that in the a4 integrins the I domainlike sequences of the 01 and 07 chains may form functional ligand-binding units through interchain associationwith the a 4 0 propeller, as well. The model of a ligand-bindmg region on the upper surface of the propeller is supported by studies showing that a function-blocking antia4 inAb maps to the upper face of 0 propeller repeats 1-4, and that mutation of residues in this same region inhibit a4 integrin binding to VCAM-1 and fibronectin (107). Mutation of analogous resides in a5 and aIIb also reduce or abolish integrin binding (108).These a 4 mutations are predicted to inhibit binding to MAdCAM-1 as well, although this has yet to be studied. In summary, the available structural data suggest that integrins use at least two distinct structural domains to interact with specific Ig family members (and other 1igands)-the “I domain,” analogous to the a chain of heterotrimeric G proteins, and the 0 propeller, a structural homolog of the G protein 0 subunit. In the 02 integrins MAC1 and LFA-1, both of these domains lie within the a chains, which are primarily responsible for determining ligand specificity. In contrast, in the a 4 integrin homing receptors, the propeller is in the a4 chain whereas I domainlike sequences are solely provided by the 07 and 01 chains. The I domain homologous sequence in 07 is clearly critical for, and is likely to be a major determinant of, integrin binding to MAdCAM-1, and may act in association with the a4 chain 0 propeller.

3. Other Molecules lnvolved in lntestinal LymphocyteEnclothelial Interactions Here we focus on the known or potential role of adhesion and homing molecules in intestinal leukocyte trafficking, comparing and contrasting

224

EUGENE C . BUTCHER et

d.

this with their participation in other sites. The reader is referred to recent reviews for additional discussions of these molecules (21,25, 26, 79, 109).

a. L-Selectin and Its HEV Ligand(s). The leukocyte L-selectin (CD62L) is a C-type calcium-dependent lectin of the selectin subfamily (110, 111).This family also includes the endothelial E-selectin (CD62E) and the platelet and endothelial P-selectin (CD62P). L-Selectin was originally identified as a peripheral lymph node homing receptor, because L-selectin helps determine the selectivity of lymphocyte recruitment to nonmucosal lymph nodes (112).HEVs in peripheral lymph nodes express extraordinarily high levels of L-selectin-binding carbohydrates, much higher than those in PP-HEVs (to be discussed below) (113, 114). These unique carbohydrate structures, which are thought to share a common sialyl Lewisxcore with functionally important sulfate capping groups (109), decorate many glycoproteins on PLN-HEVs especially,but likely not exclusively, mucins. Together, the physiologically significant L-selectin ligandpresenting molecules compose the PLN addressin (PNAd) (113).Although L-selectin appears critical to LN homing (112, 114, 115),L-selectin also plays an important if less dominant role in lymphocyte homing to PPs, where its principal but likely not exclusive HEV ligand is (L-selectinbinding, carbohydrate-modified) MAdCAM-1 (76).L-Selectin, like a4P7, is concentrated on the tips of microvilli (116), and it is highly efficient at initiating tethering and supporting rolling of lymphocytes on HEV ligands in vitro (75, 117) and on HEVs in vivo (76, 114). L-Selectin is expressed by most circulating leukocytes (20, 118),including essentially all naive lymphocytes of the adaptive immune system, and a major fraction of memory phenotype T and B cells, as well. On memory cells its expression is not mutually exclusive with that of a4p7; indeed, memory cells are divided into relatively discrete Lblood-borne c~4P7~' selectin' versus L-selectin- populations, of variable but roughly equivalent frequency (89). Although not yet experimentally established for memory cells per se, it seems likely that L-selectin expression enhances memory cell trafficking through lymph nodes, especially peripheral lymph nodes lacking MAdCAM-1' HEVs. In contrast, the trafficking of L-selectinmemory cells may be more restricted to extralymphoid sites, such as the intestinal lamina propria. On the other hand, anti-PNAd antibodies do react with venules, especially HEV-like vessels, observed in the chronically inflamed colon (90, 119), indicating that under conditions of severe chronic inflammation, Lselectin ligands can be induced to facilitate recruitment of a broader range of lymphocytes, including naive T cells. Indeed, in any setting where chronic inflammation is sufficiently prolonged, organized lymphoid tissue

LYMPHOCYTE TRAFFICKING A N D REGIONAL IMMUNITY

225

can develop de novo locally, complete with B and T zones, germinal centers, dendritic cells, and all the histologic and architectural elements of secondary lymphoid organs (20, 120). In these sites, the associated venules typically express high levels of the peripheral node addressin, just as they do in conventional lymphoid tissues arising developmentally.

b. LFA-1 (aLp2, CDllalCD18). LFA-1 is an integrin receptor for the widespread vascular and stromal cell ligands ICAM-1, ICAM-2 and ICAM3 (25, 79, 121). It is expressed by most circulating leukocytes, including the great majority of circulating lymphocytes (but not all). Like ( ~ 4 0 7its , activity can be triggered by lymphocyte activation through chemoattractant receptors. Unlike the a 4 integrins, however, LFA-1 is excluded from microvilli, and is distributed instead to the planar cell body of leukocytes (122). This topographic distribution may help explain the observation that LFA1 plays no detectable role in initiating lymphocyte contact with high endothelium in vivo, but instead appears almost exclusively limited in its function to post-activation-events (76, 114, 11 7, 123)-activation-triggered arrest, and probably subsequent transendothelial migration. c. Vascular Selectins and Other Potential Molecular Participants. Vascular E- and P-selectins appear to play little or no role in lymphocyte trafficking to normal intestinal sites-either to intestinal lymphoid tissues or to the gut lamina propria. Neither selectin can be detected immunohistologically, for example, in the normal mouse small intestine or colon. This contrasts with important roles for vascular selectins in lymphocyte homing to certain extraintestinal sites. For example, E-selectin plays a critical role in cutaneous inflammation as a skin vascular “addressin,” acting as a ligand for the cutaneous lymphocyte antigen on specialized skin homing memory T cells (22). However, E-selectin and P-selectin can be up-regulated in mucosal tissues during inflammation, especially in the acute phase during which they may facilitate neutrophil recruitment from the blood. In this setting, they may contribute to tethering of lymphocyte subsets as well (see also discussion of altered trafficking in inflammation, below). The P7 integrin chain can be expressed as a heterodimer with ae as well as a 4 (124).aep7 is highly expressed by gut intraepithelial lymphocytes and supports their interactions with intestinal epithelial cells by binding to epithelial or E-cadherin (125, 126). Unlike the bulk of lamina propria lymphocytes, which are a4j37+, IELs are predominantly a4/37- (94, 127). [In this regard, it is interesting that mouse IELs can bind to PP-HEVs in frozen sections in vitro (128),an early observation in need of‘ molecular explanation.] A small subset of circulating lymphocytes, especially CD8t (but also some CD4t) T cells, expresses aeP7 and a407 together (89):

226

EUGENE C . BUTCHER et al

the significance of this subset in the blood is unclear, because aep7 does not appear to participate in lymphocyte-EC interactions or extravasation per se (86). Instead, aep7 is thought to be induced on IEL precursors entering the lamina propria, perhaps in response to local TGF-0, thus helping regulate and target their microenvironmental homing and interactions with the epithelium (129). aep7 is also expressed by IELs in the bronchial respiratory mucosa, and by mononuclear cells in the squamous epithelium of skin (130, 131).In the mouse, a major fraction of CD8’ T cells and a subset of (predominantly memory phenotype) CD4’ T cells in LNs and spleen are also a e + (94; L. Rott, M. Williams, and E. C. Butcher, unpublished observations). CD43 is a leukocyte sialomucin, expressed by most leukocytes, including T cells, but not (or only poorly) by B cells. I n vitro studies suggest a role for this highly charged and extended molecule as an “antiadhesion” or “barrier” molecule, and also suggest its participation in T cell signaling events. A monoclonal anti-CD43 antibody inhibits lymphocyte homing from the blood into all lymph nodes, PPs, and spleen, defining CD43 as a potential target for modulating lymphocyte trafficking (132). On the other hand, lymphocyte homing appears normal or even enhanced in CD43-deficient mice (133):thus unlike the other molecules discussed here, CD43 does not appear to be required for efficient homing. It is reasonable to hypothesize that the inhibitory effects of anti-CD43 antibodies may be mediated indirectly, perhaps through interference with signaling pathways involved in integrin triggering during lymphocyte homing. Although not yet confirmed by in vivo studies, it is likely that additional molecules will prove to participate in lymphocyte trafficking to the gut and its associated lymphoid tissues. In this regard, vascular adhesion protein1 (VAP-1) may prove of particular interest. This glycoprotein has been implicated in lymphocyte-endothelial interactions in inflamed joints, but is also up-regulated by vessels in inflamed skin and intestinal lamina propria (134); it has also been implicated in binding and tethering of subsets of lymphocytes to HEVs in inflamed human lymph nodes (135).Similarly, a vascular monocyte adhesion-associated glycoprotein, VMAP-1 (136),is expressed on subsets of mucosal venules, and may prove to participate not only in monocyte but also in specialized lymphocyte subset trafficking events. Finally, the hyaluronate receptor CD44, a clear participant in lymphocyte interactions with hyaluronate on cytokine-stimulated endothelid cells in vitro, has recently been implicated in recruitment of activated T cells to the inflamed peritoneum (137). Hermes-3, a unique antibody to a proximal stalk region of CD44 [not part of the hyaluronate binding domain (A. Ge and E. C. Butcher, unpublished observations)], blocks lymphocyte binding to Peyer’s patch- and appendix-HEVs in in vitro

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

227

Stamper-Woodruff frozen section binding assays (138),and the lymphocyte isoform of CD44 and MAdCAM-1 can associate in solution (139); but the mechanisms and significance of these in vitro observations to in vivo lymphocyte homing to intestinal sites is uncertain. Importantly, there is no evidence to suggest that lymphocyte interactions with uninflamed HEVs require CD44 in its role as a hyaluronate receptor. Anti-CD44 MAbs that significantly inhibit hyaluronate binding, for example, do not inhibit lymphocyte adhesion to HEV in vitro (138, 140). Moreover, in early studies intravenous injection of hyaluronidase had little effect on lymphocyte homing to lymphoid organs ( 1 4 1 ) . B. INTEGRIN-TRIGGERING AND CHEMOTACTIC FACTORS IN LYMPHOCYTE RECRUITMENT FROM THE BLOOD: A ROLEFOR CHEMOKINES Gai-linked receptors trigger the arrest of rolling lymphocytes in PPand PLN-HEVs (114, 142) and it is reasonable to expect that similar activation-dependent signals are required for arrest of rolling memory T and B cells in sites of inflammation. (Preactivated lymphocytes such as immunoblasts, among the most efficient subsets at homing to extralymphoid tissues, may express high levels of active integrins on entering the blood, and thus may potentially bypass the requirement for venuleassociated adhesion triggering factors.) However, although several of chemoattractants have been shown to trigger arrest of rolling granulocytes in vivo, the physiologic factors involved in triggering lymphocyte arrest on endothelium are only now yielding their identities, based in large part on discovery of novel chemokines through randomly sequenced messages. Several recently identified members of the chemoattractant cytokine (chemokine) family have the potential to regulate lymphocyte arrest: in particular, SDF-1, MIP3a (also h o w n as Exodus or LARC), MIP3P (Exodus-2, ELC), and the secondary lymphoid chemokine SLC (also known as 6-C-kine, Exodus-3, or TCA4, and hereafter referred to as SLC/ 6Ckine) can each trigger robust adhesion of blood lymphocytes to ICAM1 (33, 34). Moreover, consistent with the rapidity of in vivo lymphocyte arrest on HEVs, triggered adhesion is extremely rapid, often occurring within a second in in vitro flow assays (33). MIP3P, SLC/GCkine, and SDF-1 induce rapid binding of the majority of circulating CD4' T cells, including naive cells. In contrast, MIP3a triggers adhesion of memory but not of naive CD4+ T cells, confirming the potential for subset-selective regulation of adhesion by chemokines (33). SDF-1 is a widely, almost ubiquitously, expressed chemokine, and for this reason it is perhaps unlikely to contribute directly to the specificity of lymphocyte subset targeting from the blood (143,144).Its known receptor CXCR4, an HIV coreceptor, is widely expressed as well, and many cells

228

EUGENE C. BUTCHER et al

in the body (including endothelial cells) can respond to SDF-1, as can most leukocytes. In particular, SDF-1 attracts monocytes, blood cells that do not normally circulate via uninflamed HEVs. In contrast, the 0 chemokines MIPS0 and SLC/GCkine are highly and indeed preferentially (albeit not exclusively)expressed in secondary lymphoid organs, at least as assessed at the message level by Northern and in situ hybridization analyses (145-147). SLC/GCkine is of particular interest because it is expressed by PP- and PLN-HEVs, both at the message level as assessed by in situ hybridization (34) and at the lumenal surface as assessed by fluorescent anti-SLC/GCkine (TCA4)localization (R. Warnock, M. Dorf, J. Campbell, and E. C. Butcher, unpublished observations). Interestingly, SLC/GCkine and MIPS0 share a common receptor on a majority of lymphocytes: each desensitizes lymphocytes to the other (a feature that strongly suggests that they act predominantly through the same receptor), and SLC/GCkine competes for MIP3P receptors on lymphocytes in direct binding assays (148). In fact, both MIPSP and SLC/ 6Ckine signal through the lymphocyte-expressed receptor CCR7 (also known as EBIl or BLRZ) (149, 150). Importantly, both of these lymphoid organ chemokines act preferentially on lymphocytes. Thus the available data suggest that CCR7, triggered in part by HEV-associated SLC/GCkine (and potentially by MIPS/$, may be a prominent lymphocyte chemokine “homing receptor” for secondary lymphoid tissues, including intestinal PPs and the mesenteric LNs. Consistent with this thesis, desensitizing preincubation of lymphocytes with SLWGCkine, but not with SDF-1, dramatically inhibits lymphocyte arrest and accumulation on PP-HEVs within the first 10 min after intravenous injection (R. A. Warnock, L. McEvoy, J. Campbell, and E. C. Butcher, personal observations). Antibody inhibition studies or analyses of gene-targeted SLC/GCkine-deficient mice will be required to confirm the critical role of this chemokine in triggering integrin-dependent lymphocyte arrest on HEVs in Peyer’s patches and lymph nodes. As for many other chemokines, SLC/GCkine can signal through more than one receptor, including CCR7, but also CXCR3 (151) and potentially others. In the context of lymphoid organ physiology, it may be desirable for an HEV-associated chemokine or chemokines to be able to trigger arrest of a broad range of circulating lymphocyte subsets-including memory as well as naive B and T cells (see below). It will not be surprising if yet additional lymphocyte-specific receptors for SLC/GCkine are identified. Moreover, HEVs probably express or display additional chemokines, as well. Consistent with this, although T cell homing to LNs is dramatically reduced in DDD/1 mice, a strain that is deficient in SLC/GCkine and expresses reduced levels of MIPS0 (M. Gunn, personal communication),

229

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

B cell homing into lymph nodes appears normal (152),suggesting involvement of additional HEV-associated chemokines or other activating factors selective for circulating B cells. Such a chemokine may, like SLC/GCkine, be synthesized by HEVs, or may be recruited from the surrounding tissue and presented on glycosaminoglycans at the HEV surface (153). One candidate for a “backup” chemokine for B cell recruitment into PPs and LNs is BCAl (also known as ADEC or BLCl), which is highly expressed at the mRNA level in B cell follicles (34, 154). Thus, it is likely that additional chemokines involved in lymphocyte-HEV interactions remain to be identified. Moreover, we anticipate that distinct endothelial-associated chemokines in different organs and tissues will help target the arrest of specialized memory lymphocyte subsets, especially in extralymphoid sites of tissue inflammation. In summary, it is attractive to propose that the differential expression of adhesion-triggering chemokines will prove as important as adhesion molecules in controlling lymphocyte subset trafficking in vivo. Identifying the physiologic factors responsible for activating firm adhesion and arrest of memory lymphocytes homing to the gastrointestinal lamina propria (and indeed determining whether such activation is or is not required for lymphocyte recruitment via lamina propria venules), and analysis of the extent to which this activation step contributes to the selectivity of intestinal lymphocyte trafficking, are important areas for future investigation. Finally, lymphocyte subset recruitment into mucosal tissues is undoubtedly also regulated at the level of transendothelial migration and entry into tissues. Chemoattractants, especially members of the expanding chemokine family, are likely to be important in this regard as well, but as yet the physiologic factors involved in lymphocyte diapedesis across HEVs or lamina propria postcapillary venules remain to be defined. Importantly, early studies of Lamm and Czinn (155)suggest the existence of chemoattractants that can selectively recruit mesenteric versus peripheral lymph node immunoblasts, reinforcing the need for further molecular studies in th’is arena. IV. Subset Specificity and Mechanisms of lymphocyte Targeting to Intestinal Tissues in Wvo

A.

SELECTIVE

HOMING OF dP7-DEFINED T CELL SUBSETS IN

VIVO

As mentioned above and as illustrated in Fig. 6, a407 is expressed on discrete subsets of circulating memory phenotype B and T lymphocytes (74, 89, 94, 95, 156). Studies have confirmed the hypothesis that these a4j37’licells are uniquely capable (among memory cells) of binding to MAdCAM-1 (89),and of trafficking to the intestinal lymphoid tissues (37).

230

EUGENE C. BUTCHER et al.

CD4+ T cells

B cells

$1

;.

Anti-IgD

1%-

anti-CDURA

FIG.6. Circulating memory T cells and B cells are subdivided into ~i4fi7~' and a4/37subsets. In contrast, presumptive naive lymphocytes (identified by CD45RA expression among CD4 cells, and predominating among IgD' CD19' B cells) are relatively homogeneous, expressing intermediate levels of cu4fi7. Human peripheral blood lymphocytes were stained by three-color immunofluorescence (anti-fi7 vs. IgD vs. CD19; or anti-P7 vs. CD45RA vs. CD4); two-color fluorescence-activated ceU sorting plots of gated CD4 or CD19+ cells are shown. Except on the small population of ae/37+ cells (which represent -1% of circulating lymphocytes), expression of P7 correlates with that of a4P7 (89).

Figure 7 illustrates the selective homing of a4P7-expressing memory T cells to PPs in vivo: a4P7' memory T cells and naive cells home well to PPs, but a4P7- T cells are excluded. In contrast, naive cells home -lOX better to peripheral lymph nodes compared to ~ 4 P 7memory ~' cells, and -5X better compared to a4P7- memory cells. Interestingly, all three subsets home well to the spleen, which may serve as a site of mixing, communication, and potentially homeostatic competition between various homing subsets. These studies establish that a407 expression correlates with extravasation into intestinal lymphoid tissues. Short-term homing experiments in the rat demonstrate that subsets of small memory T cells (but not naive T cells) can home efficiently to the intestinal lamina propria, as well: this process is a 4 integrin-dependent but VCAM-1-independent, and thus presumably mediated by a4P7 and MAdCAM-1 (157).Importantly, independent studies in the sheep confirm a dramatic enrichment in a4P7' memory cells among lymphocytes recirculating through intestinal lymphatics, compared with a relative paucity of cells in afferent lymph draining cutaneous and pulmonary tissues (54, 55).Together, these findings suggest that differential a4P7 expression is a major determinant of lymphocyte trafficking through intestinal versus extraintestinal tissues. In the next section, we discuss how a4P7 and MAdCAM-1 mediate selective trafficking, and how these molecuIes act in coordination with L-selectin and LFA- 1to allow differential targeting of naive versus memoryjeffector intestinal lymphocyte subsets to lymphoid versus extralymphoid (lamina propria) compartments.

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

Tz

5

Injected T cells

%-5 25

231

P

Peyer's patches

Peripheral lymph node

Spleen

Homed T cells FIG.7. Differential recruitment of a4B7 homing receptor-defined T cell subsets to lymphoid tissues in uiuo. The a4P?"memory (CDMh')T cells localize as well as naive (CD44")T cells to Peyer's patches whereas a4p7- memory cells are excluded. Both memory subsets enter peripheral lymph nodes, but poorly compared to naive cells. All three subsets home well to the spleen. Similar results were obtained both in short-term (2.5 hr) and overnight homing studies (37). Methods: Spleen cells from C57BLJ6 Thy1.2' mice were injected intravenously into congenic C57BU6 Thyl.I+ mice. Cells from peripheral lymph node, mesenteric lymph node, Peyer's patch, and spleen of recipient mice were stained with mAbs to Thyl.2, a4p7, and CD44 (a marker elevated on previously activated "memory" cells) and analyzed by fluorescence-activated cell sorting (FACS).Three regions are delineated on each plot: a4p7'"naive (CD44'"),a4P7h'memory (CD44"), and a4P7- memory (CD44'").(Top)Injected population: Representative FACS plot of injected cells illustrates a@7 versus CD44 stain of gated Thy1.2' small lymphocytes. (Bottom)Homed populations: Representative FACS plots of locahed Thy1.2' (donor phenotype) cells in peripheral and mesenteric lymph nodes, Peyer's patches, and spleen. Relative localization ratios for each subset (a measure of the efficiency of its homing following intravenous injection, relative to that of naive phenotype cells whose relative localization ratio defines unity) are given.

B. DIFFERENT DECISION PROCESSES DETERMINE NAIVEVERSUS MEMORY~EFFECTOR LYMPHOCYTE HOMING TO INTESTINAL SITES Naive Crp T cells and virgin B cells, which constitute the essential precursor pool for the adaptive immune system, require help to respond to their cognate antigens and to differentiate into memory and effector cells. Such help is normally provided by the specialized antigen processing and presenting microenvironments of secondary lymphoid tissues, and the efficiency of this process is enhanced by the remarkable efficiencywith which naive lymphocytes traffic and recirculate through the PPs, LNs, and spleen.

232

EUGENE C. BUTCHER et al.

On the other hand, although subsets of memory and effector cells can also traffic to lymphoid tissues, these previously stimulated cells (unlike their naive precursors) also circulate well through “tertiary” or extralymphoid tissues in sites of inflammation. Direct in situ analyses (76) have revealed differences in the molecular mechanisms controlling homing of resting LN lymphocytes (primarily naive T and B cells) versus preactivated or a4P7h‘ lymphoid cells (models of gut-homing immunoblasts and possibly memory cells). How do the molecules involved in lymphocyte homing interactions target these populations differentially into lymphoid tissues versus the extralymphoid intestinal lamina propria?

1. Naive Lymphocyte Adhesion Cascade Naive T and B cells express L-selectin and moderate levels of a4P7 and LFA-1. These cells interact with PP-HEVs through a complex series of events mediated by sequential but overlapping engagement of L-selectin, a4P7, and LFA-1 (76).L-Selectin interacting with carbohydrate-modified MAdCAM-1 dominates the initiation of primary contact (“tethering”). a4P7 can also initiate adhesion of naive cells, but at lower efficiency. Both L-selectin and a4P7 participate in activation-independent rolling of lymphocytes on MAdCAM-1, with a4P7 playing an apparently essential role in increasing the avidity and reducing the velocity of rolling cells on HEVs. Engagement of a4P7 is required for activation-dependent adhesion strengthening and arrest, which also requires involvement of LFA-1. As outlined above, stimulation by chemokines can trigger rapid robust adhesion and arrest of rolling lymphocytes by stimulating LFA- 1-dependent binding to ICAM-1 (see above, and Ref. 33), and studies suggest that HEVassociated chemokines (in particular SLUGCkine, but probably others as well) are involved as adhesion-triggering agonists in naive lymphocyte homing to PPs (as well as to PLNs and probably to the spleen). One of the striking features of this multistep decision cascade is the overlap in functions of the adhesion receptors involved. Although each component is required for efficient homing, overlapping roles permit significant residual interaction even when one of the components becomes inoperative. This overlap between sequential functions may explain in part the incomplete inhibition of lymphocyte homing to PPs by antibodies to any one of the various receptors involved (36). The ability of a4P7 to initiate tethering, albeit inefficiently, may help explain the population of PPs by lymphocytes in L-selectin-deficient mice (125). Overlapping roles may also be important evolutionarily,in a general sense, permitting significant trafficking even in the context of alterations or deficiencies in individual homing receptor-ligand pairs.

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

233

a4P7 occupies a unique position in the process of naive lymphocyte homing to PPs. It appears to be required in a “bridging” role for efficient conversion of L-selectin-mediated primary interaction to LFA-l-mediated arrest, thus ensuring its participation in the “decision” process leading to lymphocyte recruitment into PPs. The related a4Pl integrin (VLA4) may play a parallel bridging role between selectin-mediated tethering and P2 integrin-mediated arrest in other settings-e.g., in eosinophil or monocyte recruitment, or in conversion of CLA-E-selectin-mediated tethering to LFA-l-dependent arrest of subsets of skin-homing lymphocytes. The requirement for a4P7 participation in rolling in PPs also contrasts with the mechanism of naive cell homing to PLNs. In this site, HEVs express high levels of PNAd (ligands for L-selectin), and in situ studies in the mouse suggest that L-selectin-dependent rolling can convert to activation- and LFA-l-dependent arrest directly, as for neutrophils (114). a4 integrins are not required for lymphocyte arrest on PLN-HEVs, although involvement of additional adhesion molecules such as VAP-1 [recently implicated in Lselectin-independent rolling of subsets of human lymphocytes on PLNHEV in ex vivo assays (135)] has not been excluded. 2. Distinct Behavior of €‘reactivated a4/37hiLymphocytes The molecular basis of vascular adhesion of preactivated lymphocytes memory T cells) is quite (or of a4P7‘ lymphoid cells as a model of distinct from that of the majority of naive lymphocytes. As mentioned above, a4P7, like L-selectin, is presented on microvilli, and when expressed at high levels a4P7 can mediate direct selectin-independent interactions of blood-borne lymphocytes with MAdCAM-1, not only in PP-HEVs but also in venules in the small intestinal lamina propria in situ (81). Of particular interest is the fact that, although naive lymphocytes fail to interact detectably with lamina propria venules (which are MAdCAM-1’ but lack L-selectin-binding glycotopes),even naive lymphocytes can bind efficiently to lamina propria venules when their a4P7 integrins are preactivated experimentally. The dramatic effect of integrin triggering on the role of a@7, with loss o f requirement for L-selectin, underscores the potential of lymphocyte activation to regulate both the mechanisms and sites of lymphocyte homing. In particular, the ability of a4@7expressed at high functional levels to initiate and consummate adhesion may help explain the distinctive gut-homing properties of mucosal immunoblasts, which traffic so efficiently to the intestinal lamina propria (1, 8, 9, 12, 14, 158; reviewed in Refs. 19 and 20). It may also underlie the selective trafficking of a4PYhigut-recirculating memory T cells (55),a significant subset of which lack L-selectin. On the other hand, the inefficiency of contact initiation by

234

EUGENE C. BUTCHER et al

a 4 integrins on naive lymphocytes may help to ensure that naive cells traffic preferentially through L-selectin-binding lymphoid organ HEVs. C. COMPARISON WITH DECISION CASCADES TARGETING LYMPHOCYTES TO NONINTESTINAL TISSUES The observations just summarized illustrate how modest differences in expression of a limited number of receptors can confer differential homing of specialized lymphocyte subsets. This theme becomes even more cogent when we consider the decision cascades involved in lymphocyte trafficking to other tissue sites (see Fig. 8). To summarize, naive lymphocytes home to PLNs in a cascade involving L-selectin-dependent contact and rolling and LFA-1-dependent arrest, thus maintaining the role of L-selectin in targeting cells to HEV-bearing lymphoid tissues, while distinguishing PLN from PP homing by differential requirement for a4P7 (114). The HEVassociated chemokine SLC/GCkine (not illustrated in Fig. 8) may be an important adhesion-triggering chemokine for lymph nodes as well as for PP-HEVs: if so, the adhesion-triggering step may contribute little to differences in lymphocyte homing to these organized lymphoid tissues. (Instead, it may distinguish these lymphoid tissues from tissue inflammatory sites, which preferentially recruit memory vs. naive cells.) A unique CLA+a4P7memory lymphocyte population homes to juxtaepidermal sites in the skin, in a process thought to involve sequential CLA-E-selectin tethering and rolling, a variable requirement for a4P1-VCAM-1, and LFA-1 (22, 159). [The involvement of a4/31-VCAM-1 in skin homing may depend on the levels of CLA expression: circulating CLA'"' T cells are a4Pl' and may require VCAM-1, but many CLAbnghtmemory cells are a 4 0 l - (L. Picker, personal communication.)] a4P1-VCAM-1 and LFA-1 are thought to mediate lymphocyte homing to inflamed CNS, with no apparent selectin requirement (160, 161). Finally, the mechanisms of lymphocyte homing to inflamed joints, and to pulmonary inflammatory sites, although not as well defined yet, appear to involve distinctive decision cascades as well (54-56; reviewed in Refs. 21 and 22), thus reinforcing the segregation of intestinal immune responses.

D. ALTERATIONS IN LYMPHOCYTE RECRUITMENT DURING INTESTINAL INFLAMMATION Immunohistologic studies as well as limited physiologic analyses of lymphocyte trafficking indicate that gut inflammationinduces dramatic changes in the extent of lymphocyte recruitment to the gut wall, and also in the selectivity of recruitment. Although peripheral lymph node immunoblasts, which traffic well to the inflamed skin, are largely excluded from shortterm homing into the normal intestinal wall (4, 7, 9, 12-14, 162, 163),

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY Lymphocytes

CiABlhi LFA-~++

87-

L-selectP

Endotheliel cells

-

ecrivorion

ContBcW rolllng

non-mucosal inflamed 77 I memory cells or CNS, heart, VCAM-1 blasts other sites? ICAM'~ I

I

arrest

23.5 * dlapsdarls

-

FIG 8. A schematic summary of the adhesiodactivatioddecision cascades implicated in lymphocyte trafficking. The schematic emphasizes the sequential but overlapping roles of L-selectin, a407, and LFA-1 that allow efficient naive lymphocyte homing through Peyer's patch high endothelial venules (PP-HEVs), and the potentially self-sufficient involvement 7~' homing to PPs and to lamina propria venules of a407 in preactivated ~ ~ 4 0lymphoblast of the intestinal wall. To underscore the unique decision processes involved in these guthoming events, the algorithms involved in selecting lymphocytes into nonmucosal peripheral lymph nodes, and hypothesized to control homing of skin-associatedlymphocytes to cutaneous and other inflammatory sites, are also presented schematically. Quantitative and qualitative regulation of receptors are critical to the degree and specificity of recruitment. Overlap of bars in the figure emphasizes overlapping functions in particular settings, and the vertical widths of bars (and font sizes) reflect the relative expression level and functional importance of each component in the cascade, which may of course be variable. For example, VCAM1 may be required for homing of some but not all CLA+ T cells to inflamed skin (see text). SLC/GChne is a strong candidate as an adhesion-triggering chemokine involved in arrest of T cells in PP- and PLN-HEVs (not shown). Physiologic activating signals in intestinal lamina propria (and in other extralymphoid sites of inflammation) remain to be identified, and may not be required for arrest of immunoblasts expressing preactivated integrins. Questionable or potentially variable involvement of unidentified pathways, or pathways whose in situ contributions remain to be confirmed (e.g., VAP-1 in joints and inflamed lymph nodes), are indicated by question marks. It should be emphasized that tissue-selective cascades can be altered both developmentally and as a function of the severity and nature of local inflammatory stimuli (see text).

experimental inflammation of the intestine allows access to (although not retention of) activated LN lymphocytes (9, 164). This effect is probably largely mediated at the level of the vascular endothelium (41, 90, 101). In general, severe inflammation tends to induce additional recruitment mechanisms such as enhanced or de no00 PNAdL-selectin ligand expres-

236

EUGENE C. BUTCHER et QZ.

sion in the gut, as discussed above, as well as superinduction of MAdCAM1 and VAP-1. Up-regulation of chemokines and other activatinghhemoattractant factors is likely to contribute as well, so that the net effect of severe chronic inflammation is to induce the local recruitment of more lymphocytes, and of a broader range of lymphocyte subsets as well (e.g., naive cells). To date, however, there is no evidence of loss of intestinal exclusion of memory lymphocyte subsets targeted to other tissues: for example, VCAM-1, a key mediator of vascular recruitment of a4P7(a4Plh',nonintestinal) memory lymphocyte subsets, is rarely if ever observed on venules involved in lymphocyte recruitment into the intestinal wall, even during inflammation ( 2 1 ) . The extent and characteristics of alterations in lymphocyte recruitment depend, of course, on the intensity and nature of the inflammatory stimulus. In inflammatory bowel disease, for example, recruitment of T cells appears to remain largely selective for a4P7"j cells and memory phenotypes, whereas in severe colonic inflammation induced by repeated oral administration of dextran sulfate solution, there is massive recruitment even of naive lymphocytes (D. Picarella, E. C. Butcher, and D. Ringler, personal observations) in association with the development of lymphoid tissue architecture. Nonetheless, as implied above, even in this setting (as in PPs) recruitment remains in large part dependent on a4P7 and MAdCAM-1, so that a4P7- circulating memory cells (e.g., CLA+ skin homing lymphocytes) would remain excluded. Although expression of additional vascular ligands and chemokines undoubtedly contributes to a broader range of lymphocyte recruitment in the setting of inflammation, it is also possible that lymphocytes initially responding to antigen display less selective (more promiscuous) patterns of homing receptor expression than do the majority of circulating blood memory and effector lymphocytes. [Remember in this regard the ability of immunoblasts from PLNs to home to inflamed intestines (9, 164).] Picker and colleagues (165, 166) have demonstrated that activated CD4+ T cells undergoing the naive to memory transition in different lymphoid organs in humans (tonsils, appendix, peripheral lymph nodes) display characteristic differences in the frequency of expression of appropriate homing receptors, but that in spite of this, many cells undergoing the transition display overlapping homing receptor patterns that might be expected to allow them to migrate to multiple tissue sites, such as the skin and the gastrointestinal tract. This probably represents a transient phenotype, however, and it is not yet known whether such cells are released into the blood prior to more complete specialization. [In this context, it is worth emphasizing that studies of the trafficking of immunoblasts taken from lymphoid organs, instead of from lymph or from blood, must be interpreted with caution.] However, if such immunoblasts do enter the lymph and

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

237

blood, they may well have a greater capacity for general dissemination than fully developed homing receptor-polarized recirculating memory cells.

E. DISTINCT TRAFFICKING PATTERNS IN THE MUSCULARIS AND SEROSA It should be emphasized that, in humans at least, the rnuscularis propria and serosa appear to be immunologically distinct from the intestinal mucosa. MAdCAM-1 can be observed on submucosal venules, especially those adjacent to the lamina propria, but the muscularis, especially the muscularis propria and serosa, should not be considered part of the intestinal trafficking network. Indeed, in both the intestinal wall, and in the skin, the unique features of the local immune responses (including the expression of specialized tissue-selective trafficking molecules-vascular addressins and specific homing receptor patterns on infiltrating lymphocytes) appear to be a function of proximity to the epithelial surface. Whether the connective tissues of the extramucosal intestinal wall, and by analogy the deep tissues of the dermis, display their own selective recruitment mechanisms or instead are part of a common systemic connective tissue trafficking network remains to be studied. V. a4p7 and the Segregation of Intestinal from Systemic Memory

A. EXPRESSION OF a4p7 Is CHARACTERISTIC OF ANTIGENS MEMORY CELLSFOR INTESTINAL Central to models of the physiologx relevance of tissue-selective lymphocyte homing is the hypothesis that memory and effector cells targeted to intestinal or other sites in fact embody immunity to antigens encountered in those sites. This intuitive paradigm received early support from studies demonstrating that specific antibody-producing cells (plasmablasts), containing antibody to antigens presented experimentally in the intestines, in fact home efficiently to the intestinal wall (19, 158, 167). This selective recruitment from the blood into the intestinal lamina propria is independent of the presence of specific antigen. (Subsequent retention, survival, and proliferation of antigen-specific cells is dramatically influenced by cognate antigen; but initial recruitment is not.) Critical evaluation of the paradigm, however, has only become possible with the advent of monoclonal antibodies allowing isolation of selectively homing memory and effector lymphocyte subsets. Antibodies to the skin lymphocyte homing receptor CLA, for example, allowed isolation of skin-homing CD4' memory T cells from human blood: these cells were dramatically enriched in cells responsive to cutaneous contact allergens (159, 168). Similarly, antibodies to a4P7 have allowed functional characterization of memory responses of gut-homing T and B cell populations. For example, in studies of circulating

238

EUGENE C. BUTCHER et al.

B cell responses to experimentally administered antigens, lymphocytes capable of secreting antigen-specific antibodies were enriched by positive selection with anti-a4/37 following oral and nasal but not systemic or cutaneous inoculation (169, 1 70). One of the most physiologically relevant models of mucosal immunity studied is the response to rotavirus infection. Rotavirus, a common cause of diarrhea in infants, is a rheovirus that almost exclusively infects and proliferates in intestinal epithelial cells, thus representing a natural intestine-selective immunogen. In humans, blood-borne CD4+T cell memory to rotavirus in naturally exposed individuals is dramatically enriched in the a4P7himemory (CD45RO') subset (Fig. 9A) (171). In a mouse model of chronic rotavirus infection in immunodeficient hosts, transfer of immune CD8+ cells confers partial immunity and clears virus from the recipient (172),and the CD8' T cells capable of clearing virus are predominantly c ~ 4 P 7(Fig. ~ 9B) (173).a4P7- memory cells as well as naive phenotype CD8' cells are, by comparison, largely ineffective, even after prolonged periods in the infected host, during which alterations in homing receptor phenotype can potentially occur. Finally, intestinal humoral immunity also segregates with homing receptor expression: transfer of a4P7hi IgD- presumptive memory B cells from the spleen of immune donors can confer rotavirus-specific secretory IgA responses associated with viral clearance and protective immunity (see Fig. 9C). Interestingly, this a4P7' memory is long-lived: ~ 4 P 7B~ cells ' from donors orally challenged with rotavirus 7 months previously were still effective (see Fig. 9 legend). a4P7hi B cells also transfer significant serum IgG responses. In contrast, a4P7memory B cells can confer rotavirus-specific serum IgG in adoptive recipients only when transferred early (1-2 months) but not late (7 months) after last rotavirus exposure, and they are unable to generate detectable rotavirus-specific IgA or to provide rotavirus immunity (174).[The ability of a4P7- splenic B cells taken early after immunization to generate an adoptive IgG response may represent an initial physiologic systemic response to mucosal viral proliferation, or to viral antigen disseminated to the spleen. Alternatively, these splenocytes may represent germinal center cells (which are a4pl+P7- even in mucosal lymphoid organs) rather than mature, migration-competent cells.] Importantly, preliminary flow cytometric studies of the phenotype of rotavirus-specific B cells (identified by direct binding of fluoresceinated rotavirus antigens) are consistent with and complement these functional studies: during the initial rotavirus response, many rotavirus antigen-binding B cells within Peyer's patches, mesenteric lymph nodes, and spleen are a4/37-, often displaying the phenotype of germinal center cells (K. Youngman, M. Franco, H. B. Greenberg, and E. C. Butcher, unpublished observations). The frequency of

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

A

naive

Circulating CD4 T cell immunity

p! 1

' &p;Bg naive

239

2

P c 0.m vs hisub801

,

3

4

I

I

5

6

c

lntestinaf IgA response

In vitm stimulation index (rotavirus response) CD8-mediated immunity (in vivo)

1 0

Days after cell transfer none

non; 20

I

I

I

,

40

60

80

100

% Recipients deared of rotavirus

FIG.9. a4P7"' but not a4P7- lymphocytes confer intestinal immunity in an infectious model. Rotavirus selectively infects intestinal enterocytes. (A) In humans, the CD4+ T cell proliferative response to rotavirus following natural exposure is enriched among a4P7'" memory cells (171). (B) CD8 immunity to rotavirus is enriched in the a4/37h' memory subset. In a mouse model of rotavirus infection, 10,000 sorted lymphocytes from previously immunized (orally infected) C57BU6 donor mice were resuspended in sterile saline and injected intraperitoneally into immunodeficient recipient RAG-2 mice chronically infected with rotavirus. Viral shedding was monitored by enzyme-linked immunoassay (ELISA) of stool samples (173).By day 20 (illustrated), virus was no longer detectable in recipients of ~24P7~' CD8+ immune cells, whereas rr4p7- memory CD44'"-phenotype CD8' cells or naive phenotype (CD44'") CD8+ cells were ineffective ( N = 5-6 mice per group). ( C ) Humoral (secretory) immunity to rotavirus is conferred by c14/37~' memory cells (174);10,000 sorted IgD' (predominantly naive), or a4P7'" or a4P7- IgD- (predominantly memory) B220+ splenocytes from rotavirus-challenged (immunized) C57BU6 donor mice were injected intraperitoneally (along with immune CD4+ helper cells) into chronicallyinfected recipients as above, and levels of anti-rotavirus IgA were monitored by ELISA of stool samples. Only a4/37h'presumptive memory ( IgD-) B cells cleared recipients of virus (not shown), and transferred secretory IgA responses to intestinal rotavirus. a4P7' memory to rotavirus is long-lived: in the experiment illustrated, sorted cells were from donors orally challenged with rotavirus 7 months previously. Similar results were obtained at earlier timepoints (1-2 months after last rotavirus exposure).

rotavirus-specific B cells increases over time, and long-lived antigenbinding memory cells, which are enriched in PPs, are predominantly a4P7hi nine months after rotavirus exposure. Together, these results indicate that a4/37hilymphocytes compose the effective circulating pool for intestinal immunity. This undoubtedly reflects

240

EUGENE C . BUTCHER et d.

both their ability to traffic efficiently to target sites in the intestine, and the fact that they are dramatically enriched in memory cells specific for intestinal antigens. These studies thus provide strong experimental support for the paradigm of tissue-selective traffickinghoming receptor expression by lymphocyte populations specific for regionally presented anti ens. The extent of such enrichment of intestinal memory among the a4P7 subsets,

5

and of exclusion of intestinal antigen-reactive cells from circulating a4P7subpopulations, may depend on the nature as well as, certainly, the tissue selectivity of the eliciting stimulus; and may vary during the evolution of immune responses. It is worth discussing another possible contributing factor to the dominance of cu4/37' memory in intestinal immune responses. Preliminary studies suggest a correlation (albeit imprecise) of a4P7 with expression of other functional attributes of intestinal immune responses: for example, flow cytometric analyses of human blood B cells demonstrate that approximately two-thirds of surface IgA' memory B cells are a4P7' whereas only onethird of circulating surface IgG' B cells express a4P7 (L. Rott and E. C. Butcher, unpublished observations). In the normal evolution of the intestinal immune response, therefore, up-regulation or selection for a4P7 expression, and for recirculation through intestines, may be induced coordinately with other specialized properties of memory cells that would facilitate or support their contribution to mucosal immunity. Our bias, however, is that such coordinate regulation is probably not hard-wired, but rather may reflect parallel microenvironmental exposure to combinations of factors controlling homing receptor expression, and other factors controlling isotype expression or other functional attributes. B. TISSUE-SELECTIVE RECIRCULATION AS A FUNCTION (AND REGULATOR) OF MEMORY CELLHOMEOSTASIS As discussed above, during the initial immune system frenzy associated

with the effort to initiate an effective primary response, or to clear particularly serious resurgent insults, newly generated antigen-specific lymphocytes may initially be heterogeneous in their patterns of homing receptor expression, and in their physiologic homing properties. On the other hand, tissue-selective lymphocyte homing and segregation are prominent during the established immune response and especially during surveillance by memory cells. Current paradigms suggest that immunologic memory is retained on a competitive basis (22):clones of T cells specific for a given antigen must fight for survival with clones of other specificities (or of different avidities)through competition for supportive microenvironmental niches defined in part by residual (retained) antigen. In this model, memory cells that circulate preferentially through tissue sites of antigen retention

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

24 1

are expected to have a significant survival advantage. For example, memory T cells for intestinal antigens displaying gut-specific homing would encounter more supportive survival conditions during their passage through intestinal tissues and mucosal lymphoid organs, whereas similar memory cells trafficking selectively through cutaneous sites would be expected to compete poorly with T cells specific for skin antigens. In this model, the longterm maintenance of tissue-specific homing properties on antigen-specific memory cells may be a function of antigen-mediated homeostatic mechanisms operating at the level of cellular competition. This Darwinian system of immunologic “survival of the fittest” would select for tissue-preferential homing properties on long-term memory cells, while at the same time enforcing enhanced efficiency of immune surveillance and ensuring effective anamnestic responses in the target tissues. VI. Clinical Significance and Therapeutic Opportunities

A. a4P7 ASSOCIATION WITH INTESTINAL METASTASIS OF LYMPHOID MALIGNANCIES Recent studies of B and T lineage non-Hodgkin’s lymphomas suggest that expression of a4P7 is also a characteristic of neoplastic lymphocytes that disseminate to the intestines (175, 176). In a particularly interesting example, primary neoplastic gastric mucosa-associated lymphoid tissue (MALT) lymphomas associated with Helicobacter pylori-induced gastritis are predominantly L-~electin’”a4/37’”~ ( I 76). However, cells in intestinal metastases are characteristically L-selectiniaWtu4/37‘”, suggesting that 4 / 3 7 expression may be important to the intestinal dissemination, and that these tumors may undergo a developmental sequence paralleling that of antigenstimulated B cells contributing to intestinal immune responses.

B. IMPLICATIONS FOR VACCINE DEVELOPMENT The studies outlined above suggest that cw4/37 expression is required for efficient lymphocyte entry into intestinal sites, and thus may be essential for efficient participation in intestinal immunity. It follows that vaccination protocols that induce immunity among c~4/37~’ T and B cell subsets may be most effective at providing intestinal immunity. Antigen-specific memory T cells can be identified by flow cytometry (177):determination of their a4/37 levels might serve as a clinically significant indicator of the efficiency of intestinal protection expected from a given vaccination protocol. Moreover, although administering vaccines through oral immunization is a reasonable and powerful approach to inducing intestinal immunity, oral administration may not always be feasible. In this context, the ability of cytokines to regulate homing receptor expression in vitro (159) suggests

242

EUGENE C. BUTCHER et al.

that it may eventually be possible to manipulate specific homing receptors (i.e.,including a4P7 expression) even during intravascular or subcutaneous vaccination. The value of such an approach will of course depend on the importance of preexisting trafficking pathways to anamnestic responses on antigen reexposure, and on the stability or malleability of homing receptor expression by antigen-specific subsets, issues that remain to be explored experimentally. C. BLOCKADE OF a4P7-MADCAM-1 INTERACTION: THERAPEUTIC IMPLICATIONS FOR AUTOIMMUNE AND INFLAMMATORY DISEASES OF THE GASTROINTESTINAL TRACT The discovery of tissue-selective trafficking mechanisms offers the potential to regulate inflammation in a selective manner by interfering with regional lymphocyte homing. The critical association of a4P7 expression with memory and effector cells responsible for intestinal immunity, and the preferential gut lymphoid stromal cell expression of MAdCAM-1, suggest that interfering with a4P7-MAdCAM-1 interaction in particular could modulate intestinal immune and inflammatory responses at multiple levels, including but not limited to blockade of lymphocyte recruitment from the blood. In fact, recent studies have confirmed that interfering with a4P7MAdCAM-1 interactions not only inhibits lymphocyte recruitment and intestinal inflammation, but can also dramatically and rapidly improve clinical symptoms in animal models of inflammatory bowel disease (101, 178) (Fig. 10).Antibodies to MAdCAM-1 and to a4P7 also decrease the incidence and severity of islet inflammation and of insulin deficiency in the nonobese diabetic mouse (42). Inhibitors of a4P7-MAdCAM-1 inter-

GI 0

Diarrhea 2.0 Semi-solid 1. O i

b

Solid 0.0 t- Rx-

1 1

0

10 Time (days)

20

FIG.10. Therapeutic effects of antibody to a4P7 in the cotton-top tamarin model of inflammatory bowel disease. Administration of a4P7 ( 0 )but not of control antibody (0) induces rapid, sustained improvement in stool consistency. Antibody treatment was from day 0 to day 7, as indicated. (Adapted with permission from Ref. 178.)

LYMPHOCYTE TRAFFICKING A N D REGIONAL IMMUNITY

243

actions, and indeed of other molecular mediators of intestinal lymphocyte recruitment, thus have exciting potential as therapeutic agents in inflammatory and autoimmune diseases of the gastrointestinal tract. ACKNOWLEDGMENTS Thanks to the many laboratory members who have contributed to our efforts to understand regional lymphocyte trafficking. Special thanks to L. J. Picker and S. Jalkanen for critical reading of the manuscript, S. Grossman for administrative and secretarial assistance, and J. Jang, J. Twelves, and B. Rouse for helping produce many of the antibody reagents to mucosal addressins and homing receptors. Studies from our laboratory were supported in part by NIH Grants AI37832, GM37734, AI37319, A119957, and HL57492; by an award from the Department of Veterans Affairs; by the Stanford Digestive Disease Center under DK38707; and by fellowship support from the Arthritis Foundation (MW); and NIH Immunology Training Grant CA09302-19 (KY).

REFERENCES 1. Gowans, J., and Knight, E. (1964). The route of recirculation of lymphocytes in the rat. Proc. A. SOC.London Ser. B . 159, 257. 2. Sprent, J. (1973).Circulating T and B lymphocytes of the mouse. I. Migratory properties. Cell. Immunol. 7, 10. 3. Sprent, J. (1977). Recirculating lymphocytes. In “The Lymphocyte: Structure and Function” (J. Marchalonis, ed.), Chap. 2. Dekker, Inc., New York. 4. Hall, J., Hopkins, J., and Orland, E. (1979). Studies on the lymphocytes of sheep. 111. Destination of lymph-borne immunoblastsin relation to their tissue of origin. Eur. J. lmmunol. 7,30-37. 5. Mackay, C. R. (1993). Immunological memory. Adu. lmmunol. 53, 217-265. 6. Mackay, C. R. (1993). Homing of naive, memory, and effector lymphocytes. C u r . @in. Immunol. 5, 423-427. 7. Smith, M., Martin, A., and Ford, W. (1980). Migration of lymphoblasts in the rat: Preferential localization of DNA-synthesizing lymphocytes in particular lymph nodes and other sites. Monogr. Allergy 16, 203-232. 8. Parrot, D., and Ferguson, A. (1974). Selective migration of lymphocytes within the mouse small intestine. lmmunology 26, 571-588. 9. Rose, M., Parrot, D., and Bruce, R. (1976). Migration of lymphoblasts to the small intestine. 11. Divergent migration of mesenteric and peripheral immunoblasts to sites of inflammation in the mouse. Cell Immunol. 27, 36-46. 10. Chin, W., and Hay, J. (1980).A comparisonof lymphocyte migration through intestinal lymph nodes, subcutaneous lymph nodes, and chronic inflammatory sites of sheep. Gastroenterology 79, 1231-1242. 11. Cahill, R., Heron, I., Poskitt, D., Frost, H., and Tmka, A. (1980).Lymphocyte recirculation in the sheep fetus. Ciba Found. 71, 145-166. 12. Guy-Grand, D., Griscelli, C., and VassiUi, P. (1974). The gut-associated lymphoid system: Nature and properties of the large dividing cells. Eur.J. lmmunol. 4,435-443. 13. McDermott, M., and Bienenstock,J. (1979).Evidence for a common mucosd immunologic system. I. Migration of B immunoblasts into intestinal, respiratory, and genital tissues. J. Immunol. 122, 1892-1898. 14. Griscelli, C., Vassdi, P., and McCluskey, R. (1969).The distribution of large dividing lymph node cells in syngeneic recipient rats after intravenous injection. J. Ezp. Med. 130,1427-1451.

244

EUGENE C. BUTCHER et (11.

15. Gabor, M., Godfrey, D., and Scollay, R. (1997). Recent thymic emigrants are distinct from most medullary thymocytes. Eur. /. Immunol. 27, 2010-2015. 16. Cahill, R., Poskitt, D., Frost, H., and Tmka, A. (1977).Two distinct pools of recirculating T lymphocytes: Migratory characteristics of nodal and intestinal T lymphocytes.]. Exp. Med. 145,420-428. 17. Sprent, J., and Tough, D. (1994). Lymphocyte life-span and memory. Science 265, 1395- 1400. 18. Salmi, M., Granfors, K., Leirisalo-Repo, M., Hamalainen, M., MacDermott, R., Leino, R., Havia, T., and Jdkanen, S. (1992). Selective endothelial binding of interleukin-2dependent human T cell lines derived from different tissues. Proc. Nutl. Acad. Sci. U.S.A. 89, 11436-11440. 19. Butcher, E. C. (1986). The regulation of lymphocyte traffic. Cum Top. Microbiol. Immunol. 128,85-122. 20. Picker, L. J,, and Butcher, E. C. (1992) Physiological and molecular mechanisms of lymphocyte homing. Annu. Rev. Immunol. 10,561-591. 21. Salmi, M., and Jalkanen, S. (1997). How do lymphocytes know where to go: Current concepts and enigmas of lymphocyte homing. Adv. Immunol. 64, 139-218. 22. Butcher, E. C., and Picker, L. J. (1996).Lymphocyte homing and homeostasis. Science 272,60-66. 23. Butcher, E. C., Scollay, R. G., and Weissman, I. L. (1980). Organ specificity of lymphocyte migration: Mediation by highly selectivelymphocyteinteraction with organspecific determinants on high endothelial venules. Eur. ]. Immunol. 10, 556. 24. Jalkanen, S., Steere, A. C., Fox, R. I., and Butcher, E. C. (1986).A distinct endothelial cell recognition system that controls lymphocyte traffic into inflamed synovium. Science 233,556-558. 25. Imhof, B. A., and Dunon, D. (1995).Leukocyte migration and adhesion.Adu. Immunol. 58,345-416. 26. Kraal, G., and Mebius, R. E. (1997). High endothelial venules: Lymphocyte traffic control and controlled traffic. Adv. Immunol. 65, 347-395. 27. Girard, J. P., and Springer, T. A. (1995).High endothelial venules (HEVs): Specialized endothelium for lymphocyte migration. Immunol. Todzy 16,449-457. 28. Jeurissen, S., Duijvestijn, A., Sontag, Y., and Kraal, G. (1987). Lymphocyte migration into the lamina propria of the gut is mediated by specialized HEV-like blood vessels. immunology 62,273-277. 29. Streeter, P., Berg, E., Rouse, B., Bargatze, F., and Butcher, C. (1988).A tissue-specific endothelial cell molecule involved in lymphocyte homing. Nature 331, 41-46. 30. Weisz-Carrington, P., Emancipator, S., and Keleman, P. (1991). Specific attachment of mesenteric IgA lymphoblasts to specialized endothelium of specialized endothelium of intestinal mucosa lamina propria capillaries. Cell Immunol. 132, 494-504. 31. Butcher, E. C. (1991). Leukocyte-endothelid cell recognition: Three (or more) steps to specificity and diversity. Cell 67, 1033-1036. 32. Shimizu, Y., Newman, W., Tanaka, Y., and Shaw, S. (1992). Lymphocyte interactions with endothelial cells. Immunol. Today 13, 106-112. 33. Campbell, J., Hedrick, J., Zlotnik, A., Siani, M., Thompson, D., and Butcher, E. (1998). Chemokines and the arrest of lymphocytes rolling under flow conditions. Science 279, 381-384. 34. Gum, M., Tangemann, K., Tam, C., Cyster, J., and Rosen, S. (1998). A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 95, 258-263.

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

245

35. Holzmann, B., McIntyre, B., and Weissman, I. (1989). Identification of a murine Peyer’s patch-specific lymphocyte homing receptor as an integrin molecule with an (Y chain homologous to human VLA-4a. Cell 56, 37-46. 36. Hamann, A., Andrew, D. P., Jablonski-Westrich, D., Holzmann, B., and Butcher, E. C. (1994). Role of cr4-integrins in lymphocyte homing to mucosal tissues in uioo. 1. Immunol. 152, 3282-3293. 37. Williams, M. B., and Butcher, E. C. (1997). Selective homing of naive and memory T lymphocyte subsets to Peyer’s patches, lymph nodes, and spleen. 1.Immunol. 59, 1746-1752. 38. Wagner, N., Lolder, J., Kunkel, E., Ley, K., Leiing, E., Krissansen, G., Rajewsky, K., and Muller, W. (1996). Critical role for p7 integrins in formation of the gut-associated lymphoid tissue. Nature 382, 366-370. 39. Briskin, M. J., McEvoy, L. M., and Butcher, E. C. (1993). MAdCAM-1has homology to immunoglobulin and mucin-like adhesion receptors and to IgAl. Nature 363,461-464. 40. Shyjan, A., Bertagnolli, M., Kenney, C., and Briskin, M. (1996). Human mucosal addressin cell adhesion molecule-1 (MAdCAM-I) demonstrates structural and functional similarities to the (~4p7-integrinbinding domains of murine MAdCAM-1, but extreme divergence of mucin-like sequences. I. Inimunol. 156, 285-2857, 41. Briskin, M., Winsor-Hines, D., Shyjan, A,, Cochran, N., Bloom, S., Wilson, J., McEvoy, L. M., Butcher, E. C., Kassam, N., Mackay, C. R., Newman, W., and Ringler, D. J. (1997). Human mucosal addressin cell adhesion molecule-1 is preferentially expressed in intestinal tract and associated lymphoid tissue. Am. J. Puthol. 151, 97-110. 42. Yang, X., Sytwu, H.-K., McDevitt, H. O., and Michie, S. A. (1997). Involvement of p7 integrin and mucosal addressin cell adhesion molecule-1 (MAdCAM-1) in the development of diabetes in nonobese diabetic mice. Diabetes 46, 1542-1547. 43. Hanninen, A., Taylor, C., Streeter, P., Stark, L., Sarte, J., Shizuru, J., Simell, O., and Michie, S. (1993). Vascular addressins are induced on islet vessels during insulitis in nonobese diabetic mice and are involved in lymphoid cell binding to islet endothelium. 1. Clin. Invest. 92, 2509-2515. 44. San Gabriel-Mason, C. (1992). Adhesion of lymphocytes to the lactating mammary gland in the mouse. Ph.D. Thesis. The Pennsylvania State University, Philadelphia, Pennsylvania. 45. Michie, S., Streeter, P., Butcher, E., and Rouse R. (1995).L-Selectin and a4b7 integrin homing receptor pathways mediate peripheral lymphocyte traffic to AKR mouse hyperplastic thymus. Am. ]. Puthol. 147, 412-421. 46. Thomas, P., Spencer, J., Neluferk, J., Brettman, L., Briskin, M., Ringler, D., and Ciclitara, P. (1998). Increased expression of MAdCAM-1, a receptor for lymphocytes and gut homing potential, in the mucosa of patients with coeliac disease. (in preparation) 47. Kruse, A,, Merchant, M. J., Hallmann, R., and Butcher, E. C. (1998). Specialized leukocyte-vascular homing interactions at the matemaVfetal interface: Potential role in inaternaufetal tolerance. (submitted) 48. Kraal, G., Schornagel, K., Streeter, P. R., Holzmann, B., and Butcher, E. C. (1995). Expression of the mucosal vascular addressin, MAdCAM-1, on sinus-lining cells in the spleen. Am. 1.Puthol. 147, 763-771. 49. Tanaka, H., Hataba, Y., Saito, S., Fukushima, O., and Miyasaka, M. (1996). Phenotypic characteristics and significance of reticular meshwork surrounding splenic white pulp of mice. J. Electron Microsc. (Jpn.) 45, 407-416. SO. Steffen, B., Breier, G., Butcher, E., Schulz, M., and Engelhardt, B. (1996). ICAM1, VCAM-1 and MAdCAM-1 are expressed on choroid plexus epithelium but not

246

EUGENE C. BUTCHER et al.

endothelium and mediate binding of lymphocytes in vitro. Am. 1.Puthol. 148, 18191838. 51. Sikorski, E. E., Hdmann, R., Berg, E. L., and Butcher, E. C. (1993). The Peyer’s patch high endothelial receptor for lymphocytes, the mucosal vascular addressin, is induced on a murine endothelial cell line by tumor necrosis factor-alpha and IL-1. J. Immunol, 151,5239-5250. 52. Schaible, U. E., Vestweber, D., Butcher, E. C., Stehle, T., and Simon, M. M. (1994). Expression of endothelial cell adhesion molecules in joints and heart during Borrelia burgdorferi infection of mice. Cell Adhesion Commun. 2, 465-479. 53. O’Neill, J. K., Butter, C., Baker, D., Gschmeissner, S. E., Kraal, G., Butcher, E. C., and Turk, J. L. (1991). Expression of vascular addressins and ICAM-1 by endothelial cells in the spinal cord during chronic relapsing experimental allergic encephalomyelitis in the Biozzi AB/H mouse. Immunology 72, 520-525. 54. Abitorabi, M. A., Mackay, C., Jerome, E. H., Osorio, O., Butcher, E. C., and Erle, D. J. (1996). Differential expression of homing molecules on recirculating lymphocytes from sheep gut, peripheral, and lung lymph. J. Immunol. 156,3111-3117. 55. Mackay, C. R., Andrew, D. F., Briskin, M., Ringler, D. J., and Butcher, E. C. (1996). Preferential, and migration properties of three major subsets of tissue homing T cells in sheep. Eur. J. Immunol. 26,2433-2439. 56. Picker, L. J., Martin, R. J., Trumble, A., Newman, L. S., Collins, P. A., Bergstresser, P. R., and Leung, D. Y. (1994). Differential expression of lymphocytehoming receptors by human memory/effector T cells in pulmonary vs. cutaneous immune effector sites. Eur. 1.Immunol. 24,1269-1277. 57. McGhee, J., Mestecky, J., Dertzbaugh, M., Eldridge, J., Hirasawa, M., and Kiyono, H. (1992). The mucosal immune system: From fundamental concepts to vaccine development. Vaccine 10, 75-88. 58. Kelly, K., and Rank, R. (1997). Identification of homing receptors that mediate the recruitment of CD4 T c e h to the genital tract following intravaginal infection with Chlamydia trachomatis. Infect. Immun. 65, 5198-5208. 59. Perry, L., Feilzer, K., Portis, J., and Caldwell, H. (1998). Distinct homing pathways direct T lymphocytes to the genital and intestinal mucosae in Chlamydia-infected mice. 1.Immunol. 160,2905-2914. 60. Mebius, R., Streeter, P., Michie, S., Butcher, E., and Weissman, I. (1996).A developmental switch in lymphocyte homing receptor and endothelial vascular addressin expression regulates lymphocyte homing and permits CD4+3- cells to colonize lymph nodes. Proc. Nutl. Acud. Sci. U.S.A. 93, 11019-11024. 61. Mebius, R. E., Rennert, P., and Weissman, I. L. (1997) Developing lymph nodes collect CD4+CD3- LTB+ cells that can differentiate to APC, NK cells, and follicular cells but not T or B cells. Immunity 7, 493-504. 62. Weissman, I. (1994).Developmental switchesin the immune system. Cell 76,207-218. 63. Mebius, R., Streeter, P., Brev6, J., Duijvestijn, A. M., and Kraal, G. (1991). The influence of afferent lymphatic vessel interruption on vascular addressin expression. /. Cell Biol. 115, 85-95. 64. Szabo, M. C., Butcher, E. C., and McEvoy, L. M. (1997). Specialization of mucosal follicular dendritic cells revealed by mucosal addressin-cell adhesion molecule-1 display. J. Immunol. 158, 5584-5588. 65. Sampaio, S., Li, X., Takeuchi, M., Mei, C., Francke, U., Butcher, E., and Briskin, M. (1995). Organization, regulatory sequences, and alternatively spliced transcripts of the mucosal addressin cell adhesion molecule-1 (MAdCAM-1) gene. J. Immunol. 155,2477-2486.

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

247

66. Leung, E., Greene, J.. Ni, J., Raymond, L., Lehnert, K., Langley, R., and Krissansen, G . (1996).Cloning of the mucosal addressin MAdCAM-1 from human brain: Identification of novel alternatively spliced transcripts. Immunol. Cell Biol. 74, 490-496. 67. Briskin, M. J., Rott, L., and Butcher, E. C. (1996). Structural requirements for mucosal vascular addressin binding to its lymphocyte receptor a4b7: Common themes among integrin-Ig family interactions. J. Immunol. 156, 719-726. 68. Tan, K., Cassasnovas, J.. Liu, J.-H., Briskin, M., Springer, T., and Wang, J. (1998). The structure of immunoglobulin domains 1 and 2 of MAdCAM-1 reveals novel features important for integrin recognition. Structure 6, 793-801. 69. Newham, P., Craig, S. E., Seddon, G. N., Schofield, N. R., Rees, A., Edwards, R. M., Jones, E. Y., and Humphries, M. J. (1997). a 4 integrin binding interfaces on VCAM1 and MAdCAM-1. Integrin binding footprints identify accessory binding sites that play a role in integrin specificity. J. Biol. Chem. 272, 19429-19440. 70. Viney, J., Jones, S., Chiu, H., Lagrimas, B., Renz, M., Presta, L., Jackson, D., Hillan, K., Lew, S., and Fong, S. (1996). Mucosal addressin cell adhesion molecule-1: A structural and functional analysis demarcates the integrin binding motif. 1.Immunol. 157,2488-2497. 71. Harpaz, Y., and Chothia, C. (1994). Many of the immunoglobulin super family domains in cell adhesion molecules and surface receptors belong to a new structural set which i s close to that containing variable domains. J. Mol. Biol. 238, 528-539. 72. Shroff, H., Schwender, C., Dottavio, D., Yang, L.-L., and Briskin, M. (1996). Small peptide inhibitors of a4p7-mediated MAdCAM-1 adhesion to lymphocytes. Biorg. Med. Chem. 6,2495-2500. 73. Wang, J.-H., Pepinsky, R., Steele, T., Liu, J.-H., Karpusas, M., Browning, B., and Oshorn, L. (1995). The crystal structure of an N-terminal two-domain fragment of vascular cell adhesion molecule 1 (VCAM-1): A cyclic peptide based on the domain 1C-D loop can inhibit VCAM-l-a4 interaction. Proc. Nutl. Acud. Sci. U.S.A.92,57145718. 74. Schweighoffer, T., Tanaka, Y., Tidswell, M., Erle, D. J., Horgan, K. J., Luce, G. E., Lazarovits, A. I., Buck, D., and Shaw, S. (1993). Selective expression of integrin a4p7 on a subset of human CD4’ memory T cells with Hallmarks of gut-trophism. J. Immunol. 151,717-729. 75. Berg, E. L., McEvoy, L. M., Berlin, C., Bargatze, R. F., and Butcher, E. C. (1993). L-Selectin-mediated lymphocyte rolling on MAdCAM-1. Nature 366, 695-698. 76. Bargatze, R. F., Jutila, M. A,, and Butcher, E. C. (1995). Distinct roles of L-selectin and integrins a4p7 and LFA-1 in lymphocyte homing to Peyeis patch-HEV in situ: The multi-step model confirmed and refined. Immunity 3, 99-108. 77. Hu, M. C., Holzmann, B., Crowe, D. T., Neuhaus, H., and Weissman, I. L. (1993). The Peyer’s patch homing receptor. Curt-. Top. Microbiol. 184, 125-138. 78. Hynes, R. 0. (1992). Integrins: Versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25. 79. Springer, T. A. (1995). Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu. Rev. Physiol. 57, 827-872. 80. Campbell, J., Qin, S., Bacon, K., Schall, T., Mackay, C., and Butcher, E. (1996). The biology of chemokine and classical chemoattractant receptors: Differential requirements for adhesion-triggering vs. chemotactic responses in lymphoid cells. J Cell Biol. 134,255-266. 81. Berlin, C., Bargatze, R. F., Campbell, J. J., von Andrian, U. H., Szabo, M. C . , Hasslen, S. R., Nelson, R. D., Berg, E. L., Erlandsen, S. L., and Butcher, E. C. (1995). a 4

EUGENE C. BUTCHER et d.

248

integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80,413-422. 82. Alon, R., Kassner, P., Carr, M., Finger, E., Hemler, M., and Springer, T. (1995). The integrin VLA-4 supports tethering and rolling in flow on VCAM-1.1. Cell Biol. 128, 1243-1253. 83. von Andrian, U. H., Hasslen, S. R., Nelson, R. D., Erlandsen, S. L., and Butcher, E. C. (1995).A central role for microvillous receptor presentation in leukocyte adhesion under flow. Cell 82,989-999. 84. Ruegg, C., Postigo, A. A,, Sikorski, E. E., Butcher, E. C., Pytela, R., and Erle, D. J. (1992). Role of integrin a4P7/a4PP in lymphocyte adherence to fibronectin and VCAM-1 and in homotypic cell clustering. 1.Cell Biol. 117, 179-189. 85. Ahevogt, P., Hubbe, M., Ruppert, M., von Hoegen, P., Sammar, M., Andrew, D. P., McEvoy, L. M., Humphries, M. J., and Butcher, E. C. (1995). The a4 integrin chain is a ligand for a4P7 and a4pl. 1.Exp. Med. 182,345-355. 86. Strauch, U. G., Lifka, A., Gosslar, U., Kilshaw, P. J., Clements, J., and Holzmann, B. (1994). Distinct binding specificities of integrins a4P7 (LPAM-l),a 4 P l (VLA-4),and Int. Immunol. 6, 263-275. aIEL@7. 87. Andrew, D. P., Berlin, C., Honda, S., Yoshino, T., Hamann, A., Holzmann, B., Kilshaw, P. J., and Butcher, E. C. (1994). Distinct but overlapping epitopes are involved in a4P7-mediated adhesion to vascular cell adhesion molecule-1, mucosal addressin-1, fibronectin, and lymphocyte aggregation. J. Immunol. 153,3847-3861. 88. Chan, B., Elices, M., Murphy, E., and Hemler, M. (1992). Adhesion to vascular cell adhesion molecule 1 and fibronectin. Comparison of a4Pl (VLA4) and a4p7 on the human B cell line JY. J. Biol. Chem. 267, 8366-8370. 89. Rott, L., Briskin, M., Andrew, D., Berg, E., and Butcher, E. (1996). A fundamental subdivision of circulating lymphocytes defined by adhesion to mucosal addressin cell adhesion molecule- 1: Comparison with vascular cell adhesion molecule-1 and correlation with p7 integrins and memory differentiation. J. Immunol. 156, 3727-3736. 90. Salmi, M.. Granfors, K., MacDermott, R., and Jalkanen, S. (1994). Aberrant binding of lamina propria lymphocytesto vascular endothelium in inflammatorybowel diseases. Gastroenterology 106, 596-605. 91. Ager, A. (1993). Lymphocyte-vascular endothelial cell interactions in the immune response. Clin. Exp. Immunol. 93, 5-6. 92. Brezinschek, R., Brezinschek, H., Lazarovits, A,, Lipsky, P., and Oppenhiemer-Marks, N. (1996). Expression of the P7 integrin by human endothelial cells. Am. 1. Puthol. 149, 1651-1660. 93. Erle, D., Briskin, M. J., Butcher, E. C., Garcia-Pardo, A,, Lazarovits, A,, and Tidswell, M. (1994). Expression and function of the MAdCAM-1 receptor, integrin alpha 4 beta 7, on human leukocytes. 1,Immunol. 153,517-528. 94. Andrew, D., Rott, L., Kilshaw, P., and Butcher, E. (1996). Distribution of a4p7 and aep7 integrins on thymocytes, intestinal epithelial lymphocytes and peripheral lymphocytes. J , immunol. 26,897-905. 95. Kilshaw, P., and Murant, S. (1990). Eur. J. ImmunoZ. 20, 2201-2207. 96. Bochner, B., Sterbinsky,S., Briskin, M., Saini, S., and MacGlashan, D. (1996).Counterreceptors on human basophils for endothelial adhesion molecules. ]. Immunol. 157, 844-850.

97. Palecanda, A., Briskin, M. J., and Issekutz, T. B. (1997). Rat mast cell lines bind to the vascular cell adhesion molecule-1 (VCAM-1)and the mucosal addressin cell adhesion molecule-1 (MAdCAM-1).J. Immunol. 158, 2904-2910.

LYMPHOCmE TRAFFICKING AND REGIONAL IMMIJNITY

249

98. Walsh, G., Symon, F., Lazarovits, A., and Wardlaw, A. (1996). Integrin a4p7 mediates human eosiiiophil interaction with MAdCAM-1,VCAM-1and fibronectin. Immirnology 89, 112-119. 99. Yang, Y., Harrison, J.. Print, C . , Lehnert, K., Sammar, M., Lazarovits, A,, and Krissansen, C. (1996). Interaction of inonocytoid cells with mucosal addressin MAdCAM1 via the integrins VLA-4 and LPAM-I. Irrnnunol. Cell. Biol. (Austrnl.)74, 383-393. 100. Farstad, I., Hdstensen, T., Kvale, D., Fausa, O., and Brandtzaeg, P. (1997). Topographic distribution of homing receptors on B and T cells in human gut-associated lymphoid tissue: Relation of L-selectin and integrin a4p7 to naive and memory phenotypes. Am. J. Patho[. 150, 187-199. 101. Picarella. D., Hurlbut, P., Rottinan, J., Shi, X., Bntcher, E., and Ringler, D. (1997). Monoclonal antibodes specific for P7 integrin and mucosal addressiii cell adhesion molecule-1 (MAdCAM-1)reduce inflammation in the colon of scid mice reconstjtutecl with CD45HB’”CD4+ T cells. J. Imnwnol. 158, 2099-2106. 102. Tidswell, M., Pachynski, R., Wu, S., Qiu, S.-Q., Dunhain, E., Brish~i,M., Cochran, N., Kilshaw, P., Lazarovits, A., Andrew, D., Butcher, E., Yednock, T., and Erle, D. (1997).Structure/function analysis of the integrin p7 subunit: Identification of domains involved in adhesion to MAdCAM-1.J . Iminunol. 157, 1497-1505. 103. Feng, Y., Chung, D.. Garrard, L., McEnroe, G., Lirn, D., Scardina, J., McFadden, K., Guzzetta, A,, Lam, A,, Abraham, J., Liu, D., and Eudeinann, G. (1998). Peptides derived from the complementarity-detennining regions of anti-Mac-1 antibodies block intercellular adhesion molecule-1 interaction with Mac-1. J. Biol. Cheni. 273, 56255630. 104. Knorr, R., and Dustin, M . (1997). The lymphocyte function-associated antigen 1 I domain is ii transient binding module for intercellular adhesion molecule (ICAM)-l and ICAM-3 in hydrodynamic flow. J . Exp. Med. 186, 719-7313. 105. Loftus, J., and Liddington. R. (1997). New insights into integiin-ligand interaction. J. Clin. Incest. 100, S77-S81. 106. Springer, T. (1997). Folding of the N-terminal ligand-hinding region of integrin asubunits into a P-propeller domain. Pmc. Nntl. Acntl. Sci. U.S.A. 94, 65-72. 107. Irie, A,, Kainata, T., Puzon-McLaughlin, W., and Takacla, Y. (1995). Critical amino acid residues for ligand binding are clustered in a predicted P-turn of the third Nterminal repeat in the integrin a 4 and a 5 subunits. E M B O J . 14, 5550-5556. 108. Kamata, T.. Irie, A,, Tokuhira, M., a i d Takada, Y. (1996). Critical residues of integrin aIIb subunit for binding of aIIb /33 glycoprotein IIb-IIIa) to fibrinogen and ligandmimetic antibodies (PAC-1, OP-G2, and LJ-CP3).J. Biol. Cliein. 271, 18610-18615. 109. Rosen, S. D., Hwang, S. T., Gihlin, P. A., and Singer, M. S. (1997). High-endothelialvenule ligands for L-selectin: Identification and functions. Biochein. SOC. Trans. 25, 428-433. 11 0. Lasky, L. A. (1992). Selectins: Interpreters of cell-specific carbohydrate information during inflammation. Science 258, 964-969. 1 1 2 . Tedder, T. F., Steeher, D. A,, Chen. A,, and Engel, P. (1995). The selectIns: Vascular adhesion molecules. FASEB J . 9, 866-873. 112. Gallatin, W. M., Weissinan, I. L., and Butcher, E. C. (1983). A cell-snrface inoleciile involved in organ-specific homing of lymphocytes. Nature 303, 30-34. 113. Streeter, P. R., House. B. T. N., and Butcher, E. C. (1988). Immunohistologic and functional characteiization of a vascular addressin involved in lyniphocyte homing into peripheral lymph nodes. J. Cell B i d . 107, 1853-1862. 114. Waniock, R. A,, Askari, S., Butcher, E. C., and von Andrian, U. H. (1998). Molecular mechanisms of lymphocyte homing to peripheral lymph nodes. J . Exp. Med. 187, 205-216.

250

EUGENE C. BUTCHER et al.

115. Arbones, M. L., Ord, D. C., Ley, K., Ratech, H., Maynard-Curry, C., Otten, G., Capon, D. J., and Tedder T. F. (1994).Lymphocyte homing and leukocyte rolling and migration are impaired in L-selectin-deficient mice. Immunity 1, 247-260. 116. Picker, L. J., Warnock, R. A,, Bums, A. R., Doerschuk, C. M., Berg, E. L., and Butcher, E. C. (1991). The neutrophil selectin LECAM-1 presents carbohydrate ligands to the vascular selectins ELAM-1 and GMP-140. Cell 66, 921-933. 117. Lawrence, M., Berg, E., Butcher, E., and Springer, T. (1995). Rolling of lymphocytes and neutrophils on peripheral node addressin and subsequent arrest on ICAM-1 in shear flow. Eur. J. Immunol. 25, 1025-1031. 118. Lewinsohn, D., Bargatze, R., and Butcher, E. C. (1987). Leukocyte-endothelial cell recognition: Evidence of a common molecular mechanism shared by neutrophils, lymphocytes, and other leukocytes.1.Immunol. 138,4313-4321. 119. Michie, S., Streeter, P., Bolt, P., Butcher, E., and Picker, L. (1993). The human peripheral lymph node vascular addressin: An inducible endothelial antigen involved in lymphocyte homing. Am. J. Puthal. 143, 1688-1698. 120. Picker, L. J., and Siegelman, M. H. (1993). Lymphoid tissues and organs. In “Fundamental Immunology” (W. E. Paul, ed.), pp. 145-197. Raven Press, New York. 121. Carlos, T. M., and Harlan, J. M. (1994). Leukocyte-endothelial adhesion molecules. Blood 84,2068-2101. 122. Erlandsen, S. L., Hasslen, S. R., and Nelson, R. D. (1993). Detection and spatial distribution of the /32 integrin (Mac-1) and L-selectin (LECAM-1)adherence receptors on human neutrophils by high-resolution field emission SEM. J. Histochem. Cytochem. 4 1,327-333. 123. Jones, D. A., Smith, C. A,, McIntire, L. V., and Picker, L. J. (1994). A two-step adhesion cascade for T celVendothelial cell interactions under flow conditions. J. Clin. Invest. 94, 2443-2450. 124. Cerf-Bensussan, N . , Begue, B., Gagnon, J., and Meo, T. (1992).The human intraepithelial lymphocyte marker HML-1 is an integrin consisting of a /37 subunit associated with a distinctive a chain. Eur. J. Immunol. 22,273-277. 125. Cepek, K., Shaw, S., Parker, C., Russell, G., Morrow, J., Rimm, D., and Brenner, M. (1994).Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the aEP7 integrin. Nature 372, 190-193. 126. Karecla, P., Bowden, S., Green, S., and Kilshaw, P. (1995). Recognition of E-cadherin on epithelial cells by the mucosal T cell integrin a M290 /37 (aEP7). Eur. 1,Immunol. 25,852-856. 127. Farstad, I., Halstensen, T., Lien, B., Kilshaw, P., Lazarovits, A., and Brandtzaeg, P. (1996). Distribution of /37 integrins in human intestinal mucosa and organized gutassociated lymphoid tissue. Immunology 89, 227-237. 128. Schmitz, M . , Nunez, D., and Butcher, E. (1988). Selective recognition of mucosal lymphoid high endothelium by gut intraepithelial leukocytes. &troenterology 94, 576-581. 129. Shaw, S. K., and Brenner, M. B. (1995). The /37 integrins in mucosal homing and retention. Semin. Immunol. 7, 335-342. 130. Walton, L., Thornhill, M., Macey, M., and Farthing, P. (1997). Cutaneous lymphocyte associated antigen (CLA) and aeB7 integrins are expressed by mononuclear cells in skin and oral lichen planus. 1. Oral Puthol. Med. 26, 402-407. 131. Brenan, M., and Rees, D. (1997). Sequence analysis of rat integrin a E l and a E 2 subunits: Tissue expression reveals phenotypic similaritiesbetween intraepithelial lymphocytes and dendritic cells in lymph. Eur. J. Immunol. 27,3070-3079.

LYMPHOCmE TRAFFICKING AND REGIONAL IMMUNITY

25 1

132. McEvoy, L. M., Sun, H., Frelinger, J. G., and Butcher, E. C. (1997). Anti-CD43 inhibition of T cell homing. J. Exp. Med. 185, 1493-1498. 133. Stockton, B., Cheng, G., Manjunath, N., Ardman, B., and von Andrian, U. (1998). Negative regulation of T cell homing by CD43. Immunity 8, 373-381. 134. Salmi, M., Kalimo, K., and Jalkanen, S. (1993). Induction and function of vascular

adhesion protein-1 at sites of inflammation.J. Exp. Med. 178, 2255-2260. 135. Sdmi, M., Tohka, S., Berg, E. L., Butcher, E. C., and Jalkanen, S. (1997). Vas-

cular adhesion protein 1 (VAP-1) mediates lymphocyte subtype-specific, selectinindependent recognition of vascular endothelium in human lymph nodes. J. Exp. Med. 186,589-600. 136. McEvoy, L. M., Sun, H., Tsao, P. S., Cooke, J. P., Berliner, J. A., and Butcher, E. C. (1997). Novel vascular molecule involved in monocyte adhesion to aortic endothelium in models of atherogenesis. J. Exp. Med, 185, 2069-2077. 137. DeGrendele, H., Estess, P., and Siegelman, M. (1997). Requirement for CD44 in activated T cell extravasation into an inflammatory site. Science 278, 672-675. 138. Jalkanen, S., Bargatze, R., de 10s Toyos, J., and Butcher, E. (1987). Lymphocyte recognition of high endothelium: Antibodies to distinct epitopes of an 85-95 kD glycoprotein antigen differentially inhibit lymphocyte binding to lymph node, mucosal, or synovial endothelial cells. /. Cell Biol. 105, 983-990. 139. Picker, L., Nakache, M., and Butcher, E. (1989). Monoclonal antibodies to human lymphocyte homing receptors define a novel class of adhesion molecules on diverse cell types. 1.Cell Biol. 109, 927-937. 140. Culty, M., Miyake, K., Kincade, P. W., Sikorski, E., Butcher, E. C., Underhill, C., and Silorski, E. (1990).The hyaluronate receptor is a member of the CD44 (H-CAM) family of cell surface glycoproteins. J. Cell Biol. 111, 2765-74 [published erratum (Feb., 1991) appears in J. Cell B i d . 122(3)following 513.1 141. Ford, W., Smith, M., and Andrews, P. (1978). Possible clues to the mechanism underlying the seIective migration of lymphocytes from the blood. In “Cell-Cell Recognition” (A. Curtis ed.) Vol. 22, pp. 359-392. Cambridge Univ. Press, London and New York. 142. Bargatze, R. F., and Butcher, E. C. (1993). Rapid G protein-regulated activation event involved in lymphocyte binding to high endothelial venules.J. Exp. Med. 178,367-372. 143. Baggiolini, M., Dewald, B., and Moser, B. (1997). Human chemokines: An update. Annu. Rev. Immunol. 15,675-705. 144. Rollins, B. (1997). Chemokines. Blood 90, 909-928. 145. Hromas, R., Kim, C., Klemsz, M., Krathwohl, M., Fife, K., Cooper, S., SchnizleinBick, C., and Broxmeyer, H. (1997). Isolation and characterization of Exodus-2, a novel C-C chemokine with a unique 37-amino acid carboyl-terminal extension. J. Immunol. 159,2554-2558. 146. Nagira, M., Imai, T., Hieshima, K., Kusuda, J., Ridanpaa, M., Takagi, S., Nishimura, M., Kakizaki, M., Nomiyama, H., and Yoshie, 0. (1997). Molecular cloning of a novel human CC chemokine secondary lymphoid-tissue chemokine that is a potent chemoattractant for lymphocytes and mapped to chromosome 9p13. J . B i d . Chem. 272, 19518-19524. 147. Hedrick, J. A., and Zlotnik, A. (1997). Identification and characterization of a novel P chemokine containing six conserved cysteines. 1.Immunol. 159, 1589-1593. 148. Campbell, J., Hedrick, J., Zlotnik, A,, Siani, M., Thompson, D., and Butcher, E. (1998).6-C-kine (SLC), a lymphocyte adhesion-triggering chemokine expressed in high endothelium, is an agonist for the MIP3P receptor CCR7.J. Cell Bid. 141,1053-1059. 149. Campbell, J., Bowman, E., Murphy, K., Youngman, K., Siani, M., Thompson, D., Wu, L., Zlotnik, A., and Butcher E. (1998). 6-C-kine, a lymphocyte adhesion-triggering

252

EUGENE C. BUTCHER et al

chemokine expressed by high endothelium, is an agonist for the MIP-30 receptor CCR7. J. Cell Biol. 141, 1053-1059. 150. Yoshie, O., Imai, T., and Nomiyama, H. (1997). Novel lymphocyte-specific CC chemokines and their receptors. 1.Leukoc. Biol. 62, 634-644. 151. Soto, H., Wang, W., Strieter, R., Copeland, N., Gilbert, D., Jenkins, N., Hedrick, J., and Zlotnik, A. (1998).The CC chemokine 6Ckine binds the CXC chemokine receptor CXCR3. Proc. Natl. Acad. Sci. U.S.A. 95, 8205-8210. 152. Nakano, H., Tamura, T., Yoshimoto, T., Yagita, H., Miyasaka, M., Butcher, E. C., Nariuchi, H., Kakiuchi, T., and Matsuzawa, A. (1997). Genetic defect in T lymphocytespecific homing into peripheral lymph nodes. Eur. ]. lmmunol. 27,215-221. 153. Middleton, J., Neil, S., Wintele, J., Clark-Lewis, I., Moore, H., Lam, C., Auer, M., Hub, E., and Rot, A. (1997).Transcytosis and surface presentation of IL-8 by venular endothelial cells. Cell 91, 385-395. 154. Legler, D., Loetscher, M., Roos, R., Clark-Lewis, I., Baggiolini, M., and Moser, B. (1998).B cell-attracting chemokine 1, a human CXC chemokine expressed in lymphoid tissues, selectively attracts B lymphocytes via BLRVCXCR5. J. Exp. Med. 187, 655-660. 155. Czinn, S. J., and Lamm, M. E. (1986). Selectivechemotaxis of subsets of B lymphocytes from gut-associated lymphoid tissue and its implications for the recruitment of mucosal plasma cells. 1.Immunol. 136, 3607-3611. 156. Erle, D., Ruegg, C . , Sheppard, D., and Pytela, R. (1991). Complete amino acid sequence of an integrin 0 subunit (p7) identified in leukocytes. 1.Biol. Chem. 266, 11009. 157. Bell, E. B., Sparshott, S. M., and Ager, A. (1995). Migration pathways of CD4 T cell subsets in vivo: The CD45RC-subsets enters the thymus via a 4 integrin-VCAM-1 interaction. Int. Immunol. 7, 1861-1871. 158. Husband, A., and Gowans, J. (1978). The origin and antigen-dependent distribution of IgA-containing cells in the intestine. J. Erp. Med. 148, 1146-1160. 159. Picker, L. J. (1994). Control of lymphocyte homing. Cum. Opin. Immunol. 6,394-406. 160. Steffen, B. J., Butcher, E. C., and Engelhardt, B. (1994). Evidence for involvement of ICAM-1 and VCAM-1 in lymphocyte interaction with endothelium in experimental autoimmune encephalomyelitis in the central nervous system in the SJUJ mouse. Am. J. Puthol. 145, 189-201. 161. Engelhardt, B., Conley, F. K., and Butcher, E. C. (1995). Lymphocytes infiltrating the CNS during inflammation display a distinctive phenotype and bind to VCAM-1 but not to MAdCAM-1. lnt. lmmunol. 7, 481-491. 162. McWilliams,M., Phillips-Quagliata,J., and Lamm, M. (1975).Characteristics of mesenteric lymph node cells homing to gut-associated lymphoid tissue in syngeneic mice. J Immunol. 115, 54-58. 163. Guy-Grand, D., Griscelli, C., and Vassilli, P. (1978). The mouse gut T-lymphocyte, a novel type of T cell. Nature, origin and traffic in mice in normal and graft-versus-host conditions. J. Erp. Med. 148, 1661-1667. 164. Rose, M., Parrott, D., and Bruce, R. (1978). The accumulation of immunoblasts in extravascular tissues including mammary gland, peritoneal cavity, gut and skin. Immunology 35, 415-423. 165. Picker, L. J., Treer, J. R., Fergusson-Darnell, B., Collins, P., Buck, D., and Terstappen, L. W. M. M. (1993). Control of lymphocyte homing in man: I. Differential regulation of the peripheral lymph node homing receptor L-selectin on T cells during the virgin to memory transition. J. Immunol. 150, 1105-1121.

LYMPHOCYTE TRAFFICKING AND REGIONAL IMMUNITY

253

166. Picker, L. J., Treer, J. R., Fergusson-Damell, B., Collins, P., Bergstresser, P., and Terstappen, L. W. M. M. (1993).Control of lymphocyte homing in man: 11. Differential regulation of the T cell skin-selective homing receptor CLA. J. Immunol. 150, 11221136. 167. Husband, A. (1982). Kinetics of extravasation and redistribution of IgA-specific antibody-containing cells in the intestine. 1. Inamunol. 128, 1355-1359. 168. Santamaria Babi, L., Picker, L., Perez Soler, M., Drzimalla, K., Flohr, P., Blaser, K., and Hauser, C. (1995). Circulating allergen-reactive T cells from patients with atopic dermatitis and allergic contact dermatitis express the skin-selective homing receptor, the cutaneous lymphocyte-associatedantigen. J . Exp. Med. 181, 1935-1940. 169. Quiding-Jarbrink, M., Nordstrom, I., Granstrom, G., Kilander, A,, Jertborn, M., Butcher, E. C., Lazarovits, A,, Holmgren, J,, and Czerkinsky, C. (1997). Differential expression of tissue-specificadhesion molecules on human circulating antibody-forming cells after systemic, enteric and nasd immunizations. J. Clin. bluest. 99, 1281-1286. 170. Kantele, A,, Kantele, J. M., Savilahti, E., Westerholm, M., Arvilommi, H., Lazarovits, A,, Butcher, E. C., and Makela, P. H. (1997). Homing potentials of circulating lymphocytes in humans depend on the site of activation: Oral, but not parented, typhoid vaccination induces circulating antibody-secreting cells that all bear homing receptors directing them to the gut. J. Inzzunol. 58, 574-579. 171. Rott, L. S., RosC, J. R., Bass, D., Williams, M. B., Greenberg, H. B., and Butcher, E. C. (1997). Expression of mucosal homing receptor a4P7 by circulating CD4+ cells with memory for intestinal rotavirus. J. Clin. Iwznzunol. 100, 1204-1208. 172. Feng, H., Franco, M. A,, and Greenberg, H. B. (1997). Murine model of rotavirus infection. Adv. Exp. Med. Biol. 412, 233-240. 173. RosC, J. R., Williams, M. B., Rott, L. S., Butcher, E. C., and Greenberg, H. B. (1998). Expression of the mucosal homing receptor a4P7 correlates with the ability of CDW memory T cells to clear rotavirus infection. J. Virol. 72, 726-730. 174. Williams, M., Ros&,J., Rott, L., Franco, M., Greenberg, H., and Butcher, E. C. (1998). The memory B cell subset responsible for the mucosal IgA response and humoral immunity to rotavirus expresses the intestinal mucosal homing receptor, a4P7. J . lnzmunol. 161, 4227-4235. 175. Drillenburg, P., van der Voort, R., Koopman, G., Dragusics, B., van Krieken, J., muin, P., Meenan, J., L.dzarovits, A,, Radaszkiewicz, T., and Pals, S. (1997). Preferential expression of the mucosal homing receptor uitegrin a4P7 in gastrointestinal nonHodgkin's lymphomas. Am. J. Pathol. 150, 919-927. 176. Dogan, A,, Du, M., Koulis, A,, Briskin, M., and Isaacson, P. (1997). Expression of lymphocyte homing receptors and vascular addressins in low-grade gastric B-cell lymphomas of mucosal-associated lymphoid tissue. Am. J. Pathol. 151, 1361-1369. 177. Waldrop, S.,Pitcher, C., Peterson, D., Maino, V., and Picker, L. (1997).Determination of antigen-specific memory/effector CD4' T cell frequencies by flow cytometry: Evidence for a novel, antigen-specifichomeostatic mechanism in HIV-associated immunodeficiency. 1.Clin. Invest. 99, 1739-1750. 178. Hesterberg, P. E., Winsor-Hines, D., Briskin, M. J., Soler-Ferran, D., Merrill, C., Mackay, C. R., Newman, W., and Ringler, D. J. (1996). Rapid resolution of chronic colitis in the cotton-top tamarin with an antibody to a gut-homing integrin a4P7. Gastroenterology 111, 1373-1380.

This article was accepted for publication on July 10, 1998.

This Page Intentionally Left Blank

ADVANCES IN IMMUNOLOGY, VOL. 72

Dendritic Cells DIANA BEU;

JAMES

w. YOUNG)

AND JACQUESBANCHEREAU'

'Baylar Institute far lmmundogy Research, Sammons Cancer Cenkr, W a s , Texas 75246; and tMBmorial~oanK-*W cancer Center, come// university M i c a 1 Cdiege, New York, New Yark 10021

1. Introduction

Dendritic cells (DCs) were first identified in the epidermis in 1868, and were termed Langerhans cells (Langerhans, 1868).Their presence in other tissues was identified a century later in 1973 (Steinman and Cohn, 1973). DCs are now recognized as an integral part of the lymphohematopoietic system, and function as sentinels of the immune system, initiating immune responses. DCs are found in the interstitiurn of most organs (excluding brain) at a frequency so low that this has posed a major impediment to their study. The cells can usually be identified by their characteristic and unusual morphology, as well as their high-level expression of class I1 MHC molecules. To launch immune responses, DCs have to capture small amounts of antigen efficiently and present it to rare antigen-specific T cells to initiate their expansion and maturation (Fig. 1).These two key functions of DCs segregate in time and space. The soluble or particulate antigedpathogen that invades tissues is efficiently captured by tissue DCs. This triggers DC migration into the proximal secondary lymphoid organ, where they mature into a developmental state that allows the selection and activation of antigen-specific T cells. In particular, DCs support the generation of not only lymphokine-secreting helper T cells, but also effector cytotoxic T lymphocytes (CTLs),which subsequently migrate to the site of initial injury to eliminate virally infected cells or tumor cells. This capacity of activating not only memory, but also naive T cells, is a property not shared by other antigen-presenting cells (APCs). Hence, DCs are in fact professional APCs. Knowledge of DC physiology has progressed considerably because of the discovery of culture techniques, in the early 199Os, that support the in vitro generation of large numbers of DCs from hematopoietic progenitors. DCs comprise three distinct subsets, including two within the myeloid lineage, Langerhans cells and interstitial DCs, and one within the lymphoid lineage, the so-called lymphoid DC subset. There are three stages of development, i.e., precursor DCs (DC,,) patrolling through blood and lymphatics, immature DCs (DC,,,) residing within virtually every tissue 255

Copynght 6 1999 by Academic Press All nghts of repduction in any form reserved 0065-277&'99 $30 W

256

DIANA BELL d a!.

fc====l

bone marrow

ANnGEN \

\

PROLIFERATINGDC PROGENITOR

/ IMMATURE

INFLAMED VESSELS

LYMPHOIDTISSUE, BONE MARROW,

MATURE DCD IN T CELL AREAS

ACTIVATED1CELLS IN EFFERENTLYMPHATICS aecondery lymphoid organ

FIG.1. The lye of a dendritic cell (DC)or the capture of antigens and their presentation to selected antigen-specific lymphocytes. Circulating precursor DCs enter peripheral tissues as immature DCs, where they are poised to capture antigens ( e g , microbial products). Loaded immature DCs leave the tissues and migrate to lymphoid organs, where, after maturation, they display antigen-derived peptides on their MHC molecules, which, in turn select rare circulating antigen-specific lymphocytes. These reactive T cells become activated and further induce terminal DC maturation, which supports lymphocyte expansion and differentiation. Activated T lymphocytes migrate back to the injured tissue, because they can selectively traverse inflamed epithelium. Helper T cells secrete lymphokines,and cytotoxicT cells eventually lyse the infected cells. Activated B cells differentiate into B lymphoblasts after contact with T cells and DCs, and then migrate into various areas, where they mature into plasma cells and produce antibodies that will eventually neutralize the initial pathogen.

in ambush to capture pathogens, and mature DCs (DC,,) residing temporarily within secondary lymphoid organs. In addition to being involved in the initiation of immunity, DCs also appear to play an important role in the induction of immunological tolerance. In particular, thymic DCs present endogenous self-peptides to newly generated thymocytes, thereby allowing the deletion of self-reactive T cells. These thymic DCs may indeed originate from a precursor cell that also gives rise to lymphocytes and natural killer

DENDRITIC CELLS

257

(NK) cells, and have thus been called lymphoid DCs. There is also evidence of a role for DCs in the development of peripheral tolerance. Recent studies further indicate that DCs can directly modulate B cell and NK cell functions. Molecular genetic approaches are also ascribing to DCs new molecules, such as chemokines and chemokine receptors, proteases, and antiproteases, lectinlike receptors for antigen uptake, new members of the TNF/TNF receptor family, as well as killer inhibitory receptors. It is hoped that this will increase understanding of the biological functions of DCs, selectively identify immature and mature DCs, and explain DC development at the signaling and transcriptional levels. DC research is further fueled by the hope that cultured DCs will lead to the development of cellular vaccines for use in cancer therapy and the treatment of various infectious diseases. II. Features of Dendritic Cells

A. MORPHOLOGY Figure 2 (see color plate) illustrates the unusual shape that gives rise to the term “dendritic cell.” In situ, as in the skin and lymphoid organs, immature and mature DCs have a stellate shape. Many fine dendrites are displayed when DCs are isolated and spun onto slides; DCs extend large sheetlike processes or veils in many directions from the cell body. The processes are long (10 pm) and thin, either fine or sheetlike. Actin cables are scarce (Winzler et al., 1997). The shape and motility of DCs suit their functions, initially the efficient capture of antigen and subsequently the selection of rare antigen-specific lymphocytes.

B. PRECURSOR DENDRITIC CELLS, IMMATURE DENDRITECELLS, AND MATUREDENDRXTIC CELLS All tissues, with the possible exception of brain and testis, contain DCs that are immature (DCim,,,,), capable of capturing antigens but not yet possessing the panel of accessory molecules required for potent T cell stimulation. Antigens able to drive an immune response are those that efficiently initiate the maturation of DCs. In uiuo, transplantation (Larsen et al., 1990a,b; 1994) and contact allergens (Enk et al., 1993a,b; SilberbergSinakin et al., 1976) are among the most powerful immunologic stimuli for DC maturation. The best studied DCirnm is certainly the epidermal Langerhans cell (LC), which was shown to be derived from hematopoietic progenitors using bone marrow reconstitution experiments (Katz et al., 1979). LCs, identified by expression of the CDla antigen (Fig. 3) and the presence of Birbeck granules (cytoplasmic structures formed by double membrane joinings),

258

DIANA BELL et a1 Migration receptors CE49d. E-tadharln. CD44 varlanla CCR1,5.8.7, CD88

Recepton for antigen uptake

Adheelon and costlmuletion

MMR, DEC-205, FCcR, F C F (CDS2, CDE4). ASQPR

CDSO. CDS4, CD68

Signalling of DCs TNF-R, CD40,TRANCE’RA Cyioklne-R: QM-CSF, IL-I, IL-10,

Other molecules CDZ, CD9. CD25, CDBS, CW5

u

ChamOkinN

FIG.3. Molecules expressed by dendritic cells. Illustrated are the key features used in combination to identify DCs. At the present time there is not a single molecule that permits unambiguous assignment of a given cell to the DC family. The combination of several markers, however, defines a dendritic cell subpopulation and its stage of maturation.

are localized to the basal and suprabasal layers of the epidermis (Katz et al., 1979; Romani et al., 1985). The CD1 antigens are nonpolymorphic cell surface proteins noncovalently associated with µglobulin and bear some structural similarity to major histocompatibility complex (MHC) molecules. CD1 molecules have been shown to present peptides as well as microbial, nonpeptide, lipid-containing antigens to T cells (Maher and Kronenberg, 1997; Porcelli et al., 1992).The CDla antigen is also a cortical thymocyte marker that disappears at later stages of T cell maturation (McMichael et al., 1979). Other members of the CD1 family have also been identified on LCs. LCs express variable amounts of CDlc (Davis et al., 1988), and higher percentages of CDlb’ cells are present among dermal and migrating LCs (Richters et al., 1996). Interstitial DCs in most organs and tissues, such as lung (Gong et al., 1992; Havenith et al., 1993; Holt, 1993; Schon-Hegrad et al., 1991; Xia et al., 1995), heart and kidney (Austyn et al., 1994), and dermis (Nestle et al., 199813, 1993), represent an important reservoir of DCi,,. These cells differ from Langerhans cells in that they lack Birbeck granules and do not always express CD1 antigens. After antigen exposure or inflammatory stimuli, DCimm migrate via afferent lymph as “veiled DCs,” to the draining lymph nodes where they localize to the T cell areas as mature interdigitating

259

DENDRITIC CELLS

DCs (IDCs). IDCs are also present in other secondary lymphoid organs such as tonsils and the white pulp of spleen (Bjorck et al., 199%; Hart and McKenzie, 1988; Steinman, 1991). Before the availability of DC cultures, LCs provided the most suitable experimental model for studying maturation. Freshly isolated LCs express antigens and detectable Fcy receptors (CD3WcyRII and CD64A?cyRI), as well as high-affinity IgE receptors (FcsRI) that contribute to antigen capture. However, LCs are not particularly potent APCs for the mixed lymphocyte reaction (MLR). In contrast, LCs maintained in culture for several days resemble DC,, in phenotype and function, including their capacity to initiate T cell responses to alloantigen in the MLR (Romani et al., 1989; Schuler and Steinman, 1985; Teunissen et al., 1990). During maturation DCs undergo major changes in phenotype and function (Fig. 4).The new phenotype distinguishes DCmatfrom DCimmbased on critical epitopes such as CD83, CD80, and CD86. The CD83 antigen (a 186-aa single-chain glycoprotein, member of the immunoglobulin superfamily) is presently one of the most useful markers for identification of DCmat(Zhou et aZ., 1992; Zhou and Tedder, 199513). CD83' cells express the highest levels of MHC class I1 molecules, when compared with other leukocyte lineages, and immunohstologic analysis reveals that CD83 is found mainly on DCs within T lymphocyte areas of lymphoid organs. A DC

IMMATURE DC (DClmm)

MATUREDC (DCmat)

Hlph IntromllularYHCll (MIICs) Endocylmls, lncludlnp FcR Hlph CCR1. CCRS. CCRB L& CCR7 Low CD54,sB. 0a. 86 Low CD40. IL-12 LOW cD83

Hlph sudau MHCll Low wdocytosls and FcR Low CCRl, CCR5. CCRB Hlgh CCR7 Hlgh CD54,58.80,86 Hlgh CD40. IL-12 Hlgh CD83

FIG.4. Stages of dendritic cell maturation. DC precursors, which originate from CD34" bone marrow progenitors, circulate in the blood as nonlymphoid mononuclear cells or monocytes, identifiable as a class I1 MHC-positive CDllc+ DC,,, or CDllc- DC,,. These precursors migrate into tissues to become resident DC,,,, and this may increase in response to inflammatory cytokines. After antigen capture, DC,,", undergo maturation during migration to secondary lymphoid organs. Maturation is completed after the selection, activation, and interaction with antigen-specific T sells. In simple terms, maturation transforms an antigen-capturing sell into an antigen-presenting, lymphocyte-activating cell.

260

DIANA BELL et 02.

that is a novel member of the lysosome-associated membrane glycoprotein (LAMP) family, homologous to the lysosomal marker CD68, has been cloned by screening a cDNA library of in uitro-generated DCs. It is not expressed on interstitial DCs but is uniquely expressed by DC,, as shown by specific staining of interdigitating DCs within secondary lymphoid organs (De Saint Vis et al., 1998). The antigen recognized by the CMRF44 monoclonal antibody, most likely a glycolipid, is expressed at high density on mature DCs and its expression increases very quickly on blood DC,, cultured in uitro (Fearnley et al., 1997; Hock et al., 1994). During maturation, several other molecules are up-regulated, including class I1 MHC antigens (Fearnley et al., 1997; Hock et al., 1994; Said et al., 1997; Schuler and Steinman, 1985; Witmer-Pack et al., 1988), ICAM-1 (CD54), LFA-3 (CD58), C D l l d c , CD40, CD80 (Inaba and Steinman, 1984; Larsen et al., 1992; Lenz et al., 1993; Young et al., 1992), and CD86 (Cam et al., 1994c; Inaba et al., 1995). The actin-bundling protein p55, fascin, a molecule involved in the organization of the actin cytoskeleton that supports the formation of dendritic processes (Mosialos et al., 1996; Ross et al., 1998), also increases with differentiation. In contrast, Fc receptor expression decreases substantially during DC maturation. Some chemokine receptors are also down-regulated, whereas others are up-regulated, thus supporting the appropriate homing of DCs at their various stages of differentiation (Dieu et al., 1998; Sozzani et al., 1998). Within blood and lymphoid organs, two populations of cells with some characteristics of DCs have been identified (Fig. 5).These cells are distinguished from other lymphoid and myeloid cells by their high levels of class I1 MHC and lack of CD3, CD19, CD14, and CD56. One population, CD4+,CDllc', CD13', and CD33', mostly found within germinal centers but also in the circulation, displays a morphology of immature DCs and quickly matures in vitro (Grouard et al., 1996; O'Doherty et al., 1993; Thomas et al., 1993). The other population, CD4', CDllc-, CD13-, CD33-, and CD123+,resembles the morphologyof plasma cells and corresponds to the enigmatic plasmacytoid T cells that are restricted to the T cell-rich areas of secondary lymphoid organs (Grouard et al., 1997; Olweus et al., 1997). Interestingly, this population undergoes very rapid apoptosis in culture unless rescued by IL-3. These CDllc- CD123' cells differentiate into cells with DC characteristics in response to IL-3 and CD40L.

FIG. 5 . CDllc' and C D l l c - dendritic cell precursors from the peripheral blood. Blood mononuclear cells display -1% HLA-DRhlghlineage"%(CD3, CD14, CD16, CD19, CD56) cells (A), which can be enriched to as high as 30% after bead depletion (B). A third fluorochrome (C) identifies additional epitopes expressed by CDllc' and CDllc- DC,,.

-La

I"

n

02

wm9 51

i

01

n.

s

sol

I

a-~wnwwv

01

E

fi Jz

z

.

ool-

0

. . . ,..., . . . . ,..... . . . ,..... . .: . ..o

o =

%'

m

262

DIANA BELL et a!.

These DCs lack the myeloid cell markers CD13 and CD33 and may thus be the human lymphoid DCs. Importantly, these mature DCs appear to induce naive T cells to differentiate specifically along the Th2 pathway, as demonstrated by the secretion of IL-4 and the lack of IFN-.)/ (Y. J. Liu, personal communication). A working hypothesis is that these two cell populations correspond to patrolling precursor DCs (DC,,) that home to sites of injury, from which DCimmhave earlier fled in their migration toward draining lymphoid organs. This influx of DC,, can be measured within 30 min, whereas accumulation of neutrophils requires 4 hr (McWilliam et al., 1996). C. PHENOTYPE Figure 3 illustrates important molecules that distinguish subpopulations and stages of maturation of dendritic cells.

1. Antigen Capture and Presentation by Immature DCs a. Antigen Capture. DCimm can efficiently internalize a diverse array of antigens for processing and loading onto class I1 MHC molecules, as a consequence of high endocytic activity levels. Antigen uptake by DC,, can occur via four distinct mechanisms: (1)macropinocytosis, (2) receptormediated endocytosis through Fcy and FCEreceptors (Maurer et al., 1996; Sallusto and Lanzavecchia, 1994), (3) receptor-mediated endocytosis through the mannose receptor (Sallusto et al., 1995) and C-type lectin receptor DEC205 (Jiang et al., 1995), and (4) engulfment of apoptotic bodies through the vitronectin receptor a(v)P3(Albert et al., 1998; Rubartelli et al., 1997). 1. Macropinocytosis. Macropinocytosis is a cytoskeleton-dependent type of fluid-phase endocytosis mediated by membrane ruffling and the formation of large vesicles (1-3 pm). In DCs, macropinocytosisis constitutive, and enables a single cell to take up a very Iarge volume of fluid (half the cell’s volume per hour) (SaIlusto et al., 1995). 2. FCEand Fcy receptors. Human epidermal LCs, but not other epidermal cells, express FcERI (Kraft et al., 1998; Rieger et al., 1992; Wang et al., 1992) and use this receptor to maximize antigen uptake via specific IgE for subsequent presentation to T cells (Bieber, 1997). The FcERI on DCs is a multimeric receptor composed of the a and y chains initially identified on basophils, but lacking the @ chain (Maurer et al., 1996). LCs also express the low-affinity FcERII, CD23, which may have a role in the pathogenesis of atopic eczema as well as in the regulation of IgE synthesis (Bieber et al., 1989). In response to maturation stimuli, immature DCs down-regulate their Fc receptors for IgG, FcyRI (CD64) (Fanger et al., 1996), and Fcy RII (CD32) (Thomas et al., 1993), thereby reducing their antigen capture by this mechanism.

DENDRITE CELLS

263

3. The mnnose receptor and C-type lectin receptor DEC-205. DCs express high levels of the mannose receptor, which contains multiple carbohydrate-binding domains and is involved in the internalization of a variety of glycoproteins. Whereas Fc receptors are degraded together with their cargo, the mannose receptor releases its ligand at endosomal pH and is recycled. This allows uptake and accumulation of many ligands by a small number of receptors (Engering et al., 1997; Lanzavecchia, 1996). The mannose receptor may play a critical role in phagocytosis of particles and microbes (Inaba et al., 1983b; Moll et al., 1993; Reis e Sousa et al., 1993; Svensson et al., 1997). Another endocytic receptor is DEC-205, an integral membrane protein homologous to the mannose receptor. DEC205 and its antigenic ligand are rapidly taken up by means of coated pits and vesicles, then delivered to a multivesicular endosomal compartment that resembles the class I1 MHC-containing vesicles implicated in antigen presentation (Geuze, 1998a; Jiang et al., 1995). 4. Engulfment of apoptotic bodies. DCs are able, in vitro and in uivo, to capture and engulf apoptotic cells (Albert et al., 1998; Rubartelli et al., 1997). Immature DCs appear to be more efficient than mature DCs in capturing apoptotic bodies as a means of antigen uptake (M. L. Albert and N. Bhardwaj, personal communication). Although macrophages engulf apoptotic bodies using multiple surface molecules (CD14, CD36, phosphatidylserine receptor) (Devitt et al., 1998; Rubartelli et al., 1997), DCs may preferentially use the vitronectin receptor ayP3 and the CD36khrombospondin receptor (J. Banchereau, unpublished observations). The engulfment of apoptotic bodies induces a rise in intracellular free calcium concentration [Ca2+li,which is essential for the engulfment to occur (Rubartelli et al., 1997).Apoptosis, but not necrosis, is required for the generation and packaging of immunogenic material for delivery to DCs. In particular, DCs loaded with apoptotic bodies, derived from either macrophages or HeLa cells infected with influenza virus, can stimulate the proliferation of influenza specific T cells and the generation of class I MHC-restricted, influenza-specific CD8+ CTLs (Albert et al., 1998; Huang et al., 1994). This pathway is likely to account for the in vivo phenomenon of “crosspriming” (Bevan, 1977),whereby antigens derived from tumor cells (Inaba et al., 1998) or transplants (Fossum and Rolstad, 1986) are presented by host APCs. Tolerance to tissue-restricted self antigens may also depend on apoptotic cell death, as occurs during deveIopment and normal cell turnover. The specifics are not established, but this could be followed by antigen presentation by DCs (Kurts et al., 1996, 1997b), with a resultant nonproliferative or anergic T cell response. Interestingly, while macrophages can also engulf apoptotic bodies, they are unable to stimulate

264

DIANA BELL et a1

specific CTLs. Furthermore, they even prevent DC-mediated CTL generation by this route by sequestering antigen (Albert et al., 1998).

b. Antigen Presentation-MHC Class II and MHC Class I Molecules. 2. MHC class II Zoading. MHC class I1 loading is critical for CD4 T cells. In addition to efficient antigen capture, DCs fulfill other requirements for antigen presentation by synthesizing and expressing high levels of class I1 MHC (Kleijmeer et al., 1994, 1995; Young et al., 1992). Considerable evidence indicates that late endosomes (which develop from the vacuolar parts of the early endosomes network) and their lysosomal derivatives play a crucial role in class I1 MHC-mediated antigen presentation (Geuze, 1998b; Pierre and Mellman, 1998). In APCs, and most particularly DCs, the majority of intracellular class I1 MHC is found in late endocytic structures with numerous internal membrane vesicles and sheets, collectively designated MIICs (MHC class I1 compartments). A minor compartment is represented by early endosomes that contain mature class I1 MHC molecules, which are internalized from the cell surface and rapidly recycled (Harding and Unanue, 1989; Reid and Watts, 1990). The major compartment (MIIC) contains newly synthesized class I1 MHC molecules that are targeted to this structure by the invariant chain (Ii). It also contains HLADM molecules that remove the Ii-derived class II-associated invariant chain peptide (CLIP) and promote the formation of stable complexes (Lanzavecchia, 1996). Cell fractionation studies have indicated the presence of class I1 MHC-positive vesicles (CIIV) that are physically and biochemically distinct from conventional endosomes and lysosomes (Pierre et al., 1997). During DC maturation, three sequential stages are identified: early DCs, in which class I1 MHC antigens are localized to lysosomal compartments; intermediate DCs that accumulate class I1 in distinctive nonlysosomal vesicles; and mature DCs, in which peptide-class I1 MHC complexes are present on the plasma membrane for long periods of time, thereby allowing the selection of rare antigen-specific T cells (Cella et al., 1997c; Pierre et at., 1997~). 2. MHC class Z loading. MHC class I loading is critical for CD8 T cells. Professional APCs can capture exogenous antigens for presentation on MHC class I molecules. This ensures an efficient generation of cytotoxic CD8+ T cells (Heemels and Ploegh, 1995; Watts, 1997), even against viral or tumor antigens that are expressed only in nonprofessional APCs. In vitro experiments suggest two fundamentally different pathways for the presentation of exogenous antigens: ( 1)one involving unconventional postGolgi loading of MHC class I (Harding and Song, 1994; Liu et al., 1995) and (2) another one involving the classical transporter associated with antigen processing (TAP) loading mechanism (Rock et al., 1986). In vitro

DENDRITE CELLS

265

cross-priming requires a functional TAP pathway (Huang et al., 1994; Norbury et al., 1997).The peptides for class I MHC on DCs can be derived from nonreplicating microbes (Svensson et al., 1997), soluble proteins (Norbury et al., 1997), or apoptotic cells (Albert et al., 1998). Experiments using a unique class I MHC+/class I1 MHC-/CD80+ dendritic cell line (80/1DC) derived from murine fetal skin have led to the conclusion that direct dogeneic class I MHC-restricted immunity can occur in the absence of class I1 expression (Kolesaric et al., 1997; Lenz et al., 1996). This mechanism has biological relevance to transplantation immunity, as well as immunity against opportunistic infections in conditions of congenital, iatrogenic, or acquired immunodeficiencies.

2. Adhesion Molecules During their migration and subsequent interaction with T cells, DCs are involved in a variety of adhesion events. Expression of cutaneous lymphocyte antigen (CLA) may allow DCs to reach the skin by interacting with E-selectin (CD62E) on activated endothelial cells (Strunk et al., 1996; 1997). LCs adhere to the surrounding keratinocytes through homotypic interactions involving E-cadherin. After antigen capture, LCs downregulate E-cadherin, losing adhesive interactions with surrounding keratinocytes and allowing migration from the skin (Tang et al., 1993). Interestingly, following epicutaneous stimulation with haptens, LCs produce type IV collagenase (MMP 9), which probably facilitates the crossing of the basement membrane (Kobayashi, 1997). Integrins and intercellular adhesion molecules contribute to DC adhesion and migration through vessel walls (Jakob et al., 1997). Immature blood DCs can enter the lymphoid organs through high endothelial venules via CD49d 6-integrin (Brown et al., 1997). ICAM-1, which together with ICAM-2 is up-regulated on DC activation and may contribute to DC migration as well as to the later phases of T lymphocyte activation. ICAM-3, the predominant LFA-1 ligand on resting blood DCs, is probably used for initial DC-T cell interactions (Hart and Prickett, 1993; Starling et al., 1995). 3. Migration of Dendritic Cells a. Patterns of Dendritic Cell Migration. An important attribute of DCs at various stages of their maturation is their mobility. This property enables DCs to move from the blood to peripheral tissues, and from peripheral tissues to lymphoid organs, where the pool of quiescent T cells recirculates. The selective migration of DCs, their residence in a given tissue, and their migratory capacity are tightly regulated events. The induced migration of DCs was first noted at the site of contact allergy (Lens et al., 1983; Silberberg-Sinakin et al., 1976). Transplantation

266

DIANA BELL et al.

of heart or skin is also accompanied by an efflux of DCs from the graft (Larsen et al., 199Oa,b, 1994). In normal lung tissue, a functionally and morphologically identical DC population exists within the epithelial lining of the conductive airways of both humans and rodents, forming a contiguous network analogous to the LC population in the epidermis (Lipscomb et al., 1995). Brief exposure to aerosolized bacterial endotoxin induces a transient increase (-50%) in the density of airway epithelial DCs for 2448 hr after exposure, suggesting active participation by DCs in the acute inflammatory response (Schon-Hegrad et al., 1991). Within the respiratory tract, inhalation of bacteria, viruses, or soluble protein antigens (McWilliam et al., 1996) rapidly recruits DC,, into the airway epithelium. The earliest detectable cellular response after inhalation of Morarella catarrhalis is the recruitment of putative class I1 MHC-bearing DC,,, into the airway epithelium, the initial wave arriving earlier than neutrophil influx. Unlike neutrophils, which rapidly transit through the epithelium and into the airway lumen, the DC,, remain within the epithelium during the acute inflammatory response. Here they differentiate and develop the dendritic morphology typical of resident DCs found in normal epithelium (McWilliam et al., 1994), subsequently migrating to the regional lymph nodes. In the intestinal lumen antigens are taken up by specialized epithelial cells (M cells) overlying the dome region of Peyer’s patches. Immature DCs, strategically located below the M cells, capture the incoming antigens (Ruedl et al., 1996) and migrate to the T cell areas of the same Peyer’s patches or draining mesenteric lymph nodes, where they present antigen to T cells (Kelsall and Strober, 1996). After intravenous injection of inert particles, particle-laden cells can be detected in the hepatic lymph (Kudo et al., 1997; Matsuno et al., 1996).These cells may represent DC,,, recently derived from monocytes, and recruited to the hepatic sinusoidsby phagocytosing Kupffer cells. These DC,, manifest temporary phagocytic activity for intravascular particles, which is in turn down-regulated on maturation and translocation from the sinusoidal area to the hepatic lymph (Cella et al., 1997~).

b. Control of Dendritic Cell Migration. Although the pathways of DC migration are relatively well characterized, the molecular mechanisms that control recruitment and migration of DCs are far less well defined. Chemotactic factors released by the target tissue and surface adhesins are involved in these processes (Girolomoni and Ricciardi-Castagnoli, 1997). Several approaches have demonstrated that IL-1 and TNF-a are involved in the activation and mobilization of Langerhans cells (LCs). In particular, contact allergens that induce emigration of Langerhans cells induce an accumulation of IL-1 and TNF within the epidermis (Enk et al., 1993a,b), and

DENDRITIC CELLS

267

antibodies to IL-1 and TNF inhibit contact allergen-induced sensitization and LC redistribution (Cumberbatch and Kimber, 1995). More directly, dermal injection of IL-1 or TNF induces a decrease in LCs within the epidermis, together with an increase in DCs in the draining lymph nodes (Cumberbatch et al., 1992,1994, 1997; Cumberbatch and Kimber, 1992). These cytokines act by down-regulating the surface expression of Ecadherin on LCs, thereby loosening their interactions with keratinocytes (Blauvelt et al., 1995; Jakob and Udey, 1998; Schwarzenberger and Udey, 1996; Tang et al., 1993). DCs likewise migrate from the kidney and heart in response to IL-1 and TNF (Roake et al., 1995). DCs can both produce and respond to chemokines, e.g., IL-8 (Zhou and Tedder, 1995a), MIP-la and MIP-1P, RANTES (Sozzani et al., 1995), and MIP-ly (Mohamadzadeh et al., 1996). In particular, DCs express high levels of mRNA for CCRl (receptor for RANTES), CCR2 (receptor shared by MCP-1 and MCP-3), CCR3 (receptor for eotaxin) (Rubbert et al., 1998), CCR5 (receptor for MIP-la, MIP-lP, and RANTES) (Sozzani et al., 1995),and CCR6 (receptor for MIP-3a) (Greaves et al., 1997; Power et al., 1997). CCR1, CCR5, and CCR6, which are expressed on DCimm, are down-regulated during maturation (Sozzani et al., 1998; Dieu et al., 1998). Conversely, CCR7, a receptor for MIP-3P, is lacking on DCi,, but is induced upon activation (Dieu et al., 1998; Sozzani et al., 1998). Importantly, MIP-3a is preferentially produced at sites enriched with DCi,, whereas MIP-3P is preferentially expressed within the paracortex of secondary lymphoid organs where DCmatmigrate (Dieu et al., 1998). Thus the coordinated expression of distinct chemokine receptors may play a critical role in the migration of DCs at various stages of maturation. The migration of DCs induced by bacteria is likely due to the capacity of LPS to stimulate many cell types to secrete cytokines and chemokines that modulate DC movement and maturation. These include M-CSF (Heufler et al., 1988; Witmer-Pack et d., 1987), TNF-a (Sallusto et al., 1995; Sallusto and Lanzavecchia, 1994), IL-1 (Koide et al., 1987), MIP-la, -lP, and -1y (Mohamadzadeh et al., 1996; Sozzani et al., 1996). Another heterogeneous multifunctional molecule involved in DC trafficking is CD44 ( Weiss et al., 1997). CD44 is a receptor for the extracellular matrix component hyaluronate, which is involved in lymphocyte homing and activation as well as spreading of tumor metastases. CD44 is encoded by a total of 20 exons, 7 of which form the invariant extracellular region of the so-called standard form (CD44s). By alternative splicing, up to 10 variant exons (CD44vl-v10) can be inserted into the cell membrane (Herrlich et al., 1993).The CD44 isoforms play an essential role in LC and DC functions, the CD44 isoforms being differentially modulated during the LC-dependent sensitization phase of contact hypersensitivity, LC activa-

268

DIANA BELL et a[,

tion, and migration from the skin, and DC adhesion to the paracortical T cell zones of peripheral lymph nodes. During their migration to peripheral lymph nodes, LCs and DCs up-regulate pan-CD44 epitopes and sequences encoded by CD44 variant exons CD44v4, v5, v6, and v7 (Weiss et al., 1997). 4. Costimulatoy Molecules The most reliable functional assessment of histocompatibility remains the mixed lymphocyte reaction (MLR), in which T cells proliferate in response to allogeneic antigen-presenting cells (APCs). DCs are at least 30- to 100-fold more efficient than other APC populations, including B cells and macrophages, in inducing the MLR (Steinman and Witmer, 1978; Van Voorhis et al., 1983). Numerous cytokines, including IL-12, IL-4, and IFN--y are released when DCs stimulate T cells in the MLR. Although CD4’ cells account for much of the T cell proliferation during the MLR, DCs can also stimulate CD8’ T cells without CD4’ help, although higher antigen-presenting cell doses are needed (Inaba et al., 1987; Young and Steinman, 1990).This implies that either antigen-presenting cells are killed during the course of the response or that stimulation is simply less efficient in the absence of CD4 help. DCs are also 10- to 50-fold more potent than monocytes or B cells in inducing T cell responses to fentomolar concentrations of superantigens (Bhardwaj et al., 1992, 1993). However, the unique and most critical function of DCs is their ability to prime naive T cells to proteins that require processing into peptides (Christinck et al., 1991; Croft et al., 1992). Antigen-loaded DCs and antigen-specific T cells form aggregates that constitute a microenvironment optimal for the development of an immune response (Flechner et al., 1988;Inaba and Steinman, 1984).The interaction between DCs and T cells is coordinated by several molecules. “Signal one” is represented by MHC-peptide complexes and is recognized by antigenspecific TCRs. The availability of TCR transgenic mice has allowed investigators to prove that the capacity of DCs to induce a primary antigenspecific T cell response to soluble antigens in vitro is 100- to 300-fold more efficient than that of any other APC (Croft et al., 1992; Macatonia et al., 1995). High levels of adhesins ICAM-1 (CD54), ICAM-3 (CDSO), LFA-3 (CD58), and & integrin (CD29), and cell binding and homing molecules LFA-1 (CDlla), LFA-2 (CD2),and LFA-3, enhance adhesion and signaling (Caux et al., 1994c; Freudenthal and Steinman, 1990; Larsen et al., 1992; Lenz et al., 1993;Young et al., 1992).A variety of accessory molecules, coexpressed on DCs (B7.1/CD80, B7.2/CD86, CD40) and interacting with ligands and counterreceptors on T cells, together constitute “signal two,” which is required to initiate T lymphocyte activation. Studies with antibod-

DENDRITIC CELLS

269

ies using human and mouse DCs have shown that CD86 on DCs is so far the most critical molecule for amplification of T cell responses (Caw et al., 1994c; Inaba et al., 1995). The interaction between CTLA-4/CD28 on T cells, and CD80-lCD86 on DCs, also appears to play a role in the regulation of type 1versus type 2 T cell development. In particular, B7.U CD80 rather orients toward type 1 responses, whereas B7.2/CD86 ligation rather skews toward type 2 responses (Freeman et al., 1995; Kuchroo et al., 1995).

5. Signaling of DCs Members of the TNFITNF Receptor Families. 1. TNF and TNF-R. The effects of TNF on DC progenitors were identified in the early 1990s (Caux et al., 1992a; Reid et al., 1992; SantiagoSchwarz et al., 1992). TNF enhances DC development through several mechanisms. In particular, TNF allows primitive hematopoietic progenitor cells to respond to IL-3 and GM-CSF following up-regulation of the P chain common to the IL-3/IL-5/GM-CSF receptor (Caux et al., 199213, 1993; Sato et al., 1993). Furthermore, TNF inhibits granulopoiesis (Caux et al., 1993),possibly by decreasing G-CSF-R expression. TNF-(r is particularly important in the final maturation of these cells and the effects appear to be mostly mediated through TNF-Rl/p55/CD120a (Lardon et al., 1997), although TNF-R2/p75/CD120b has been identified on DCs (McKenzie et al., 1995). 2. CD40KD4OL. The CD40 molecule, a member of the TNF-R family, is found on the surface of B lymphocytes, dendritic cells, hematopoietic progenitor cells, epithelial cells, and carcinomas (reviewed in Banchereau et al., 1994; Grewal et al., 1997; Van Kooten and Banchereau, 1996). The natural ligand for CD40 (CD40UCD154) is expressed on the surface of activated CD4' and CD8' T cells, basophils, and B cells, as well as on DCs. Although the DC-T interaction has been traditionally viewed as a one-way interaction whereby DCs activate T cells, there is now evidence that T cells may play an important role in activating DCs via CD40L-CD40 interactions (Fig. 6). This further enhances the T cell stimulatory capacity of DCs. Ligation of CD40 also increases DC viability (Caux et d., 199413; Ludewig et al., 1995) and induces DC maturation manifested by increased expression of CD80, CD83, and CD86 (Caux et al., 1994b; Sallusto and Lanzavecchia, 1994). Following CD40 ligation DCs produce numerous cytokines, including IL-1, TNF, chemokines, and, importantly, IL-12, a key cytokine for the generation of T h l responses (Cella et al., 1996; Macatonia et al., 1995). It is commonly accepted that macrophages represent the main source of IL-12 during immune responses to pathogens (Caux et al., 1993; Skeen et al., 1996; Takahashi et al., 1993; Trinchieri, 1995),

270

DIANA BELL et al.

FIG.6. The dialogue between dendritic cells and T cells. CD4+T cells recognize peptide presented by class I1 MHC on dendritic cells. Adhesion moleculesstrengthenthe interaction. This results in up-regulation of CD40 ligand on T cells. Triggering of CD40 on DCs permits cytokine production and up-regulation of CD80/CD86 (B7). The secreted cytokines further activate T cells and support their proliferation. The increased CD80/CD86 expression on DCs triggers CD28 and/or CTLA-4 on T cells. The T cells then secrete cytokines in turn, which will either further activate the DCs or act as autocrine T cell growth factors.

but studies with Toxoplasma and Leishmania suggest that DCs may indeed be the first cells to make IL-12 during an immune response (Gorak et al., 1998; Sousa et al., 1997). The production of chemokines may be important to recruit other antigen-specific cells (e.g.,CD8' T cells or B cells), whereas TNF may induce the apoptosis of nonspecific bystander T and B cells or act as an autocrine agent to keep the DCs in an activated state. Recently, CD40L-activated DCs were found to express decysin, a novel member of the disintegrin metalloproteinases, which include the enzymes that cleave the transmembrane TNF precursor into soluble TNF (Black et al., 1997; Moss et al., 1997). Interestingly, decysin appears to be expressed by the mature myeloidhonlymphoid DCs in germinal centers (Grouard et al., 1996) but not by those in the T cell areas. A few molecules are down-regulated in response to CD40 activation. Among these are CDla and the recently isolated DORA, a member of the CD8 family of receptors whose function on DC remains to be determined (Bates, 1998). The importance of CD40-dependent activation of DCs is illustrated in the hyper-IgM syndrome of humans and mice with congenital and experimental alterations of CD40L, respectively. These individuals display a syndrome more suggestive of a primary T cell deficit than a primary B cell deficit. In particular, they show considerably altered T cell priming that results in increased susceptibility to numerous pathogens (e.g., Leishmania, Pneumocystis) (Grewal et al., 1997).

DENDRITIC CELLS

271

The CD40-activated DCs can trigger T killer responses in vitro and in vivo in the absence of helper T cells (Bennett et al., 1998; Ridge et al., 1998; Schoenberger et al., 1998). As stated by Ridge et al. (1998), a conditioned DC can be a temporal bridge between a CD4+ T helper and a T killer cell. However, CD40 activation of DCs can be bypassed by inflammatory agents, as provided by an adjuvant (Bennett et al., 1998) or by viral infection (Ridge et al., 1998). Interestingly, LCs from CD40L-’mice fail to emigrate to the draining lymph node on antigen sensitization, resulting in decreased contact hypersensitivity ( Flores-Romo, personal communication). Of note, CD40L has also been identified on activated DCs (Pinchuk et al., 1996), though its role remains to be determined. 3. OX4OlOX4OL. Mature DCs constitutively express OX40L, the ligand for 0x40, which is another member of TNF-R family present on activated peripheral CD4+ T cells and a subset of CD8+ T cells. Ligation of OX40L on monocyte-derived DCs, which are at an intermediate and reversible stage of maturation, markedly enhances their development into typical mature DCs (Ohshima et al., 1997). Engagement of T cell CD40 promotes the development of anti-CD3-stimulated naive T cells into Th2 effectors producing high levels of IL-4, IL-5, and IL-13, but little IFNy. Conversely, blocking OX4OlOX4OL interaction in primary cultures containing naive T cells and allogeneic DCs, using anti-OX40L monoclonal antibodies, inhibits the development of IL-ML-5-secreting T cells (Oshima and Delespesse, Santa Fe, 1998). 4. RANK-WTRANCEIODF and RANWTRANCE-Wosteoprotegerin. A new member of TNF-R family derived from dendritic cells, RANK (receptor activator of NF-KB)/TRANCE-R (TNF-related activationinduced cytokine), and its ligand RANK-LSTRANCE, have been isolated and characterized (Anderson et al., 1997; Wong et al., 1997). RANK-L’ TRANCE expression is restricted to lymphoid organs and T cells (Wong et al., 1997). High levels of RANmRANCE-R are detected on mature DCs but not on freshly isolated B cells, T cells, or macrophages. FUNK/ TRANCE-R signals via TNF receptor-associated factor 2 (TRAF2) and increases DC survival by up-regulating bcl-x, expression, thereby providing another tool to enhance DC activity by prolonging viability (Wong et al., 1997). RANmRANCE-R augments the ability of DCs to stimulate naive T cell proliferation in the MLR and increases the survival of RANK-L‘ TRANCE-positive T cells generated with IL-4 and TGF-/3 (Anderson et al., 1997; Wong et al., 1997). More recently, the osteoclast differentiation factor (ODF) was found to be identical to TRANCE/RANK-L (Yasuda et al., 1998). This cytokine is present on the surface of stromal cells and is responsible for osteoclast differentiation. Osteoprotegerin (OPG), a molecule of the TNR-R family

272

DIANA BELL et nl.

that suppresses bone resorption (Simonet et al., 1997), binds to TRANCE/ RANK-IJODF, thereby inhibiting osteoclast differentiation (Suda et al., 1995). Thus TRANCE/RANK-IJODF appears to bind to two distinct molecules of the same family: RANK and OPG. However, distinct from the other agonist receptor-ligand pairings in this family (TNF, LTa, LTP and TNF-R1, TNF-RS), OPG acts as a soluble competitive inhibitor of the transmembrane receptor RANK. 5. FaslFasL. Fas/Apo 1(CD95) is expressed on human DCs generated in vitro by culturing CD34' HPCs with GM-CSF and TNF-a, and on Fas ligation DCs undergo apoptosis (Bjorck et al., 1997a). Surprisingly and in contrast to B cells (Garrone et al., 1995), fully mature DCs obtained after CD40 ligation are fully resistant to Fas ligation, possibly as a consequence of up-regulated bcl-2 expression. Parallel experiments with mature, interdigitating DCs isolated from tonsils have revealed that IDCs express Fas but do not enter apoptosis after Fas ligation, a finding correlating with their high level of bcl-2 (Bjorck et al., 1997a). Other mechanisms should therefore be pursued to explain the in vivo disappearance of antigenloaded, mature DCs during an immune response (Ingulli et al., 1997). Studies in mice have also shown that DCs of the lymphoid lineage express a FasL (Lu et al., 1997; Suss and Shortman, 1996), which may be distinct from the classical one (K. Shortman, personal communication). 6. Enzymes Because of their potent antigen-presenting capacity, DCs are expected to express an enzymatic armamentarium tailored to the degradation of virtually any antigen into peptides. These antigens include not only proteins but also large particles such as viruses, bacteria, mycobacteria, parasites, and apoptotic bodies. Therefore, their processing undoubtedly requires a very diverse set of enzymes, but little has been published to date on this topic. Cathepsin D, an asparagyl protease, has been identifiedwithin human and murine DCs (Lutz et al., 1997; Sallusto et al., 1995). Furthermore, a novel member of the disintegrin metalloproteinases, decysin, has been identified using cDNA substraction libraries (Mueller et al., 199713). Although absent from DC,, and DCimm, decysin is induced to high levels following spontaneous and CD40-induced maturation. In vivo, decysin appears restricted to germinal center dendritic cells (Grouard et al., 1996), but its functions remain unknown. As discussed earlier, type IV collagenase, identified in Langerhans cells, facilitates the migration of these cells across basement membranes. Genomic analysis of DC libraries has also permitted the identification of numerous protease inhibitors. In particular, several cystatins, which are inhibitors of cystein proteases, have been identified (S. Lebecque, C. Caw, and G. Zurawski, personal communication). A

DENDRITIC CELLS

273

serine protease inhibitor (serpin) has also been identified from subtractive cDNA libraries (Mueller et al., 1997a). This serpin is absent from monocytes, B cells, and T cells, but is expressed in CD40-activated DCs. Additional studies demonstrate that proteases and protease inhibitors are also important in the presentation of antigens, most particularly in the routing of the class I1 MHC antigens within DCs. DCs express cathepsin S, an enzyme that has been shown to play a role in the processing of the invariant chain in B cell lines. Indeed, blocking cathepsin S with the specific and irreversible inhibitor LHVS (Riese et al., 1996;Villadangos et al., 1997) results in a significantly decreased export of class I1 MHC to the DC surface, while the total content remains unchanged. In DCimm, inefficient Ii chain cleavage due to low cathepsin S activity leads to the transport of class I1 MHC-Ii chain complexes to lysosomes. In contrast, elevated Cathepsin S activity in DC,,, results in efficient transport of class I1 MHC molecules to the cell surface. The increased cathepsin S activity observed following DC maturation is not due to its increased transcription but to the decreased presence of its specific inhibitor, cystatin C, a cystein protease inhibitor (Henskens et al., 1996; Pierre and Mellman, 1998); Maurer and Sting, personal communication).

7. Natural Killer Phenotype of Dendritic Cells Rat spleen and thymus dendritic cells express low levels of the natural killer cell receptor protein 1 (NKR-P1) (Josien et al., 1997). NKR-P1, a disulfide-linked homodimer expressed by all NK cells and a small subset of T cells, belongs to group V of the C-type lectin superfamily. This superfamily also includes the CD69, Ly-49, and CD54 molecules (Lanier, 1997; Moretta and Moretta, 1997).The rat NKR-P1 molecule is an activation receptor that leads to stimulation of granule exocytosis. The expression of NKR-P1 on DCs is strongly up-regulated after overnight culture. In addition to expressing this typical NK cell marker, rat spleen DCs, but not thymus DCs, are able to kill the NK cell-sensitivetarget YAC-1. Human dendritic cells generated in vitro by culturing monocytes or CD34+ HPCs also express NKR-PI, ligation of which results in Ca" fluxes and IL-12 secretion (Poggi et al., 1997). It is not presently known whether human DCs express any functional NK activity.

8. Calcium Channels of Dendritic Cells Monocyte-derived DCs display L-type calcium channels that mediate the influx of extracellular Ca2+(Poggi et al., 1998a). These Ca" channels are composed of three transmembrane subunits (alC, 7,and a26 complex) and one cytoplasmic chain (the pl chain) (Catterall and Striessnig, 1992), comparable to those of skeletal and cardiac muscle. The dihydropyridine

274

DIANA BELL et al.

derivative nifedipine, which specifically binds to the a l C chain, prevents apoptotic body engulfment and IL-12 secretion by DCs (Poggi et al., 1998b).Importantly, HIV-1 Tat also blocks these two DC functions (Zocchi et al., 1997) by acting on these Ca2' channels (Poggi et al., 1998a). This may explain the altered function of circulating DCs in AIDS patients (Macatonia et al., 1989). 111. Ontogeny of Dendritic Cells

Studies of DCs have been greatly hampered in the past by difficulties in isolating these cells from tissues or blood in substantial numbers and with purity. Great progress has been achieved since the establishment of several procedures for the in vitro generation of murine and human DCs from progenitors in bone marrow, placental and umbilical cord blood, and cytokine-mobilized peripheral blood (Cam et al., 1992a, 1996a; FloresRomo et al., 1997; Inaba et al., 1992a,b; Santiago-Schwarz et al., 1992; Strunk et al., 1996; Szabolcs et al., 1995,1996; Young et d.,1995). Methods have also been developed to generate DCs from blood monocytes (Bender et al., 1996; Reddy et al., 1997; Romani et al., 1994, 1996; Sallusto and Lanzavecchia, 1994). The current understanding of DC ontogeny is summarized in Fig. 7. A. GENERATION OF MOUSEDENDRITIC CELLLINES

Addition of GM-CSF to mouse blood or bone marrow results in the formation of DC aggregates that originate from Ia-negative nonadherent cells (Inaba et al., 1992b, 1993). Long-term dendritic cell lines have also been generated from fetal tissues using either GM-CSF (Winder et al., 1997) or stromal cell culture supernatants (Takashima et al., 1995). B. GENERATION OF DENDRITE CELLSFROM CD34' HEMATOPOIETIC PROGENITOR CELLS 1. TNF, in Association with GM-CSF or ZL-3, Znduces Development of DCs from CD34' HPCs CD34' HPCs, isolated from cord blood or bone marrow mononuclear cells, can be induced to proliferate in vitro in response to several cytokines in combinations. TNF strongly potentiates the proliferation of CD34' HPCs induced by either IL-3 or GM-CSF (Caw et al., 1990, 1992a; reviewed in Cam and Banchereau, 1996). After 12 days, a majority of cells express CDla and acquire typical DC features according to morphology, phenotype (CD40, CD4, CD54, CD80, CD86; high levels of class I1 MHC, lack of CD64 and CD35), presence of Birbeck granules (specific for LCs) in 20% of cells, and potent capacity to induce proliferation of naive T cells

DENDRITE CELLS

DCpre

275

DCimm DCmat

-- - - - - - - - - - - - - - -* [LYMPHOCYTES]

'-

FIG.7 . Dendritic cell ontogeny. The pluripotent stem cell gives rise to myeloid and lymphoid progenitor cells. The lymphoid progenitor may give rise to the CDllc- DC precursor that may differentiate into thymic dendritic cells, as well as a subpopulation of parafollicular, interdigitating dendritic cells in the T cell-rich areas of secondary lymphoid tissues. The myeloid DC progenitor differentiates in uitro into a CDla+ precursor that yields Langerhans cells and a CD14+ precursor that may yield germinal center DCs. The in uitro-generated CD14' precursor may be the equivalent of circulating monocyte precursors of DCs, although IL-4 has not proved to be as critical in the DC differentiation of the former.

to alloantigen and of CD4' T cells to soluble antigen (Caw et al., 1992a,b, 1995, 1996b). TNF is also required for the clonogenic growth of pure human DC colonies in the additional presence of CM-CSF (Younget al., 1995). Addition of stem cell factor (SCF, c-kit ligand) andl'or Flt-3L increases the yield of DCs but does not directly affect DC differentiation in vitro (Siena et al., 1995; Strobl et al., 1997; Young et al., 1995). For cultures performed under serum-free conditions, TGF-P may be required for generation of DCs with characteristics of LCs, e.g., Birbeck granules and Lag antigen (Riedl et al., 1997; Strobl et al., 1997). The maturation of CD34+HPCs into DCs also involvesprotein kinase C-mediated signaling and can be partly induced by phorbol esters alone (Davis et al., 1998).

276

DIANA BELL et a1

2. CD34+ HPCs, Cultured in the Presence of GM-CSF + TNF, Differentiate along Two Independent DC Pathways Many candidate DCs are CDla+CD14- during the later stages of culture (approximately days 12-14) from CD34+ HPCs in FCS-containing medium; CDla expression is lost with final maturation, however, just as CD83 expression increases. When examined earlier, however, two DC subsets emerge independently by days 5-7, as defined by the exclusive expression of CD la and CD14. Both precursor populations eventually mature into TNF: CDla+ CD14- cells give rise to DCs in response to GM-CSF CDlat CD14- LCs (Birbeck granules, Lag', E-cadherin'), while CDlaCD14t intermediates develop into interstitial (dermal) DCs that are also CDla+ CD14- (lack of Birbeck granules, Lag-, E-cadherin-, CD2+, CD9+,CD68+,Factor XIIIa' ). The CDla- CD14' intermediate is bipotential, however, in that it can alternatively differentiate into CDla- CD14' macrophages on reculture without exogenous cytokines; M-CSF can enhance viability and support possibly one additional round of cell division (Cauxet al., 1996a; Szabolcs et al., 1996).Primitive CD34' CD38- hematopoietic progenitors can also develop into interstitial DCs when cultured over thymic stromal monolayers in the absence of exogeneous cytokines (Miralles et al., 1998). The commitment to either pathway may have already occurred at the level of the CD34' HPCs. For example, a minor population of CD34+ HPCs, which can be increased by exposure to TNF, coexpresses CD86, and this CD34' HPC subset exhibits bipotential differentiation capacity into macrophages or dendritic cells (Ryncarz and Anasetti, 1998). CD34+ CLA+ cells also reportedly give rise in vitro to Langerhans cells, whereas CLA- precursors yield interstitial DCs (Strunk et al., 1997). Although the two populations are equally potent in stimulating naive CD45RA+ cord blood T cells, each also displays specific activities (Caux et al., 1997). In particular, interstitial DCs demonstrate a potent and long-lasting antigen uptake activity (FITC-dextran or peroxidase) that is about 10-fold higher than that of Langerhans cells and is mediated by mannose receptors. The high efficiency of antigen capture by interstitial DCs correlates with the expression of nonspecific esterase activity, a tracer of the lysosomal compartment that is not observed in Langerhans cells. A striking difference between the two populations is also the unique capacity of interstitial DCs to induce naive B cells to differentiate into IgM-secreting cells in response to CD40 ligation and IL-2. Thus, although T cell priming is accomplished by both DC populations, one can envision that the two different pathways of DC development are preferentially specialized: (1)the Langerhans cell type, which would be mainly involved in cellular immune responses, and

+

DENDRITIC CELLS

277

(2)the interstitial DC, which would be dedicated to the initiation of cognate T cell help and humoral immune responses by B cells. C. GENERATION OF DENDRITIC CELLSFROM BLOODMONONUCLEAR CELLS Using GM- CSF and IL-4 (Cella et al., 1997a; Chapuis et al., 1997; Pickl et al., 1996; Porcelli et al., 1992; Roinani et al., 1994, 1996; Sallusto and Lanzavecchia, 1994; Zhou and Tedder, 1996) or GM-CSF and IL-13 (Piemonti et d., 1995), lymphocyte-depleted, adherent blood mononuclear cells as well as purified CD14' monocytes yield DCs that can be maintained in culture for weeks in the presence of cytokines and FCS-containing medium. The resulting DCs display features of DCi,, (low levels of CD80, CD86, intracytoplasmic expression of MHC class 11, an efficient antigencapturing and -processing capacity, and a weak capacity to prime naive T cells) (reviewed in Caw and Banchereau, 1996). When stimulated by T cell signals such as CD40L (Sallustoet al., 1995; Sallusto and Lanzavecchia, 1994) or signals activating ceramide mediators, such as LPS, TNF, and IL-1 (Sallustoetal., 1996),these DCsundergo the phenotypicand functional changes of a maturation process. The irreversible maturation of these cells after culture in media with human serum or plasma necessitates addition of macrophage-conditioned medium that contains at least IL-1, IL-6, TNF, and IFN-a (Bender et al., 1996; Reddy et al., 1997; Romani et aE., 1996), and undoubtedly other as yet undefined inflammatory mediators such as prostaglandm (Jonuleit et al., 1997a). Addition of TGF-P to cultures of monocytes with GM-CSF + IL-4 results in the generation of DCs with properties of Langerhans cells (Geissmann et al., 1998).Thus lymphocytedepleted, peripheral blood mononuclear cells, i.e., monocytes, represent a considerable pool of circulating precursors of DCs and macrophages. Surprisingly, highly purified neutrophil granulocyte-committed precursors can also be driven to acquire DC characteristics when cultured in the presence of GM-CSF + TNF-a + IL-4 (Oehler et al., 1998). This indicates that cells from the innate immune system can be reprogrammed to become inducers of the adaptive immune system even at a penultimate stage of terminal differentiation. PATHWAY OF DENDRITIC CELLDEVELOPMENT D. A LYMPHOID The myeloid or nonlymphoid model of DC development does not apply to the thymus, where DCs are indeed present to induce death of selfreactive thymocytes (Ardavin, 1997; Brocker et al., 1997). Murine thymic DCs express a peculiar phenotype, with lymphoid cell markers such as the CD8 aa homodimer, CD2, and BP1 (Vremec et al., 1992; Wu et al., 1995). A subgroup of these DCs is found in spleen (50% of DCs) as well

278

DIANA BELL el ol

as lymph node (Inaba et al., 1997). These cells originate from a progenitor cell that can also give rise to T cells and B cells (Ardavin et al., 1993; Wu et al., 1996). The differentiation of this precursor in vitro is independent of GM-CSF and can be achieved by combining TNF, IL-1, IL-3, IL-7, SCF, Flt-3L, and CD40L (Saunders et al., 1996). Human CD34' HPCs have also been identified that can give rise to T cells, B cells, and DCs in vitro (Galy et al., 1995; Res et al., 1996). A mature human lymphoid DC, however, has not been identified to date. The function of these lymphoid DCs is discussedwith regard to immune tolerance in Sections V,D andV,E.

E. FLT-3 LICANDAND DENDRITIC CELLS The search for new receptor tyrosine kinases led to the discovery of a murine gene termed fetal liver kinase 2 (FLK-2) and of a human gene termed FMS-like tyrosine kinase 3 (FLT-3),which are homolog. FLT-3 also has substantial homology to c-kit, c-fms, and PDG receptor genes, which play a central role in hematopoiesis. FLT-3 is expressed on early, nonerythroid hematopoietic progenitor cells, as well as more mature hematopoietic cells (Lyman and Jacobsen, 1998). Both human and murine ligands for FLT-3 were cloned and have been shown to encode a type I transmembrane protein and a soluble protein following alternative splicing rather then proteolysis (Lyman and Williams, 1995). Flt-3L is found on various stromal cells and in a variety of tissues. In uitro, Flt-3L acts in concert with other cytokines to induce proliferation of early progenitors (Jacobsen et al., 1995; Muench et al., 1995; Shah et al., 1996), but it has never been shown to have a differentiating capacity alone ex uiuo. In contrast, in viuo administration of Flt-3L results in the blood recirculation of CD34+ HPCs, and a striking enlargement of spleen, lymph nodes, and liver. The organs show increased levels of B cells but the most notable feature is an accumulation of dendritic cells (Maraskovsky et al., 1996; Pulendran et al., 1997; Shurin et al., 1997). F. CURRENT VIEW OF THE PATHWAYS OF DENDRITIC CELLDEVELOPMENT Although DCs derive from proliferating CD34' progenitor cells, three stages of DC differentiation are being distinguished, namely, patrolling DC,,, tissue-residing DCimm, and DCmatfrom secondary lymphoid organs. DCs are also composed of distinct subpopulations, in many cases related to distinct precursors (Fig. 7). These precursors include CD4' CD14+ monocytes, and CD4' CD14- CDllc- as well as CD4' CD14- CDllc+ cells. Monocytes are primarily identified in the blood, whereas CDllcand CDllc+ precursors can be identified in blood and secondary lymphoid organs. CDllc- cells remain localizedwithin T cell-rich areas, but CDllc'

DENDRITIC CELLS

279

cells migrate into B cell follicles as germinal center DCs (not to be confused with follicular DCs). We currently think that the recirculating DC,, eventually colonize tissues to become DCi,,. Although DCs are found in very low numbers in virtually every tissue, there is emerging evidence that epithelial DCs (Langerhans cells) and interstitial DCs represent alternative pathways of differentiation (Cam et al., 1996a). The relationship between the above three stages of DC differentiation and these two populations of immature cells remains unresolved. It is possible that CDllc- precursors eventually differentiate into DCs within thymus and secondary lymphoid organs, where they establish and maintain tolerance. Blood and lymph also contain a very minor population of maturing DCs on their way from tissues to secondary lymphoid organs (reviewed in Banchereau and Steinman, 1998). IV. Maturation of Dendritic Cells

A. STIMULATORS OF MATURATION During migration after antigen loading, DCs undergo changes in phenotype and function as part of their maturation. This represents a control point for the onset of immunity. As discussed earlier, maturation includes a coordinate series of changes, which include down-regulation of macropinocytosis and Fc receptors, transition of the class I1 MHC-rich intracellular compartments to cell surface MHC peptide complexes, and the upregulation of accessory molecules (Cella et al., 1997b; HeuAer et at., 1988; Pierre et al., 1997; Sallustoet al., 1995;Witmer-Packet al., 1987;Yamaguchi et al., 1997). A variety of agents contribute to DC maturation. These include cytokines such as IL-1, GM-CSF, and TNF-a, released by a variety of cell types, e.g., keratinocytes, mast cells, macrophages, or T cells, as well as other T cell products such as IL-2, and bacterial products such as LPS. Some viruses, e.g., influenza virus, can also directly induce the maturation of DCs ((Ridge et al., 1998);Lanzavecchia, personal communication), Phagocytosed bacteria also induce DC maturation with an increased synthesis of MHC class I and class I1 molecules. In particular, bacteria stabilize MHC class I complexes and allow efficient loading of MHC class I molecules (Rescigno et al., 1998). Intramembrane diffusible mediators such as ceramides, involved in transducing signals that originate from a variety of cell surface receptors, down-modulate antigen capture and thus mimic one step of DC maturation (Sallusto et al., 1996). In this context, the potent DC maturation ability of LPS may be related to its structural similarity to ceramides. The transcription factors Rel/NF-KB proteins (p50, p52, p65, c-Rel, Rel-B) play an important role in the biology of DCs, from their ontogeny to their maturation. Physiologically high levels

280

DIANA BELL et al

of p50, p52, and Rel-B are restricted to accessory cells of the immune system, which include DCs and macrophages in the T cell zones (Carrasco and Bravo, 1993; Feuillard et al., 1996). Studies have localized Rel-B to interdigitating DCs in lymph nodes as well as scattered germinal center cells, but not to undifferentiated DCs in normal skin (Pettit et al., 1997). Active nuclear Rel-B has been detected by supershift assay only in differentiated DCs derived froin either blood precursors or monocytes, and in B cells, implying that Rel-B may specifically transactivate genes within the nucleus that are critical for APC function (Pettit et al., 1997). Rel-B knockout mice have no DCs in their altered lymphoid organs, although Langerhans cells are present (Burkly et al., 1995; Salomon et al., 1994). It is not clear whether the lack of DCs within secondary lymphoid organs results from altered cell migration, cell survival, or cell maturation. B. IL-10 AS AN INHIBITOR OF DENDRITIC CELLMATURATION Early studies have shown that IL-10 inhibits the antigen-presenting capacity of monocytes/macrophages (de Waal Malefyt et al., 1992; Moore et al., 1993). Subsequently, IL-10 was shown also to inhibit the APC functions of in uitro-generated DCs (Caw et al., 1994a; Steinbrink et al., 1997; Thomssen et al., 1995) as well as freshly isolated tonsillar DCs and epidermal LCs (Caw et al., 1994a; Peguet-Navarro et al., 1994), through mechanisms that have not been fully established. Some studies indicate an inhibition of CD80 and CD86 expression (Buelens et aE.,1995; Mitra et al., 1995; Ozawa et al., 1996; Steinbrink et al., 1997). Others fail to identify any alteration of CD80/CD86 or class I1 MHC peptide expression (Morel et al., 1997). The lack of consensus is likely to stem from major differences in experimental protocols and differential sensitivity of DCs to IL-10 with regard to stage of maturation. In this respect, IL-10 inhibits DC expression of CD83 and CD86, as well as secretion of IL-8 and TNF when DCs have been activated with LPS; IL-10 does not have the same effect when DCs are terminally matured by exposure to CD40L (Buelens et al., 1997a). Furthermore, the assays used to measure DC function should be carefully assessed, because the alterations may be very subtle. For example, IL-10-treated DCs appear to induce the differentiation of naive T cells toward the Th2 pathway (Allavena et al., 1998; De Smedt et al., 1997; Liu et al., 1997a). One of the most critical points of action by IL-10 on DCs concerns their ontogeny: IL-10 has been shown to inhibit the IL-4 + GM-CSF-induced proliferation of monocytes into DCs (Buelens et al., 199%; More et al., 1997),to the benefit of macrophages (Allavenaet al., 1998). In unpublished studies (F. Rousset, C. Caw, and J. Banchereau) we also found that IL10 could prevent the GM-CSF + TNF-dependent generation of DCs

DENDRITIC CELLS

28 I

from CD34' HPCs. IL-10 acts as an antagonist to TNF, facilitating the generation of granulocytes and inducing the apoptosis of emerging DCs. IL-lO-induced apoptosis of DCs has indeed been described with freshly isolated LCs (Ludewig et al., 1995).This IL-10 inhibition of DC generation has been further demonstrated in vivo, where IL-lO-expressing tumor cells blocked GM-CSF-induced accumuIation of DCs (Qin et al., 1997). V. Interactionsof Dendritic Cells with T Cells

A. ASSOCIATION OF DENDRITE CELLSWITH T CELLSIN Vwo Substantial numbers of DCs are found in the T cell areas of secondary lymphoid tissues, where they are termed interdigitating DCs ( IDCs). These IDCs form a network through which T cells continually recirculate. These DCs, which express mature epitopes that include CD80, CD83, and CD86, are nevertheless heterogeneous, as best illustrated in studies of mouse spleen and lymph nodes. Here the T cell areas are enriched in CD8aa' lymphoid DCs, and the CD8- myeloid DCs are mostly localized within the marginal zone (Pulendran et al., 1997). New observations that DCs within the T cell areas also express high levels of self-antigens and functional Fas-ligand capable of inducing CD4 T cell death suggest the presence of at least two sets of DCs in the T cell areas: (1) a migratory myeloid pathway that brings antigens from the periphery and induces immunity and (2)a lymphoid pathway that presents self-antigens and maintains tolerance (Steinman et al., 1997).The heterogeneity of the DC population in animals is also illustrated in TGF-P-'- mice that lack Langerhans cells and a subpopulation of Ep-Cam' DCs within lymph nodes (Borkowski et al., 1996). Similarly clear distinctions have not yet been identified in human tissues. DCs in the periphery acquire antigens and migrate to the T cell areas to initiate immunity. Although many in vitro and in vivo experiments argue strongly for the critical role of DCs in initiating immune responses, formal in vivo evidence has only recently been discovered. For example, proliferating T cells have been identified in contact with the DCs of the T cell-rich areas of secondary lymphoid organs after injection of either allogeneic cells (Kudo et al., 1997), superantigens (Luther et al., 1997), or protein antigen (Ingulli et al., 1997). This last study used adoptive transfer of fluorochrome-labeled, ovalbumin-loaded DCs and T cells expressing a receptor specific for an OVA peptide-MHC complex. This interaction results in the expansion of antigen-specific T cells that peaks at 96 hr, even though antigen-pulsed DCs disappear after 48 hr. The likely elimination of antigen-loaded DCs represents an efficient way to limit the development of T cell responses.

282

DIANA BELL d al.

The survival of mature CD4 T cells is also dependent on the presence of MHC class II-positive DCs (Brocker, 1997). After grafting of MHC

class II-positive embryonic thymic tissue depleted of bone manow-derived cells, an accumulation of CD4 T cells in the blood and secondary lymphoid organs can be observed only in mice expressing class I1 MHC on DCs but not in mice completely deficient in class I1 MHC (Brocker, 1997). *

B. DENDRITIC CELLSEXPRESS CYTOKINES THATAITRACTT CELLS To attract and select antigen-specific T cells, DCmatsecrete multiple chemokines, including RANTES, MIP-ld&, and IL-8. Novel chemokines are presently being identified using DC cDNA libraries. In particular, human dendritic cells present in the germinal center and T cell areas of secondary lymphoid organs express high levels of DC-CK1, which, in contrast to RANTES, MIP-la, and IL-8, preferentially attracts naive CD45M' T cells (Adema et al., 1997). DC,, from the T cell-rich areas secrete MIP-SP, which attracts naive CD4' T cells and CD8 cells (Ngo et al., 1998). Thymic dendritic cells also express TECK, a novel CC chemokine that may be involved in T cell development (Vicari et al., 1997). Finally, DCs also secrete IL-15, which is able to chemoattract T lymphocytes (Jonuleit et al., 199713). C. DENDRITIC CELLSCANDIRECTLY PRIMECD8' T CELLS DCs can stimulate an MLR from highly purified CD8+ T cells, though higher numbers of APCs are needed when compared to the response of CD4' T cells (Inaba et al., 1997; Young and Steinman, 1990). Because allospecific CTLs are generated rapidly during these cultures, the need for higher APC numbers may indicate that the APCs are killed during the course of the response. Indeed there is now ample evidence that DCs represent excellent CTL targets. Alternatively and as discussed earlier, this lower efficiency may be due to a suboptimal maturation of the DCs because of the lack of helper T cell activation, mostly dependent on CD40 ligation (Bennett et aZ., 1998; Ridge et al., 1998; Schoenberger et al., 1998). The availability of DCs devoid of class I1 MHC has facilitated confirmation of this unique functional property of DCs to prime CD8' cells independently of CD4' help. CD8' T cells specific for alloantigens, as well as for tumor and viral antigens, can be generated by DCs. In particular, class I1 MHC' class I1 MHC- epidermal Langerhans cell lines derived from fetal skin can activate allogeneic CD8' T cells in vitro (Elbe et al., 1994), as well as prime the immune system against transplantation antigens (Lenz et al., 1996) and exogenous hepatitis B (Bohm et aZ., 1995)in uiuo. More recently, skin- and bone marrow-derived DCs obtained from MHC class I P C57BY6 mice and pulsed with dinitrofluorobenzene (DNFB) induced

DENDRITIC CELLS

283

contact sensitivity and generated hapten-specific CD8+ T cells in lymphoid organs (Krasteva et al., 1998). Class 1' 11- DCs are as potent as class It 11; DCs in priming for contact sensitivity, further demonstrating that activation of effector CD8' T cells can occur independently of CD4+ T cell help. Conversely, class I- 11' DCs cannot immunize for DNFB contact sensitivity,whereas they can induce a delayed-type hypersensitivityreaction to proteins. Importantly, hapten-loaded class I- II+ DCs down-regulate hapten-induced inflammatory responses through the induction of CD4' regulatory T cells. TOLERANCE D. DENDRITIC CELLSAND CENTRAL The majority of dendritic cell studies in the field have focused on immune responsiveness, but central and peripheral tolerance may also require DCs. T cells bearing receptors with high affinity for self-antigens are responsible for the generation of autoimmune diseases. Therefore, potentially autoreactive thymocytes must be eliminated or inactivated in normal individuals. Induction of tolerance in thymocytes occurs by negative selection controlled by the thymic stroma, and in particular by the thymic DCs. In MHC-disparate thymic grafts depleted of bone marrow-derived APCs, these T cells show only limited tolerance to the MHC antigens of the thymic grafts (Von Boehmer and Schubiger, 1984).Second, purified splenic DCs are tolerogenic when injected into APC-depleted thymi (Matzinger and Guerder, 1989), findings that have been extended to parents of F1 bone marrow chimeras (Gao et al., 1990). Third, intrathymic inoculation of autoantigen (myelin basic protein) (Goss et al., 1994) or dloantigens (Oluwole et al., 1995) reduces an autoimmune reaction or allograft rejection, respectively. Finally, transgenic mice in which genes are specifically expressed on medullary DCs using a C D l l c promoter demonstrate that negative, but not positive, selection can be induced by DCs in vivo (Brocker et al., 1997). Within the thymus, DCs do not behave as classic APCs. Effective interaction between thymic DCs and thymocytes does not induce the same series of T cell activation events that occur in the periphery. The initial interaction determines the negative selection of autoreactive thymocytes and leads to the generation of T cell tolerance (Ardavin, 1997). The biology of thymic DCs is not fully understood. Located mainly in the corticomedullary border and medulla, human thymic DCs express high levels of class I and class I1 MHC molecules (Crowleyet d., 1989; Guillemot et al., 1984; Kyewski et al., 1986), CD45 and C D l l c (Agger et al., 1992; Sotzik et al., 1994), intercellular adhesion molecules ICAM-1 and CD44 (Lafontaine et al., 1992; Sotzik et al., 1994), LFA-1, and costimulatory molecules CD40 and CD86 (Lenschow et aZ., 1996).Although some mouse

284

DIANA BELL et al.

thymic DCs express high levels of CD8 but low levels of CD4, some human thymic DCs express low levels of CD8 but high levels of CD4 (Ardavin, 1997; Sotzik et al., 1994).

E. PERIPHERAL TOLERANCE Peripheral selection, dependent on TCR-ligand interactions, differs from thymic selection with regard to specificity and mechanism, requiring binding of antigen to the TCR and induction of T cell clonal expansion. In contrast, tolerance to self-antigens that are restricted to the periphery may occur through the anergy of self-reactive T cells as a consequence of down-regulation of the Cup TCR and CD8 (Rocha andvon Boehmer, 1991). Protection from myelin basic protein (MBP)-induced encephalitis can be induced by intravenous injection of thymic DCs either pulsed with the immunodominant peptide of MBP or isolated from thymi inoculated in viwo with MBP (Khoury et al., 1995). Such a property is not restricted to thymic DCs, however, because intravenous administration of antigenpulsed LCs or splenic DCs can selectively suppress delayed-type hypersensitivity responses (Morikawa et al., 1992, 1993). As discussed earlier, DCs may be rendered tolerogenic after incubation with IL-10 (Enk et al., 1993a,b) or by exposure to UV irradiation, which induces DNA damage and perturbs the expression of CD80/CD86 (Simon et al., 1991; Vink et al., 1996, 1997). In both cases immature DCs seem more susceptible to exogenous factors that can render them tolerogenic, than are fully mature DCs (Buelens et ul., 1997b; Young et al., 1993). Interestingly, the induced tolerance may represent a skewing of the immune response toward the type 2 pathway (Morikawa et al., 1995). Such skewing may also explain (1)the immune privilege of the anterior ocular chamber, where high levels of TGF-P2 may alter the function of local DCs (Streilein, 1997), and (2) oral tolerance induced by low doses of antigens (Weiner, 1997).In contrast, oral tolerance induced by high doses of antigen appears to depend on T cell deletion and anergy. Two distinct DC populations isolated from mouse spleen and lymph node, CD8at and CDBa-, may explain the induction of tolerance versus immunity. Those DCs that bear CD8a- express Fas ligand and restrict peripheral CD4 T cell responses by initiating Fas-mediated apoptosis (Lu et al., 1997; Suss and Shortman, 1996), whereas CD8a- DCs induce a vigorous proliferative response in CD4' T cells. The proliferative response of CD8 T cells is markedly less on stimulation by CD8' DC than by conventional CD8- DCs, but this reduced proliferation occurs without involving FasL-induced apoptosis, and is completely reversed by the addition of exogenous IL-2 (Kronin et al., 1996).

DENDRITE CELLS

285

VI. Interactions of Dendritic Cells with B Lymphocytes

A. AN ABBREVIATED VIEWOF A B LYMPHOCYTE LIFE Naive B lymphocytes, generated within the bone marrow, migrate into the secondary lymphoid organs where they either die or are recruited into the recirculating B cell pool. Without antigenic encounter, these naive B cells recirculate through the blood, secondary lymphoid organs (tonsils, lymph nodes, spleen, and mucosal-associated lymphoid tissues), lymph, and back to the blood. However, once antigendpathogens are localized within a secondary lymphoid organ, the recirculating naive B lymphocytes (and naive T lymphocytes) bearing specific antigen receptors are retained. During T cell-dependent immune responses, naive B cells with specific antigen receptors are activated in association with antigen-specific T cells and interdigitating dendritic cells within the extrafollicular areas. The activated B blasts undergo either terminal differentiation toward plasma cells or become germinal center (GC) founder cells that will migrate into primary follicles or the dark zone of established germinal centers in secondary follicles. The GC founder cells undergo clonal expansion and differentiation into proliferating centroblasts that form and sustain the dark zones. At this level, point mutations are introduced into the immunoglobulin variable (IgV) region genes, in a stepwise fashion. Three types of mutants can be generated, including high-affinity, low-affinity, and autoreactive mutants, which compose the basal light zone of the GC. The survival of these somatic mutants depends on their binding to the low levels of antigenantibody immune complexes on the surface of follicular dendritic cells (FDCs). High-affinity mutants capture antigen, process it, and present it to GC T cells. Autoreactive mutant clones and low-affinity mutants are deleted. The selected high-affinity centrocytes present processed antigen to antigen-specific T cells, which are induced to express CD40 Iigand (CD40L) and secrete cytokines, including IL-4 and IL-10. These are all key elements for B cell survival, proliferation, and isotype switching. This cognate T cell-B cell interaction results in the expansion and isotype switching of high-affinity centrocytes. Finally, the high-affinity isotypeswitched centrocytes differentiate into memory B cells in the presence of prolonged CD40L signaling or into plasma cells when CD40L signaling is removed. During secondary humoral immune responses, recirculating memory B cells can be activated in extrafollicular areas, giving rise to plasma cells and GC founder cells. (see also reviews in Liu and Banchereau, 1996b; Kelsoe, 1996; MacLennan, 1994; and Zrniizunologicd Reviews 156 (1997),which is dedicated to the anatomy of antigen-specific immune responses).

286

DIANA BELL et al

B. FOLLICULAR DENDRITIC CELLSAND GERMINAL CENTER DENDRITIC CELLS Thirty years ago, the follicular dendritic cell was identified as a new cell type within both primary and secondary follicles, based on its ability to trap antigens in the form of immune complexes on the surface of complicated dendritic processes (Nossal et al., 1968a,b; Said et al., 1997; Szakal and Hanna, 1968).The origin of these FDCs, hematopoietic versus nonhematopoietic, has been a much debated issue. However, it is now clear that there are two very distinct populations: (1)the follicular dendritic cell of mesenchymal origin (fibroblast-like) (Matsumoto et al., 1997) and (2) the germinal center dendritic cell (Grouard et al., 1996)or antigen-transporting cell (Szakal et al., 1989) of hematopoietic origin. 1. Follicular Dendritic Cells Human FDCs display a fibroblast-like morphology together with extensive cytoplasmic extensions and foldings (Fig. 8). FDCs also contain one to several large round nuclei with dispersed chromatin and clear nucleoli. The phenotype of human FDCs is better characterized than that of mouse or rat FDCs (Dijkstra and Van den Berg, 1991; Schriever and Nadler, 1992; Tew et al., 1990). All FDCs express the monocyte marker CD14, the three types of complement receptors (CRUCD35, CRUCD21, CR3/ CDllb), and the Ig Fcy receptor (CD32). FDCs specifically express the longer form of CDUCD21 that has 16 short-chain consensus repeats versus the 15 short-chain consensus repeats of B cells (Liu et al., 199%). A subset of FDCs in the GC light zone expresses the low-affinity receptor for IgE

FIG.8. Phenotype of follicular dendritic cells, showing molecules that they express.

DENDRITIC CELLS

287

(FCERII, CD23>, which also represents one of the ligands for CD21. Thus, these complement receptors and Fc receptors confer FDCs with an efficient mechanism for trapping Ag-Ab-C3 complexes. FDCs express a wide range of adhesion molecules, including ICAM-1/ CD54, VCAM-1, VLA-3, -4, -5, -6, and VLA p chain. Experiments in vitro have concluded that adhesion between B cells and FDCs is mediated by the interaction between ICAM-KD54 and LFA-l/CDlla, as well as between VCAM-1 and VLA-4 (Freedman et al., 1990; Koopman et al., 1991). Interestingly, purified ICAM-1 molecules appear to deliver survival signals to human GC B cells through LFA-1 molecules, indicating that these types of molecules may perform other functions in addition to adhesion (Koopman et al., 1993). FDCs express CD40, and activated human T cells can induce a human FDC-like cell line to proliferate in a CD40Ldependent fashion (Kim et al., 1994), suggesting that CD40/CD40L interactions may be involved in FDC-T cell interactions in viuo. The expression of Fc receptors by FDCs renders their phenotypic analysis difficult, because of increased background staining. Accordingly, the expression on human FDCs of the B cell markers CD19, CD20, and CD24, the panleukocyte antigen CD45, and class I1 MHC antigens remains controversial. In the mouse, adoptive transfer of B cells from class I1 MHC IEtransgenic mice into congenic mice has suggested that the host FDCs do not synthesize MHC class I1 antigens but rather capture the donor class I1 MHC IE molecules shed by surrounding donor GC B cells (Gray et al., 1994). The FDCs organize the primary follicles as evidenced by the lack of FDCs and follicles in TNF knockout mice (Liu and Banchereau, 1996b; Pasparakis et al., 1996). It also seems that FDCs may enhance the growth and differentiation of activated B cells. Human FDC clusters promote moderate and short-term autologous B cell proliferation. FDCs also mediate a powerful stimulatory effect on the secretion of IgG, IgA, and IgM by CD40-activated B cells, most particularly when cells are cultured with IL-2 and IL-10 (Grouard et al., 1995).There is also evidence that suggests that FDCs inhibit apoptosis in GC B cells by rapid inactivation of preexisting endonuclease, using a mechanism distinct from CD40 ligation (Lindhout et al., 1995). Development of the follicle requires a pre-FDC of mesenchymal origin that expresses TNF-RI (Matsumoto et al., 1997) and B cells that produce lymphotoxin a (LT-a) (Fu et al., 1998; Gonzalez et al., 1998). 2. Germinal Center Dendritic Cells

CD4+ CDllc' germinal center dendritic cells (GCDCs) have been found among CD4+ CD3- cells within the germinal center of human

288

DIANA BELL et al.

tonsils. They represent less than 0.5-1% of GC cells and are distributed in both the dark and light zones, These GCDCs express all Fcy receptors (CD16, CD32, CD64) as well as the three complement receptors (CR1, CR2, CR3), accounting for their efficient binding of immune complexes. Isolated GCDCs display (1) poor uptake of soluble FITC-dextran or phagocytosis of FITC-latex beads, (2) potent induction of allogeneic naive CD4' T cell proliferation, and ( 3 ) a strong capacity to enhance B cell growth and differentiation (Dubois and Briere, personal communication). The current hypothesis is that these GCDCs represent the mature form of the CD ll c+ precursors in circulating blood (O'Doherty et al., 1993; Thomas et al., 1993) and may correspond to the interstitial DCs generated in vitro from CD34' HPCs. These GCDCs also likely correspond to the previously described antigen-transporting cells (Szakal et al., 1985), which coordinate the generation of memory T and B cells that share specificity for a given antigen. C. DENDRITE CELLAND B CELLDIALOGUES Several in vitro and in vivo observations have suggested the importance of DCs in the establishment of humoral responses (Cebra et al., 1994; Flamand et al., 1994; Francotte and Urbain, 1985; Inaba et al., 1983a; Inaba and Steinman, 1985; Schrader et al., 1990; Sornasse et al., 1992; Spalding and Griffin, 1986), but it is a common understanding that DCs act to select and activate antigen-specific resting T cells that subsequently induce B cell responses. More specifically, on priming by DCs, activated T cells express CD40 ligand (CD40L), which in turn interacts with CD40expressing B cells to form a cellular triad. Activated T cells promote B cell survival (Liu, 1989),proliferation (Banchereau et al., 1991),differentiation, and isotype switching (Defrance et al., 1992; Jabara et al., 1990; Malisan et al., 1996) through cytokines CD40 and CD70 (Jacquot et al., 1997). However, there is now evidence that DCs directly interact with B cells to regulate humoral responses.

1. Dendritic Cells and Humoral Responses in Vivo In murine models, the requirement for splenic adherent cells in primary antibody synthesis (Mosier, 1967) led to the discovery of a key role played by DCs in such responses (Inaba et al., 1983a). Using hapten-carrier conjugates, DCs can sensitize carrier-specific T cells, which in turn interact with hapten-specific B cells that then proliferate and differentiate (Inaba and Steinman, 1985).The in vivo T cell priming obtained by administration of antigen-pulsed DCs (Inabaet al., 1990a,b)is followed by the appearance of antigen-specific immunoglobulin in serum (Berg et al., 1994; Flamand et al., 1994; Francotte and Urbain, 1985; Sornasse et al., 1992). Ig levels

DENDRITIC CELLS

289

become detectable after a second challenge with soluble antigen, a few days after DC injection (Liu and MacPherson, 1993; Sornasse et al., 1992). Such studies unfortunately cannot confirm the presumed direct interaction of DCs with B cells. Immunohistochemical studies have, however, shown that human tonsillar interdigitating dendritic cells from T cell areas colocalize with naive B cells (Bjorck et al., 199%). Furthermore, DC-B cell clusters have been observed in vivo in rat lymph (Kushnir et al., 1998).

2. Dendritic Cells and Humoral Responses in Vitro To analyze the possible interactions between DCs and B cells in a T cell-dependent context, in vitro-generated DCs have been cocultured with allogeneic B cells using a CD40 ligand-transfected cell line with or without cytokines (Dubois et al., 1997). Such a CD40L-expressing cell line can serve as an experimental surrogate of activated T cells. a. Dendritic Cells Enhance the Proliferation of Activated

B Cells. Through the release of uncharacterized soluble factors (different from sgp80, IL-12, and IL-lo), in uitro-generated DCs increase the yield of viable human naive and memory B cells activated solely through their CD40 antigen. The proliferation of B cells activated with particles of Staphylococcus aureus Cowan I (Banchereau and Rousset, 1992) is also enhanced by DCs. DCs can further enhance the considerable proliferation of CD40-activated B cells that occurs in response to IL-4, IL-13, and IL-10. Furthermore, DCs allow CD40-activated B cells to proliferate in response to IL-2 (Dubois et al., 1997; Fayette et al., 1997). The induction by DCs of IL-%mediated B cell proliferation necessitates CD40 activation of DCs and involves both IL-12 and sgp80 (Dubois, 1998).

b. Dendritic Cells Induce B Cell Diferentiation. 1. Naive B cells can secrete 1gM in response to IL-2. Addition of DCs allows CD40-activated naive B cells to produce IgM in response to IL-2. IL-12 represents the critical DC-derived molecule, secreted following CD40 engagement, that permits the differentiation of naive B cells into IgM-secreting plasma cells (Arpin et al., 1997; Dubois, 1998). This further establishes the underestimated role of IL-12 on B cells, previously shown to enhance (1)the proliferation and polyclonal Ig secretion of BCR-activated human peripheral blood B cells cultured in the presence of IL-2 (Jelinek and Braaten, 1995) and (2) the antigen-specific antibody response by peripheral blood mononuclear cells (Clerici et al., 1993; Luzzati et al., 1997; Uherova et al., 1996). Furthermore, the primary humoral response in uivo to a microbial antigen in SCID mice engrafted with human PBLs was shown to be IL-12-dependent (Westerink et al., 1997). IL-12-treated

290

DIANA BELL et al

mice respond to proteins and haptens with increased IgGz, and decreased IgGl antibodies through mechanisms that are both dependent and independent of IFN-y (Buchanan et al., 1995; Germann et al., 1995; McKnight et al., 1994; Metzger et al., 1996). Thus, in addition to priming T cells toward Thl development, DC-secreted IL-12 may directly signal naive B cells during the initiation of the immune response. 2. Memory B cells differentiate without exogenous cytokines. In the absence of exogenous cytokines, DCs potentiate the differentiation of CD40-activated memory B cells into IgG- and IgA-secreting cells (Dubois et al., 1997). Although the effect is IL-12 independent, endogenous IL-6 represents the major factor responsible for the observed differentiation (Dubois, 1998). This is consistent with the critical role of IL-6 in B cell differentiation (Burdin et al., 1996; Kishimoto, 1985; Kishimoto et al., 1984). DCs also secrete soluble IL-6Ra chain (sgpBO), which allows the formation IL-6hgp80 complexes that bind with high affinity to the IL-6R transducing chain, gp130 (Peters et al., 1997), thus resulting in enhanced IL-6 action. 3. Skewing of isotype switching toward ZgA, and ZgA,. Provided that naive B cells are activated through CD40, DCs induce isotype switching toward IgA in the absence of exogenous cytokines (Fayette et al., 1997). Induction of surface IgA-expressing B cells is quantitatively comparable to that obtained with the combination of IL-10 and TGF-P (Defrance et al., 1992).The DC-induced expression of sIgAf B cells is partially mediated by TGF-P (Fayette et al., 1997). Although DCs allow CD40-activated naive B cells to express surface IgA, IL-10 is necessary for their differentiation into IgA-secreting cells. In the presence of IL-10 and TGF-P, naive B cells secrete both IgAl and IgAz subclasses (Fayette et al., 1997). These observations extend earlier studies with mouse B cells (Cebra et al., 1994; Schrader and Cebra, 1993; Schrader et al., 1990) and pre-B cell lines (Spalding and Griffin, 1986), which were shown to secrete high levels of IgA in the presence of a combination of polyclonally activated T cells or Th2 clones and DCs. Thus, it is tempting to speculate that DCs generated in vitro possibly share an important role in the regulation of mucosal humoral responses with mucosal DCs (Kelsall et al., 1996). Studies in rats have also shown that DCs can skew the antibody responses toward the Th2 type (Wykes et al., 1998). This study further indicates that DCs can capture and retain unprocessed antigen in vitro and in vivo and transfer it to naive B cells. 4. Distinct subpopulations of dendritic cells diflerentially regulate B cell responses. DCs and monocytes display a comparable ability to enhance CD40-activated B cell proliferation, whereas DCs are more efficient than monocytes in inducing memory B cells to secrete IgC and IgA in the absence of cytokines (Dubois et al., 1997). DCs derived from either CD34'

DENDRITIC CELLS

29 1

hematopoietic progenitors or monocytes, but not monocytes themselves, induce surface IgA expression on CD4O-activated naive B cells in the absence of cytokines. Both the interstitial DCs and LCs are able to enhance the proliferation of CD40-activated B cells and to induce the differentiation of memory B cells, but only interstitial DCs can induce naive B cells to differentiate into IgM-secreting cells in response to CD40 ligation and IL2 (Caw et al., 1997).This suggests that dermal DCs rather than epidermal LCs could be critical in launching primary B cell responses. VII. Dendritic Cells in Clinical Disease States

A. AUTOIMMUNITY 1. Rheumatoid Arthritis

Synovial fluid contains cells that are comparable in function, phenotype, and structure to blood DCs, although the frequency (1-5%) is 10-fold greater. The reason for DC accumulation in the articular cavity is unknown, as is any role of DCs perpetuating the joint inflammation characteristic of this disease (Thomas et al., 1994; Zvaifler et al., 1985). Phenotypic analyses suggest that synovial DCs are not fully activated, however, because they express low levels of CD80 and CD86. Fluids from affected joints also contain modulators of DC maturation, e.g., IL-10 (Summers et al., 1995a,b, 1996).

2. Psoriasis Local activation of T lymphocytes is regarded as an important immunological component of psoriatic skin lesions. Within psoriatic plaques large numbers of dermal (interstitial) DCs are surrounded by T cells (Nestle et al., 1994). Psoriatic DCs are more active stimulators of autologous T cell proliferation than are either psoriatic blood-derived or normal skin-derived DCs. These psoriatic DCs are not more potent in supporting superantigeninduced T cell proliferation, however, which suggests that the autostimulatory potency of psoriatic skin DCs may be a critical alteration leading to the skin lesion (Nestle et al., 1994). In contrast to normal skin DCs, psoriatic DCs express high levels of C D l b and CDlc. Whether this represents a marker of the activation status of psoriatic DCs or an explanation for the enhanced autostimulatory capacity ( Fivenson and Nickoloff, 1995) remains to be established.

B. TRANSPLANTATION 1. Dendritic Cells and Transplantation Immunity Interstitial DCs were originally suspected to be the passenger leukocytes that led to the primary allograft reaction (Hart, 1997; Hart and Fabre,

292

DIANA BELL et al.

1981; Hart et al., 1981). Indeed, DCs have been shown to migrate from cardiac (Larsen et al., 1990b) or liver allografts (Qian et al., 1994) to the T cell areas of recipient spleens, where they effectively prime antigenspecific immune responses. The depletion of DCs from solid organ grafts such as kidney (McKenzie et al., 1984), heart (McKenzie et al., 1984), Langerhans islets (Faustman et al., 1984), and thyroid (Iwai et al., 1989) prolongs graft survival. Clinical trials aimed at depleting donor kidney DCs have also shown some beneficial effects (Brewer et al., 1989). Furthermore, MHC-incompatible tissue devoid of DCs only provokes responses comparable to those induced by minor histocompatibility differences (Lechler and Batchelor, 1982a,b)).Very little is known about the role of DCs in graft-versus-host disease, but they likely play a role because all the involved sites are populated by DCs. DCs, which are radioresistant, theoretically contribute to direct donor T lymphocyte allosensitization and prime for the donor immune reactivity that results in the clinical syndrome of graftversus-host disease.

2. Dendritic Cells and Transplantation Tolerance The spontaneous acceptance of transplanted livers in mice despite MHC mismatch suggests the existence of tolerance induction pathways that can be exploited especially by this organ (Qian et al., 1994). Inasmuch as liver represents an early site of hematopoiesis, it has been hypothesized that DC precursors are seeded from the liver graft to recipient lymphoid tissue after transplantation. Supporting evidence derives from the identification of donor-derived cells in recipient bone marrow, or spleen, whereas such cells are not observed in marrows of mice rejecting heart allografts (Lu et al., 1995b). Microchimerism has also been detected in the tissues or blood of human kidney or liver transplants studied 2 to 30 years postoperatively (Starzl et al., 1992, 1993). Some of the donor cells appear to have been candidate DCs. Although it can be argued that this microchimerism is merely a consequence of long-term allografting (Starzl et al., 1997; Thomson et al., 1995), it is equally plausible that microchimerism actively supports induction of transplantation tolerance (Starzl et al., 1996). For example, costimulatory molecule-deficient DC progenitors (class I1 MHC, B7.1"", B7.2-) grown in low concentrations of GM-CSF alone fail to stimulate a primary MLR and induce donor-specific T cell anergy (Lu et al., 1995a).Administering costimulatory molecule-deficient DC precursors to normal mice also allows the subsequent engraftment of vascularized cardiac allografts (Fu et al., 1996, 1997). Thus in addition to having a role in central tolerance, DCs are now regarded as potential modulators of peripheral immune responses, offering a new approach to the immunosup-

DENDRITIC CELLS

293

pressive therapy of allograft rejection or autoimmunity (Steptoe and Thomson, 1996). C. CONTACT ALLERGY Contact sensitivity (CS) is a T cell-mediated immune reaction occurring after cutaneous immunization and challenge with low molecular weight chemicals (haptens) that covalently bind to self- or exogenous proteins. Hapten-modified proteins are then processed by APCs (Langerhans cells) that subsequently migrate to draining lymph nodes to initiate immune responses (Girolomoni et al., 1995; Macatonia et al., 1986,1987; Sullivan et al., 1985; Toews et al., 1980).Unlike classical delayed-type hypersensitivity (DTH) to proteins or cellular antigens, mediated primarily by MHC class II-restricted CD4+ T cells (Cher and Mosmann, 1987),the T cell response to haptens appears more complex and may involve CD4+ T cells and/or CD8+ T cells, depending on the hapten and the mouse strain (Grabbe and Schwarz, 1998). Responses to dinitrofluorobenzene in C57BU6 mice are mediated by MHC class I-restricted CD8+ effector T cells that can be primed by class I MHCf, class I1 MHC- DCs. The response is downregulated by CD4 regulatory T cells that are primed by class I1 MHC+, class 1 MHC" (Bour et al., 1995; Krdsteva et d., 1998). IL-10 is released during the induction phase of contact sensitivity and was shown in prior functional studies to convert LCs from potent inducers of primary immune responses specifically to tolerizing cells in vitro. Data indicate that in vivo application of IL-10 before allergen exposure induces antigen-specific tolerance in mice and that IL-10 might act via inhibition of proinflammatory cytokines (Enk et al., 1994). APCS IN ASTHMA D. DENDRITECELLSAS IMPORTANT IgE plays an important role in asthma, with total serum IgE levels closely related to both clinical expression of the disease and airway hyperresponsiveness. IgE binds to a high-affinity cell surface receptor ( FceRI), which is present not only on mast cells but also on cutaneous DCs (Maurer et al., 1996; Stingl and Maurer, 1997; Stingl et al., 1977) and, by extension, on DCs of the airway epithelium especially in asthmatics (Semper and Hartley, 1996; Tunon-De-Lara et al., 1996). T lymphocytes, secreting Th2 cytokines such as IL-4 and IL-5 in response to inhaled antigen, play a major role in the pathogenesis of allergic bronchial asthma (Robinson et al., 1992). The network of airway DCs in the lung is particularly well developed to capture inhaled Ag (Gong et al., 1992; Holt et al., 1990; Schon-Hegrad et al., 1991). On encountering inhaled Ag, aimay DCs migrate to the draining lymph nodes of the lung and induce primary immune responses (Havenith et al., 1993; Masten et al., 1997; Xia et nl.,

294

DIANA BELL et al.

1995). DCs are also important for presenting inhaled Ag to previously primed Th2 lymphocytes in the lung, leading to chronic eosinophilic airway inflammation (Lambrecht et al., 1998).In contrast, B cells do not play an important role in the induction of airway inflammation (Kosgren, 1997), and alveolar macrophages appear to suppress the activity of other APCs (Holt et al., 1985; Thepen et al., 1992). The number of DCs is significantly higher in the airways of asthmatics compared with control subjects, as is the proportion of DCs expressing FcsRI-a (Semper and Hartley, 1996; Tunon-De-Lara et al., 1996). Thus DCs may play a significant role in the onset and perpetuation of allergic asthma, and targeting DCs may represent an important new approach to the treatment of asthma. Indeed the therapeutic benefit of steroids in this disease may be due to an alteration of DC functions (Nelson et al., 1995).

E. DENDRITIC CELLSAND BACTERIA Immature DCs phagocytose dead and live bacteria, including CalmetteGuerin organisms, the attenuated strain of Mycobacterium bovis used as a vaccine against tuberculosis (Inaba et al., 1993), Mycobacterium tuberculosis (Larsson et al., 1997), Chlamydia trachomutis (the agent that results in blindness) (Larsson et al., 1997), Salmonella typhimurium (Svensson et al., 1997), Listeria mnocytogenes (MacLean et al., 1996), Escherichia coli (Eloranta et al., 1997; Svensson et al., 1997), Bordetella bronchoseptica (Guzman et al., 1994a,b), and Borrelia burgdo$eri, the agent of Lyme disease (Filgueira et al., 1996). Listeria is able to kill DCs, possibly through the production of listeriolysin. DCs process the live bacteria for peptide presentation by class I and class I1 MHC molecules. Bacterial infections of DCs result in their activation as demonstrated by the increased expression of surface costimulatory molecules (CD54, CD40, CD80, CD83, CD86) and the secretion of multiple cytokines, including TNF, IL-1, IL-12, and IFN-a and IFN-fl (Thurnher et al., 1997). Bacterial-induced maturation is in turn associated with a decreased antigen capture capacity. F. DENDRITIC CELLSAND PARASITES Human infections with Leishmania parasites range from self-healing cutaneous to uncontrolled, diffuse cutaneous disease, and from subclinical to fatal visceral disease. Immature DCs can phagocytose the organism in vitro, and LCs infected by Leishmania major are present in the dermal infiltrate of lesional skin (Blank et aE., 1993). DCs restrain intracellular parasite replication through uncharacterized mechanisms (Moll et al., 1993).Leishmania-infected LCs can migrate into the draining lymph nodes, where they mature and activate resting and memory T cells with specificity for Leishmania. Macrophages are unable to elicit primary responses and

DENDRITIC CELLS

295

are poorly efficient in stimulating secondary responses (Moll et al., 1995; Will et al., 1992). Lymph node DCs carry persistent parasites that may result in the sustained stimulation of memory T cells and allow the maintenance of protective immunity. After intravenous administration of Leishmania donovani (the species responsible for visceral leishmaniosis), the parasites are found within the marginal metallophil macrophages, where they are degraded. However, a small proportion of the parasites localizes to DCs within the periarteriolar lymphocyhc sheath, where they persist and stimulate production of IL-12 (Gorak et al., 1998). In humans, infections with Toxoplmma gondii are largely asymptomatic, although fetal contamination results in malformations that can be extremely severe. Moreover, life-threatening systemic toxoplasmosis can occur in AIDS patients and other conditions associated with profound immune suppression. Toxoplmma antigens induce the redistribution of DCs to T cell areas and activate the secretion of IL-12 by DCs but not by macrophages (Sousa et al., 1997). It remains to be determined whether the Toxoplmma parasites that invade the gut are directly taken up by DCs or whether macrophages capture and process them (Johnson and Sayles, 1997). Toxoplmmu has been shown to infect human DCs (T. Curiel, personal communication). G. DENDRITECELLSAND VIRUSES The role of DCs as potentiatorshnitiators of antiviral immune responses has been well documented in murine systems. In particular, DCs are the most efficient AF’Cs in stimulating recall CTL responses against Sendai viruses (Kast et al., 1990), Herpes simplex virus (Hengel et al., 1987), and influenza virus (Nonacs et al., 1992). However, viruses s t i l l survive and replicate despite the pressures exerted by the immune response and most particularly by CTLs (Koup, 1994). These include reduced expression of critical antigenic epitopes, genetic variation of MHC class I-restricted CTL epitopes, clonal exhaustion of CTLs, down-regulation of class I MHCpeptide complex expression, production of “immunosuppressive” cytokines such as IL-10, and down-regulation of critical cytokines such as IL-12. DCs represent a cellular target of choice for viruses for multiple reasons. Because of the critical role of DCs in initiating immune reactions, it is very advantageous for the viruses to affect DC viability and biological functions. Furthermore, because of the distribution of DCs throughout body surfaces such as skin and mucosae, DCs provide a means of accessing other cells, such as T cells. Findy, sequestration within the DCs may provide a very efficient strategy for viruses not to be identified by the immune system. As summarized below, evidence is now accumulating that

296

DIANA BELL et al

viruses target DCs for their own benefit, thus antagonizing the function of DCs as initiators and potentiators of antiviral immune responses.

1 Herpesvirus Since the discovery of Kaposi’s sarcoma-associated herpesvirus, or Herpesvirus-8 (KSHV or HHV8) (Chang et al., 1994), this virus has been shown to be associated with human diseases, including Kaposi’s sarcoma (Chang et al., 1994), systemic Castleman’s disease (Cesarman et al., 1995), and primary effusion or body-cavity-based lymphoma (Cesarman et al., 1995; Gao et al., 1996). The virus has been localized to malignant cells, although its role in disease pathogenesis remains controversial. HHV8 DNA as well as viral IL-6 RNA transcripts have been detected in CD83+, fascin/p55+, CD68’ cells from cultured bone marrow stromal cells in 15 out of 15 myeloma patients and from 2 out of 8 patients with monoclonal gammapathy of undetermined significance, a precursor to myeloma (Rettig et at., 1997).This initial finding sparked controversy because results could not be reproduced by several groups (Cottoni and Uccini, 1997; Masood et al., 1997; Parravicini et al., 1997; Whitby et al., 1997; Yi et al., 1998), but were corroborated by other investigators (Brouss et al., 1997). HHV8 has been demonstrated by in situ hybridization within the bone marrow of myeloma patients (Said et al., 1997). It has been proposed that HHV8 or KSHV, if it is actually present in DCs, may stimulate and maintain abnormal plasma cell proliferation in myeloma through alterations in the bone marrow microenvironment and production of viral IL-6 (vIL-6). Nevertheless, the initial genetic alterations that lead to plasma cell transformation remain to be identified. ~

2. Cytomegalovirus Cytomegalovirus (CMV) is a ubiquitous pathogen that is a major cause of morbidity and mortality in immunocompromised individuals, including patients with AIDS or those who have undergone bone marrow or solid organ transplantation (Britt and Mach, 1996). CMV is also associated with the development of chronic rejection in organ transplant patients (Grattan et al., 1989; MeInick et at., 1995), and chronic graft-versus-host disease in bone marrow transplant recipients (Lonnqvist et al., 1984; Soderberg et al., 1996). Similar to other herpesviruses, CMV establishes lifelong latency in the host after primary infection, which is characterized by persistence of the viral genome without production of infectious virus. However, transmission of latent CMV can occur through blood transfusion and allografts of bone marrow or solid organs. In long-term cultures of allogeneically stimulated, adherent, monocyte-derived macrophages, human CMV reactivates (Soderberg-Naucler et al., 1997). CD33’ progenitors of dendritic

DENDRITIC CELLS

297

cells and monocyte-macrophages are also important reservoirs of latent CMV, whereas T cells, B cells, and CD33- mature granulocytes are not (Hahn et al., 1998; Kondo et al., 1994, 1996). 3. Dendritic Cells and Influenza Virus

Virtually all DCs are infected on exposure to influenza virus, as demonstrated by expression of the viral proteins hemagglutinin and nonstructural protein 1 (Bhardwaj et al., 1994; Ridge et al., 1998). Infected cells remain viable for more than 2 days, however, and produce little infectious virus. This contrasts with macrophages, which produce infectious virus while undergoing apoptosis (Fesq et al., 1994; Hofmann et al., 1997).Infected DCs, but not infected macrophages or B cells, can induce recall CTL responses by CD8' T cells without an absolute requirement for CD4' T cell help (Bhardwaj et al., 1994). Several experimental findings account for this specialized function of DCs. First of all, very few infectious virus particles and very small numbers of DCs stimulate a powerful CTL response, as is true of other T cell responses elicited by DCs. DCs also stimulate strong CTL responses after infection with influenza virus inactivated by heat or UV radiation, which almost completely abrogates active viral protein synthesis but apparently maintains viral binding and access to the DC cytoplasm (Bender et al., 1996). Perhaps most relevant to in vivo biology, DCs can acquire influenza antigens from vinis-infected apoptotic cells and subsequently stimulate MHC class I-restricted CD8+ CTLs (Albert et aE., 1998). This may explain the phenomenon of crosspriming, whereby donor cell antigens are presented by host bystander cells (Bevan, 1977; Fossum and Rolstad, 1986; Huang et al., 1994), as well as the induction of tolerance to tissue-restricted self-antigens (Kurts et al., 1996, 1997a,b). 4. Measles Virus a. Immunosuppression in Measles. Measles virus causes a profound immunosuppression that is responsible for the high morbidity and mortality induced by secondary infections (Oldstone, 1996). The mechanism of immune suppression is poorly understood, but it is widely accepted to be the consequence of virus replication within leukocytes, especially within the lymphoid system (Griffin, 1995; Griffin et al., 1994). Infected T cells and monocytes die by apoptosis, particularly within syncyba (EsoIen et al., 1995) identifiable in viva in the submucosal areas of tonsils and pharynx (Warthin, 1931) once viral replication has begun, after virus has started replicating. Marked and prolonged alterations of cell-mediated immunity have been noted as a consequence of measles virus infection: T lymphocytopenia, inhibition of delayed-type hypersensitivity responses, and suppres-

298

DIANA BELL et

01.

sion of antibody responses (McChesney et al., 1986) despite a skewing of T cell responses toward the Th2 pathway (Ward and Griffin, 1993). Interestingly, cutaneous anergy is also observed in response to measles vaccines (Starr, 1964). Three recent studies have highlighted the pathogenic effects of measles virus on human DCs (Fugier-Vivier et al., 1997; Grosjean et al., 1997; Schnorr et al., 1997).

b. Measles Virus Replicates in DCs. Wild-type measles virus as well as the Edmonston and Halle vaccine strains can infect human DCs isolated from skin (Langerhans cells) or blood or generated in vitro by culturing either CD34' HPCs with GM-CSF TNF (DC,,,) or blood monocytes with GM-CSF + IL-4 (DC,,). This infection results in the surface expression of hemagglutinin on a large proportion of DCs and the generation of giant syncytia. Infectious virions are produced, and DCs eventually undergo apoptosis. The production of virions by DCimm is enhanced followingcontact with T cells in a CD40-dependent fashion as observed with HIV (Pinchuk et al., 1994).

+

c. Measles Virus Interferes with Dendritic Cell Stimulation of T Cells. DCs infected by measles virus show reduced IL-12 production (FugierVivier et al., 1997), as previously reported for monocytes (Karp et al., 1996), and are unable to stimulate proliferation by alloreactive T cells ( Fugier-Vivier et al., 1997). Measles virus-infected DCs can also block the allostimulatory capacity of uninfected DCs, even when the infected cells are present at very low numbers (Grosjean et al., 1997). This inhibitory effect is in part due to the release of viable viral particles. However, addition of UV-treated, paraformaldehyde-fixed measles virus-infected DCs also inhibits the allogeneic DC-T cell MLR (Grosjean et al., 1997),supporting an active virus-independent immunosuppression, the mechanisms for which remain to be determined. The intense immunosuppression induced by measles virus can be explained by a major cytopathic effect on DCs. It is therefore unclear how immunity against measles is ever established. One possibility is that unaffected DCs may acquire measles virus-induced apoptotic bodies, as occurs with influenza (Albert et al., 1998), and subsequently initiate CTL responses. Alternatively, measles virus may differentially affect the various DC subsets or maturational stages, as evidenced by the fact that measles virus-infected immature DCs induce T cell death, whereas T cell viability is not altered by infected mature DCs. 5. Dendritic Cells and Retroviruses The interaction of retroviruses with DCs is best exemplified by human immunodeficiency virus (HIV), the causative agent of AIDS (Fauci, 1996;

DENDRITIC CELLS

299

Wyatt and Sodroski, 1998). Interestingly, HIV interacts with both the mesenchymal FDCs and the DCs of hematopoietic origin. a. Follicular Dendritic Cells and H N . In the 1980s, several groups noted the presence of large numbers of HIV particles on the dendritic processes of FDCs within the germinal centers of secondary lymphoid organs of infected individuals (Armstrong and Horne, 1984; Biberfeld et al., 1988; Fox and Cottler-Fox, 1992; Le Tourneau et al., 1985; TennerRacz et al., 1988). Such retention of C-type retroviruses had in fact already been recognized in the early days of FDC research (Hanna and Szakal, 1968; Hanna et al., 1970; Szakal and Hanna, 1968). Virus trapping is most likely due to the formation of immune complexes that bind to Fc and complement receptors on FDCs. The HIV particles trapped within the FDC processes remain infectious for protracted periods of time (Heath et al., 1995). The lymphadenopathy characteristic of the early stages of HIV infection is followed by the disappearance of the FDC network, which is in turn followed by follicle lysis and generalized immunosuppression (Fauci, 1996). Susceptibility to HIV replication cannot explain the disappearance of the FDC network because FDCs lack virus receptor and coreceptors and are therefore not permissive for HIV infection. The loss of FDCs may therefore be due to (1)the activation of CD8' CTLs that lyse FDCs in a bystander fashion or (2) the lack of T cell-dependent FDC growth and/or survival as a consequence of T cell exhaustion (Kapasi et al., 1993). These issues are being addressed in vivo in the mouse MAIDS model induced by the murine leukemia retrovirus (Burton et al., 1997).

b. Dendritic Cells and H N . Because DCs express CD4, the receptor for HIV, early studies analyzed whether DCs would essentially act as (1)transporters of the virus, initially deposited on the mucosa, to activated T cells in secondary lymphoid organs (Cameron et al., 1992) or (2) permissive sites for virus replication (Fauci, 1996; Langhoff et al., 1991;Macatonia et al., 1989, 1990; Weissman and Fauci, 1997; Weissman et al., 1997). These studies eventually led to the finding that explosive HIV replication occurs when DCs and resting T cells are cocultured (Pinchuk et al., 1994; Pope et al., 1994, 1995). Although resting T cells, as opposed to activated T cells, are unable to support a productive infection, DCs can support low levels of virus replication consistent with their expression of multiple chemokine coreceptors (Ayehunie et al., 1997; Granelli-Piperno et al., 1996; Zaitseva et al., 1997). Infection and transmission may also vary with the maturational stage of DCs (Granelli-Piperno et al., 1998).When immature and mature populations of DCs were generated from blood monocytes (using GM-CSF IL-4 to provide DCimmcells, followed by

+

300

DIANA BELL et al

LPS to provide DC,",), the DCi,, replicated M-tropic but not T-tropic HIV-1 whereas DC,, replicated both types of viruses but only in concert with T cells, and not as populations depleted of T cells. Most of the viral production from these DC-T cell cocultures occurs within syncytia that are heterokaryons of DCs and T cells. Each cell type brings a specific transcription factor allowing viral genome expression. Specifically DCs provide high levels of active NF-KB whereas T cells provide the Spl transcription complex (Granelli-Piperno et al., 1995). In accordance with these in vitro studies, HIV-expressing syncytia have been found in vivo at the surfaces of mucosal lymphoid tissues such as tonsils and adenoids (Frankel et al., 1996, 1997). Chemokine receptors, ordinarily considered most pertinent to immune cell trafficing and inflammation, have also proved critical to certain infectious disease processes. In the case of HIV, CR5 as well as CXCR4 can act as coreceptors for the virus (Wyatt and Sodroski, 1998). It has been demonstrated that HIV-1 can infect DCs in vitro through interactions with CCR5 and CXCR4 receptors (Ayehunie et al., 1997; Rubbert et al., 1998; Zaitseva et al., 1997). Conflicting results revolve around the function of circulating DCs in AIDS patients. A deficit of circulating DCs observed early in infection (Macatonia et al., 1989) may explain the early loss of CD4" memory T cells (Knight et al., 1997), because the memory T cell pool in v i m has been shown to depend highly on the presence of functional DCs (Brocker et al., 1997). It is hoped that an improved understanding of the pathogenic role of HIV in the DC system will facilitate the use of DCs to establish long-term immunity against HIV. H. DENDRITIC CELLSAND TUMORS 1 . General Considerations Regarding Tumor Immunity The immune system has the potential to reject tumors as evidenced by occasional spontaneous remission of various types of cancer, e.g., renal cell carcinomas and melanomas (Boon et al., 1994; Houghton, 1994). Tumor regression occurs when CTLs recognize class I MHC peptide complexes on the tumor cell surface. For this to occur, antigen-presenting cells (and more specifically DCs) should first home into the tumor, capture tumor antigens, then migrate to secondary lymphoid organs to initiate T lymphocyte responses against the tumor-associated antigens (TAAs). Numerous studies over the past decade have now identified a large number of TAAs that can be categorized as (1)antigens encoded by genes that are completely silent in most normal tissues but activated in tumors (e.g., the MAGE, BAGE, GAGE genes that are expressed in most melanomas and many other tumors, but in normal tissue only in placenta and/or

DENDRITIC CELLS

301

testis), (2) differentiation antigens present within a tumor cell as well as its normal counterpart [e.g., tyrosinase of melanocytes, or carcinoembryonic antigen (CEA)], (3) antigens resulting from mutations that are truly tumorspecific antigens that affect a wide variety of proteins, including those involved in cell growth (e.g., Her Wneu, a growth factor receptor overexpressed in breast and ovarian carcinomas but present at low levels in some normal tissue), (4) overexpressed tumor antigens, and (5) viral antigens derived from oncogenic viruses [e.g., E7 oncoprotein of human papilloma virus (HPV) 16 found in most cervical carcinomas]. The final or efferent step of the antitumor immune response occurs when the primed TAA-specific CTLs leave the secondary lymphoid organs and return to the tumor to kill the inalignant cells. Why then do cancers develop despite the immune system? What is the role of DCs in cancer development and/or regression? 2. Tumors with increased Dendritic Cell Numbers

Have a Better Prognosis Immunohistological analysis performed in the late 1980s and early 1990s using S l O O staining as a marker for DCs demonstrated that an increased number of DCs located within tumors was associated with better prognosis. This has been described for colorectal adenocarcinoma (Ambe et al., 1989), adenocarcinoma of the lung (Furukawa et al., 1985; Fox, 1989), papillary carcinoma of the thyroid (Schroder et al., 1988),as well as gastric (Tsujitani et al., 1990), esophageal (Imai and Yamakawa, 1993), and nasopharyngeal (Nomori et al., 1986) carcinomas.

3. Developing Tumors Contain Dendritic Cells with an immature Phenotype Colon carcinomas display a heavy infiltrate of macrophages and/or DCs that express high levels of class I1 HLA antigens. However, these DCs marginally express CD80 and CD86 (Chaux et al., 1996). Similar findings have been reported for basal cell carcinomas, wherein only 1-2% of intratumor and 5-10% of peritumor APCs expressed CD80 and CD86 (Nestle et al., 1997) as well as reduced levels of CD40 (Viac et al., 1997). Consistently, DCs isolated from basal cell carcinomas display low allostimulatory capacity as one indicator of altered immunogenicity. A recent detailed functional analysis of infiltrating DCs in responding versus progressing melanoma metastases in the same patient showed that DCs infiltrating the responding metastases have the characteristics of mature DCs, with potent allostimulatoryproperties and high levels of CD80, CD83, and CD86 (Enk et al., 1997). In contrast, DCs within progressing melanoma metastases display reduced CD83 and almost no CD86, and they exert fivefold less

302

DIANA BELL et 02.

stimulation of allogeneic T cells than do DCs from the regressing metastases (Enk et al., 1997). Importantly, in vitro assays measuring tolerance induction show that DCs from progressing metastases induce T cell anergy, whereas DCs from regressing metastases do not. The alteration of dendritic cell functions appears to go beyond the tumor site, because blood DCs from patients suffering from stage I11 and IV breast cancer show decreased allostimulatory capacity and decreased CD80-CD86 expression (Gabrilovich, 1997). DCs with altered functions have also been found in the spleens of tumor-bearing animals (Gabrilovich et al., 1994). Cancer cells also secrete factors that alter DC functions as well as development. Among these, IL-10 appears to play a critical role, as evidenced by IL-10 production by progressing melanoma metastases (Engering et al., 1997) and by the absence of DC infiltration in experimental tumors secreting IL-10 in vivo (Qin et al., 1997). Vascular endothelial cell growth factor ( W G F ) , which is produced by nearly all tumor cells, represents another candidate that affects the development of DCs from hematopoietic progenitors (Gabrilovich et al., 1996). VEGF offer tumors the additional advantage of inducing endothelial cell growth and angiogenesis. 4. Mature Dendritic Cells Presenting Tunwr-Associated Antigen Can Cure Most Experimental Mouse Models of Cancer Experiments over the past few years have demonstrated the feasibility of eradicating tumors in mice with DCs loaded with tumor-associated antigens. Initial studies, performed with tissue-derived DCs, concentrated on antitumor responses that were essentially MHC class I1 dependent (Cohen et al., 1994; Flamand et al., 1994; Grabbe et al., 1991). However, potent MHC class I-restricted CD8 responses can also be induced in vivo by administration of Ag-pulsed DCs obtained from either tissues (Takahashi et al., 1993) or cultured bone marrow cells (Porgador and Gilboa, 1995). DCs were initially loaded with Ag by pulsing defined peptides of known sequence (Celluzzi et al., 1996; Mayordomo et al., 1995) or undefined peptides isolated by acid elution from tumor cell lines (Zitvogel et al., 1996). Genetically modified DCs have been shown to induce strong MHC-restricted CTL responses, resulting in considerable antitumor effects. Genetic modification has been performed either at the bone marrow precursor level using retroviral vectors (Specht et al., 1997) or at the mature stage using replication-deficient, recombinant adenoviral vectors (Song et al., 1997). In most experimental models tested to date, the afferent sensitization arm of the response has required concomitant presentation of a xenogeneic peptide, e.g., OVA-peptide, in order for the elicited CTLs to recognize a

DENDRITIC CELLS

303

parental tumor expressing only the tumor-specific antigen. For repeated vaccinations with DCs, the induction of antiviral or antivector immunity may aIso represent an important limiting step. In this regard, the successful loading of DCs with whole tumor cell-derived RNA (Boczkowski et al., 1996) represents an interesting approach that would render feasible the presentation by DCs of both cytolytw and helper antigenic epitopes from small tumor samples. Autoantigens might also be simultaneouslypresented, however. Fusing DCs with tumor cells has also yielded antitumor responses in mice (Gong et al., 1997; Gong, 1998), but this approach is difficult to implement in human therapy. Indeed, fusion per se may not be necessary, because physical contact between DCs and tumor cells may produce an immunogen that induces tumor protection and therapeutic tumor rejection (Celluzzi and Falo, 1998). Whether this results in the capture of tumorderived apoptotic bodies remains to be determined. Interestingly, DC viability may not even be required for immunity to occur, as demonstrated by the ability of tumor peptide-pulsed DC-derived exosomes to prime specific CTLs in vivo and eradicate or suppress growth of established murine tumors in a T cell-dependent manner (Zitvogel et al., 1998).

5. Pilot Clinical Trials Indicate the Safety of Dendritic Cell Adm~ni~t~ation to Humans Significant clinical responses have been observed in pilot trials using blood-derived dendritic cells loaded with lymphoma idiotype (Hsu et al., 1996). Peptide-pulsed antigen-presenting cells generated by culturing monocytes with GM-CSF alone have also elicitedin vivo immune responses (Mukherji et al., 1995). Some clinical responses have also been observed in prostate cancer using DCs generated by culturing monocytes with GMCSF + IL-4, then pulsed with prostate-specific membrane antigen peptide (Murphy et al., 1996; Tjoa et al., 1995, 1996, 1997). Melanoma peptidepulsed DCs, also generated by culturing monocytes with GM-CSF + IL4,induced clinical regression in 5 of 16 patients treated, two of the patients showing a complete response of all evaluable disease (Nestle et al., 1998a). Longevity of the responses, as well as real variation of the observed responses from the natural history of the tumors or from the effects of other adjuvants used with DC immunizations, are outstanding unknowns. Transposing to human cancer the encouraging results observed in mice after DC immunotherapy will require significant efforts for multiple reasons. First, cancer in humans is in no way comparable to the reproducible, well-defined, cell line-based animal models. Second, the complexity of the DC lineage, with diverse subsets, stages of maturation, and methods of generation, necessitates that each step be tested independently. Furthermore, the nature of the tumor antigens, and the optimal method for loading

304

DIANA BELL et al

DCs with those tumor antigens, represent additional parameters for careful analyses. Strategies that introduce antigen into DCs, but allow the DCs to select and tailor peptides for presentation on available MHC molecules, would circumvent the need to identify tumor-specific peptides with known HLA restrictions a priori. Such approaches would also offer the theoretical advantage of introducing both helper and cytolytic antigenic epitopes for the generation of effective CTLs. Route of administration, intravenous versus intracutaneous versus intranodal, the dose of DCs, and the frequency of injections also need to be established. Assuming successful induction of strong antitumor CTL activity in patients after DC immunization, there are still caveats to the long-term success of DC-based immunotherapy of cancer. CTLs may not readily migrate to the tumor site. Tumor variants may lose the class I MHC expression required for CTL recognition (Jager et al., 1997). Tumor variants may also lose expression of critical tumor antigens, or express surface molecules such as FasL (Walker et al., 1997), or secrete cytokines such as IL-10 (Chen et al., 1994) that inactivate CTLs. Patients may experience either tumor-related or drug-induced immune suppression that would render CTL priming inefficient in vivo, in which case CTL priming may best be accomplished in vitro, followed by adoptive transfer to the diseased host. In spite of all these potential pitfalls, the prospects are bright for immunotherapy of human cancer and very probably other diseases, using in vitrogenerated DCs. Accordingly, numerous investigators are embarking on studies in this arena. This level of scientific investigation should facilitate rapid answers to many important unknowns, especially whether ex wiwo manipulation of DCs represents the “holy grail” of tumor immunology. An alternative approach may be to increase, directly, the levels of DCs in viwo that are capable of capturing tumor antigens and turning in specific immune responses. Accordingly, administration of Flt-3L to mice challenged with methylcholanthrene-induced fibrosarcoma has been shown to induce complete tumor regression in a significant proportion of mice and decreased tumor growth in the remaining mice (Lynch et al., 1997). There is, however, some evidence that this effect may not be due to the generation of specific CTLs, but rather to the activation of NK cells by the Flt-3Lelicited DCs (L. Zitvogel, personal communication). The systemic administration of Flt-3L may also break tolerance to tumors based on a study showing that administration of Flt-3L to animals breaks tolerance induced by systemic administration of soluble ovalbumin (Pulendran et al., 1998). The complexity of Flt-3L effects in uivo, however, is revealed by the enhanced induction of oral tolerance (Viney et al., 1998), which can be observed for very low doses of Ag that are ineffective in controls. Such a tolerizing effect of Flt-3L has, however, not been reported for tumors. In

305

DENDRITE CELLS

fact, in this context, DCs have been able to break tolerance to tumors that has been induced by tumor peptides administered with adjuvants (Toes et al., 1996a,b, 1998). VIII. Concluding Remarks

It is now clear that DCs can no longer be considered the parent pauvre (poor relative, or black sheep) of the antigen-presenting cell family. DCs form a complex population of cells with the potential to engage in functions as contrasting as the induction of immunity versus the induction of tolerance. Much remains to be learned about these cells. In particular, the mechanisms regulating the balance between immunizing and tolerizing DCs must be investigated. The cellular and molecular events involved in T cell activation by DCs are becoming better established, but there are enormous deficits in the knowledge of how DCs could induce tolerance, especially in the periphery. Answers to these questions will permit the therapeutic manipulation of the DC system. Initially, defined DC populations generated in vitro will be administered to patients to induce either immunity (as required in cancer and infectious diseases) or tolerance (as required in allergy, autoimmunity, and transplantation). Finally, one may directly target DCs in vivo using specific pharmacologic agents. Although single agents such as steroids (Kitajima et al., 1996) or Flt-3L exert effects on DCs in experimental models, more sophisticated strategies targeting various DC subpopulations and various stages of maturation will probably be necessary to enhance or inhibit specific immune responses with precise control. Although the tasks are immense, considerable means from academic, government, private, and industrial sources are now being devoted to DC research. It should not be long before DC-targeted therapy becomes part of numerous medical interventions.

ACKNOWLEDGMENTS The authors acknowledge support of their respective laboratories through grants R01AI-26875 (JWY), Pol-CA-23766 (JWY), and Pol-CA-59350 from the National Institutes of Health; LSA 6124-99, from the Leukemia Society of America the ; from Cap CURE foundation (JB); DeWitt Wallace Clinical Research Fund ( J W Y ) award as well as a grant from the Baylor Research Institute (JB). We also appreciate the assistance of Elizabeth Kraus for the data of Fig. 5. This detailed review represents an extension of the brief review written by Jacques Banchereau with Ralph Steinman, whose suggestions have been invaluable in the production of the present review. The authors wish to thank Karolina Palucka for helpful comments on the manuscript.

(JWY),

(JWY);

REFERENCES Adema, C . 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). Nature 387,713-717.

306

DIANA BELL et a1

Agger, R., Witmer-Pack, M., Romani, N., Stossel, H., Swiggard,W. J., Metlay, J. P., Storozynsky, E., Freimuth, P., and Steinman, R. M. (1992).J. Leukoc. Biol. 52, 34-42. Albert, M. L., Sauter, B., and Bhardwaj, N. (1998). Nature 392, 86-89. Allavena, P., Piemonti, L., Longoni, D., Bernasconi, S., Stoppacciaro, A., Ruco, L., and Mantovani, A. (1998). Eur. J. lmmunol. 28,359-69. Ambe, K., Mori, M., and Enjoji, M. (1989). Cancer 63, 496-503. Anderson, D. M., Maraskovsky, E., BilIingsley, W. L., Dougall, W. C., Tometsko, M. E., Roux, E. R., Teepe, M. C., DuBose, R. F., Cosman, D., and Galibert, L. (1997).Nature 390, 175-179. Ardavin, C. (1997). lmmunol. Today 18, 350-361. Ardavin, C., Wu, L., Li, C. L., and Shortman, K. (1993). Nature 362, 761-763. Armstrong, J, A,, and Home, R. (1984). Lancet 2, 370-372. Arpin, C., Banchereau, J., and Liu, Y. J. (1997). J. Exp. Med. 186, 931-940. Austyn, J. M., Hankins, D. F., Larsen, C. P., Morns, P. J., Rao, A. S., and Roake, J. A. (1994).J. lmmunol. 152, 2401-10. Ayehunie, S., Garcia-Zepeda, E. A,, Hoxie, J. A,, Horuk, R., Kupper, T. S., Luster, A. D., and Ruprecht, R. M. (1997). Blood 90, 1379-86. Banchereau, J., and Rousset, F. (1992).Adv. Immunol. 52, 125-262. Banchereau, J., and Steinman, R. M. (1998). Nature 392, 245-252. Banchereau, J., de Paoli, P., Valle, A., Garcia, E., and Rousset, F. (1991).Science 251,70-72. Banchereau, J., Bazan, F., Blanchard, D., Briere, F., Galizzi, J. P., van Kooten, C., Liu, Y. J., Rousset, F., and Saeland, S. (1994). Annu. Rev. lmmunol. 12, 881-922. Bates, E. E., Dieu, M. C., Ravel, O., Zurawski, S. U., Patel, S., Bridon, J. M., Ait-Yahia, S., Vega, F., Jr.. Banchereau, J., and Lebeque, S. (1998). Mol. lmmunol. 35,513-524. Bender, A., Sapp, M., Schuler, G., Steinman, R. M., and Bhardwaj, N. (1996).J. lmmunol. Methods 196, 121-135. Bennett, S. R., Carbone, F. R., Karamalis, F., Flavell, R. A,, Miller, J. F., and Heath, W. R. (1998). Nature 393,478-480. Berg, S . F., Mjaaland, S., and Fossum, S. (1994). Eur. J. lmmunol. 24, 1262-1268. Bevan, M. J. (1977).J. lmmunol. 118, 1370-1374. Bhardwaj, N., Friedman, S. M., Cole, B. C., and Nisanian, A. J. (1992). J. Exp. Med. 175,267-273. Bhardwaj, N., Young, J. W., Nisanian, A. J., Baggers, J., and Steinman, R. M. (1993).J. Exp. Med. 178,633-642. Bhardwaj, N., Bender, A., Gonzalez, N., Bui, L. K., Garrett, M. C., and Steinman, R. M. (1994).J. Clin. Invest. 94, 797-807. Biberfeld, P., Ponvit, A., Biberfeld, G., Harper, M., Bodner, A., and Gallo, R. (1988). Cancer Detect. Prevent. 12, 217-224. Bieber, T. (1997). lmmunol. Today 18, 311-313. Bieber, T., Rieger, A., Neuchrist, C., Prinz, J. C., Rieber, E. P., Boltz-Nitulescu, G., Scheiner, O., Kraft, D., Ring, J., and Stingl, G. (1989).J. Exp. Med. 170,309-314. Bjorck, P., Banchereau, J., and Flores-Romo, L. (1997a).lnt. lmmunol. 9, 365-372. Bjorck, P., Flores-Romo, L., and Liu, Y. J. (199713). Eur. J. lmmunol. 27, 1266-1274. Black, R. A., Rauch, C. T., Kozlosky, C. J., Peschon, J. J., Slack, J. L., Wolfson, M. F., Castner, B. J., Stocking, K. L., Reddy, P., Srinivasan, S., Nelson, N., Boiani, N., Schooley, K. A., Gerhart, M., Davis, R., Fitzner, J. N., Johnson, R. S., Paxton, R. J., March, C. J., and Cerretti, D. P. (1997). Nature 385, 729-733. Blank, C., Fuchs, H., Rappersberger, K., Rollinghoff, M., and Moll, H. (1993).J. Infect. Dis. 167, 418-425. Blauvelt, A., Katz, S. I., and Udey, M. C. (1995).J. Invest. Dermutol. 104, 293-296.

DENDRITIC CELLS

307

Boczkowski, D., Nair, S. K., Snyder, D., and Gilboa, E. (1996).J. Exp. Med. 184,465-472. Bohm, W., Schirmbeck, R., Elbe, A., Melber, K., Diminky, D., Kraal, G., van Rooijen, N., Barenholz, Y., and Reimann, J. (1995).J. Immunul. 155, 3313-3321. Boon, T., Cerotthi, J. C., Van den Eynde, B., van der Bruggen, P., and Van Pel, A, (1994). Annu. Reu. Immunol. 12,337-365. Borkowski, T. A,, Nelson, A. J., Farr, A. G., and Udey, M. C. (1996). Eur. J. Immunol. 26, 110-114. Bour, H., Peyron, E., Gaucherand, M., Garrigue, J. L., Desvignes, C., Kaiserlian, D., Revillard, J. P., and Nicolas, J. F. (1995). Eur. J. Immunol. 25, 3006-3010. Brewer, Y., Bewick, M., Palmer, A., Sevem, A,, Welsh, K., and Taube, D. (1989).Transplant. Proc. 21, 1772-1773. Britt, W. J., and Mach, M. (1996). Znteroirology 39, 401-412. Brocker, T. (1997).J. Exp. Med. 186, 1223-1232. Brocker, T., Riedinger, M., and Kajdainen, K. (1997).J. Exp. M e . 185,541-550. Brousset, P., Meggetto, F., Attal, M., and Delsol, G. (1997). Science 278, 1972; discussion 1972-1973. Brown, K. A., Bedford, P., Macey, M., McCarthy, D. A., Leroy, F., Vora, A. J., Stagg, A. J., Dumonde, D. C., and Knight, S. C. (1997). Clin. Exp. Immunol. 107, 601-607. Buchanan, J. M., Vogel, L. A,, Van Cleave, V. H., and Metzger, D. W. (1995).Int. Immunol. 7, 1519-1528. Buelens, C., Willems, F., Delvaux, A,, Pierard, G., Delville, J. P., Velu, T., and Goldman, M. (1995). Eur. 1.Immunol. 25, 2668-2672. Buelens, C., Verhasselt, V., De Groote, D., Thielemans, K., Goldman, M., and Willems, F. (1997a). Eur. J. Immunol. 27, 1848-1852. Buelens, C., Verhasselt, V., De Groote, D., Thielemans, K., Goldman, M., and Willems, F. (1997b). Eur. J. Immunol. 27, 756-762. Burdin, N., Galibert, L., Garrone, P., Durand, I., Banchereau, J., and Rousset, F. (1996). J. Immunol. 156,4107-4113. Burkly, L., Hession, C., Ogata, L., Reilly, C., Marconi, L. A., Olson, D., Tizard, R., Cate, R., and Lo, D. (1995). Nature 373, 531-536. Burton, G. F., Masuda, A., Heath, S. L., Smith, B. A,, Tew, J. G., and Szakd, A. K. (1997). Immunol. Reu. 156, 185-197. Cameron, P. U., Freudenthal, P. S., Barker, J. M., Gezelter, S., Inaba, K., and Steinman, R. M. (1992). Science 257, 383-387. Carrasco, D., and Bravo, R. (1993). Cell Growth Difler. 4, 849-859. Catterall, W. A., and Striessnig, J. (1992). Trends Pharmucol. Sci. 13, 256-262. Caux, C., and Banchereau, J. (1996). “Blood Cell Biochemistry,” Vol. 7, pp. 263-290. Plenum, New York. Caw, C., Saeland, S., Favre, C., Duvert, V., Mannoni, P., and Banchereau, J. (1990).Blood 75,2292-2298. Caux, C., Dezutter-Dambuyant, C., Schmitt, D., and Banchereau, J. (1992a). Nature 360,258-261. Caw, C., Moreau, I., Saeland, S., and Banchereau, J. (1992b). Blood 79, 2628-2635. Cam, C., Durand, I., Moreau, I., Duvert, V., Saeland, S., and Banchereau, J. (1993). J. Exp. Med. 177, 1815-1820. Caw, C., Massacrier, C., Vanbervliet, B., Barthelemy, C., Liu, Y. J., and Banchereau, J. (1994a). Znt. Immunol. 6, 1177-1185. Caw, C., Massacrier, C., Vanbervliet, B., Dubois, B., Van Kooten, C., Durand, I., and Banchereau, J. (1994b).1.Exp. Med. 180, 1263-1272.

308

DIANA BELL et nl

Caw, C., Vanbervliet, B., Massacrier, C., Azuma, M., Okumura, K., Lanier, L. L., and Banchereau, J. (1994~). J. Exp. Med. 180, 1841-1847. Caw, C., Massacrier, C., Dezutter-Dambuyant, C., Vanbervliet, B., Jacquet, C., Schmitt, D., and Banchereau, J. (1995).J. lmmunol. 155,5427-5435. Caw, C., Vanbervliet, B., Massacrier, C., Dezutter-Dambuyant, C., de Saint-Vis,B., Jacquet, C., Yoneda, K., Imamura, S,, Schmitt, D., and Banchereau, J. (1996a).J. Exp. Med. 184, 695-706. Caw, C., Vanbervliet, B., Massacrier, C., Durand, I., and Banchereau, J. (1996b). Blood 87,2376-2385. Caux, C., Massacrier, C., Vanbervliet, B., Dubois, B., Durand, I., Cella, M., Lanzavecchia, A., and Banchereau, J. (1997). Blood 90, 1458-1470. Cebra, J. J., Bos, N. A,, Cebra, E. R., Cuff, C. F., Deenen, G. J., Kroese, F. G., and Shroff, K. E. (1994).Adv. Exp. Med. Biol. 355, 255-259. Cella, M., Scheidegger, D., Palmer-Lehmann, K.,Lane, P., Lanzavecchia, A,, and Alber, G. (1996). 1,Ezp. Me&. 184, 747-752. Cella, M., Dohring, C., Samaridis, J., Dessing, M., Brockhaus, M., Lanzavecchia, A., and Colonna, M. (1997a).J. Exp. Med. 185, 1743-1751. Cella, M., Engering, A,, Pinet, V., Pieters, J., and Lanzavecchia, A. (1997b). Nature 388, 782-787. Cella, M., Sallusto, F., and Lanzavecchia, A. (1997~). Cum. +in. lmmunol. 9, 10-16. Celluzzi, C. M., and Falo, L. D., Jr. (1998).J. Immunol. 160, 3081-3085. Celluzzi, C. M., Mayordomo, J. I., Storkus, W. J., Lotze, M. T., and Falo, L. D., Jr. (1996). J. Exp. Med. 183,283-287. Cesarman, E., Chang, Y., Moore, P. S., Said, J. W., and Knowles, D. M. (1995). N . Engl. J. Med. 332, 1186-1191. Chang, Y., Cesarman, E., Pessin, M. S., Lee, F., Culpepper, J., Knowles, D. M., and Moore, P. S. (1994). Science 266, 1865-1869. Chapuis, F., Rosennvajg, M., Yagello, M., Ekman, M., Biberfeld, P., and Gluchan, J. C. (1997). Eur. J. Immunol. 27, 431-441. Chaux, P., Moutet, M., Faivre, J,, Martin, F., and Martin, M. (1996). Lab. Invest. 74, 975-983. Chen, Q., Daniel, V., Maher, D. W., and Hersey, P. (1994). lnt. J. Cancer 56, 755-760. Cher, D. J., and Mosmann, T. R. (1987).J. Immunol. 138, 3688-3694. Christinck, E. R., Luscher, M. A., Barber, B. H., and Williams, D. B. (1991). Nature 352, 67-70. Clerici, M., Lucey, D. R., Berzofsb, J. A., Pinto, L. A., Wynn, T. A., Blatt, S. P., Dolan, M. I., Hendrix, C. W., Wolf, S. F., and Shearer, G. M. (1993). Science 262,1721-1724. Cohen, P. A,, Cohen, P. J., Rosenberg, S. A., and Mule, J. J. (1994). Cancer Res. 54,10551058. Cottoni, F., and Uccini, S. (1997). Science 278, 1972; discussion 1972-1973. Croft, M., Duncan, D. D., and Swain, S. L. (1992).]. Exp. Med. 176, 1431-1437. Crowley, M., Inaba, K., Witmer-Pack, M., and Steinman, R. M. (1989). Cell Immunol. 118, 108-125. Cumberbatch, M., and Kimber, I. (1992). lmmunology 75, 257-263. Cumberbatch, M., and Kimber, I. (1995). Zmmunology 84, 31-35. Cumberbatch, M., Gould, S. J., Peters, S. W., Basketter, D. A., Dearman, R. J., and Kimber, I. (1992).J. Invest. D e m t o l . 99, 107s-108s. Cumberbatch, M., Fielding, I., and Kimber, I. (1994). Immunology 81,395-401. Cumberbatch, M., Dearman, R. J., and Kimber, I. (1997). ~ ? ~ m ~ n o l92,388-395. ogy Davis, A. L., McKenzie, J. L., and Hart, D. N. (1988). Immunology 65,573-581.

D E N D R I T E CELLS

309

Davis, T. A., Saini, A. A., Blair, P. J., Levine, B. L., Craighead, N., Harlan, D. M., June, C. H., and Lee, K. P. (1998).J. Immunol. 160, 3689-3697. Defrance, T.,Vanbervliet, B., Briere, F., Durand, I., Rousset, F., and Banchereau, J. (1992). J. Exp. Med. 175, 671-682. De Saint Vis, B., Vincent, J., Vandenabeele, S., Vandervliet, B., Pin, J. J., Ait-Yahia, S., Patel, S., Mattei, M. G., Banchereau, J., Zurawski, S., Davouet, J., Cam, C., and Lebeque, S. (1998). Immunity 9, 325-336. De Smedt, T., Van Mechelen, M., De Becker, G., Urbain, J., Leo, O., and Moser, M. (1997). Eur. 1. Immunol. 27, 1229-1235. Devitt, A., Moffatt, 0. D., Raykundalia, C., Capra, J. D., Simmons, D. L., and Gregory, C. D. (1998). Nature 392,505-509. de Wad Mdefyt, R., Yssel, H., Roncarolo, M. G., Spits, H., and de Vries, J. E. (1992). Cum. +in. Immunol. 4, 314-320. Dieu, M. C., Vanbervliet, B., Vicari, A., Bridon, J. M., Oldham, E., Ait-Yahia, S., Briere, F., Zlotnik, A., Lebecque, S., and Caw, C. (1998).J. Exp. Med. 188, 373-386. Dijkstra, C. D., and Van den Berg, T. K. (1991). Res. Immunol. 142, 227-231. Dubois, B., Vanbervliet, B., Fayette, J., Massacrier, C., Van Kooten, C., Briere, F., Banchereau, J., and Caw, C. (1997).J. Exp. Med. 185, 941-951. Dubois, B., Massacrier, C., Vanbervliet, B., Fayette, J., Briere, F., Banchereau, J., and Caw, C. (1998).J. Immunol. 161,2223-2231. Elbe, A., Schleischitz, S., Strunk, D., and Stingl, G. (1994).J. Immunol. 153, 2878-2889. Eloranta, M. L., Sandberg, K., Ricciardi-Castagnoli,P., Lindahl, M., and Alm, G. V. (1997). Scund. J. Immunol. 46, 235-241. Engering, A. J., Cella, M., Fluitsma, D., Brockhaus, M ., Hoefsmit, E. C., Lanzavecchia, A,, and Pieters, J. (1997). Eur. J. Immunol. 27, 2417-2425. Enk, A. H., Angeloni, V. L., Udey, M. C., and Katz, S. I. (1993a).J.Immunol. 150,3698-3704. Enk, A. H., Angeloni, V. L., Udey, M. C., and Katz, S. I. (199313).J. Immunol. 151,23902398. Enk, A. H., Sdoga, J., Becker, D., B, P. m. M., and Knop, J. (1994).J.Exp. Med. 179,13971402. Enk, A. H., Jonuleit, H., Sdoga, J., and Knop, J. (1997). Int. J. Cancer 73, 309-316. Esolen, L. M., Park, S. W., Hardwick,J. M., and Griffin, D. E. (199S).J.Virol.69,3955-3958. Fanger, N. A,, Wardwell, K., Shen, L., Tedder, T. F., and Guyre, P. M. (1996).J. Immunol. 157,541-548. Fauci, A. S. (1996). Nature 384, 529-534. Faustman, D. L., Steinman, R. M., Gebel, H. M., Hauptfeld, V., Davie, J. M., and Lacy, P. E. (1984). Proc. Nutl. Acud. Sci. U.S.A. 81, 3864-3868. Fayette, J., Dubois, B., Vandenabeele, S., Bridon, J. M., Vanbervliet, B., Durand, I., Banchereau, J,, Caw, C., and Briere, F. (1997).J. Exp. Med. 185, 1909-1918. Fearnley, D. B., McLellan, A. D., Mannering, S. I., Hock, B. D., and Hart, D. N. (1997). Blood 89, 3708-3716. Fesq, H., Bacher, M., Nain, M., and Gemsa, D. (1994). Immunobiobgy 190, 175-182. Feuillard, J., Korner, M., Israel, A,, Vassy, J., and Raphael, M. (1996). Eur. J. Immunol. 26,2547-2551. Filgueira, L., Nestle, F. O., Rittig, M., Joller, H. I., and Groscurth, P. (1996).J. Immunol. 157,2998-3005. Fivenson, D. P., and Nickoloff, B. J. (1995).J. Cutun. Puthol. 22, 223-228. Flamand, V., Sornasse, T., Thielemans, K., Demanet, C., Bakkus, M., Bazin, H., Tielemans, F., Leo, O., Urbain, J., and Moser, M. (1994). Eur. J. Immunol. 24, 605-610.

310

DIANA BELL et a1

Flechner, E. R., Freudenthal, P. S., Kaplan, G., and Steinman, R. M. (1988).Cell Immunol. 111, 183-195. Flores-Romo, L., Bjorck, P., Duvert, V., van Kooten, C., Saeland, S., and Banchereau, J. (1997).J. Exp. Med. 185, 341-349. Fossum, S., and Rolstad, B. (1986). Eur. J. Immunol. 16, 440-450. Fox, C. H., and Coder-Fox, M. (1992). Zmmunol. Today 13, 353-356. Fox, S. B., Jones, M., Dunnill, M. S., Gatter, K. C., and Mason, D. Y. (1989).Histopathology 14,269-275. Francotte, M., and Urbain, J. (1985). Proc. Natl. Acad. Sci. U.S.A. 82, 8149-8152. Frankel, S. S., Wenig, B. M., Burke, A. P., Mannan, P., Thompson, L. D., Abbondanzo, S. L., Nelson, A. M., Pope, M., and Steinman, R. M. (1996). Science 272, 115-117. Frankel, S. S., Tenner-Racz, K., Racz, P., Wenig, B. M., Hansen, C. H., Heffner, D., Nelson, A. M., Pope, M., and Steinman, R. M. (1997).Am. J. Pathol. 151, 89-96. Freedman, A. S., Munro, J. M., Rice, G. E., Bevilacqua, M. P., Morimoto, C., McIntyre, B. W., Rhynhart, K., Pober, J. S., and Nadler, L. M. (1990). Science 249, 1030-1033. Freeman, G. J., Boussiotis, V. A., Anumanthan, A,, Bernstein, G. M., Ke, X. Y., Rennert, P. D., Gray, G. S., Gribben, J. G., and Nader, L. M. (1995). Immunity 2, 523-532. Freudenthal, P. S., and Steinman, R. M. (1990).Proc. Natl. Acad. Sci. U.S.A.87,7698-7702. Fu, F., Li, Y., Qian, S., Lu, L., Chambers, F., Starzl, T. E., Fung, J. J., and Thomson, A. W. (1996). Transplantation 62, 659-665. Fu, F., Li, Y., Qian, S., Lu, L., Chambers, F. D., Starzl, T. E., Fung, J. J., and Thomson, A. W. (1997). Transplant. Proc. 29, 1310. Fu, Y. X.,Huang, G., Wang, Y., and Chaplin, D. D. (1998).J. Exp. Med. 187, 1009-1018. Fugier-Vivier, I., Servet-Delprat, C., Rivailler, P., Rissoan, M. C., Liu, Y. J., and RabourdinCombe, C. (1997).J. Exp. Med. 186, 813-823. Furukawa, T., Watanabe, S., Kodama, T., Sato, Y., Shimosato, Y., and Suemasu, K. (1985). Cancer 56, 2651-2656. Gabrilovich, D. I., Patterson, S., Harvey, J. J., Woods, G. M., Elsley, W., and Knight, S. C. (1994). Cell Immunol. 158, 167-181. Gabrilovich, D. I,, Chen, H. L., Girgis, K. R., Cunningham, H. T., Meny, G. M., Nadaf, S., Kavanaugh, D., and Carbone, D. P. (1996). Nature Med. 2, 1096-1103. Gabrilovich, D. I., Corak, J., Ciernik, I. F., Kavanaugh, D., and Carbone, D. P. (1997). Clin. Cancer Res. 3, 483-490. Galy, A,, Travis, M., Cen, D., and Chen, B. (1995). Immunity 3,459-473. Gao, E. K., Lo, D., and Sprent, J. (1990).J. Exp. Med. 171, 1101-1121. 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). Nature Med. 2, 925-928. Garrone, P., Neidhardt, E. M., Garcia, E., Galibert, L., van Kooten, C., and Banchereau, J. (1995).J. Exp. Med. 182, 1265-1273. Geissmann, F., Prost, C., Monnet, J. P., Dy, M., Browse, N., and Hermine, 0. (1998). J. Exp. Med. 187, 961-966. Germann, T., Bongartz, M., Dlugonska, H., Hess, H., Schmitt, E., Kolbe, L., Kolsch, E., Podlaski, F. J., Gately, M. K., and Rude, E. (1995). Eur. J. Immunol. 25,823-829. Geuze, H. J. (1998). Immunol. Today 19, 282-287. Girolomoni, G., and Ricciardi-Castagnoli, P. (1997). Immunol. Today 18, 102-104. Girolomoni, G., Lutz, M. B., Pastore, S., Assmann, C. U., Cavani, A., and Ricciardi-Castagnoli, P. (1995). Eur. J. Zmmunol. 25, 2163-2169. Gong, J. (1998). Proc. Nutl. Acad. Sci. U.S.A. 95, 6279-6283.

DENDRITIC CELLS

311

Gong, J. L., McCarthy, K. M., Telford, J., Tamatani, T., Miyasaka, M., and Schneeberger, E. E. (1992).]. Exp. Med. 175, 797-807. Gong, J., Chen, D., Kashiwaha, M., and Kufe, D. (1997). Nature Med. 3, 558-561. Gonzalez, M., Mackay, F., Browning, J. L., Kosco-Vilbois, M. H., and Noelle, R. J. (1998). J. Exp. Med. 187,997-1007. Gorak, P. M., Engwerda, C. R., and Kaye, P. M. (1998). Eur. J . Immunol. 28, 687-95. Goss, J. A., Nakafusa, Y., Roland, C. R., Hickey, W. F., and Flye, M. W. (1994).J. Immunol. 153,3890-3898. Grabbe, S., and Schwarz, T. (1998). Immunol. Today 19, 37-44. Grabbe, S., Bruvers, S., Gallo, R. L., Knisely, T. L., Nazareno, R., and Granstein, R. D. (1991).I. Immunol. 146, 3656-3661. Granelli-Piperno, A,, Pope, M., Inaba, K., and Steinman, R. M. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 10944-10948. Granelli-Piperno, A., Moser, B., Pope, M., Chen, D., Wei, Y., Isdell, F., O’Doherty, U., Paxton, W., Koup, R., Mojsov, S., Bhardwaj, N., Clark-Lewis, I., Baggiolini, M., and Steinman, R. M. (1996).J . Exp. Med. 184, 2433-2438. Granelli-Piperno, A., Delgado, E., Finkel, V., Paxton, W., and Steinman, R. M. (1998). 1.Virol. 72, 2733-2737. Grattan, M. T., Moreno-Cabral, C. E., Starnes, V. A., Oyer, P. E., Stinson, E. B., and Shumway, N. E. (1989).JAMA 261,3561-3566. Gray, D., Dullforce, P., and Jainandunsing, S. (1994).J. Exp. Med. 180, 141-155. Greaves, D. R., Wang, W., Dairaghi, D. J., Dieu, M. C., Saint-Vis, B., Franz-Bacon, K., Rossi, D., Caux, C., McClanahan, T., Gordon, S., Zlotnik, A., and Schall, T. J. (1997). I. Exp. Med. 186,837-844. Grewal, I. S., Borrow, P., Pamer, E. G., Oldstone, M. B., and Flavell, R. A. (1997). Curt-. +in. Immunol. 9, 491-497. GriMin, D. E. (1995). Cum. Top. Microbid. Immunol. 191, 117-134. Griffin, D. E., Ward, B. J., and Esolen, L. M. (1994).J.Infect. Dis. 170, (Suppl. l),S24-S31. Grosjean, I., Caux, C., Bella, C., Berger, I., Wild, F., Banchereau, J., and Kaiserlian, D. (1997).J. Exp. Med. 186, 801-812. Grouard, G., de Bouteiller, O., Banchereau, J., and Liu, Y. J. (1995).]. Immunol. 155,33453352. Grouard, G., Durand, I., Filgueira, L., Banchereau, J., and Liu, Y. J. (1996). Nature 384, 364-367. Grouard, G., Rissoan, M. C., Filgueira, L., Durand, I., Banchereau, J., and Liu, Y. J. (1997). 1.Exp. Med. 185, 1101-1111. Guillemot, F. P., Oliver, P. D., Peault, B. M., and Le Douarin, N. M. (1984).1.Exp. Med. 160, 1803-1819. Guzman, C. A., Rohde, M., Bock, M., andTimmis, K. N. (1994a).Infect. Immun. 62,55285537. Guzman, C. A., Rohde, M., and Timmis, K. N. (1994b). Infect. Immun. 62,5538-5544. Hahn, G., Jores, R., and Mocarski, E. S. (1998).Proc. Natl. Acad. Sci. U.S.A. 95,3937-3942. Hanna, M. G., Jr,, and Szakal, A. K. (1968).J. Immunol. 101, 949-962. Hanna, M. G., Jr., Szakal, A. K., and Tyndall, R. L. (1970). Cancer Res. 30, 1748-1763. Harding, C. V., and Song, R. (1994).I. Immunol. 153,4925-4933. Harding, C. V., and Unanue, E. R. (1989).J. Immunol. 142, 12-19. Hart, D. N. (1997). Blood 90, 3245-3287. Hart, D. N., and Fabre, J, W. (1981).]. Exp. Med. 154, 347-361. Hart, D. N., and McKenzie, J. L. (1988).]. Exp. Med. 168, 157-170. Hart, D. N., and Prickett, T. C. (1993). Cell Immunol. 148,447-454.

312

DIANA BELL et al

Hart, D. N., Fuggle, S. V., Williams, K. A., Fabre, J. W., Ting, A,, and Moms, P. J. (1981). Transplantation 31, 428-433. Havenith, C. E., van Miert, P. P., Breedijk, A. J., Beelen, R. H., and Hoefsmit, E. C. (1993). Am. J. Respir. Cell. Mol. Biol. 9, 484-488. Heath, S. L., Tew, J. G., Tew, J. G., Szakal, A. K., and Burton, G. F. (1995). Nature 377, 740-744. Heemels, M. T., and Ploegh, H. (1995).Annu. Rev. Biochem. 64,463-491. Hengel, H., Lindner, M., Wagner, H., and Heeg, K. (1987).J. Immunol. 139,4196-4202. Henskens, Y. M., Veerman, E. C., and Nieuw Amerongen, A. V. (1996).Biol. Chem. Hoppe Seyler 377, 71-86. Herrlich, P., Zoller, M., Pals, S. T., and Ponta, H. (1993). Immunol. Today 14, 395-399. Heufler, C., Koch, F., and Schuler, G. (1988).J. Exp. Med. 167, 700-705. Hock, B. D., Starling, C. C., Daniel, P. B., and Hart, D. N. (1994).Immunology 83,573-581. Hofmann, P., Sprenger, H., Kaufmann, A., Bender, A., Hasse, C., Nain, M., and Gemsa, D. (1997).J. Leukoc. Bid. 61, 408-414. Holt, P. G. (1993). Ado. Exp. Med. Biol. 329, 557-562. Holt, P. G., Degebrodt, A., OLeary, C.,Krska, K., and Plozza, T. (1985).Clin. Exp. Immunol. 62,586-593. Holt, P. G., Schon-Hegrad, M. A., and McMenamin, P. G. (1990). Int. Rev. Immunol. 6, 139-149. Houghton, A. N. (1994).J. Exp. Med. 180, 1-4. Hsu, F. J., Benike, C., Fagnoni, F., Liles, T. M., Czenvinski, D., Taidi, B., Engleman, E. G., and Levy, R. (1996). Nature Med. 2,52-58. Huang, A. Y., Golumbek, P., Ahmadzadeh, M., Jaffee, E., Pardoll, D., and Levitsky, H. (1994). Science 264, 961-965. Imai, Y., and Yamakawa, M. (1993). In Viuo 7, 239-248. Inaba, K., and Steinman, R. M. (1984).J. Exp. Med. 160, 1717-1735. Inaba, K., and Steinman, R. M. (1985). Science 229, 475-479. Inaba, K., Granelli-Piperno, A., and Steinman, R. M. (1983al.J.Exp. Med. 158,2040-2057. Inaba, K., Steinman, R. M., Van Voorhis, W. C., and Muramatsu, S. (198313).Proc. Natl. Acad. Sci. U.S.A. 80, 6041-6045. Inaba, K., Young, J. W., and Steinman, R. M. (1987).J. Exp. Med. 166, 182-194. Inaba, K., Metlay, J. P., Crowley, M. T., and Steinman, R. M. (1990a). J. Exp. Med. 172, 631-640. Inaba, K., Metlay, J. P., Crowley, M. T., Witmer-Pack, M., and Steinman, R. M. (1990b). Int. Rev. Immunol. 6, 197-206. Inaba, K., Inaba, M., Romani, N., Aya, H., Deguchi, M., Ikehara, S., Muramatsu, S., and Steinman, R. M. (1992a).J. Exp. Med. 176, 1693-1702. Inaba, K., Steinman, R. M., Pack, M. W., Aya, H., Inaba, M., Sudo, T., Wolpe, S., and Schuler, G. (1992b).J. Exp. Med. 175, 1157-1167. Inaba, K., Inaba, M., Naito, M., and Steinman, R. M. (1993).J. Exp. Med. 178,479-488. Inaba, K., Inaba, M., Witmer-Pack, M., Hatchcock, K., Hodes, R., and Steinman, R. M. (1995).Adv. Exp. Med. Biol. 378, 65-70. Inaba, K., Pack, M., Inaba, M., Sakuta, H., Isdell, F., and Steinman, R. M. (1997).J. Exp. Med. 186, 665-672. Inaba, K., Turley, S., Yamaide, F., Iyoda, T., Mahnke, K., Inaba, M., Pack, M., Subklewe, M., Sauter, B., Sheff, D., Albert, M., Bhardwai, N., Mellman, I., and Steinman, R. M. (1998).J. Exp. Med. 188, 2163-2173. InguIli, E., Mondino, A., Khoruts, A., and Jenkins, M. K. (1997).J.Exp. Med. 185, 21332141.

DENDRITIC CELLS

313

Iwai, H., Kuma, S., Inaba, M. M., Good, R. A., Yamashita, T., Kumazawa, T., and Ikehara, S. (1989). Transplantation 47,45-49. Jabara, H. H., Fu, S. M., Geha, R. S., andVercelB, D. (199O).J.Exp. Med. 172, 1861-1864. Jacobsen, S. E., Okkenhaug, C., Myklebust, J., Veiby, 0. P., and Lyman, S. D. (1995). 1.Exp. Med. 181, 1357-1363. Jacquot, S., Kobata, T., Iwata, S., Morimoto, C., and Schlossman, S. F. (1997).J. Zmmunol. 159,2652-2657. Jager, E., Ringhoffer, M., Altmannsberger, M., Arand, M., Karbach, J., Jager, D., Oesch, F., and Knuth, A. (1997). Znt. Cancer 71, 142-147. Jakob, T., Saitoh, A., and Udey, M. C. (1997).J. Zmmunol. 159, 2693-2701. Jelinek, D. F., and Braaten, J. K. (1995).]. Zmmunol. 154, 1606-1613. Jiang, W., Swiggard, W. J., Heufler, C., Peng, M., Mirza, A , , Steinman, R. M., and Nussenzweig, M. C. (1995). Nature 375, 151-155. Johnson, L. L., and Sayles, P. C. (1997).]. Exp. Med. 186, 1799-1802. Jonuleit, H., Kuhn, U., Mdler, G., Steinbrink, K., Paragnik, L., Schmitt, E., Knop, J., and Enk, A. H. (1997a). Eur. J. Zmmunol. 27, 3135-3142. Jonuleit, H., Wiedemann, K., Muller, G., Degwert, J., Hoppe, U., Knop, J., and Enk, A. H. (1997b).1.Zmmunol. 158, 2610-2615. Josien, R., Heslan, M., Soulillou, J. P., and Cuturi, M. C. (1997).J.Exp. Med. 186,467-472. Kapasi, Z. F., Burton, G. F., Shultz, L. D., Tew, J. G., and Szakal, A. K. (1993).]. Zmmunol. 150,2648-2658. Karp, C. L., Wysocka, M., Wahl, L. M., Ahearn, J. M., Cuomo, P. J., Sherry, B., Trinchieri, G., and Griffin, D. E. (1996). Science 273, 228-231. Kast, W. M., Bluestone, J. A., Heemskerk, M. H., Spaargaren, J., Voordouw, A. C., Ellenhom, J. D., and Melief, C. J. (1990).J . Zmmunol. 145, 2254-2259. Katz, S. I., Tamaki, K., and Sachs, D. H. (1979). Nature 282, 324-326. Kelsall, B. L., and Strober, W. (1996).1. Exp. Med. 183, 237-247. Kelsall, B. L., Stuber, E., Neurath, M., and Strober, W. (1996). Ann. N.Y. Acud. Sci. 795, 116-126. Kelsoe, G. (1996). Zmmunity 4, 107-111. Khoury, S. J., Gallon, L., Chen, W., Betres, K., Russell, M. E., Hancock, W. W., Carpenter, C. B., Sayegh, M. H., and Weiner, H. L. (1995).J. Exp. Med. 182, 357-366. Kim, H. S., Zhang, X., and Choi, Y. S. (1994).J. Immunol. 153, 2951-2961. Kishimoto, T. (1985).Annu. Reu. Zmmunol. 3, 133-157. Kishimoto, T., Yoshizaki, K., Kimoto, M., Okada, M., Kuritani, T., Kikutani, H., Shimizu, K., Nakagawa, T., Nakagawa, N., Mild, Y., et al. (1984). Zmmunol. Reo. 78, 97-118. Kitajima, T., Ariizumi, K., Bergstresser, P. R., and Takashima, A. (1996).J. C h . Znuest. 98, 142-147. Kleijmeer, M. J., Oorscbot,V. M., and Geuze, H. J. (1994).]. Znuest. Demtol. 103,516-523. Kleijmeer, M. J., Ossevoort, M. A., van Veen, C. J., van Hellemond, J. J., Neefjes, J. J., Kast, W. M., Melief, C. J., and Geuze, H. J. (1995).J. Zmmunol. 154, 5715-5724. Knight, S. C., Elsky, W., and Wang, H. (1997).J. Leukoc. Biol. 62, 78-81. Kobayashi, Y. (1997). Zmmunobgy 90, 496-501. Koide, S. L., Inaba, K., and Steinman, R. M. (1987).]. Exp. Med. 165, 515-530. Kolesaric, A., Stingl, G., and Elbe-Burger, A. (1997). Znuest. D e m t o l . 109, 580-585. Kondo, K., Kaneshima, H., and Mocarski, E. S. (1994). Proc. Natl. Acud. Sci. U.S.A. 91, 11879-11883. Kondo, K., Xu, J., and Mocarski, E. S. (1996).Proc. Natl. Acud. Sci. U.S.A. 93,11137- 11142. Koopman, G., Parmentier, H. K., Schuurman, H. J., Newman, W., Meijer, C. J., and Pals, S. T. (1991).J. Exp. Med. 173, 1297-1304.

I.

I.

3 14

DIANA BELL et al.

Koopman, G., Keehnen, R. M., and Pals, S. T. (1993).Adv. Exp. Med. Biol. 329,387-392. Korsgren, M., Erjefdt, J. S., Korsgren, O., Sunder, F., and Persson, C. G. (1997).J. Exp. Med. 185, 885-892. Koup, R. A. (1994).J. Ex?. Med. 180, 779-782. Kraft, S., Wessendorf, J. H., Hanau, D., and Bieber, T. (1998).J.lmmunol. 161,1000-1006. Krasteva, M., Kehren, J., Horand, F., Akiba, H., Choquet, G., Ducluzeau, M. T., Tedone, R., Garrigue, J. L., Kaiserlian, D., and Nicolas, J. F. (1998).J. Immunol. 160,1181-1190. Kronin, V., W d e l , K., Suss, G., Kelso, A., Heath, W., Kirberg, J.. von Boehmer, H., and Shortman, K. (1996).J. Immunol. 157, 3819-3827. Kuchroo, V. K., Das, M. P., Brown, J. A., Ranger, A. M., Zarnd, S. S., Sobel, R. A,, Weiner, H. L., Nabavi, N., and Glimcher, L. H. (1995). Cell 80, 707-718. Kudo, S., Matsuno, K., Ezald, T., and Ogawa, M. (1997).1. Exp. Med. 185, 777-784. Kurts, C., Heath, W. R.,Carbone, F. R., Allison, J., Miller, J. F., and Kosaka, H. (1996). J. Exp. Med. 184,923-930. Kurts, C., Carbone, F. R., Bamden, M., Blanas, E., Allison, J., Heath, W. R., and Miller, J. (1997a).J. Exp. Med. 186, 2057-2062. Kurts, C., Kosaka, Carbone, F. R., Miller, J. F., and Heath, W. R. (199713).J. Exp. Med. 186,239-24 Kushnir, N., Liu, L., and MacPherson, G. G. (1998).J. Immunol. 160, 1774-1781. Kyewski, B. A., Fathman, C. G., and Rouse, R. V. (1986).J. Exp. Med. 163,231-246. Lafontaine, M., Landiy, D., and Montplaisir, S. (1992). Cell Immunol. 142, 238-251. Lambrecht, B. N., Salomon, B., Klatzrnann, D., and Pauweh, R. A. (1998).J. lmmunol. 160,4090-4097. Langerhans, P. (1868).Virchows Arch. Anat. 44, 325. Langhoff, E., Tetwilliger, E. F., Bos, H. J., Kalland, K. H., Poznansb, M. C., Bacon, 0. M., and Haseltine, W. A. (1991). Proc. Nutl. Acad. Sci. U.S.A. 88, 7998-8002. Lanier, L. L. (1997). C u r . @in. Immunol. 9, 126-131. Lanzavecchia, A. (1996). Cum. @in. Immunol. 8, 348-354. Lardon, F., Snoeck, H. W., Bememan, Z. N., Van Tendeloo, V. F., Nijs, G., Lenjou, M., Henckaerts, E., Boeckxtaens, C. J., Vandenabeele, P., Kestens, L. L., Van Bockstaele, D. R., and Vanham, G. L. (1997). Immunology 91,553-559. Larsen, C. P., Morns, P. J., and Austyn, J. M. (1990a).J. E3ip. Med. 171,307-314. Larsen, C. P., Steinman, R. M., Witmer-Pack, M., Hankins, D. F., Morns, P. J., and Austyn, J. M. (1990b).J. Exp. Med. 172, 1483-1493. Larsen, C. P., Ritchie, S. C., Pearson, T. C., Linsley, P. S., and Lowly, R. P. (1992).J. Exp. Med. 176, 1215-1220. Larsen, C. P., Ritchie, S. C., Hendrix, R., Linsley, P. S., Hathcock, K. S., Hodes, R. J., Lowry, R. P., and Pearson, T. C. (1994).J. lmmunol. 152,5208-5219. Larsson, M., Majeed, M., Emst, J. D., Magnusson, K. E., Stendahl, O., and Forsum, U. (1997). Immunology 92,501-511. Lechler, R. I., and Batchelor, J. R. (1982a).I. Exp. Med. 156, 1835-1841. Lechler, R. I., and Batchelor, J. R. (1982b).J. Exp. Med. 155, 31-41. Lens, J. W., Drexhage, H. A., Benson, W., and Bdfour, B. M. (1983). lmmunology 49, 415-422. Lenschow, D. J., Wdunas, T. L., and Bluestone, J. A. (1996). Annu. Rev. lmmunol. 14, 233-258. Lenz, A., Heine, M., Schuler, G., and Romani, N. (1993).J. Clin. Invest. 92, 2587-2596. Lenz, P., Elbe, A., Stingl, G., and Bergstresser, P. R. (1996). J. Inuest. Demnatol. 107, 844-848.

DENDRITIC CELLS

315

Le Tourneau, A,, Audouin, J., Aubert, J. P., Denis, J., Baufine-Ducrocq, H., Duterque, M., and Diebold, J. (1985).Ann. Pathol. 5, 137-142. Lindhout, E., Lakeman, A., and de Groot, C. (1995).]. Exp. Med. 181, 1985-1995. Lipscornb, M. F., Bice, D. E., Lyons, C. R., Schuyler, M. R., and Wilkes, D. (1995). Adu. Immunol. 59,369-455. Liu, Y. J.. Joshuo, D. E., Williams, G. T., Smith, C. A,, Gordon, J., and MacLennan, I. C. (1989). Nature 342, 929-931. Liu, Y. J., and Banchereau, J. (1996a).]. Exp. Med. 184, 1207-1211. Liu, Y. J., and Banchereau, J. (1996b). Immunologist 4, 55-66. Liu, L. M., and MacPherson, G. G. (1993).I. Exp. Med. 177, 1299-1307. Liu, T., Zhou, X., Orvell, C., Lederer, E., Ljunggren, H. G . , and Jondal, M. (1995). J. Immunol. 154,3147-3155. Liu, L., Rich, B. E., Inobe, J., Chen, W., and Weiner, H. L. (1997a).Adv. Exp. Med. Biol. 417, 375-381. Liu, Y. J., Xu, J., de Bouteiller, O., Parham, C. L., Grouard, G., Djossou, O., de Saint-Vis, B., Lebecque, S., Banchereau, J., and Moore, K. W. (1997b).]. Exp. Med. 185,165-170. Lonnqvist, B., Ringden, O., Wahren, B., Gahrton, G., and Lundgren, G. (1984).Transplantation 38, 465-468. Lu, L., Rudert, W. A., Noyola, H., Qian, S., Fu, F., Li, Y., Rao, A. S., Demetris, A. J., Fung, J. J., Trucco, M., et al. (1995a). Transplant. Proc. 27, 191-193. Lu, L., Rudert, W. A,, Qian, S., McCaslin, D., Fu, F., Rao, A. S., Trucco, M., Fung, J. J., Stard, T.E., and Thomson, A. W. (1995b).]. Exp. Med. 182, 379-387. Lu, L., Qian, S., Hershberger, P. A., Rudert, W. A,, Lynch, D. H., and Thomson, A. W. (1997).1.Immunol. 158,5576-5684. Ludewig, B., Graf: D., Gelderblom, H. R.,Becker, Y., Kroczek, R. A., and Pauli, G. (1995). Eur. 1.Immunol. 25, 1943-1950. Luther, S. A., Gulbranson-Judge, A., Acha-Orbea, H., and MacLennan, I. C. (1997).]. Exp. Med. 185,551-562. Lutz, M. B., Rovere, P., Kleijmeer, M. J., Rescigno, M., Assmann, C. U., Oorschot, V. M., Geuze, H. J., Trucy, J., Demandoh, D., Davoust, J., and Ricciardi-Castagnoli,P. (1997). J. Immunol. 159,3707-3716. Luzzati, A. L., Giordani, L., and Giacomini, E. (1997). Eur. 1.Immunol. 27, 2696-2701. Lyman, S. D., and Jacobsen, S. E. (1998). Blood 91, 1101-1134. Lyman, S. D., and Williams, D. E. (1995). Curt-. Opin. Hematol. 2, 177-181. Lynch, D. H., Andreasen, A,, Maraskovsky, E., Whitmore, J., Miller, R. E., and Schuh, J. C. (1997). Nature Med. 3, 625-631. Macatonia, S. E., Edwards, A. J., and Knight, S. C. (1986). Immunology 59, 509-514. Macatonia, S. E., Knight, S. C., Edwards, A. J., Griffiths, S., and Fryer, P. (1987).]. Exp. Med. 166,1654-1667. Macatonia, S. E., Patterson, S., and Knight, S. C. (1989). Immunology 67, 285-289. Macatonia, S. E., Lau, R., Patterson, S., Pinching, A. J., and Knight, S. C. (1990).Immunology 71,38-45. 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’Garra, A. (1995). 1. Immunol. 154,5071-5079. MacLean, J. A,, Xia, W., Pinto, C. E., Zhao, L., Liu, H. W., and Kradin, R. L. (1996).Am. 1.Pathol. 148, 657-666. MacLennan, I. C. (1994).Annu. Rev. Immunol. 12, 117-139. Maher, J. K., and Kronenberg, M. (1997). Curt-. +in. Immunol. 9, 456-461.

316

DIANA BELL et al.

Malisan, F., Briere, F., Bridon, J. M., Harindranath, N., Mills, F. C., Max, E. E., Banchereau, J., and Martinez-Valdez, H. (1996).J. Exp. Med. 183, 937-947. Maraskovsky, E., Brasel, K., Teepe, M., Roux, E. R., Lyman, S. D., Shortman, K., and McKenna, H. J. (1996).J. Exp. Med. 184, 1953-1962. Masood, R., Zheng, T., Tupule, A., Arora, N., Chatlynne, L., Handy, M., and Whitman, J., Jr. (1997). Science 278, 1970-1971; discussion 1972-1973. Masten, B. J., Yates, J. L., Pollard Koga, A. M., and Lipscomb, M. F. (1997).Am. J. Respir. Cell. Mol. Biol. 16, 335-342. Matsumoto, M., Fu, Y. X., Molina, H., Huang, G., Kim, J., Thomas, D. A., Nahm, M. H., and Chaplin, D. D. (1997).J. Exp. Med. 186, 1997-2004. Matsuno, K., Ezaki, T., Kudo, S., and Uehara, Y. (1996).J. Exp. Med. 183, 1865-1878. Matzinger, P., and Guerder, S. (1989). Nature 338, 74-76. Maurer, D., Fiebiger, S., Ebner, C., Reininger, B., Fischer, G. F., Wichlas, S., Jouvin, M. H., Schmitt-Egenolf, M., Kraft, D., Kinet, J. P., and Stingl, G. (1996). J. lmmunol. 157, 607-616. Mayordomo, J. I., Zorina, T., Storkus, W. J., Zitvogel, L., Celluzzi, C., Falo, L. D., Melief, C. J., Ildstad, S. T., Kast, W. M., Deleo, A. B., andetal. (1995).Nature Med. 1,1297-1302. McChesney, M. B., Fujinami, R. S., Lampert, P. W., and Oldstone, M. B. (1986).J. Exp. Med. 163, 1331-1336. McKenzie, J. L., Beard, M. E., and Hart, D. N. (1984). Transplant. Proc. 16, 948-951. McKenzie, J. L., Calder, V. L., Starling, G. C., and Hart, D. N. (1995). Bone Marrow Transplant 15, 163-171. McKnight, A. J., Perez, V. L., Shea, C. M., Gray, G. S., and Abbas, A. K. (1994).J.lmmunol. 152, 5220-5225. McMichael, A. J., Pilch, J. R., Galfre, G., Mason, D. Y., Fabre, J. W., and Milstein, C. (1979). Eur. 1. lmmunol. 9,205-210. MeWilliam, A. S., Nelson, D., Thomas, J. A., and Holt, P. G. (1994).J.Exp. Med. 179,13311336. MeWilliam, 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).J. Exp. Med. 184, 2429-2432. Melnick, J. L., Adam, E., and DeBakey, M. E. (1995). BioEssays 17, 899-903. Metzger, D. W., Buchanan, J. M., Collins, J. T., Lester, T. L., Murray, K. S., Van Cleave, V. H., Vogel, L. A., and Dunnick, W. A. (1996).Ann. N.Y. Acad. Sci. 795, 100-115. Miralles, G. D., Smith, C. A., Whichard, L. P., Morse, M. A., Haynes, B. F., and Patel, D. D. (1998).J. lmmunol. 160, 3290-3298. Mitra, R. S., Judge, T. A., Nestle, F. O., Turka, L. A., and Nickoloff, B. J. (1995).J.Immunol. 154, 2668-2677. Mohamadzadeh, M., Poltorak, A. N., Bergstressor, P. R., Beutler, B., and Takashima, A. (1996).J. lnimunol. 156, 3102-3106. Moll, H., Fuchs, H., Blank, C., and Rollinghoff, M. (1993).Eur. J. lmmunol. 23,1595-1601. Moll, H., Flohe, S., and Rollinghoff, M. (1995). Eur. J. lmmunol. 25, 693-699. Moore, K. W., OGarra, A., de Waal Malefyt, R., Vieira, P., and Mosmann, T. R. (1993). Annu. Rev. lmmunol. 11,165-190. Morel, A. S., Quaratino, S., Douek, D. C., and Londei, M. (1997).Eur.J. Immunol. 27,26-34. Moretta, A,, and Moretta, L. (1997). Curr. Opin. lmmunol. 9, 694-701. Morikawa, Y., Furotani, M., Kuribayashi, K., Matsuura, N., and Kakudo, K. (1992).lmmunology 77, 81-87. Morikawa, Y., Furotani, M., Matsuura, N., and Kakudo, K. (1993). Cell lmmunol. 152, 200-210.

DENDRITIC CELLS

317

Morikawa, Y., Tohya, K., Ishida, H., Matsuura, N., and Kakudo, K. (1995). Inznwnology 85, 575-581. Mosialos, G., Birkenbach, M., Ayehunie, S., Matsuniura, F., Pinkus, G. S., Kieff, E., and Langhoff, E. (1996). Am. J. Pathol. 148, 593-600. Mosier, D. E. (1967). Science 158, 1573-1575. Moss, M. L., Jin, S. L., Milla, M. E., Bickett, D. M., Burkhart, W., Carter, H. L., Chen, W. J., Clay, W. C., Didsbury, J. R., Hassler, D., et al. (1997). Nature 385, 733-736. Mueller, C. G., Ho, S., Massacrier, C., Lebecque, S., and Liu, Y. J. (1997a). Eur./. Immunol. 27, 3130-3134. Mueller, C. G., Rissoan, M. C., Salinas, B., Ait-Ydua, S., Ravel, O., Bridon, J. M., Briere, F., Lebecque, S., and Liu, Y. J. (1997b)./. Exp. Med. 186, 655-663. Muench, M. O., Roncarolo, M. G., Menon, S., Xu, Y., Kastelein, R., Zurawski, S., Hannum, C. H., Culpepper, J., Lee, F., and Narnikawa, R. (1995). Blood 85, 963-972. Mukheji, B., Chakraborty, N. G., Yamasaki, S., Okino, T., Yamase, H., Spom, J. R., Kurtzman, S. K., Ergin, M. T., Ozols, J., Meehan, J., and & a[. (1995). Proc. NatZ. Acud. Sci. U.S.A. 92, 8078-8082. Murphy, G., Tjoa, B., Ragde, H., Kenny, G., and Boynton, A. (1996).Prostate 29,371-380. Nelson, D. J., McWilliam, A. S., Haining, S., and Holt, P. G. (1995). Ani. J. Respir. Crit. Care. Med. 151, 475-481. Nestle, F. O., Zheng, X. G., Thompson, C. B., Turka, L. A,, and NickolofT, B. J. (1993). J. Immunol. 151, 6535-6545. Nestle, F. O., Turka, L. A,, and Nickoloff, B. J. (1994). J. Clin. Inuest. 94, 202-209. Nestle, F. O., Burg, G., Fah,J,, Wrone-Smith, T., and Nickoloff, B. J. (1997).Am. J. Pathol. 150,641-651. Nestle, F. O., Alijagic, S., Gilliet, M., Sun, Y., Grabbe, S., Dummer, R., Burg, G., and Schadendorf, D. (1998a). Nature Med. 4, 328-332. Nestle, F. O., Filgueira, L., Nickoloff, B. J., and Burg, G . (1998b). J. Inuest. Denrintol. 110,762-766. Ngo, V. N., Tang, H. L., and Cyster, J. G. (1998).J. Exp. Med. 188, 181-191. Nomori, H., Watanabe, S., Nakajima, T., Shimosato, Y., and Kameya, T. (1986). Cancer 57, 100-105. Nonacs, R., Humborg, C., Tam, J. P., and Steinman, R. M. (1992).J. Exp. Med. 176,519-529. Norbury, C. C., Chambers, B. J., Prescott, A. R., Ljunggren, H. G., and Watts, C. (1997). Eur. J. bnmunol. 27, 280-288. Nossal, G. J., Abbot, A., and Mitchell, J. (1968a). J. Exp. Med. 127, 263-276. Nossal, G. J., Abbot, A., Mitchell, J., and Lummus, Z. (196%). J. Exp. Med. 127,277-290. O’Doherty, U., Steinman, R. M., Peng, M., Cameron, P. U., Gezelter, S., Kopeloff, I., Swiggard, W. J., Pope, M., and Bhardwaj, N. (1993).J. Enp. Med. 178, 1067-1076. Oehler, L., Majdic, O., Pickl, W. F., Stockl, J., Riedl, E., Drach, J.. Rappersberger, K., Geissler, K., and Knapp, W. (1998).J. Exp. Men. 187, 1019-1028. Ohshima, Y., Tanaka, Y., Tozawa, H., Takahashi, Y., Maliszewski, C., and Delespesse, G. (1997). J. Immunol. 159, 3838-3848. Oldstone, M. B. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 12756-12758. Oluwole, S. F., Jin, M. X., Chowdhury, N. C., Engelstad, K., Ohajekwe, 0. A., and James, T. (1995). Cell. Immunol. 162, 33-41. Olweus, J,, BitMansour, A,, Warnke, R., Thompson. P. A,, Carballido, J., Picker, L. J., and Lund-Johansen, F. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 12551-12556. Ozawa, H., Aiba, S., Nakagawa, and Tagarni, H. (1996). Eur. J. Immunol. 26,648-652. Parravicini, C., Lauri, E., Bddini, L., Neri, A., PoIi, F., Sirchia, G., Moroni, M., Gdli, M., and Corbellino, M. (1997). Science 278, 1969-1970; discussion, 1972-1973.

318

DIANA BELL d al

Pasparakis, M., Alexopoulou, L., Episkopou, V., and Kollias, G. (1996). J. Exp. Med. 184, 1397-1411. Peguet-Navarro, J., Moulon, C., Caux, C., Dalbiez-Gauthier, C., Banchereau, J., and Schmitt, D. (1994). Eur. J. Immunol. 24, 884-891. Peters, M., Schirmacher, P., Goldschmitt, J., Odenthal, M., Peschel, C., Fattori, E., Ciliberto, G., Dienes, H. P., Meyer zum Buschenfelde, K. H., and Rose-John, S. (1997).J. Exp. Med. 185, 755-766. Pettit, A. R., Quinn, C., MacDonald, K. P., Cavanagh, L. L., Thomas, G., Townsend, W., Handel, M., and Thomas, R. (1997).J. Immunol. 159, 3681-3691. Pickl, W. F., Majdic, O., Kohl, P., Stockl, J., Riedl, E., Scheinecker, C., Bello-Fernandez, C., and Knapp, W. (1996).J. Immunol. 157, 3850-3859. Piemonti, L., Bernasconi, S., Luini, W., Trobonjaca, Z., Minty, A., AUavena, P., and Mantovani, A. (1995). Eur. Cytokine Netw. 6, 245-252. Pierre, P., and Mellman, I. (1998). 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). Nature 388, 787-792. Pinchuk, L. M., Polacino, P. S., Agy, M. B., Klaus, S. J., and Clark, E. A. (1994). Immunity 1,317-325. Pinchuk, L. M., Klaus, S. J., Magaletti, D. M., Pinchuk, G. V., Norsen, J. P., and Clark, E. A. (1996).J. Immunol. 157,4363-4370. Po@, A., Rubartelli, A., Moretta, L., and Zocchi, M. R. (1997).Eur. J. Zmmunol. 27,29652970. Poggi, A., Rubartelli, A,, and Zocchi, M. R. (1998a).J. B i d . Chem. 273, 7205-7209. Poggi, A., Costa, P., Tomasello, E., and Moretta, L. (1998b).Eur.J.Immunol.28,1611-1616. Pope, M., Betjes, M. G., Romani, N., Hirmand, H., Cameron, P. U., Hoffman, L., Gezelter, S., Schuler, G., and Steinman, R. M. (1994). Cell 78, 389-398. Pope, M., Gezelter, S., Gallo, N., Hoffman, L., and Steinman, R. M. (1995).J. Exp. Med. 182,2045-2056. Porcelli, S., Morita, C. T., and Brenner, M. B. (1992). Nature 360, 593-597. Porgador, A., and Gilboa, E. (1995).I. Exp. Med. 182, 255-260. Power, C. A., Church, D. J., Meyer, A,, Alouani, S., Proudfoot, A. E., Clark-Lewis, I., Sozzani, S., Mantovani, A,, and Wells, T. N. (1997).J. Exp. Med. 186, 825-835. Pulendran, B., Lingappa, J., Kennedy, M. K., Smith, J., Teepe, M., Rudensky, A., Maliszewski, C. R., and Maraskovsky, E. (1997).1.Immunol. 159,2222-2231. Pulendran, B., Smith, J. L., Jenkins, M., Schoenborn, M., Marakowsky, E., and Maliszewski, C. R. (1998)./. Exp. Med. 188, 2075-2082. Qian, S., Demetris, A. J., Murase, N., Rao, A. S., Fung, J. J., and Starzl, T. E. (1994). Hepatology 19, 916-924. Qin, Z., Noffz, G., Mohaupt, M., and Blankenstein, T. (1997).J. Immunol. 159, 770-776. Reddy, A., Sapp, M., Feldman, M., Subklewe, M., and Bhardwaj, N. (1997).Blood 90,36403646. Reid, P. A., and Watts, C. (1990). Nature 346, 655-657. Reid, C. D., Stackpoole,A,, Meager, A., andTikerpae, J. (1992).J.Immunol. 149,2681-2688. Reis e Sousa, C., Stahl, P. D., and Austyn, J. M. (1993).]. Exp. Med. 178, 509-519. Res, P., Martinez-Caceres, E., Cristina Jaleco, A,, Staal, F., Noteboom, E., Weijer, K., and Spits, H. (1996). Blood 87, 5196-5206. Rescigno, M., Citterio, S., Thery, C., Rittig, M., Medaglini, D., Pozzi, G., Amigorena, S., and Ricciardi-Castagnole, P. (1998). Proc. Nutl. Acad. Sci. U.S.A. 95, 5229-5234. Rettig, M. B., Ma, H. J., Vescio, R. A,, Pold, M., Schiller, G., Belson, D., Savage, A., Nisbikuho, C., Wu, C., Fraser, J., Said, J. W., and Berenson, J. R. (1997). Science 276, 1851-1854.

DENDRITIC CELLS

319

Richters, C. D., Reits, E. A., Van Pelt, A. M., Hoekstra, M. J., Van Baare, J., Du Pont, J. S., and Kamperdijk, E. W. (1996). Clin. Exp. Zmmunol. 104, 191-197. Ridge, J. P., Di Rosa, F., and Matzinger, P. (1998). Nature 393,474-478. Riedl, E., Strobl, H., Majdic, O., and Knapp, W. (1997).J. Zmmunol. 158, 1591-1597. Rieger, A., Wang, B., Kilgus, O., Ochiai, K., Mauerer, D., Fodinger, D., Kinet, J. P., and Stingl, G. (1992)./. Invest. Dermutol. 99, 30s-325. Riese, R. J., Wolf, P. R., Bromme, D., Natkin, L. R., Villadangos, J. A., Ploegh, H. L., and Chapman, H. A. (1996). Zmmunity 4, 357-366. Roake, J. A., Rao, A. S., Moms, P. J.. Larsen, C. P., Hankins, D. F., and Austyn, J. M. (1995).J. Erp. Med. 181, 2237-2247. Robinson, D. S., Hamid, Q., Jacobson, M., Ying, S., Kay, A. B., and Durham, S. R. (1993). Springer Semin. Immunopathol. 15, 17-27. Rocha, B., and von Boehmer, H. (1991). Science 251, 1225-1228. Rock, K. L., Yeh, E. T., Gramm, C. F., Haber, S. I., Reiser, H., and Benacerraf, B. (1986). J. Exp. Med. 163,315-333. Romani, N., Stingl, G., Tschachler, E., Witmer, M. D., Steinman, R. M., Shevach, E. M., and Schuler, G. (1985).J. Exp. Med. 161, 1368-1383. Romani, N., Koide, S., Crowley, M., Witmer-Pack, M., Livingstone, A. M., Fathman, C. G., Inaba, K., and Steinman, R. M. (1989).J. Exp. Med. 169, 1169-1178. Romani, N., Gruner, S., Brang, D., Kampgen, E., Lenz, A., Trockenbacher, B., Konwalinka, G., Fritsch, P. O., Steinman, R. M., and Schuler, G. (1994).J. Exp. Med. 180, 83-93. Romani, N., Reider, D., Heuer, M., Ebner, S., Kampgen, E., Eibl, B., Niedenvieser, D., and Schuler, G. (1996).J. Zmmunol. Methods 196, 137-151. Ross, R., Ross, X. L., Schwing, J., Langin, T., and Reske-Kunz, A. B. (1998). 1.Zmmunol. 160,3776-3782. Rubartelli, A., Po@, A,, and Zocchi, M. R. (1997). Eur. J. Immunol. 27, 1893-1900. Rubbert, A., Combadiere, C., Ostrowski, M., Arthos, J., Dybul, M., Machado, E., Cohn, M. A,, Hoxie, J. A,, Murphy, P. M., Fauci, A. S., and Weissman, D. (1998).J. Zmmunol. 160,3933-3941. Rued, C., Rieser, C., Bock, G., Wick, G., and Wolf, H. (1996).Eur. J. Zmnzunol. 26, 18011806. Ryncarz, R. E., and Anasetti, C. (1998). Blood 91, 3892-3900. Said, J. W., Rettig, M. R., Heppner, K., Vescio, R. A,, %hiller, G., Ma, H. J.. Belson, D., Savage, A., Shintaku, I. P., KoeMer, H. P., Asou, H., Pinkus, G., Pinkus, J., Schrage, M., Green, E., and Berenson, J. R. (1997). Blood 90,4278-4282. Sallusto, F., and Lanzavecchia, A. (1994).J. Exp. Med. 179, 1109-1118. Sdlusto, F., Cella, M., Danieli, C., and Lanzavecchia,A. (1995).]. Exp. Med. 182,389-400. Sallusto, F., Nicolo, C., De Maria, R., Corinti, S., and Testi, R. (1996). J. Exp. Med. 184, 2411-2416. Salomon, B., Lores, P., Pioche, C., Racz, P., Jami, J., and Klatzmann, D. (1994).]. Zmmunol. 152,537-548. Santiago-Schwarz, F., Belilos, E., Diamond, B., and Carsons, S. E. (1992).J. Leukoc. Biol. 52, 274-281. Sato, N., Caux, C., Kitamura, T., Watanabe, Y., Arai, K., Banchereau, J. (1993). B h d 82, 752-761. Saunders, D., Lucas, K., Ismaili, J., Wu, L., Maraskovsky, E., Dunn, A., and Shortman, K. (1996).1. Exp. Med. 184, 2185-2196. Schnorr, J. J., Xanthakos, S., Keikavoussi, P., Kampgen, E., ter Meden, V., and SchneiderSchaulies, S. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 5326-5331.

320

DIANA BELL et al.

Schoenberger, S. P., Toes, R. E., van der Voort, E. I., Offringa, R., and Melief, C. J. (1998). Nature 393, 480-483. Schon-Hegrad, M. A., Oliver, J,, McMenamin, P. G., and Holt, P. G. (1991).J. Exp. Med. 173,1345-1356. Schrader, C. E., and Cebra, J. J. (1993).Adu. Exp. Med. Biol. 329, 59-64. Schrader, C. E., George, A,, Kerlin, R. L., andcebra, J. J. (1990).Int. Immunol. 2,563-570. Schriever, F., and Nadler, L. M. (1992).Adv. Immunol. 51, 243-284. Schroder, S., Schwarz, W., Rehpenning, W., Loning, T., and Bocker, W. (1988). Am. J. Clin. Pathol. 89, 295-300. Schnler, G., and Steinman, R. M. (1985).J. Exp. Med. 161, 526-546. Schwarzenberger, K., and Udey, M. C. (1996).J. Invest. Dermutol. 106, 553-558. Semper, A. E., and Hartley, J. A. (1996). Clin. Exp. Allergy 26, 485-490. Shah, A. J., Srnogorzewska,E. M., Hannum, C., and Crooks, G. M. (1996).Blood 87,35633570. Shurin, M. R., Pandharipande, P. P., Zorina, T. D., Haluszczak, C., Subbotin, V. M., Hunter, O., Bnimfield, A,, Storkus, W. J., Maraskovsky, E., and Lotze, M. T. (1997).Cell. Immunol. 179, 174-184. Siena, S., Di Nicola, M., Bregni, M., Mortarini, R., Anichini, A., hmbardi, L., Ravagnani, F., Parmiani, G., and Gianni, A. M. (1995). Exp. Hemutol. 23, 1463-1471. Silberberg-Sinakin, I., Thorbecke, G. J., Baer, R. L., Rosenthal, S. A., and Berezowsky, V. (1976). Cell. Immunol. 25, 137-151. Simon, J. C., Tigelaar, R. E., Bergstresser, P. R., Edelbanm, D., and Cruz, P. D., Jr. (1991). J. Imrnunol. 146, 485-491. Simonet, W. S., Lacey, D. L., Dunstan, C. R., Kelley, M., Chang, M. S., Luthy, R., Nguyen, H. Q., Wooden, S., Bennett, L., Boone, T., Shimamoto, G., DeRose, M., Elliott, R., Colombero, A., Tan, H. L., Trail, G., Sullivan, J., Davy, E., Bucay, N., Renshaw-Gegg, L., Hughes, T. M., Hill, D., Pattison, W., Campbell, P., Boyle, W. J., et al. (1997). Cell 89,309-319. Skeen, M. J., Miller, M. A., Shinnick, T. M., and Ziegler, H. K. (1996). J. Immunol. 156, 1196-1206. Soderberg, C., Larsson, S., Rozell, B. L., Sumitran-Karuppan, S., Ljungman, P., and Moller, E. (1996). Transplantation 61, 600-609. Soderberg-Naucler, C., Fish, K. N., and Nelson, J. A. (1997). Cell 91, 119-126. Song, W., Kong, H. L., Carpenter, H., Torii, H., Granstein, R., Rafii, S., Moore, M. A., and CrystaI, R. G. (1997). J. Exp. Med. 186, 1247-1256. Sornasse,T., Flamand, V., De Becker, G., Bazin, H., Tielemans, F., Thielemans, K., Urbain, J., Leo, O., and Moser, M. (1992).J. Exp. Med. 175, 15-21. Sotzik, F., Rosenberg, Y., Boyd, A. W., Honeyman, M., Metcalf, D., Scollay, R., Wu, L., and Shortman, K. (1994).J. Immunol. 152, 3370-3377. Sonsa, C. R., Hieny, S., Scharton-Kersten, T., Jankovic, D., Charest, H., Germain, R. N., and Sher, A. (1997).J. Exp. Med. 186, 1819-1829. Sozzani, S., Sallusto, F., Luini, W., Zhou, D., Piemonti, L., Allavena, P., Van Damme, J., Valitutti, S., Lanzavecchia, A,, and Mantovani, A. (1995).J. Immunol. 155, 3292-3295. Sozzani, S., Locati, M., Allavena, P., Van Damme, J., and Mantovani, A. (1996). Int. J. Clin. Lab. Res. 26,69-82. Sozzani, S., Allavena, P., D’Amico, G., Luini, W., Bianchi, G., Kataura, M., Imai, T., Yoshie, O., Bonecchi, R., and Mantovani, A. (1998).J. Immunol. 161, 1083-1086. Spalding, D. M., and Griffin, J. A. (1986). Cell 44, 507-515. Specht, J. M., Wang, G., Do, M. T., Lam, J. S., Royal, R. E., Reeves, M. E., Rosenberg, S. A., and Hwu, P. (1997).J. Exp. Med. 186, 1213-1221.

DENDRITIC CELLS

32 1

Starhg, G. C., McLellan, A. D., Egner, W., Sorg, R. V., Fawcett, J., Simmons, D. L., and Hart, D. N. (1995). Eur. J. Immunol. 25, 2528-2532. Starr, S. (1964). N. Engl. J. Med. 270, 386-391. S t a d , T. E., Demetris, A. J., Trucco, M., Ramos, H., Zeevi, A,, Rudert, W. A., Kocova, M., Ricordi, C., Ildstad, S., and Murase, N. (1992). Lancet 340, 876-877. Starzl, T. E., Demetris, A. J., Trucco, M., Murase, N., Ricordi, C., Ildstad, S., Ramos, H., Todo, S., Tzahs, A,, Fung, J. J., et ul. (1993). Hepatology 17, 1127-1152. Starzl, T. E., Demetris, A. J., Murase, N., Tmcco, M., Thomson, A. W., and Rao, A. S. (1996). Immunol. Today 17, 577-584; discussion, 588. Stard, T. E., Demetris, A. J.. Murase, N., Valdivia, L., Thomson, A. W., Fung, J., and Rao, A. S. (1997). Transplant. Proc. 29, 19-27. Steinbrink, K., Wolfl, M., Jonuleit, H., Knop, J., and Enk, A. H. (1997). /. Immunol. 159,4772-4780. Steinman, R. M. (1991). Annu. Reu. Immunol. 9, 271-296. Steinman, R. M., and Cohn, Z . A. (1973).]. Exp. Med. 137, 1142-1162. Steinman, R. M., and Witmer, M. D. (1978). Proc. Nut!. Acud. Sci. U.S.A. 75, 5132-5136. Steinman, R. M., Pack, M., and Inaba, K. (1997). Immunol. Rev. 156, 25-37. Steptoe, R. J., and Thomson, A. W. (1996). Clin. Exp. Immunol. 105, 397-402. Stingl, G., and Maurer, D. (1997). Int. Arch. Allergy Irnmunol. 113, 24-29. Stingl, G., Wolff-Schreiner, E. C., Pichler, W. J., Gschnait, F., Knapp, W., and Wolff, K. (1977). Nature 268, 245-246. Streilein, J. W. (1997). Eye 11, 171-175. Strobl, H., Bello-Fernandez, C., Riedl, E., Pickl, W. F., Majdic, O., Lyman, S. D., and Knapp, W. (1997). Blood 90, 1425-1434. Strunk, D., Rappersberger, K., Egger, C., Strobl, H., Kromer, E., Elbe, A., Maurer, D., and Stingl, G. (1996). Blood 87, 1292-1302. Strunk, D., Egger, C., Leitner, G., Hanau. D., and Stingl, G. (1997).J.Exp. Med. 185,11311136. Suda, T., Udagawa, N., Nakamura, I., Miyaura, C., and Takahashi, N. (1995). Bone 17, 87s-91s. Sullivan, S., Bergstresser, P. R., Tigelaar, R. E., and Streilein, J. W. (1985). J. Invest. Dennatol. 84,491-495. Summers, K. L., Daniel, P. B., O’Donnell, J. L., and Hart, D. N. (1995a). Clin. Exp. Immunol. 100,81-89. Summers, K. L., O’Donnell, J. L., Daniels, P. B., and Hart, D. N. (1995b).Ado. Exp. Mecl. B i d 378,561-563. Summers, K. L., O’Donnell, J. L., Williams, L. A,, and Hart, D. N. (1996).Arthritis Rheum. 39, 1287-1291. Suss, G., and Shortman, K. (1996).J. Exp. Med. 183, 1789-1796. Svensson, M., Stockinger, B., and Wick, M. J. (1997)./. lmmunol. 158,4229-4236. Szabolcs, P., Moore, M. A,, and Young, J. W. (1995).J. Immunol. 154,5851-5861. Szabolcs, P., Avigan, D., Gezelter, S., Ciocon, D. H., Moore, M. A,, Steinman, R. M., and Young, J. W. (1996). Blood 87, 4520-4530. Szakal, A. K., and Hanna, M. G., Jr. (1968). Exp. Mol. Puthol. 8, 75-89. Szakal, A. K., Gieringer, R. L., Kosco, M. H., and Tew, J. G. (1985).1.lmmunol. 134,13491359. Szakal, A. K., Kosco, M. H., and Tew, 1. G. (1989). Annu. Rev. Immunol. 7, 91-109. Takahashi, H., Nakagawa, Y., Yokomuro, K., and Berzofsky, J. A. (1993). Int. bnmunol. 5, 849-857.

322

DIANA BELL et al

Takashima, A., Edelbaum, D., Kitajima, T., Shadduck, R. K., Gilmore, G. L., Xu, S., Taylor, R. S., Bergstresser, P. R., and Ariizumi, K. (1995).J. Immunol. 154, 5128-5135. Tang, A., Amagai, M., Granger, L. G., Stanley, J. R., and Udey, M. C. (1993). Nature 361, 82-85. Tenner-Racz, K., Racz, P., Schmidt, H., Dietrich, M., Kern, P., Louie, A., Gartner, S., and Popovic, M. (1988). Aids 2,299-309. Teunissen, M. B., Wormmeester, J., Krieg, S. R., Peters, P. J., Vogels, I. M., Kapsenberg, M. L., and Bos, J. D. (1990).J . Invest. Dennatol. 94, 166-173. Tew, J. G., Kosco, M. H., Burton, G. F., and Szakal, A. K. (1990). Immunol. Reu. 117, 185-21 1. Thepen, T., McMenamin, C., Gim, B., Kraal, G., and Holt, P. G. (1992).Clin. Exp. Allergy 22, 1107-1114. Thomas, R., Davis, L. S., and Lipsky, P. E. (1993).J. Immunol. 150,821-834. Thomas, R., Davis, L. S., and Lipsky, P. E. (1994).J. Zmmunol. 152,2613-2623. Thomson, A. W., Lu, L., Murase, N., Demetris, A. J., Rao, A. S., and S t a d , T. E. (1995). Stem Cells (Dayt.) 13, 622-639. Thomssen, H., Kahan, M., and Londei, M. (1995). Eur. J. Immunol. 25,2465-2470. Thurnher, M., Ramoner, R., Gastl, G., Radmayr, C., Bock, G., Herold, M., Klocker, H., and Bartsch, G. (1997). lnt. J. Cancer 70, 128-134. Tjoa, B., Erickson, S., Barren, R. R., Ragde, H., Kenny, G., Boynton, A., and Murphy, G. (1995). Prostate 27, 63-69. Tjoa, B., Boynton, A,, Kenny, G., Ragde, H., Misrock, S. L., and Murphy, G. (1996).Prostate 28,65-69. Tjoa, B. A., Erickson, S. J., Bowes, V. A., Ragde, H., Kenny, G. M., Cobb, 0. E., Ireton, R. C., Troychak, M. J., Boynton, A. L., and Murphy, G. P. (1997).Prostate 32,272-278. Toes, R. E., Blom, R. J., Offringa, R., Kast, W. M., and Melief, C. J. (1996a).J. Immunol. 156,3911-3918. Toes, R. E., Offringa, R., Blom, R. J., Melief, C. J., and Kast, W. M. (1996b). Proc. Natl. Acad. Sci. U.S.A. 93, 7855-7860. Toes, R. E., van der Voort, E. I., Schoenberger, S. P., Drijfhout, J. W., van Bloois, L., Storm, G., Kast, W. M., Offringa, R., and Melief, C. J. (1998).J.Immunol. 160,4449-4456. Toews, G . B., Bergstresser, P. R., and Streilein, J. W. (1980).J. Immunol. 124,445-453. Trinchieri, G. (1995). Annu. Reu. Zmmunol. 13,251-276. Tsujitani, S., Kakeji, Y.,Watanabe, A., Kohnoe, S., Maehara, Y., and Sugimachi, K. (1990). Cancer 66,2012-2016. Tunon-De-Lara, J. M., Redington, A. E., Bradding, P., Church, M. K., Hartley, J. A., Semper, A. E., and Holgate, S. T. (1996). C h . Exp. Allergy 26,648-655. Uherova, P., Connick, E., MaWhinney, S., Schlichtemeier, R., Schooley, R. T., and Kuritzkes, D. R. (1996).]. Infect Dis. 174, 483-489. Van Kooten, C., and Banchereau, J. (1996). Adu. Immunol. 61, 1-77. Van Voorhis, W. C., Valinsky, J., Hoffman, E., Luban, J., Hair, L. S., and Steinman, R. M. (1983).J. Exp. Med. 158, 174-191. Viac, J., Schmitt, D., and Claudy, A. (1997). Anticancer Res. 17, 569-572. Vicari, A. P., Figueroa, D. J., Hedrick, J. A., Foster, J. S., Singh, K. P., Menon, S., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Bacon, K. B., and Zlotnik, A. (1997). Immunity 7,291-301. Villadangos, J. A,, Riese, R. J., Peters, C., Chapman, H. A., and Ploegh, H. L. (1997). J. Exp. Med. 186,549-560. Viney, J. L., Mowat, A. M., O M d e y , J. M., Williamson, E., and Fanger, N. A. (1998). 1.lmmunol. 160,5815-5825.

DENDRITIC CELLS

323

Vink, A. A., Strickland, F. M., Bucana, C., Cox, P. A., Roza, L., Yarosh, D. B., and Kripke, M. L. (1996).J. Exp. Med. 183, 1491-1500. Vink, A. A., Moodycliffe, A. M., Shreedhar, V., Ullrich, S. E., Roza, L., Yarosh, D. B., and Kripke, M. L. (1997). Proc. Nad. Acad. Sci. U.S.A. 94,5255-5260. Von Boehmer, H., and Schubiger, K. (1984). Eur. J. Immunol. 14, 1048-1052. Vremec, D., Zorbas, M., Scollay, R., Saunders, D. J., Ardavin, C. F., Wu, L., and Shortman, K. (1992).J. Exp. Med. 176, 47-58. Walker, P. R., Saas, P., and Dietrich, P. Y. (1997).J. Immunol. 158, 4521-4524. Wang, B., Rieger, A., Kilgus, O., Ochiai, K., Maurer, D., Fodinger, D., Kinet, J. P., and Stingl, G. (1992).J. Exp. Med. 175, 1353-1365. Ward, B. J., and Griffin, D. E. (1993). Clin. Immunol. Immunopathol. 67, 171-177. Warthin, A. S. (1931). Arch. Pathol. 11, 864-874. Watts, C. (1997).Annu. Rev. Immunol. 15, 821-850. Weiner, H. L. (1997). Annu. Rev. Med. 48,341-351. Weiss, J. M., Sleeman, J., Renkl, A. C., Dittmar, H., Termeer, C. C., Taxis, S., Howells, N., Hofmann, M., Kohler, G., Schopf, E., Ponta, H., Herrlich, P., and Simon, J. C. (1997). J. Cell Biol. 137, 1137-1147. Weissman, D., and Fauci, A. S. (1997). Clin. Microbiol. Reo. 10, 358-367. Weissman, D., Rabin, R. L., Arthos, J., Rubbert, A,, Dybul, M., Swofford, R., Venkatesan, S., Farber, J, M., and Fauci, A. S. (1997). Nature 389, 981-985. Westerink, M. A., Metzger, D. W., Hutchins, W. A., Adkins, A. R., Holder, P. F., Pais, L. B., Gheesling, L. L., and Carlone, G. M. (1997).J. Infect. Dis. 175, 84-90. Whitby, D., Boshoff, C., Luppi, M., and Torelli, G. (1997). Science 278, 1971-1972; discussion, 1972-1973. Will, A., Blank, C., Rohghoff, M., and Moll, H. (1992).Eur. J. Immunol. 22, 1341-1347. Winzler, C., Rovere, P., Rescigno, M., Granucci, F., Penna, G., Adorini, L., Zimmermann, V. S., Davoust, J., and Ricciardi-Castagnoli, P. (1997).J. Exp. Med. 185, 317-328. Witmer-Pack, M. D., Olivier, W., Valinsky, J., Schuler, G., and Steinman, R. M. (1987). J. Exp. Med. 166, 1484-1498. Witmer-Pack, M. D., Valinsky, J., Olivier, W., and Steinman, R. M. (1988). J. Invest. D e m t o l . 90,387-394. Wong, B. R., Josien, R., Lee, S. Y., Sauter, B., Li, H. L., Steinman, R. M., and Choi, Y. (1997).J. Exp. Med. 186, 2075-2080. Wu, L., Vremec, D., Ardavin, C., Winkel, K., Suss, G., Georgiou, H., Maraskovsloj, E., Cook, W., and Shortman, K. (1995). Eur. J. Immunol. 25,418-425. Wu, L., Li, C. L., and Shortman, K. (1996).]. Exp. Med. 184, 903-911. Wyatt, R., and Sodroski, J. (1998). Science 280, 1884-1888. Wykes, M., Pombo, A., Jenkins, C., and MacPherson, G. G. (1998).J.Immunol. 161,13131319. Xia, W., Pinto, C. E., and Kradin, R. L. (1995).J. Exp. Med. 181, 1275-1283. Yamaguchi, Y., Tsumura, H., Miwa, M., and Inaba, K. (1997). Stem C e h 15, 144-153. Yasuda, H., Shima, N., Nakagawa, N.,Yamaguchi, K., Kinosaki,M., Mochizuki, S.,Tomoyasu, A,, Yano, K., Goto, M., Murakami, A., Tsuda, E., Morinaga, T., Higashio, K., Udagawa, N., Takahashi, N., and Suda, T. (1998). Proc. Natl. Acad. Sci. U.S.A. 95, 3597-3602. Yi, Q., Ekman, M., Anton, D., Bergenbrant, S., Osterborg, A., Georgii-Hemming, P., Holm, G., Nilsson, K., and Biberfeld, P. (1998). Blood 92, 402-404. Young, J. W., and Steinman, R. M. (1990).J , Exp. Med. 171, 1315-1332. Young, J. W., Koulova, L., Soergel, S. A., Clark, E. A., Steinman, R. M., and Dupont, B. (1992).J. Clin. Invest. 90, 229-237. Young, J. W., Bagger, J., and Soergel, S. A. (1993). Blood 81, 2987-2997.

324

DIANA BELL et al.

Young, J. W., Szabolcs, P., and Moore, M. A. (1995).J.Exp. Med. 182, 1111-1119. Zaitseva, M., Blauvelt, A., Lee, S., Lapham, C. K., Klaus-Kovtun, V., Mostowski, H., Manischewitz, J., and Golding, H. (1997).Nature Med. 3, 1369-1375. Zhou, L. J., and Tedder, T. F. (1995a).Blood 86,3295-3301. Zhou, L. J., and Tedder, T. F. (1995b).J. Immunol. 154, 3821-3835. Zhou, L. J., and Tedder, T. F. (1996). Proc. Natl. Acad. Sci. U.S.A. 93,2588-2592. Zhou, L. J., Schwarting, R., Smith, H. M., and Tedder, T. F. (1992). J. Immunol. 149, 735-742. Zitvogel, L., Mayordomo, J. I., Tjandrawan, T., DeLeo, A. B., Clarke, M. R., Lotze, M. T., and Storkus, W. J. (1996).J. Exp. Med. 183, 87-97. Zitvogel, L., Regnault, A., Lozier, A,, Wolfers, J., Flament, C., Tenza, D., RicciardiCastagnoli, P., Raposo, G., and Amigorena, S. (1998). Nature Med. 4, 594-600. Zocchi, M. R., Po@, A., and Rubartelli, A. (1997).AIDS 11, 1227-1235. Zvaifler, N. J., Steinman, R. M., Kaplan, G., Lau, L. L., and Rivelis, M. (1985). J. Clin. Invest. 76, 789-800. This article was accepted for publication on August 28, 1998.

ADVANCES IN IMMUNOLOGY, VOL 72

lntegrins in the Immune System YOJl SHIMIZU,*.' DAWD M. ROSE, AND MARK H. GINSBERG 'Deparhnent of laboratory Medicine and hhology, Center for Imrnunobgy, Cancer Center, Universiv of Minnesota Medical School, Minneapolis, Minnesota 55455; and Deparhnent of Vascular Bidqy, The Scripps Research Institute, la Idla, Califomio 92037

1. Introduction

The remarkable information storage and retrieval capacities of the immune system employ the general machinery involved in vertebrate development. Among the components of this machinery are the integrins, a family of cell surface receptors vital to the development and functioning of multicellular animals (Hynes, 1992). Integrins are important because they play key roles in cell migration, in cell adhesion, in control of differentiation, and in critical decisions specifjmg cell proliferation and programmed cell death (Schwartz et al., 1995; Hynes, 1992). As will be elaborated below, integrin receptors are essential for the development of the immune system and in the generation of the immune response. In this article, we discuss principles governing the integrins, their capacity to recognize ligands, cellular regulation of integrin function, and aspects of integrin function in the immune system. Our particular goal is to bring these general principles into focus in understanding the immune system and to describe how perturbation of these receptors can be used to modulate immunological responses. II. Ligand-Binding Sites in Integrins

Integrins are heterodimers formed by combination of 17 cr and 8 0 subunits in humans. These subunits are products of separate genes (Ginsberg et al., 199213; Hynes, 1992; Akiyama et al., 1990; Albelda and Buck, 1990) and are generally interdependent for biosynthetic processing and surface expression (O'Toole et al., 1989). Integrins contain an extracellular domain formed by the -1000-residue a and -750-residue /3 subunits. Each subunit contains a single transmembrane segment. At least 8 integrin a subunits contain in their amino-terminal third (Hemler, 1990; Shaw et al., 1994; Camper et al., 1998) a -200-residue sequence that was inserted by exon shuffling (Fleming et al., 1993; Hughes, 1992; Larson et al., 1989; Springer, 1990). The region is referred to as the I (inserted) domain and it is homologous to the A domains of von Willebrand factor (Bonthron et

' To whom correspondence should be addressed. 325

Copyright 0 1909 by Academic Press. All rights of reproduction in any funn reserved. 0 0 6 5 - m $30.00 ~ ~

326

YOJI SHIMIZU et d.

al., 1986; Shelton-Inloes et al., 1986; Titani et al., 1986; Venveij et al., 1986; Corbi et al., 1988). In general, integrin subunits possess a short cytoplasmic C-terminal tail (Sastry and Honvitz, 1993;Hemler et at.,1994; Williams et al., 1994), with the exception being p4, whose cytoplasmic domain is more than 1000 residues (Hogervorst et al., 1990; Suzuki and Naitoh, 1990; Tamura et al., 1990). Integrins were thought to consist of 3 p subunits, which form unique heterohmers with different sets of a subunits (Ginsberg et al., 1987; Hynes, 1987). However, a subunits appear to have evolved independently of /3 subunits (Takada and Hemler, 1989) and several a subunits have multiple p partners (Kajijiet al., 1989; Cheresh et al., 1989; Hemler et al., 1989).Consequently, more complicated schemes have been required to characterize this family (Hynes, 1992). Most of the more than 20 known mammalian integrins are expressed on one or more of the many cell types involved in the immune response. Seven of these receptors are predominantly (but not exclusively) involved in immunity: five of these contain I domains in their a subunits (aEP7, aLp2, aMP2, axp2, aDP2) and two (a4p1, ~2407)lack the I domain. Electron microscopy suggests that these receptors are composed of a ligand-binding (Weisel et al., 1992) globular -10-nm head (Nermut et al., 1988; Carrel1 et al., 1985), and two extended tails that contain carboxylterminal portions of a (Weisel et al., 1992) and fl (Du et al., 1993) subunits and their transmembrane domains (Parise and Phillips, 1985). Biophysical analyses of integrins in detergent solution indicate that these are asymmetric molecules (Jennings and Phillips, 1982; Hantgan et al., 1993). In addition, disulfide bond arrangements and intersubunit contacts have been proposed in proteolytic fragments of integrin, aIIbp3 (Calvete et al., 1989a,b, 1991b,c, 1992,1994). Integrins are conformationallylabile (Parise et al., 1987; Steiner et at., 1991; Frelinger et al., 1990, 1988) and are subject to disulfide bond exchange (Phillips and Agin, 1977). Thus, the existing studies should be viewed as only a beginning in relating integrin structure to ligand-binding function. Proteolytic (Lam, 1992) and recombinant truncated (Wippler et al., 1994) fragments of integrins that contain the N-terminal half of both subunits bind ligand. Chemical cross-linking studies also indicate that ligand recognition sites reside in the N-terminal portion of both (Santoro and Lawing, 1987; Smith and Cheresh, 1988, 1990; Santoro and Lawing, 1987; D’Souza et al., 1988) subunits. In addition, high-affinity ligand recognition usually requires both subunits (Fitzgerald and Phillips, 1985; Pidard et al., 1986; Brass et al., 1985; Kunicki et al., 1981; ShattiI et al., 1985; Buck et al., 1986) and consequently may involve multiple ligand contact points. Several such potential contact points have now been identified.

INTEGRINS IN T H E IMMUNE SYSTEM

327

A conserved sequence in /33(D109-E171) is proximal to a bound peptide ligand (D’Souzaetal., 1988;Smith and Cheresh, 1988).This region is probably involved in ligand recognition because (1)point mutations here abrogate ligand-binding function (Loftus et al., 1990), (2)certain antibodies directed against this region inhibit ligand binding (Andrieux et al., 1991; Calvete et al., 1991a;Wangetal., 1992;D’Souzaet al., 1994),(3)amutation herecauses gain of ligand-binding function (Bajt et al., 1992),and (4)an isolated peptide from this region binds ligand (D’Souza et al., 1994). This region is highly conserved among integrin classes (Loftus et al., 1990), suggesting that this is a common ligand contact site. That hypothesis has been supported by the loss of ligand-binding function associated with mutations at similar residues in this region in other integrins (Takada et al., 1992; Bajt et al., 1995; Huang et al., 1993; Bajt and Loftus, 1994; Puzon-McLaughlin and Takada, 1996; Takada et al., 1992; Lin et al., 199%). This highly conserved region of integrin /3 subunits is rich in oxygenated residues whose linear spacing approximates that of the oxygenated residues in the calcium-binding loop of E F hand proteins (Kretsinger, 1980). Furthermore, mutations of these (Loftus et al., 1990; Bajt et al., 1995; Bajt and Loftus, 1994; Puzon-McLaughlin and Takada, 1996; Takada et al., 1992; Lin et al., 199%) oxygenated residues block ligand binding and can alter divalent cation-dependent conformation (Ginsberg et al., 1996), suggesting that these residues may provide coordinating ligands for divalent cations. A synthetic peptide from this region directly binds ligands and terbium, a luminescent calcium analog (D’Souza et al., 1994). Thus, these data support the hypothesis that ligands interact with divalent cations bound to this highly conserved site in the /3 subunit (Loftus et al., 1990). This idea is also supported by the presence of critical oxygenated residues in integrin ligands (Ginsberg et al., 1985; Pierschbacher and Ruoslahti, 1984a,b; Ruggeri et al., 1986; Humphries et al., 1986; Komoriya et al., 1991; Vonderheide et al., 1994; Osborn et al., 1994; Berendt et al., 1992). In addition to this conserved site, other nonconserved sites appear to contribute to ligand recognition specificity by different /3 subunits (Lin et al., 1997a; Takagi et al., 1997). Striking similarities exist between the ligand-binding region of the /3 subunit and the ligand-binding region of I domains (of the a subunit). These include a conserved motif of D ( 4)SDXSXS+,where 4 is any hydrophobic residue and X is any residue (Loftus et al., 1994). Furthermore, the similarity in hydropathy of the two regions suggested a similar global fold (Lee et al., 1995) (see below),which led to the proposal of hypothetical molecular models (Tozer et al., 1996; Lin et al., 199%). One important prediction of these models is the role of sequentially dispersed oxygenated residues in cation coordination and ligand binding. Such residues that impact ligand

328

YOJI SHIMIZU et d.

binding have been identified by site-directed mutagenesis (Tozer et al., 1996; Puzon-McLaughlin and Takada, 1996; Lin et al., 1997b) and by random mutagenesis (Baker et al., 1997). The validation of the proposed molecular models and resolution of their differences await a high-resolution structure of this region. Elegant studies from Springer’s group indicate that the folding of this region depends on its association with the a subunit (Huang et at., 1997). Consequently, a structural analysis of an a@ heterodimer or of proteolytic (Lam, 1992; Lam et al., 1989) or recombinant (McKay et al., 1996) “minidimers” will probably be required to resolve this issue. Bound ligands are also proximal to the N-terminal region of the a subunit (D’Souza et al., 1990; Smith and Cheresh, 1990). Moreover, antibodies against this region block ligand binding and synthetic peptides and fragments containing this region have been reported to bind ligands (D’Souza et al., 1991; Gulino et al., 1992; Stanley et al., 1994). Perhaps the most decisive studies have been a combination of epitope mapping, point mutagenesis, and domain exchange experiments, which have all strongly argued that repeats two to five of the seven N-terminal repeats of integrin a subunits are involved in ligand binding (Loftus et al., 1996; Irie et al., 1997; Kamata et al., 1995, 1996). An ab initio prediction proposed that the seven N-terminal repeats of integrin a subunits each fold into fourstranded @ sheets arranged in a torus around an axis of pseudosymmetry (the P propeller model) (Springer, 1997). In this model, the ligand-binding site is predicted to be on the upper surface of the propeller. To date this prediction has been borne out by mutational and immunochemical analyses (Irie et al., 1997; Oxvig and Springer, 1998).The lower surface, containing the cation-binding repeats, then could interface with the predicted I domainlike structure present in the 0 subunit. The remarkable similarity of this predicted protein-protein interaction to that between a and @ subunits of G proteins is appealing (Hamm, 1998).High-resolution structural studies of this region are anxiously awaited as a definitive test of this insightful model. Because this region’s folding seems to depend on interactions with the @ subunit (Lu et al., 1998; Huang and Springer, 1997), such structural studies will probably require analyses of either intact heterodimers or dimers of fragments. As noted above, at least eight integrin a subunits contain a -200-residue I domain that participates in ligand binding. Function-altering antibodies map to I domains of aM@2(Mac-1, CDllb/CD18, CR3) (Diamond et al., 1993; Michishita et al., 1993), alp1 (VLA-1) (Kern et al., 1994), aLP2 (LFA-1, CDlldCD18) (Randi and Hogg, 1994), and a2@l (VLA-2) (Kamata et al., 1994). Moreover, mutations in these domains block ligandbinding function (Kamata et al., 1994; Michishita et al., 1993; Kern et al.,

I N T E C R I N S I N THE I M M U N E SYSTEM

329

1994). Because these domains are protein modules (Fleming et al., 1993; Hughes, 1992; Larson et al., 1989; Springer, 1990) they fold autonomously (Perkins et al., 1994; Lee et al., 1995) as isolated proteins or in the intact a subunits (Lu et al., 1998; Huang and Springer, 1997) and bind ligands (Michishitaet al., 1993;Randi and Hogg, 1994; Kamata et al., 1994; Kamata and Takada, 1994). The I domain also binds cations (Michishita et nl., 1993) and cations seem to be important in its ligand-binding function (Michishita et al., 1993; Muchowski et al., 1994). Crystal structures of several I domains have been solved (Emsley et al., 1997; Lee et al., 1995; Qu and Leahy, 1995) and confirm the role of the DXSXS motif in coordinating bound cation. Two other coordination positions are filled by oxygenated residues near the DXSXS motif in the folded protein, but more than 50 residues distant in the primary sequence. These residues thus form the novel metal ion-dependent adhesion site (MIDAS) (Lee et al., 1995). The overall structure is that of a dinucleotide-binding fold with a central parallel p sheet surrounded by a helices. The MIDAS motif lies at the C-terminal end of the p sheet. Mutational analyses show that residues involved in ligand binding to the I domain cluster about the MIDAS face (Zhang and Plow, 1997; van Kooyk et al., 1996). A striking exception is the binding site for echovirus in a2, which maps to one edge of the p sheet (Emsley et al., 1997; King et al., 1997). The relative ligand-binding roles of the three contact sites described above remain unresolved. Clearly, in I domain-containing integrins, that domain is involved. Nevertheless, Bajt and colleagues (Bajt et al., 1995) showed that mutations of the DXSXS motif in p 2 impair ligand recognition by both aMP2 and aLp2. The effects are not likely to be due to alteration in the conformation of the autonomously folded I domain. Nor do these mutations appear to block accessibility of the I domain to certain ligands (Zhang and Plow, 1996). Furthermore, other studies implicate sites from the putative j3 propeller motif of the a! subunit in ligand binding to aLp2 (Stanley et al., 1994) and a 2 p l (Dickeson et al., 1997). Thus, in I domain integrins, other sites are probably also involved in ligand recognition. In integrins lacking the I domain, the picture is even less clear. One idea is that ligands could interact independently with spatially discrete sites in both a and p subunit. Studies using ligands derived from fibrinogen and fibronectin support this hypothesis. Peptides derived from fibrinogen y chain C terminus cross-link preferentially to aIIb, whereas RGD peptides cross-link to the p3 subunit (Santoro and Lawing, 1987). Furthermore, there is differential cross-competition between fibronectin-derived ligands lacking either the RGD sequence or the “synergy” sites and antibodies against the integrin a5 or pl subunit (Mould et al., 1997). However, the interdependence of the folding and function of a and p subunit ligand-

330

YOJI SHIMIZU et al.

binding sites (Huang and Springer, 1997; Huang et al., 1997; Lu et al., 1998; Fitzgerald and Phillips, 1985; Pidard et al., 1986; Brass et al., 1985; Kunicki et al., 1981; Shattil et al., 1985; Buck et al., 1986; Cheresh and Harper, 1987) suggests that both sites participate in the formation of a common ligand-binding pocket. Indeed, there is cross inhibition between fibrinogen y chain and RGD peptides for binding to aIIbp3 (Lam et at., 1987). Furthermore, the binding of the isolated “synergy” site to aIIb/33 is inhibited by RGD peptides (R. D. Bowditch and M. H. Ginsberg, unpublished observations). Visualization of the structure of an integrin heterodimer with bound ligand will probably be required to decipher the topology of the ligand-binding site. 111. lntegrin Ligands in the Immune System

Integrins bind to the extracellular matrix (Ginsberg et al., 1983; Pytela et al., 1985; Parise and Phillips, 1986), cell surface immunoglobulin (Ig) superfamilyreceptors (Marlin and Springer, 1987),microorganisms ( Isberg and Leong, 1990; Relman et al., 1990; Fernandez et al., 1993), and certain plasma proteins (Bennett and Vilaire, 1979; Bennett et al., 1982). Most of these interactions are divalent cation dependent, and mapping of integrin recognition sequences in ligands almost invariably identifies an important acidic residue (Ginsberg et al., 1985;Pierschbacher and Ruoslahti, 1984a,b; Ruggeri et al., 1986; Humphries et al., 1986; Komoriya et al., 1991;Vonderheide et al., 1994; Osborn et al., 1994; Berendt et al., 1992). In some protein ligands, integrin recognition sites can be assigned to short linear peptide sequences, e.g., Arg-Gly-Asp (Piershbacher and Ruoslahti, 1984a), but additional discontinuous regions of the protein may be involved (Obara et al., 1988; Bowditch et al., 1991, 1994; Aota et a!., 1994). Many integrins recognize short peptide sequences often presented in extended loops containing /3 turns (Leahy et al., 1992; Main et al., 1992; Adler et al., 1991; Taub et al., 1989; Tomiyama et al., 1992) (but see below). Other regions of protein ligands may then provide secondary interactive sites. Extracellular matrix integrin ligands are used for diverse biological processes. A subset of integrin ligands has a primary role in the immune system (Table I ) . This includes ICAMs (ligands for 0 2 integrins), VCAM1 (a ligand for a 4 P l and or4/37), and MAdCAM-1 (a ligand for or4/37). VCAM-1, originally thought to be specific to the immune system, plays an essential role in development of the heart and placenta (Kwee et al., 1995; Iademarco et al., 1992). Ligands having a more general role in integrin biology, such as the extracellular matrix proteins fibronectin, collagen, and laminin, have been the subject of excellent reviews (Hemler and Lobb, 1995).

33 1

INTEGRINS IN T H E IMMUNE SYSTEM

TABLE I INTEGRIN LIGANDS Integrin ligand

Expression

Major functions

ICAM-1

Widely expressed at low levels"

ICAM-2

Endothelium, leukocytes

ICAM-3 VCAM-1

Leukocytes Inflamed endothelium," endothelium (spleen bone marrow), reticular cells (bone marrow), follicular dendritic cells, thymic cortical epithelium

MAdCAM-1

HEVs" (Peyer's patches and mesenteric lymph nodes), endothelium (lamina propria)" Epithelium

Leukocyte recruitment during inflammation, ligand for rhinovirus and Plasmodium falcipamm, lymphocyte activation Lymphocyte recirculation, lymphocyte activation, target cell recognition Lymphocyte activation Leukocyte recruitment during inflammation, lymphocyte recirculation, retention of lymphocytes in lymphoid tissue, embryonic development (cardiac + placental) Homing of lymphocytes to gutassociated lymphoid tissues

E-Cadherin

+

Retentiadfunction of intestinal intraepithelial lymphocytes

" Inducible up-regulation by inflammatory stimuli.

A. STRUCTURE OF INTEGRIN LIGANDS In general, many of the integrin ligands involved in cell-cell interaction in the immune system contain Ig domains consisting of a sandwich of antiparallel P sheets (Fig. 1) (Leahy, 1997). Usually, the N-terminal Ig domain (domain 1) is most critical for integrin binding (Staunton et al., 1990; Li et al., 1995; Klickstein et al., 1996; Karecla et al., 1996). An exception to this rule is the interaction of aMP2 with domain 3 of ICAM1 (Diamond et al., 1991). The N-terminal Ig domain 1 of ICAM-1 is important for binding to nonintegrin ligands such as rhinovirus and Plusmodium falcipamm (Staunton et al., 1990). However, these bind to distinct sites on domain 1that only partially overlap with the integrin-binding site.

VCAM-1 contains two Ig domains that bind to integrins. The sevendomain form of VCAM-1 has a4 integrin-binding sites located in Ig domains 1 and 4 (Osborn et al., 1992),whereas the alternatively spliced sixdomain form lacks domain 4.Domain 1 and domain 4 can support cell

332

YOJI SHIMIZU et al.

A

ICAM-1 ICAM-2 ICAM-3

VCAM-1 104OSP

LD41T

CD loop

CD loop*

Critical Residue(s)

E34

E37

E40

Location

p-C

p-C

p-C*

MAdCAM-1 E-cadherin

E31 BC loop

B

I D

E

B

A

G

F

C

FIG.1. Integrin ligands of the immunoglobulin superfainily involved in immune cell-cell interactions. (A) Schematic of structures of Ig family members. Ig domains are numbered and domains important in integrin binding are shaded. The critical residues for integrin binding are listed, and the locations of the residues within the domains, as determined by X-ray crystal structure, are also shown (asterisks denote predicted location based on homology modeling). (B) Schematic representation of an Ig domain. The domain is composed of -100 amino acids organized into a sandwich of two sheets of antiparallel /3 strands (drawing is open faced). The /3 strands are connected by loop structures. Strands are labeled alphabetically from the N terminus.

adhesion independently (Osborn et al., 1992), but differences exist in cellular activation requirements for recognition of these domains (Needham et al., 1994). The a401 and a407 integrins are differentially regulated with respect to binding to domains 1 and 4 in static (Kilger et al., 1995) and in flowing (Abe et al., 1996) systems. The differing recognition of domains 1 and 4 by a4 integrins suggests that alternative splicing can be used to regulate cell adhesion in the immune response. Mutational and structural analyses have defined amino acid motifs responsible for integrin binding. In fibronectin, a prototype ligand, integrins

INTEGRINS IN THE IMMUNE SYSTEM

333

aIIbp3 and a5pl use two spatially separated sites: the RGD site in the Type I11 repeat 10 (Pierschbacher and Ruoslahti, 1984a) and the “synergy” site in repeat 9 (Bowditch et al., 1991, 1994; Aotaet al., 1991, 1994; Danen et al., 1995). Both sites are present within loops between p strands (Leahy et al., 1996).A similar theme has emerged in some Ig superfamily ligands. In VCAM-1, the amino acid motif IDSP is critical for a 4 integrin binding (Vonderheide et al., 1994; Osborn et al., 1994) to both domain 1 and domain 4. In domain 1, this motif is located on a loop projecting between p strands (Wang et al., 1995). A similar motif in MAdCAM-I, LDT, is involved in a4p7 (Viney et al., 1996) and this motif is on a loop between P strands (Tan et al., 1998). “Synergy” sites occur in these a 4 ligands as well (Newham et al., 1997; Tan et al., 1998). Glu-34 of domain 1of ICAM-1 (Staunton et al., 1990) and the homologous Glu residues on Ig domain 1 of ICAM-2 and ICAM-3 (Li et al., 1995; Klickstein et al., 1996) are important for integrin binding. Crystal structures of ICAM-1 and ICAM-2 establish that the critical Glu residue is on a flat recognition surface (Casasnovas et al., 1997, 1998) on a p strand (Fig. 2, see color plate). Furthermore, the aEP7 binding site in E-cadherin also contains a critical Glu that resides on a quasihelix connecting P strands, rather than on a projecting loop (Karecla et al., 1996). Thus, the I domain surface may form extended contacts with a “flat” surface structure on the ligand. In contrast, as noted above, I domain-lacking integrins such as aIIbp3, a5p1, and a 4 p l recognize critical residues that are within loops (Wang et al., 1995). These loops may dock into and form limited contacts within a binding pocket on the integrin. High-affinity small ligands have been readily identified for integrins that lack I domains (Cheng et al., 1994; Barbas et al., 1993; Greenspoon et al., 1993; Horton et al., 1993; Koivunen et al., 1993; Nowlin et al., 1993; Alig et al., 1992). This structural difference in ligand recognition could account for the apparently low affinity of inhibitory peptides for I domain integrins (Li et al., 1993; Ross et al., 1992; Staatz et al., 1991). Furthermore, this difference could imply that small-molecule competitive antagonists of I domain integrins might be difficult to identify. B. LOCALIZATION OF INTEGRIN LICANDS The spatial localization of integrin ligands provides positional landmarks that guide the trafficking of leukocytes. In addition, integrin ligands provide signals involved in the development and function of leukocytes. Vascular endothelium, lymphoid tissues, and leukocytes are major loci for the integrin ligands important in immune cell function. Endothelial VCAM-1 and ICAM-1 are expressed on peripheral endothelium; their expression is markedly up-regulated at sites of inflammation. The increased expression

334

YOJI SHIMIZU et al

is due in part to effects of cytokines such as TNF-a and IL-1 (Springer, 1995; Bevilacqua, 1993; Schleimer et al., 1992). Induction of these integrin ligands plays a major role in the recruitment of leukocytes to sites of inflammation (Cornejo et al., 1997; Sligh et al., 1993). In contrast to ICAM-1, ICAM-2 is constitutively expressed on vascular endothelium, and inflammatory stimuli do not seem to increase expression significantly (de Fougerolles et al., 1991). The constitutive expression on endothelium may play a role in leukocyte migration into early inflammatory sites prior to the induced expression of ICAM-1 and VCAM-1 as well as in normal lymphocyte recirculation. ICAM-3 is not normally expressed on vascular endothelium either constitutively or at sites of inflammation (Doussis-Anagnostopoulouet al., 1993; Pateyet al., 1996). However, like ICAM-2, its expression is up-regulated on endothelium at sites of tumors (Doussis-Anagnostopoulouet al., 1993; Patey et al., 1996).Thus, ICAM-3 is not thought to play a major role in leukocyte influx during inflammation but may be involved in tumor development. ICAM-3 is also expressed at high levels on lymphocytes, neutrophils, and monocytes (de Fougerolles and Springer, 1992; Fawcett et al., 1992), and ICAM-3 engagement by LFA-1 plays a role in leukocyte responses (deFougerolles et al., 1994; Cid et al., 1994; Hernandez-Caselles et al., 1993). VCAM-1 is constitutively expressed in primary and secondary lymphoid tissue and on reticular cells in the bone marrow (Jacobsen et al., 1996). Here it probably participates in adhesion, proliferation, survival, and/or differentiation of T and B lymphocytes in the bone marrow. VCAM-1/ a 4 p l is involved in the adhesion of lymphoid cells to the bone marrow stroma (Arroyo et al., 1996). In germinal centers, adhesion through VCAM1 is implicated in the prevention of B cell apoptosis (Koopman et al., 1994). VCAM-1 is also expressed on thymic cortical epithelial cells. Here it may play a role in thymocyte adhesion and development. Blocking studies suggest that a 4 p l integrins and VCAM-1 are involved in T cell proliferation and survival (van Seventer et al., 1991b; Damle et al., 1993a; Udagwa et al., 1996). However, a4-null cells exhibit apparently normal short-term T cell development in the thymus (Arroyo et al., 1996). VCAM-1 is expressed in secondary lymphoid tissue, such as spleen and lymph nodes (Koopman et al., 1994; Castro et al., 1997; Schriever et al., 1997). In secondary lymphoid tissues, ICAM-3 is the major ICAM expressed on dendritic cells (Starling et al., 1995) and may provide a counterreceptor for the interactions of T, B, and dendritic cells during antigen presentation (Kushnir et al., 1998).

c. SPECIAL INTEGRIN LIGANDSIN MUCOSAL IMMUNITY MAdCAM-1 and E-cadherin are integrin ligands that play a key role in the mucosal immune system. MAdCAM-1, a specific ligand for integrin

INTEGRINS IN THE IMMUNE SYSTEM

335

a4P7, is expressed on high endothelial venules in Peyer's patches and mesenteric lymph nodes and in venules in the lamina propria (Nakache et al., 1989). During inflammation, MAdCAM-1 expression is up-regulated in the gut, the choroid plexus (Steffen et al., 1996), and the pancreatic endothelium (Hanninen et al., 1993). The a4P7-bearing lymphocytes are selectively directed to the sites of MadCAM-1 expression (Berlin et al., 1993). Furthermore, enteric pathogens stimulate generation of a4P7 high memory T cells subsets. These cells are primarily recruited to the gut (Erle et at., 1994; Rose et al., 1998). The MadCAM-l:a4P7 interaction plays a key role in the development and maintenance of the gut immune system and in the inflammation at this site. It may also participate in inflammation in the pancreas and in the brain. The primary role of E-cadherin is in homotypic adhesion of epithelial cells and in development of epithelial sheets (Yaps et al., 1997).In addition, it is a specific ligand for aEP7 integrin (Cepek et al., 1994; Higgins et al., 1998). In contrast to MAdCAM-1, E-cadherin does not function as a ligand for the homing of lymphocytes to the gut, but instead appears to facilitate adhesion between lymphocytes and gut epithelial cells (Austrup et al., 1995). This cellular interaction could be involved in the retention of intraepithelial lymphocytes and the immune surveillance and the general health of the gut mucosa. IV. lntegrin Signaling

In addition to their adhesive function, integrins can influence gene expression and cell growth, differentiation, and survival (Schwartz et al., 1995).These influences are the consequence of biochemical signals arising from ligand-occupied and clustered integrins (Miyamoto et al., 1995; Yamada and Geiger, 1997). Examples of integrin signaling in the immune system include cosignaling in T cells (Shimizu et al., 1990a), regulation of cytokine gene expression in monocytes (Yurochko et al., 1992), and enhancement of superoxide generation (Nathan et al., 1989) and phagocytosis (Gresham et al., 1989) in neutrophils. Integrin-triggered reactions, such as activation of protein tyrosine kinases (e.g., pp6OS", p ~ 1 2 5 ~and *~, pp72'Y') and activation of phosphatidylinositol3-kinase and MAP kinases, are shared with those generated by soluble agonists (e.g., cytokines) and B and T cell receptors (Schwartz et al., 1995; Shattil et al., 1998). Integrins also mediate cell anchorage and organize elaborate signaling complexes at their cytoplasmic face. As a result, integrin signaling is always initially focused at topographicallylocalized regions of the plasma membrane where cell-cell and cell-extracellular matrix contacts take place.

336

YOJI SHIMIZU et al.

In contrast with agonist or T and B cell receptors, integrins mediate an extracellular effector response (cell anchorage) coincident with ligand engagement. For most integrins, adhesion is regulated by cellular signaling mechanisms that modulate the binding affinity and kinetics of interaction between adhesive ligands and cell surface receptors through a process referred to as integrin activation or “inside-out” signaling (Diamond and Springer, 1994; Hynes, 1992; Shattil and Ginsberg, 1997; Schwartz et al., 1995). In addition, “postoccupancy” events such as lateral diffusion of receptors (Chan et al., 1991) and interactions with and reorganization of the cytoskeleton regulate adhesion (Lotz et al., 1989; Grinnell, 1977).Cells can rapidly change integrin function both by changing integrin binding affinity for ligands (Bennett and Vilaire, 1979;Altieri and Edgington, 1988; Crowe et al., 1994; Shimizu et al., 1990b)and by influencing postoccupancy events (Danilov and Juliano, 1989; Faull et al., 1994; Ylanne et al., 1993; Kucik et al., 1996).This process adds flexibility to integrins and is indispensable in the functioning of many of them. Activation of T and B cell receptors, other costimulatory molecules, and soluble factors such as cytokines and chemokines can rapidly but transiently increase integrin function (Dustin and Springer, 1989; Hayashi et al., 1990; Campbell et al., 1998). Activation of leukocyte integrins in many instances can be blocked by inhibition of phosphatidylinositol 3-kinase (PIS-K) signaling (Shimizu et al., 1995; Kinashi et al., 1995; Zell et al., 1996, 1998; Zauli et al., 1997; Shimizu and Hunt, 1996). However, there is some specificity in the various signals that activate integrins via PISK, as illustrated by the ability of overexpression of a mutant form of the adapter protein Cbl to inhibit specifically pl integrin activation induced by CD28 stimulation (Zell et al., 1998). For LFA-1, PI3-K activation leads to membrane recruitment of cytohesin-1 (Nagel et al., 1998), a protein containing a pleckstrin homology (PH) domain and a SEC7 domain that associates with the 02 cytoplasmic domain (Kolanus et al., 1996). Inhibition of CD3-mediated activation of LFA-1 on overexpression of the PH domain of cytohesin-1 (Kolanus et al., 1996) argues for a role for PI3-K-mediated recruitment of cytohesin-1 via its PH domain in LFA-1 activation. Because cytohesin-1 associates specifically with the p2 cytoplasmic tail, the role of other PH domaincontaining effectors of PI3-K in regulating activation of other leukocyte integrins is unclear. Thus integrin signaling across the plasma membrane is a bidirectional process. A computer search of the National Library of Medicine Medline database revealed a 30-100% annual increase in papers published with the text words “integrin” and “signal transduction” every year since 1993. Consequently, an in-depth analysis of this rapidly moving area is beyond our current scope and the reader is referred to numerous reviews (Fornaro and Languino, 1997; Howe et al., 1998; Morimoto et al.,

INTEGRINS IN THE I M M U N E SYSTEM

337

1998; Shattil et al., 1998; Aota and Yamada, 1997; Chapman, 1997; Frisch and Ruoslahti, 1997; Giancotti, 1997; Guan, 1997; Kolanus and Seed, 1997; LaFlarnme et al., 1997; Marcantonio and David, 1997; Naik and Parise, 1997; Newton et al., 1997; O’Toole, 1997; Shyy and Chien, 1997; Smith, 1997; van Kooyk and Figdor, 1997; Wei et al., 1997). V. lntegrin Function and the Immune Syskm

A. INTEGRINS CONTROL THE TRAFFICKING OF CELLSINVOLVED IN IMMUNE RESPONSES A major immunological function of integrins is to move lymphocytes into and through appropriate microenvironments and to regulate proliferation and differentiation at those sites. Similar integrin-dependent processes are involved in development, leading to “seeding” of the immune system by precursor cells. Experiments with antibodies or other antagonists or with subjects genetically deficient in integrin subunits have elucidated the role of integrins in these processes. Occasional apparent discrepancies between effects of inhibitors and effects of genetic deficiency (Hynes and Wagner, 1996) may be due to compensation or background effects in the deficient individuals. Conversely, integrin blocking reagents can have unexpected trans-dominant effects (Imhof et al., 1997; Blystone et al., 1994; Zhou and Brown, 1993; Diaz-Gonzalez et al., 1996), i.e., they can produce biological effects not ascribable to the lack of integrin function. These issues should be considered in evaluating the role of these receptors in immune function. The trafficking of leukocytes through vessel walls occurs through a remarkably general sequential adhesion cascade (Butcher, 1991; Shimizu et al., 1992; Springer, 1994; Lawrence and Springer, 1991; Lawrence et at., 1995; von Andrian et al., 1991). The steps in this cascade are as follows: 1. Initial tethering followed by reversible rolling on the endothelial surface. Because this interaction involves a leukocyte in flowing blood, this binding interaction requires a rapid on rate. A rapid off rate is required to prevent indiscriminate anchorage of leukocytes to vessel walls. Consequently, the cells roll. Rolling is usually mediated by selectins interacting with glycosylated counterreceptors, but a4 integrins can also mediate rolling (Berlin et al., 1995). 2. Leukocyte stimulation, resulting in augmentation of integrin function. This stimulation is frequently transduced by heterotrimeric G-proteincoupled receptors such as those for chemokines, other chemoattractants, or lipid mediators (e.g., platelet activating factor). As noted above, this enhanced adhesion may owe either to changes in the affinity of the integrins

338

YOJI SHIMIZU et al.

(Ginsberg et al., 1992a), or to changes in integrin mobility (Yauch et al., 1997), cytoskeletal association (Kucik et al., 1996), clustering (van Kooyk et al., 1994), or other postoccupancy events. The p 2 (Lo110 et al., 1993; Altieri et al., 1988) and p 7 (Crowe et al., 1994) integrins manifest active affinity regulation. No such regulation is observed in the interaction of a 4 p l with the N-terminal two domains of VCAM-1 (Jakubowski et al., 1995; Lobb et al., 1995). This surprising result might suggest that a 4 p l affinity modulation does not occur. However, the activation dependence of ligand binding to integrins is often a function of the size of the ligands (Beer et al., 1992; Wayner and Kovach, 1992; Lam et aE., 1987). Consequently, studies with the full-length seven-domain form of VCAM-1 are probably warranted before assuming that a 4 p l is not subject to affinity modulation. 3. Integrin-mediated sticking and stable arrest on the vessel wall. This causes cessation of rolling and is the first step in transmigration. 4. Migration through the vessel wall (diapedesis). Cell migration is a complex event requiring that integrins bind ligand, transmit motive force, and detach in a spatially and temporally coordinated manner (Palecek et al., 1996). The rate of migration is a predictable function of adhesion strength, which in turn is related to integrin density, ligand density, and the affinity of ligand binding (Palecek et al., 1997). Thus, the migration of immune cells into tissues and their decision to stop and reside are tightly controlled by integrin activation. Indeed, integrins can be “too active,” resulting in arrest of migration at high ligand concentrations (Kuijpers et al., 1993; Huttenlocher et al., 1996; Weber et al., 1997). This may explain the remarkable integrin-specific temporal regulation of chemokineinduced cell adhesion (Carr et al., 1996; Weber et al., 1996a,b; Campbell et al., 1996). Integration of the combinatorial inputs of the adhesion cascade, variations in adhesion molecule repertoires, and site-specific integrin-activating signals determine the specificity of lymphocyte recirculation and leukocyte influx into inflammatory sites (Butcher and Picker, 1996; Sdmi and Jalkanen, 1997) B. SEEDING OF HEMATOPOIETIC ORGANS Beginning with its colonization by hematopoietic stem cells at embryonic day 10 (Johnson and Moore, 1975), the fetal liver serves as the major hematopoietic organ until birth. pl integrin deficiency causes early embryonic lethality (Stephens et al., 1995; Fassler and Meyer, 1995); however, chimeric mice generated with pl-negative embryonic stem cells have elucidated the role of pl integrins in the development of lymphocytes (Hirsch

INTEGRINS IN T H E IMMUNE SYSTEM

339

et al., 1996). In such chimeric animals, Pl-negative hematopoietic cells are not found in the fetal liver. These cells are found in blood islands in yolk sac tissue and can differentiate properly in in vitro colony-forming assays. Thus, loss of Pl integrin expression does not appear to affect blood cell differentiation at some sites. Consequently, the key requirement for Pl in hematopoiesis may be in migration of stem cells into fetal liver. In addition to their role in the fetal liver, integrins control homing of hematopoietic stem cells to the bone marrow. Bone marrow progenitors express the integrins a461, a561, and aLP2 (Liesveld et ul., 1993). @l integrins appear to be involved in seeding and development of progenitors in the marrow. Stem cells adhere to VCAM-1 (a4) and fibronectin (a4 and a5)via Pl integrins (Williamset al., 1991; Miyake et al., 1991; Hurley et al., 1995; Kerst et al., 1993; Levesque et al., 1995; Simmons et al., 1992). Furthermore, anti-a4 and anti-pl antibodies block the engraftment of hematopoietic stem cells (Williams et al., 1991; Papayannopoulou et al., 1995), and anti-a4 releases granulocyte-macrophage colony-forming progenitors into the bloodstream (Papayannopoulou and Nakamoto, 1993). Precursors for both T and B cells require a 4 integrins for normal development within the bone marrow (Arroyo et al., 1996). In contrast, other mononuclear cells can develop normally without these integrins. Thus, there are lymphocyte-specific requirements for a4 integrins in hematopoiesis in the bone marrow (Arroyo et al., 1996). Because lymphocyte development is normal in mice lacking P7 integrin subunits (Schmits et al., 1996; Wagner et al., 1996), a 4 P l integrin is probably the critical a 4 integrin for hematopoiesis within the marrow. In vitro, anti432 partially blocks transmigration. However, hematopoiesis is not impaired in P2-deficient humans or mice (Schmits et al., 1996; Belkin et al., 1997). Dysregulation of integrin function may contribute to some of the manifestations of leukemia. For example, excessive release of progenitor cells into the bloodstream occurs in chronic myelogenous leukemia (CML). CML progenitors adhere poorly to bone marrow stroma, despite normal levels of Pl integrin expression (Verfaillie et al., 1992, 1997; Bhatia et al., 1994). Furthermore, Pl integrin-mediated adhesion of progenitor cells to fibronectin inhibits cell proliferation (Hurley et al., 1995; Bhatia et al., 1996). Thus, loss of Pl integrin function may contribute to abnormal growth and retention of progenitors in CML. C. B CELLDEVELOPMENT IN THE BONEMARROW B cell development in the bone marrow probably depends on the physical interaction of B cell precursors with bone marrow stromal cells mediated by a4Pl binding to stromal cell VCAM-1 (Kishimoto et al., 1987; Dittel et al., 1993; Arroyo et al., 1996). In contrast, B cell precursors express low

340

YOJI SHIMIZU et al.

levels of aLp2, and anti-ICAM-1 antibodies do not inhibit B cell precursor adhesion to bone marrow stromal cells (Dittel et al., 1993). Early studies revealed varying roles for a 4 p l interactions with VCAM-1 in B cell development (Kishimoto et al., 1991; Ryan et al., 1992; Manabe et al., 1994; Roldan et al., 1992; Friedrich et al., 1996). The dramatic reduction in the number of a$-negative B cells in adult bone marrow, blood, and lymphoid organs of chimeric mice provides direct proof of the role of a 4 in B cell development. This requirement is probably age specific because a4negative fetal liver, but not a4-negative bone marrow, can reconstitute the B cell compartment of irradiated recipients (Arroyo et al., 1996). However, because a4 gene ablation blocks B cell development prior to the pro-B cell stage in a4-null chimeric mice, these animals have not clarified the role of a 4 p l integrin in B cell development subsequent to commitment of stem cells to the B cell lineage. B cell development is reportedly normal in P7-negative mice (Wagner et al., 1996),which indicates that the effects of a 4 gene ablation on B cell development are likely due to loss of a 4 p l integrin function. Loss of p2 integrin expression in mice also does not affect B cell development (Schmits et al., 1996), consistent with in vitro studies demonstrating a minimal role for p2 integrins in mediating B cell precursor adhesion to stromal cells (Dittel et al., 1993). The requirement for a 4 p l in B cell development may be simply an adhesive one. Alternatively, this integrin can generate a wide variety of biochemical signals in B cells (Freedman et al., 1993; Manie et al., 1996, 1997a-c; Astier et al., 1997; Xiao et al., 1996). These signals may play a role in driving B cell development. D. T CELLDEVELOPMENT Colonization of the thymus by T cell progenitors and their development involve integrin function. The first step in this process is the migration of T cell precursors from the yolk sac and fetal liver via the bloodstream into the thymic stroma (Dunon and Imhof, 1993, 1996). An anti-a6 antibody inhibits the adhesion of pro-T cells to cultured thymic endothelium in vitro (Imhof et al., 1997) and colonization of the thymus in vivo by bone marrow-derived T cell progenitors (Ruiz et al., 1995). Double-negative and double-positive thymocytes express both isoforms of the a6 integrin (Ruizet al., 1995)in a developmentally regulated fashion (Ruizet al., 1995). Mice lacking the a6 integrin subunit reach term (Georges-Labouesse et al., 1996), but detailed analysis of thymic development in these animals has not been reported. Similarly, cr4-null chimeric mice show no defects in T cell development at birth (Arroyo et al., 1996). It is likely that other integrin and nonintegrin (Horst et al., 1990) adhesion receptors are involved in precursor homing to the thymus.

341

INTEGRINS IN T H E IMMUNE SYSTEM

The role of integrins in development of thymocytes remains an area of active investigation. Thymocytes express differentiation state-specific repertoires of integrins (Fig. 3). a 4 p l is expressed on all thymocytes; expression is highest on double-negative thymocytes, and progressively decreases in the double-positive and single-positive stages. In addition, small subsets of double-negative and single-positive, but not doublepositive, thymocytes express p7 (Andrew et d.,1996) integrin receptors. Other pl integrins (a5p1, a3p1, a 6 p l ) are expressed by the majority of human thymocytes, with high levels of expression on double-negative cells, decreasing on double-positive cells and then increasing again on singlepositive cells (Mojcik et d.,1995). aLp2 is expressed at high levels on all thymocytes, with the highest level of expression on single-positive cells (Mojcik et al., 1995). These integrins mediate adhesion of thymocytes to

I h

Ligand Exoression

VCAM-1

conversion

horn adheswe to migratory

phmotVpe

,

lntearin xoression 1

n

.

U4!31++++* U3p1+

2;:: CD3CDCCDS double-negative

&P++'

I CIAp1+++

U3p' Usp1"

Qm+++ CD3+CD4+CDBt double-poslve

cs-1 FN

1

CD3+CD4+CD8-

CD3+CW-CDB+

singieposnive

FIG.3. Patterns of expression of integrins and integrin ligands in the thymus. Thymocyte differentiation is marked by phenotypic changes in CD3, CD4, and CD8 expression, which is accompanied by the directed movement of thymocytes during the differentiation process from the cortex to the medulla. The double-negative, double-positive, and single-positive stages of thymocyte differentiation are marked by relative changes in the expression of /3l and /3Z integrins. In addition, constitutively active a4pl integrin is expressed on doublenegative thymocytes and a major portion of double-positive thymocytes. Further phenotypic subdivision within double-positive thymocytes (not shown) identifies subsets with a fibronectin (FN)-adhesive and a FN-migratory phenotype. VCAM-1 is expressed on thymic epithelium in the cortex and corticomedullary junction of the human thymus, and FN containing the CS-1 region is preferentially expressed in the corticomedullaryjunction and the medulla.

342

YOJI SHIMIZU et nl.

their cognate ligands (Chang et al., 1995; St-Pierre et al., 1996; Salomon et al., 1994, 1997). Engagement of integrins on thymocytes facilitates thymocyte proliferation induced by CD3 stimulation (Ticchioniet al., 1995; Chang et al., 1995; Salomon et al., 1994).Thus thymocyte integrins mediate both adhesion and signahng events in these cells. In addition, thymocyte activation modulates integrin-mediated adhesion in a stage-specific manner. The majority of double-negative cells and some double-positive cells constitutively bind to fibronectin, VCAM-1, and thymic stromal cells (Utsumi et al., 1991; Sawada et al., 1992; St-Pierre et al., 1996). a 4 p l is constitutively active for adhesion of double-positive thymocytes to fibronectin and VCAM-1 (Salomon et al., 1994). Additional stage-specific regulation of integrin function on thymocytes could explain the capacity of fibronectin to stimulate migration of mature thymocytes expressing high levels of CD3 and CD69 (Crisa et al., 1996). Some of this regulation could be due to exposure to varying levels of matrix-bound instructive agonists. For example, IL-7 enhances a4fll-mediated adhesion of all thymocytes to fibronectin without altering levels of a 4 p l expression (Kitazawa et al., 1997). The presence of IL-7 on the extracellular matrix of the thymus, and the effects of ablating IL-7 or IL-7 receptor in mice, suggest that one function of IL-7 might be to regulate integrin-mediated signals critical to thymocyte survival, adhesion, or motility (Kitazawa et al., 1997; von Freeden-Jeffry et al., 1995; Peschon et al., 1994). Integrin ligands and cellular counterreceptors are expressed in the thymus. VCAM-1 is expressed at sites that are likely to be critical for selection events such as cortical thymic epithelium and the junction between the cortex and medulla (Salomon et al., 1997) and dendritic cells (Salomon et al., 1997). Laminin and its integrin receptors are also found in the thymus (Lannes-Vieira et al., 1993). Cortical immature thymocytes express abundant active a4p1, whereas more mature thymocytes that move to the medulla are induced to migrate on interaction with fibronectin (Crisa et al., 1996). Fibronectin is predominantly expressed in the medulla and contains the binding site recognized by a 4 p l (Crisa et al., 1996). Thus, integrin-fibronectin interactions may play a critical role in regulating movement of thymocytes from the cortex to the medulla of the thymus (Crisa et al., 1996). This hypothesis is consistent with (1)the proposed locations within the thymus where positive and negative selection occur and (2) the notion that maturation of T cells in the thymus requires an ordered movement of thymocytes from the cortex to the medulla (Scollay and Godfrey, 1995). Indeed, the extracellular matrix may serve as the scaffold for this “conveyor belt” movement of thymocytes through the thymus (Savino et al., 1996).

INTEGRINS IN T H E IMMUNE SYSTEM

343

Currently, the in vivo evidence that integrins play a vital role in thymocyte development is less compelling than the in vitro evidence. Although the addition of anti-aLP2 or anti-ICAM-1 antibodies inhibited the development of double-positive thymocytes in fetal thymic organ cultures (Fine and Kruisbeek, 1991), T cell development is essentially normal in mice lacking expression of the a L subunit (Shier et al., 1996; Schmits et al., 1996). T cell development is also normal in P7 knockout mice (Wagner et al., 1996). Antifibronectin antibodies and fibronectin peptides inhibit the in vitro differentiation of mouse double-negative thymocytes, presumably by interfering with a4pl-mediated adhesion of thymocytes to fibronectin on thymic stromal cells (Utsumi et al., 1991). Although T cell development is impaired in a4-negative chimeric mice, reconstitution experiments demonstrated that the T cell population could be temporarily restored with bone marrow from a4-negative chimeras (Arroyo et al., 1996). Thus, a4-negative T cells develop normally in the thymus but not in the bone marrow. The role of other integrins in thymic development, such as a 5 p l and a6p1, has yet to be analyzed with genetically modified mice.

E. DEVELOPMENT OF LYMPH NODES:ROLEOF a4P7 AND MAdCAM-1 During fetal development, a4P7 is probably involved in seeding of lymph nodes. Blocking antibodies against a4, p7, or MAdCAM-1 inhibit the population of fetal lymph nodes by 76 T cells and a unique population of CD3-CD4' T cells (Mebius et al., 1996). These cells are recruited from a small (1-2%) subpopulation of peripheral a4P7-expressing leukocytes in fetal mice (Mebius et al., 1996). Furthermore, MAdCAM-1 is expressed on fetal peripheral lymph node high endothelial venules (HEVs), the site where lymphocytes enter the node (Mebius et al., 1996; Girard and Springer, 1995). After birth, antibodies to a4p7 or MAdCAM-1 fail to block lymph node entry (Mebius et al., 1996) and the process becomes dependent on L-selectin. This switch from integrin to selectin dependence is associated with a dramatic developmental switch in counterreceptor expression in HEVs. In fetal and early neonatal life, peripheral lymph node HEVs express MAdCAM-1 and lack peripheral node addressin (PNAd) (the counterreceptor for L-selectin). Shortly after birth MAdCAM-1 disappears and there is concordant induction of PNAd expression on peripheral lymph node HEVs. This developmental switch does not require circulating lymphocytes, because it also occurs in lymphocytopenic mice (Mebius et al., 1996). Lymphocyte development is grossly normal in P7-deficient mice (Wagner et al., 1996), but a detailed analysis of lymph node ontogeny in these animals would be of interest.

344

YOJI SHIMIZU et nl.

VI. lntegrins in Lymphocyte Recirculation

A. RECIRCULATION OF NAIVET CELLS Naive T cells must circulate through the body and migrate through peripheral lymph nodes and secondary lymphoid organs, in which they encounter antigen. These cells enter peripheral lymph nodes via interactions with HEVs (Mackay et al., 1990; Girard and Springer, 1995). The general principles of the adhesion cascade model of leukocyte interactions with endothelium govern this event (Warnock et al., 1998). Studies with antibodies (Gallatin et al., 1983) as well as genetically deficient (Arbones et al., 1994; Xu et al., 1996) mice implicate L-selectin as the tethering receptor. The signal that stimulates arrest of naive T cells remains unclear. Engagement of L-selectin is one candidate (Hwang et al., 1996; Sikorski et al., 1996; Steeber et al., 1997). Alternatively, chemokines may be the factors that activate integrin function during attachment to HEVs (Tanaka et al., 1993; Campbell et al., 1998). An excellent candidate is GCkine, also known as secondary lymphoid-tissue chemokine (SLC) and Exodus2 (Hedrick and Zlotnik, 1997; Hromas et al., 1997; Gunn et al., 1998); 6Ckine manifests relatively restricted expression in the HEVs of lymph nodes and Peyer’s patches (Gunn et al., 1998). This chemokine activates p2 integrins on naive T cells (Gunn et al., 1998; Campbell et al., 1998), probably via a G-protein-coupled receptor, CCR7 (Yoshida et al., 1998). Naive T cells have a relatively uniform pattern of p2, pl, and p7 integrin expression (Fig. 4 ) (Shimizu et al., 1990b; Rott et al., 1996; Mackay et al., 1996). aLp2, expressed at high levels on these cells, probably mediates their interaction with HEVs (Warnock et al., 1998).This explains why aLdeficient mice manifest small peripheral lymph nodes (Schmits et al., 1996) and defects in responses to cutaneously applied antigens (Schmits et al., 1996). These antigens are presented to naive T cells in peripheral lymph nodes (Tedder et al., 1995; Catalina et al., 1996). Naive T cells express low levels of a 4 integrin, and lymphocyte homing to rat peripheral lymph nodes has been reported to involve a4 integrins (May et al., 1993).This is consistent with reports that VCAM-1 is expressed on rat peripheral lymph node HEVs (May et al., 1993). However, HEVs in normal mouse and human peripheral lymph nodes do not express appreciable levels of VCAM-1 or MAdCAM-1 (Szabo et al., 1997; Nakache et al., 1989; Hahne et al., 1993), and anti-a4 integrin antibodies do not affect lymphocyte stable arrest on peripheral lymph node HEVs in mice in vivo (Warnock et al., 1998). Thus, there seems to be a species-specific role for a4 in naive T cell trafficking. B. MEMORY T CELLS Activation of naive T cells results in proliferation and differentiation, resulting in the generation of “memory” or “memory/effector” T cells (Bell

345

INTEGRINS IN THE IMMUNE SYSTEM

-

Memory T Cells (CD45RO+) Non-mucosal sites

Naive T Cell (CD45RA+)

I Peripheral

lymph nodes

/ \ p7high

-

Intestinal intraepithelial lymphocyle compartment

FIG.4. Phenotypic and functional subdivision of peripheral T cells based on /3l and /37 integrin expression. Naive T cells (CD45RA') express uniform levels of L-selectin, /3l integrin, and p7 integrin, and recirculate efficiently through peripheral lymph nodes via peripheral lymph node high endothelial venules. Memory T cells (CD45RO+)are more heterogeneous in integrin phenotype than are naive T cells, and three classes of memory T cells in the human have been defined on the basis of differential expression of the /3l, /37, and crE subunits. The majority (>80%) of memory T cells are /31h'~h//37-/crE'. These memory T cells express high levels of a 4 p l and are particularly efficient at migrating into nonmucosal sites of inflammation through adhesion to endothelial VCAM-1. The remaining and a minor subset of these memory T cells also express memory T cells are /31'"w//37'"~'1, the a E subunit. High levels of /37 integrin expression allow for efficient trafficking of these memory T cells to gut-associated lymphoid tissue. The circulating aE+memory T cell subset may be involved in trafficking specifically to the intraepithelial compartment in the gut, although this has not been definitively established. Abbreviations: int, intermediate.

et al., 1998). These cells must exit the bloodstream at a wide variety of sites. The use of multiple exit sites may be facilitated by the marked heterogeneity in adhesion receptor phenotypes in these cells. Memory T cells (Fig. 4), in general, express higher levels of Pl and 02 integrins (Sanders et al., 1988; Shimizu et al., 199Ob).There is marked heterogeneity in expression of a4P7 (Schweighoffer et al., 1993; Rott et al., 1996; Abitorabi et al., 1996; Mackay et al., 1996) and these differences correlate with differences in adhesion to MAdCAM-1 (Rott et al., 1996). A small subpopulation of a407'& cells express aEP7 (Rott et al., 1996). As with naive T cells, these phenotypic characteristics are best described in the CD4 T cell lineage; limited studies of CD8' memory T cells show similar patterns, although a4P7 integrin expression is much more heterogeneous in CD8'CD45RAt naive T cells than in CD4+CD45RAf naive T cells (Rott et al., 1996).These differences in integrin expression between CD4+

346

YO11 SHIMIZU et al.

and CD8+ T cells likely reflect the different recirculation properties of these two major subsets of T cells, properties that are related to the different functions of CD4+ and CD8’ T cells (Kedl and Mescher, 1997), although definitive proof of this hypothesis is currently lacking. The gut-associated lymphoid tissue (GALT) consists of secondary lymphoid organs, such as Peyer’s patches, and effector sites, which consists of lamina propria and the intraepithelial lymphocyte compartment (Kraehenbuhl and Neutra, 1992). Homing of lymphocytes into the GALT is dependent on lymphocyte expression of the a4P7 integrin (Holzmann et al., 1989; Holzmann and Weissman, 1989).a4P7‘gh memory T cells and a4P7+ naive T cells directly extravasate into Peyer’s patches (Williams and Butcher, 1997). In contrast, a4P7-negative memory T ceIls do not home to Peyer’s patches (Williams and Butcher, 1997). Thus, there is a direct relationship between a4P7 expression and migration to Peyer’s patches in viva (Williams and Butcher, 1997). The adhesion of lymphocytes to Peyer’s patch HEVs involves L-selectin, a4P7, and aLP2 in a sequential fashion. However, a4P7 participates in both rolling and stable arrest (Bargatze et al., 1995). Like L-selectin, a4P7 is concentrated on the microvilli of lymphocytes, which mediate the initial contact with HEVs (Berlin et al., 1995). Furthermore, the activation state of the lymphocyte determines the relative role of a4P7 in mediating lymphocyte interactions with Peyer’s patch HEVs. Activated cells expressing high levels of a4P7 arrest on Peyer’s patch HEVs in an L-selectinindependent, a4P7-dependent, manner (Bargatze et al., 1995).Attachment of lymphocytes to lamina propria venules in viva is also dependent on a4/37, but independent of L-selectin (Berlin et al., 1995). Just as L-selectin-deficient mice exhibit dramatically decreased peripheral lymph nodes due to loss of influx of lymphocytes, p7 integrin-deficient mice exhibit a reduction in size and cellularity of Peyer’s patches (Wagner et al., 1996). The P7-deficient lymphocytes are unable to bind to Peyer’s patch HEVs in vitro and are unable to migrate into Peyer’s patches in viva Furthermore, loss of P7 profoundly reduced arrest and firm adhesion to Peyer’s patch HEVs (Wagner et al., 1996). Lymphocyte homing to Peyer’s patches is also defective in a4-negative chimeric mice (Arroyo et al., 1996), further supporting the preeminent role of a4P7 in mediating homing of lymphocytes to Peyer’s patches. C. THEaEP7 INTEGRINAND MUCOSAL IMMUNITY Intraepithelial lymphocytes ( IELs) comprise a unique T cell population that resides in the intestinal epithelium. Most of these T cells express the CD8 coreceptor, and can be subdivided based on expression of either an aP T cell receptor (TCR) or a y6 TCR (Goodman and Lefrancois, 1989).

I N T E G R I N S IN THE I M M U N E SYSTEM

347

IEL development differs from other T cell development in that T cell receptor rearrangement in IELs occurs outside of the thymus (Lefrancois and Olson, 1994; Lin et nl., 1993). These cells also differ in having a reduced TCR repertoire (Sim, 1995),constitutive lyhc activity (Lefrancois and Goodman, 1989), and they localize preferentially to spaces between intestinal epithelial cells. aEP7 integrin is expressed at high levels on all IELs and dendritic cells found in the intestinal epithelium, as well as on many T cells in the lamina propria (Cerf-Bensussan et al., 1992; Kilshaw and Murant, 1990; Kilshaw, 1993). This is particular1 striking, given that only a very small subpopulation of peripheral a4/37'gh, Pl'c'wmemory T cells express aEP7 (Rott et al., 1996).Transforming growth factor P induces a E integrin expression. Because this cytokine is expressed in the small intestinal epithelium (Barnard et al., 1993), it may regulate aEP7 expression. E-Cadherin, the counterreceptor for aEP7, is expressed on intestinal epithelial cells (Roberts and Kilshaw, 1993; Cepek et al., 1993, 1994; Karecla et al., 1995; Higgins et al., 1998). Divalent cations and cellular activation (Higgins et al., 1998) can regulate aEP7-mediated adhesion to E-cadherin. aEP7 binding to E-cadherin on intestinal epithelial cells may retain IELs in the intestinal epithelium. A reduction in the number of IELs in mice deficient for the a E subunit (Higgins et al., 1998) is consistent with this model. However, others have noted an increase in p7-positive IELs in mice expressing a dominant negative form of N-cadherin, which disrupts E-cadherin expression in the intestine (Hermiston and Gordon, 1995). Furthermore, there is an influx of aEP7-negative CD8+ T cells into the intestinal epithelium in graft-versus-host disease (Kilshaw and Baker, 1988). The aEP7-mediated adhesion of IELs to E-cadherin on intestinal epithelial cells may also regulate IEL effector function (Kilshaw and Karecla, 1997), because aEP7 can provide signals that enhance TCRmediated activation of T cells (Sarnacld et al., 1992; Begue et al., 1995). Furthermore, the in vitro cytotoxic activity of some, but not all, y6+ tumorinfiltrating lymphocytes isolated from patients with colorectal cancer is partially inhibited by an anti-P7 antibody (Maeurer et al., 1996). P2 integrins are involved in IEL function. Mice deficient in expression of either P2 or ICAM-1 have defects in the activation and expansion of some lamina propria T cells and IELs (Huleatt and Lefrancois, 1996). Deficiency of 0 2 integrin or ICAM-1 results in almost complete loss of a subset of ap TCR+ T cells in both intestinal epithelium and lamina propria and in a dramatic reduction of Thy-l+ y6 TCR-positive IELs. Because these Thy-l+T cells exhibit constitutive lytic activity that is dependent on normal microbial flora, loss of P2 integrin or ICAM-1 expression may disrupt p2 integrin-dependent signals required for the development of

348

YOJI SHIMIZU et d

Thy-l+ IELs. The inability of bone marrow from ICAM-1-deficient mice to restore the generation of some ap TCR IELs in radiation reconstitution experiments is consistent with this model (Huleatt and Lefrancois, 1996). An overall decrease in lymphocytes in Peyer’s patches was also noted in PZ-nuIl and ICAM-1-null mutant mice, but this phenotype may be due to diminished homing. In addition, antigen-induced stimulation of IELs in vivo leads to expression of the a m 2 integrin that may enhance IEL recognition of target cells during cytolyhc responses (Huleatt and Lefrancois, 1995). Other differences in integrin expression between ap TCR T cells and 76 TCR T cells have also been noted. High expression of a461 and a5pl on Val+ T cells correlates with increased adhesiveness to fibronectin (Nakajima et al., 1995). The increased adhesion may contribute to the accumulation and retention of V a l + T cells in intestinal epithelium (Halstensen et al., 1989).

D. DENDRITIC CELLS Dendritic cells capture antigen in the periphery and migrate to lymphoid organs, where they present the antigen to T and B cells (Banchereau and Steinman, 1998). Blood dendritic cells express pl and p2 integrins, and anti-fi2 and a n t i 4 antibodies inhibit the adhesion of blood dendritic cells to human umbilical vein endothelial cells (Brown et al., 1997). In addition, differential expression of the aXp2 integrin subdivides blood dendritic cells into two populations (O’Doherty et al., 1994) that may migrate to distinct areas of lymphoid organs (Banchereau and Steinman, 1998). The specific roles of integrins in dendritic cell migration into and within lymphoid organs, along with the factors that regulate the activity of integrins on dendritic cells, are important areas for future investigation. VII. The Role of Integrins in Immune Responses and Inflammation: Two Case Studies

As the foregoing discussion makes clear, integrins are critically involved in virtually all aspects of the development and patterning of the immune system. Consequently, manipulation of these receptors is a potential strategy for immunomodulation. A detailed review of all inflammatory diseases and immune reactions and the various integrins involved is beyond our scope. Consequently, we will discuss two representative “case studies”: the role of aLp2 in selected immune responses and the role of integrins on one autoimmune inflammatory disease, experimental autoimmune encephalomyelitis (EAE).

INTEGRINS IN THE I M M U N E SYSTEM

349

A. aLp2 AND IMMUNE RESPONSES As described in the foregoing sections, aLp2 plays a pivotal role in lymphocyte trafficking. Moreover, its signaling functions can cooperate with specific immunologic receptors in many contexts (Mobley et al., 1993). Consequently, defects in immune responses of aLp2-deficient individuals are readily understandable. These include defects in rejection of allografts (Le Deist et al., 1989) and tumors (Schmits et al., 1996; Shier et nl., 1996). Antibody inhibition experiments confirm the role of this integrin in transplant rejection (Isobe et al., 1992). Furthermore, in vitro studies show that lymphocytes from aL-null mice do not aggregate and exhibit defects in proliferation in mixed lymphocyte reactions (MLRs) and in response to concanavalin A (Schrnits et al., 1996; Shier et al., 1996). Mice with null CD18 expression exhibit similar defects in T cell allogeneic responses (Scharffetter-Kochanek et al., 1998). In addition, inhibitory anti-aL antibodies (Odum et al., 1988; Kuijpers et al., 1990) or absence of ICAM-1 on antigen-presenting cells blocks MLRs (Sligh, et al., 1993). Thus, there is strong evidence that aLp2 plays an essential role in response to allografts. T cells from CD18-null mice also fail to proliferate in response to staphylococcal enterotoxin A (Scharffetter-Kochanek et al., 1998), consistent with earlier antibody blocking studies implicating aLp2 in superantigen-dependent T cell activation (Damle et al., 199313; Nickoloff et al., 1993; van Seventer et al., 1991a). In contrast, there are discrepant reports on the effects of a L gene ablation on the in vitru cytotoxic activity of natural killer cells toward YAC-1 target cells (Schmits et al., 1996; Shier et al., 1996). Responses of aL-deficient mice in delayed-type hypersensitivity (DTH) vary depending on the nature of the sensitizing agent. The response to dinitrofluorobenzene (DNFB) is reduced in aL-negative mice (Schmits et al., 1996) or in response to anti-aL antibodies (Kondo et al., 1994). In contrast, the response to sheep red blood cells (SRBCs) is comparable to aL-positive mice (Shier et al., 1996). Antigen-specific variance in T cell priming might explain these differences (Schmits et al., 1996; Shier et al., 1996). Some humans deficient in j32 manifest normal DTH responses in vivu and are not susceptible to severe viral infections (Anderson et al., 1985). In contrast, a patient with a selective defect in “activation” of aLp2 manifests reduced DTH responses due to an inability of antigen-specific T cells to infiltrate the skin (Kuijpers et nl., 1997).This individual suggests that activation of p 2 plays a role in human DTH. Possibly, the functional defect in aLp2 evokes less compensation than its complete loss, accounting for the apparent discrepancy between the quantitative and qualitative deficits in aLp2 function. In sum, the contribution of aLp2 to DTH is

350

YOJI SHIMIZU et ul

likely to be dependent on the type of antigen-presenting cell, the quantity of activating antigen, and the nature of the microenvironment in which the activation occurs. Early antibody inhibition studies implicated aLP2 in the killing of target cells by cytotoxic T lymphocytes (CTLs) (Davignonet al., 1981a,b; SanchezMadrid et al., 1982; Shaw et al., 1986). Remarkably, aL-negative mice manifest normal CTL effector function in response to certain viruses (Schmits et al., 1996); similar findings were reported in ICAM-l-deficient mice and in mice with a partial loss of P2 integrin expression (Christensen et al., 1996). Furthermore, P2-deficient humans do not seem predisposed to viral infections (Anderson et al., 1985), consistent with normal in vivo CTL function. CTL priming to these viruses occurs in the spleen (Schmits et al., 1996) and aL-null mice manifest splenomegaly due to markedly increased splenic T cells (Schmits et al., 1996). Consequently, increased aL-negative T cells in the spleen may compensate for the lack of aLP2 in the CTL response.

B. INTEGRINS AND LYMPHOCYTE-MEDIATED INFLAMMATORY DISEASE Lymphocyte entry into inflammatory sites utilizes the four-step adhesion cascade (Carlos and Harlan, 1994). The roles of 0 2 and a 4 integrins have received the most attention because their ligands, ICAM-1, VCAM-1, and MAdCAM-1, are dramatically up-regulated on inflamed endothelium. Indeed, these integrins are candidate therapeutic targets for a wide range of chronic inflammatory diseases, including inflammatory bowel disease (Hesterberg et al., 1996), asthma (Das et al., 1995), diabetes (Hanninen et al., 1996), and rheumatoid arthritis (Postigo et al., 1992; Diaz-Gonzalez and Ginsberg, 1996; Diaz-Gonzalez and Sanchez-Madrid, 1998). Perhaps the greatest amount of data has been obtained in a model of multiple sclerosis (Swanborg, 1995), experimental autoimmune encephalomyelitis. Studies in EAE vividly illustrate the role of integrins in lymphocytemediated inflammation and will therefore be the focus of our discussion. EAE is elicited by a T cell response to myelin or to certain of its components (Swanborg, 1995),leading to extensive lymphocyte infiltration of the central nervous system and damage to myelin sheaths. Integrins play a central role in the pathogenesis of EAE. ICAM-1 and VCAM-1 expression are dramatically increased on central nervous system venules in both multiple sclerosis and EAE (Irani and Griffin, 1996; Washington et al., 1994; Sasseville et al., 1992; Barten and Ruddle, 1994; Engelhardt, 1998). Furthermore, there is a direct relationship between a 4 integrin expression and the ability of autoreactive T cell clones to induce EAE in mice (Sasseville et al., 1992; Baron et al., 1993). Anti-a4 or anti-VCAM1 antibodies delay the onset and decrease the severity of EAE (Yednock

INTEGRINS IN THE IMMUNE SYSTEM

351

et al., 1992; Baron et al., 1993; Soilu-Hanninen et al., 1997; Keszthelyi et al., 1996), possibly by preventing the influx of autoreactive T cells into the central nervous system. Autoreactive T cells that cause EAE do not express a407 (Engelhardt, 1998; Engelhardt et al., 1995). Consequently, a 4 p l is likely to be the relevant integrin. In addition to mediating adhesion, the a4plNCAM-1 interaction facilitates responses necessary for extravasation into the brain. These include metalloproteinase expression (Romanic and Madri, 1994), which may degrade the basement membrane, facilitating transmigration of T cells (Leppert et al., 1995). aLp2 does not induce metalloproteinase expression, perhaps accounting for the inability of antip2 or anti-ICAM-1 antibodies to ameliorate EAE consistently (Baron et al., 1993; Welsh et al., 1993). The integrin phenotype of the T cells changes once they enter the brain (Romanic et al., 1997).After transmigration, T cells down-regulate integrins involved in endothelial binding (e.g., a401, aLp2) and up-regulate those involved in matrix adhesion (e.g., a5p1, a 2 p l ) with resulting increased adhesion to collagen and fibronectin. These changes suggest mechanisms by which autoreactive T cells might be retained in the inflammatory site by firm adhesion to the extracellular matrix components found in tissue. Thus, a 4 p l and its ligands play a key role in the pathogenesis of EAE and are candidate therapeutic targets for multiple sclerosis. VIII. Regulation of lntegrin Ligand Expression in Inflammation

As noted above, changes in expression of integrin ligands play a role in the inflammatory response. The regulation of VCAM-1 and ICAM-1 has been most extensively studied, and many stimuli can influence their expression. Proinflammatory agents such as TNF-a, IL-lp, and LPS induce VCAM-1 as well as ICAM-1 expression on endothelium (Springer, 1995; Bevilacqua, 1993; Schleimer et al., 1992). This up-regulation appears to involve both transcriptional and posttranscriptional mechanisms. The 5' regions of the ICAM-1 and VCAM-1 genes contain several potential transcriptional regulatory elements, some of which are common to both genes. The activation of NF-KBtranscription factor and its interaction with multiple KB motifs on the VCAM-1 and ICAM-1 gene promoters is critical for TNF-a and IL-lp-induced expression (lademarco et al., 1995; Degitz et al., 1991). Other transcription factors can also modify this induction. For example, AP-1 potentiates TNF-a-induced VCAM-1 expression (Ahmad et al., 1998). Interestingly, this does not appear to involve AP-1 interaction with its cognate enhancer, but instead involves a direct interaction with NF-KB.

352

YOJI SHIMIZU et at,

Posttranscriptional mechanisms for regulating ICAM-1 and VCAM-1 expression have also been proposed. At least two regulatory regions have been defined for ICAM-1. The 3‘ untranslated regions (UTR) of ICAM1 contains multiple AUUUA motifs (Staunton et al., 1988). This motif is partially responsible for the short half-life of ICAM-1 mRNA as well as other gene products (Ohh and Takei, 1996). Removal of the AUUUA motifs increases the half-life of ICAM-1 mRNA and prevents the stabilization of mRNA induced by phorbol myristate acetate (PMA) stimulation (Ohh and Takei, 1996). Interestingly, ICAM-2 and ICAM-3, which are constitutively expressed at high levels, do not contain 3’ UTR AUUUA motifs, and thus this may partially explain the constitutive high expression. Inclusion of the 3’ UTR of ICAM-1 on the ICAM-2 gene shortens the mRNA half-life (Ohh and Takei, 1994). A second regulatory element located in the region encoding the cytoplasmic domain of ICAM-1 has also been implicated in the regulation of mRNA stability (Ohh and Takei, 1996). This region is responsible for the increased mRNA stability induced by IFN-.)I stimulation. Similarly, VCAM-1 may also be regulated by posttranscriptional mechanisms. For example, IL-4, in combination with TNF-a, seems to be a selective inducer of VCAM-1 expression on endothelial cells. IL-4 stimulation alone is a weak inducer of VCAM-1 expression; however, it acts synergistically with TNF-a to increase VCAM-1 expression dramatically (Schleimer et al., 1992; Thornhill et al., 1991; lademarco et al., 1992). This synergistic effect of IL-4 appears to occur largely through an increase in VCAM-1 mRNA stability (Iademarco et al., 1992). Other conditions can down-regulate VCAM-1 and ICAM-1 expression. Laminar shear stress on the vessel wall suppresses endothelial VCAM-1 expression (Ohtsuka et al., 1993; Korenaga et al., 1997). This appears to involve two AP-1 cis elements acting as negative regulators of transcription (Korenaga et al., 1997). Paradoxically, shear stress stimulates ICAM-1 expression on vascular endothelium (Nagel et al., 1994). This appears to involve a shear stress response element located on the promoter of ICAM1 but not on the VCAM-1 gene. Agents that elevate intracellular CAMP and activate protein kinase A, such as prostaglandins, have also been implicated in the suppression of ICAM-1 and VCAM-1 expression in both endothelial and smooth muscle cells (Pober et al., 1993; Panettieri et al., 1995; Braun et al., 1997).This suppression is at the level of gene transcription, but the precise mechanism is unclear. An additional potential mechanism for the down-regulation of ICAM1 and VCAM-1 expression involves shedding from the cell surface and release of soluble fragments. Soluble forms of ICAM-1 and VCAM-1 have been detected in the culture supernatant of human endothelial cells and

INTEGRINS IN THE IMMUNE SYSTEM

353

in human plasma (Pigott et al., 1992; Leca et al., 1995; Rothlein et al., 1991). It is speculated that the cleavage of surface VCAM-1 involves a zinc-dependent metalloprotease (Leca et al., 1995). This shedding may suppress further leukocyte migration into tissues by removing the ligand from the cell surface and providing a soluble competitive ligand inhibitor. Furthermore, soluble forms may interact with the ability of integrins to provide signals, which alters cell function. Indeed, in T cells, immobilized VCAM-1 provides a costimulatory signal, whereas binding of soluble VCAM-1 initiates an inhibitory signal (Kitani et al., 1996). In sum, integrins and their ligands play indispensable roles in the development and functioning of the immune system. They do so primarily by mediating the cellular traffic instrumental to the immune response. Furthermore, integrins play important roles in modulating signals generated by clonotypic receptors and as effectors in processes such as cytolytic killing and phagocytosis. Many of the mechanisms that govern the functioning of integrins throughout the animal kingdom have been adapted to the specialized needs of the immune system. Furthermore, these receptors are readily accessible targets for small-molecule and antibody inhibitors. Because many integrin functions are surprisingly nonredundant (Hynes and Wagner, 1996), modification of integrin function is a promising avenue for therapeutic control of certain immune responses.

ACKNOWLEDGMENTS Work in our laboratories was funded by grants from the National Institutes of Health. DMR was supported by funds from the California Breast Cancer Research Program of the University of California, Grant No. 3FB-0164.

REFERENCES Abe, Y., Ballantyne, C. M., and Smith, C. W. (1996). Functions of domains 1 and 4 of' vascular cell adhesion molecule-1 in a4 integrin dependent adhesion under static and flow conditions are differentially regulated. J. Iinrnunol. 157, 5061-5069. Abitorabi, M. A,, Mackay, C. R., Jerome, E. H., Osorio, O., Butcher, E. C., and Erle, D. J. (1996). Differential expression of homing molecules on recirculating lymphocytes from sheep gut, peripheral, and lung lymph. J. ImrnunoZ. 156,3111-3117. Adler, M., Lazarus, R. A,, Dennis, M. S., and Wagner, G. (1991).Solution structure of Kistrin, a potent platelet aggregation inhibitor and GP IIb-IIIa antagonist. Science 253,445-448. Ahmad, M., Theofanisis, P., and Medford, R. M. (1998). Role of activating protein-1 in the regulation of the vascular cell adhesion molecule-1 gene expression by tumor necrosis factor-a. J. B i d . Chern. 273, 4616-4621. Akiyama, S. K., Nagata, K., andYamada, K. M. (1990).Cell surface receptors for extracellular matrix components. Biochirn. Biophgs. Actu Reu. Bionleinbr. 1031, 91-1 10. Abelda, S. M., and Buck, C. A. (1990). Integrins and other cell adhesion molecules. FASEB J. 4,2868-2880. d i g , L., Edenhofer, A., Hadvary, P., Hurzeler, M., Knopp, D., Muller, M., Steiner, B., Trzeciak, A,, and WeIler, T. (1992). Low molecuIar weight, non-peptide fibrinogen receptor antagonists. J. Med. Chem. 35, 4393-4407.

354

YOJI SHIMIZU et al

Altieri, D. C., Bader, R., Mannucci, P. M., and Edgington, T. S. (1988). Oligospecificity of the cellular adhesion receptor MAC-1 encompasses an inducible recognition specificity for fibrinogen. J. Cell Biol. 107, 1893-1900. Altieri, D. C., and Edgington, T. S. (1988). The saturable high affinity association of factor X to ADP-stimulated monocytes defines a novel function of the Mac-1 receptor. J. Biol. Chem. 263, 7007-7015. Anderson, D. C., Schmalsteig, F. C., Finegold, M. J., Hughes, B. J., Rothlein, R., Miller, L. J., Kohl, S., Tosi, M. F., Jacobs, R. L., and Waldrop, T. C. (1985). The severe and moderate phenotypes of heritable Mac-1, LFA-1 deficiency:Their quantitative definition and relation to leukocyte dysfunction and clinical features. J. Infect. Dis. 152, 668-689. Andrew, D. P., Rott, L. S., Kilshaw, P. J., and Butcher, E. C. (1996). Distribution of a4p7 and aEP7 integrins on thymocytes, intestinal epithelial lymphocytes and peripheral lymphocytes. Eur. J. Immunol. 26,897-905. Andrieux, A., Rabiet, M. J., Chapel, A., Concord, E., and Marguerie, G. (1991). A highly conserved sequence of the Arg-Gly-Asp-binding domain of the integrin p3 subunit is sensitive to stimulation. 1.Biol. Chem. 266, 14202-14207. Aota, S., Nagai, T., and Yamada, K. M. (1991). Characterization of regions of fibronectin besides the Arg-Gly-Asp sequence required for adhesive function of the cell-binding domain using site-directed mutagenesis. J. Bid. Chem. 266, 15938-15943. Aota, S., Nomizu, M., and Yamada, K. M. (1994). The short amino acid sequence pro-his-ser-arg-asn in human fibronectin enhances cell-adhesive function. J. Biol. Chem. 269, 24756-24761. Aota, S., and Yamada, K. M. (1997). Integrin functions and signal transduction. Adu. Exp. Med. Biol. 400B, 669-682. Arbones, M. L., Ord, D. C., Ley, K., Ratech, H., Maynard-Curry, C., Otten, G., Capon, D. J., and Tedder, T. F. (1994). Lymphocyte homing and leukocyte rolling and migration are impaired in L-selectin-deficient mice. Immunity 1, 247-260. Arroyo, A. G., Yang, J. T., Rayburn, H., and Hynes, R. 0. (1996). Differential requirements for a 4 integrins in hematopoiesis. Cell 85, 997-1008. Astier, A., Avraham, H., Manie, S. N., Groopman, J., Canty, T., Avraham, S., and Freedman, A. S. (1997). The related adhesion focal tyrosine kingse is tyrosine-phosphorylated after pl-integrin stimulation in B cells and binds to pl3Ocas. J. Biol. Chem. 272, 228-232. Austrup, F., Rebstock, S., Kilshaw, P. J,, and Hamann, A. (1995). Transforming growth factor -pl-induction expression of the mucosa-related integrin a E on lymphocytes is not associated with mucosa-specific homing. Eur.J. Immunol. 25, 1487-1491. Bajt, M. L., Loftus, J. C., Gawaz, M. P., and Ginsberg, M. H. (1992). Characterization of a gain of function mutation of integrin aIIbP3 (platelet GPIIb-IIIa). J. Biol. Chem. 267,22211-22216. Bajt, M. L., Goodman, T., and McGuire, S. L. (1995). p2 (CD18) is involved in ligand recognition by I domain integrins, LFA-1 (aLp2 CDlldCD18) and MAC-1 (aMP2 CDllb/CD18).J. Biol. Chem. 270, 94-98. Bajt, M. L., and Loftus, J. C. (1994). Mutation of a ligand binding domain of p3 integrin. Integral role of oxygenated residues in all&, (GPIIb-IIIa) receptor function. J. Biol. Chem. 269,20913-20919. Baker, E. K., Tozer, E. C., Pfaff, M., Shattil, S. J., Loftus, J. C., and Ginsberg, M. H. (1997). A genetic analysis of integrin function: Glanzmann thrombasthenia in uitro. Proc. Nutl. Acad. Sci. U.S.A. 94, 1973-1978. Banchereau, J., and Steinman, R. M. (1998). Dendritic cells and the control of immunity. Nature 392, 245-252.

INTECRINS IN THE IMMUNE SYSTEM

355

Barbas, C. F., Languino, L. R., and Smith, J. W. (1993). High-affinity self-reactive human antibodies by design and selection: Targeting the integrin ligand binding site. Proc. Natl. Acad. Sci. U.S.A. 90, 10003-10007. Bargatze, R. F., Jutila, M. A., and Butcher, E. C. (1995). Distinct roles of L-selectin and integrins a4P7 and LFA-1 in lymphocyte homing to Peyer’s patch-HEV in situ: The multistep model confirmed and refined. lmmunity 3, 99-108. Bamard, J, A., Warwick, G. J., and Gold, L. I. (1993). Localization of transforming growth factor isoforms in the normal murine small intestine and colon. Gastroenterology 105,67-73. Baron, J. L., Madri, J. A., Ruddle, N. H., Hashim, G., and Janeway, C. A., Jr. (1993). Surface expression of a 4 integrin by CD4 T cells is required for their entry into brain parenchyma. 1.Exp. Med. 177, 57-68. Barten, D. M., and Ruddle, N. H. (1994). Vascular cell adhesion molecule-1 modulation by tumor necrosis factor in experimental allergic encephalomyelitis. J. Neuroimmunol. 51, 123-133. Beer, J. H., Springer, K. T., and ColIer, B. S. (1992). Immobilized Arg-Gly-Asp (RGD) peptides of v a q n g lengths as structural probes of the platelet glycoprotein IIbiIIIa receptor. Blood 79, 117-128. Begue, B., Samacki, S., Le Deist, F., Buc, H., Gagnon, J., Meo, T., and Cerf-Bensussan, N. (1995). HML-1, a novel integrin made of the 0 7 chain and of a distinctive a chain, exerts an accessory function in the activation of human IEL via the CD3-TCR pathway. Adv. Exp. Med. Biol. 371A, 67-75. Belkin, A. M., Retta, S. F., Pletjushkina, 0.Y., Balzac, F., Silengo, L., F a d e r , R., Koteliansky, V. E., Burridge, K., and Tarone, G. (1997). Muscle PlD integrin reinforces the cytoskeleton-matrix link: Modulation of integrin adhesive function by alternative splicing. 1. Cell Biol. 139, 1583-1595. Bell, E. B., Sparshott, S. M., and Bunce, C. (1998). CD4+ T-cell memory, CD45R subsets and the persistence of antigen-A unifymg concept. lmmunol. Today 19, 60-64. Bennett, J. S., Vilaire, G., and Cines, D. B. (1982). Identification of the fibrinogen receptor on human platelets by photoaffinity labeling. J. Bwl. Chem. 257, 8049-8054. Bennett, J. S., and Viaire, G. (1979). Exposure of platelet fibrinogen receptors by ADP and epinephrine. J. Clin. Invest. 64, 1393-1401. Berendt, A. R., McDowall, A,, Craig, A. G., Bates, P. A., Sternberg, M. J. E., Marsh, K., Newbold, C. I., and H o g , N. (1992). The binding site on ICAM-1 for Plasmodium fulcipumm-infected erythrocytes overlaps, but is distinct from, the LFA-1-binding site. Cell 68, 71-81. Berlin, C., Berg, E. L., Briskin, M. J., Andrew, D. P., Kilshaw, P. J., Holzmann, B., Weissman, I. L., Hamann, A,, and Butcher, E. C. (1993). a4P7 integrin mediates lymphocyte binding to the mucosal vascular addressin madcam-1. Cell 74, 185-195. Berlin, C., Bargatze, R. F., Campbell, J. J., von Andrian, U. H., Szabo, M. C., Hasslen, S. R., Nelson, R. D., Berg, E. L., Erlandsen, S. L., and Butcher, E. C. (1995). a 4 integrins mediate lymphocyte attachment and rolling under physiologic flow. Cell 80, 413-422. Bevilacqua, M. P. (1993). Endothelid-leukocyte adhesion molecules. Annu. Rev. lmmunol. 11,767-804. Bhatia, R., Wayner, E. A,, McGlave, P. B., and Verfaillie, C. M. (1994). Interferon-a restores normal adhesion of chronic myelogenous leukemia hematopoietic progenitors to bone marrow stroma by correcting impaired Pl integrin receptor function. J. Clin. lnuest. 94,384-391. Bhatia, R., McCarthy, J. B., and Verfaillie, C. M. (1996). Interferon-a restores normal Pl integrin-mediated inhibition of hematopoietic progenitor proliferation by the marrow microenvironment in chronic myelogenous leukemia. Blood 87,3883-3891.

356

YO11 SHlMlZU et nl.

Blystone, S. D., Graham, I. L., Lindberg, F. P., and Brown, E. J. (1994). Integrin m p 3 differentially regulates adhesive and phagocytic functions of the fibronectin receptor a5bl.J. Cell Biol. 127, 1129-1137. Bonthron, D., Orr, E. C., Mitsock, L. M., Ginsburg, D., Handin, R. I., and Orkin, S. H. (1986). Nucleotide sequence of pre-pro-von Willebrand factor cDNA. Nucleic Acids Res. 14, 7125-7127. Bowditch, R. D., Halloran, C. E., Aota, S., Obara, M., Plow, E. F., Yamada, K. M., and Ginsberg, M. H. (1991). Integrin aII& (platelet GPIIb-IIIa) recognizes multiple sites in fibronectin. 1.Biol. &hem. 266, 23323-23328. Bowditch, R. D., Hariharan, M., Tominna, E. F., Smith, J. W., Yamada, K. M., Getzoff, E. D., and Ginsberg, M. H. (1994). Identification of a novel integrin binding site in fibronectin: Differential utilization by 0 3 integrins. 1,Biol. Chem. 269, 10856-10863. Brass, L. F., Shattil, S. J., Kunicki, T. J., and Bennett, J. S. (1985). Effect of calcium on the stability of the platelet membrane glycoprotein IIb-IIIa complex. 1. Biol. Chem. 260, 7875-7881. Braun, M., Pietsch, P., Zepp, A., Schror, K., Baumann, G., and Felix, S. B. (1997).Regulation of TNF-a and IL-lp-induced adhesion molecule expression in human vascular smooth muscle cells by CAMP.Arterioscler. Thromb. Vasc. Biol. 17, 2568-2365. Brown, K. A,, Bedford, P., Macey, M., McCarthy, D. A,, Leroy, F., Vora, A. J., Stagg, A. J., Dumonde, D. C., and Knight, S. C. (1997). Human blood dendritic cells: Binding to vascular endothelium and expression of adhesion molecules. Clin. Exp. Immunol. 107,601-607. Buck, C. A., Shea, E., Duggan, K., and Horwitz, A. F. (1986). Integrin (the CSAT antigen): Functionality requires oligomeric integrity. 1. Cell Biol. 103, 2421-2428. Butcher, E. C. (1991). Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell 67, 1033-1036. Butcher, E. C., and Picker, L. J. (1996). Lymphocyte homing and homeostasis. Science 272,60-66. Calvete, J. J., Alvarez, M. V., Rivas, G., Hew, C. L., Henschen, A,, and Gonzalez-Rodriguez, J. (1989a).Interchain and intrachain disulphide bonds in human platelet glycoprotein IIb. Localization of the epitopes for several monoclonal antibodies. Biochem.1,261,551-560. Calvete, J. J., Henschen, A,, and Gonzalez-Rodriguez, J. (198913).Complete localization of the intrachain disulphide bonds and the N-glycosylation points in the a-subunit of human platelet glycoprotein IIb. Biochem. J. 261, 561-568. Calvete, J. J., Arias, J., Alvarez, M. V., Lopez, M. M., Henschen, A., and Gonzalez-Rodriguez, J. (1991a).Further studies on the topography of the N-terminal region of human platelet glycoprotein IIIa. Localization of monoclonal antibody epitopes and the putative fibrinogen-binding sites. Biochem. 1.274, 457-463. Calvete, J. J., Arias, J., Alvarez, M. V., Lopez, M. M., Henschen, A., and Gonzalez-Rodriguez, J. (1991b).Further studies on the topography of human platelet glycoprotein IIb. Biochem. 1. 273,767-775. Calvete, J. J., Henschen, A., and Gonzalez-Rodriguez,J. (1991~). Assignment of disulphide bonds in human platelet GPIIIa. A disulphide pattern for the P-subunits of the integrin family. Biochem. 1.274,63-71. Calvete, J. J., Mann, K., Aharez, M. V., Lopez, M. M., and Gonzdez-Rodriguez, J. (1992). Proteolytic dissection of the isolated platelet fibrinogen receptor, integrin CPIIbflIIa. Biochem. 1. 282,523-532. Calvete, J. J., Mann, K., Schafer, W., Fernandez-Lafuente, R., and Guisan, J. M. (1994). Proteolytic degradation of the RGD-binding and non-RGD-binding conformers of human

INTEGRINS IN T H E IMMUNE SYSTEM

357

platelet integrin glycoprotein IIbnIIa: Clues for identification of regions involved in the receptor’s activation. Biochem. J. 298, 1-7. Campbell, J. J., Qin, S., Bacon, K. B., Mackay, C. R., and Butcher, E. C. (1996). Biology of chemokine and classical chemoattractant receptors: Differential requirements for adhesion-triggering versus chemotactic responses in lymphoid cells. 1. Cell Biol. 134, 255-266. Campbell, J. J., Hedrick, J., Zlotnik, A,, Siani, M. A,, Thompson, D. A,, and Butcher, E. C. (1998). Chemokines and the arrest of lymphocytes rolling under flow conditions. Science 279, 381-384. Camper, L., Hellman, U., and Lundgren-Akerlund, E. (1998). Isolation, cloning, and sequence analysis of the integrin subunit a10, a Bl-associated collagen binding integrin expressed on chondrocytes. 1.Biol. Chem. 273, 20383-20389. Carlos, T. M., and Harlan, J. M. (1994). Leukocyte-endothelial adhesion molecules. Blood 84,2068-2101. Carr, M. W., Alon, R., and Springer, T. A. (1996).The C-C chemokine MCP-1 differentially modulates the avidity of Bl and /32 integrins on T lymphocytes. lmmunity 4, 179-187. Carrell, N. A., Fitzgerald, L. A,, Steiner, B., Erickson, H. P., and Phillips, D. R. (1985). Structure of human platelet membrane glycoproteins IIb and IIIa as determined by electron microscopy. J. Biol. Chem. 260, 1743-1749. Casasnovas, J. M., Springer, T. A., Lm,J.-H., Hamson, S. C., and Wang, J.-H. (1997). Crystal structure of ICAM-2 reveals a distinctive integrin recognition surface. Nature 387, 312-315. Casanovas, J. M., Stehle, T., Liu, J.-H., Wang, J.-H., and Springer, T. A. (1998). A dimeric crystal structure for the N-terminal two domains of intercellular adhesion molecule-1. Proc. Natl. Acad. Sci. U.S.A. 95, 4134-4139. Castro, A,, Bono, M. R., Simon, V., Vargas, L., and Rosemblatt, M. (1997). Spleen-derived stromal cells, adhesion molecules expression and lymphocyte adhesion to reticular cells. Eur. J. Cell Bid. 74, 321-328. Catalina, M. D., Carroll, M. C., Arizpe, H., Takashima, A., Estess, P., and Siegelman, M. H. (1996). The route of antigen entry determines the requirement for L-selectin during immune responses. 1.Exp. Med. 184,2341-2351. Cepek, K. L., Parker, C. M., Madara, J. L., and Brenner, M. B. (1993). Integrin aEP7 mediates adhesion of T lymphocytes to epithelial cells. J. Immunol. 150,3459-3470. Cepek, K. L., Shaw, S. K., Parker, C. M., Russell, G. J., Morrow, J. S., Rmm, D. L., and Brenner, M. B. (1994). Adhesion between epithelial cells and T lymphocytes mediated by E-cadherin and the aEP7 integrin. Nature 372, 190-193. Cerf-Bensussan, N., B’egue, B., Gagnon, J., and Meo, T. (1992). The human intraepithelid lymphocyte marker HML-1 is an integrin consisting of a P7 subunit associated with a distinctive a chain. Eur. 1,Immunol. 22, 273-277. Chan, P., Lawrence, M. B., Dustin, M. L., Ferguson, L. M., Golan, D. E., and Springer, T. A. (1991). Influence of receptor lateral mobility on adhesion strengthening between membranes containing LFA-3 and CD2. /. Cell Biol. 115, 245-255. Chang, A. C., Salomon, D. R., Wadsworth, S., Hong, M. J., Mojcik, C. F., Otto, S., Shevach, E. M., and Coligan, J. E. (1995). A3pl and a 6 p l integrins mediate laminidmerosin binding and function as costimulatory molecules for human thymocyte proliferation. 1.Immunol. 154,500-510. Chapman, H. A. (1997). Plasminogen activators, integrins, and the coordinated regulation of cell adhesion and migration. Cum. Opin. Cell B i d . 9, 714-724. Cheng, S., Craig, W. S., Mullen, D., Tschopp, J. F., Dixon, D., and Pierschbacher, M. D. (1994). Design and synthesis of novel cyclic RGD-containing peptides as highly potent and selective integrin a,,& antagonists. J. Med. Chem. 37, 1-8.

358

YOJI SHIMIZU et al.

Cheresh, D. A., Smith, J. W., Cooper, H. M., and Quaranta, V. (1989). A novel vitronectin receptor integrin (a&) is responsible for distinct adhesive properties of carcinoma celIs. Cell 5 7 , 5 9 4 9 . Cheresh, D. A,, and Harper, J. H. (1987). Arginine-Glycine-Asparticacid recognition by a cell adhesion receptor requires its 130-kDa a subunit. 1.Biol. Chem. 262, 1434-1437. Christensen, J. P., Marker, O., and Thomsen, A. R. (1996). T-cell-mediated immunity to lymphocytic choriomeningitis virus in /32-integrin (CD18)- and ICAM-1 (CD54)-deficient mice. I. Virol. 70, 8997-9002. Cid, M. C., Esparza, J., Juan, M., Miralles, A,, Ordi, J., Vilena, R., Urbano-Marquez, A., Gaya, A,, Vives, J., and Yague, J. (1994). Signaling through CD50 (ICAM-3) stimulates T lymphocyte binding to human umbilical vein endothelial cells and extracellular matrix and /32 integrin function. Eur. 1.Immunol. 24, 1377-1382. proteins via an increase in @l Corbi, A. L., Kishimoto, T. K., Miller, L. J., and Springer, T. A. (1988).The human leukocyte adhesion glycoprotein Mac-1 (Complement receptor type 3, CD1 Ib) a subunit: Cloning, primary structure, and relation to the integrins, von Willebrand factor and factor B. 1.Biol. Chem. 263, 12403-12411. Cornejo, C. J., Winn, R. K., and Harlan, J. M. (1997).Anti-adhesion therapy. Adu. Pharmucol. 39,99-142. Crisa, L., Cirulli, V., Ellisman, M. H., Ishii, J. K., Elices, M . J., and Salomon, D. R. (1996). Cell adhesion and migration are regulated at distinct stages of thymic T cell development The roles of fibronectin, V U 4 , and VLA5.I. Exp. Med. 184, 215-228. Crowe, D. T., Chiu, H., Fong, S., and Weissman, I. L. (1994). Regulation of the avidity of integrin Q& by the /3, cytoplasmic domain. J. Biol. Chem. 269, 14411-14418. DSouza, S. E., Ginsberg, M. H., Burke, T. A,, Lam, S. C., and Plow, E. F. (1988). Localization of an Arg-Gly-Asp recognition site within an integrin adhesion receptor. Science 242, 91-93. D’Souza, S. E., Ginsberg, M. H., Burke, T. A,, and Plow, E. F. (1990). The ligand binding site of the platelet integrin receptor GPIIb-IIIa is proximal to the second cdcium binding domain of its LY subunit. 1.Biol. Chem. 265, 3440-3446. D’Souza, S. E., Ginsberg, M. H., Matsueda, G. R., and Plow, E. F. (1991). A discrete sequence in a platelet integrin is involved in ligand recognition. Nature 350, 66-68, DSouza, S. E., Haas, T. A., Piotrowicz, R. S., Byers-Ward, V., McGrath, D. E., Soule, H. R., Cierniewski, C., Plow, E. F., and Smith, J. W. (1994). Ligmd and cation binding are dual functions of a discrete segment of the integrin subunit: Cation displacement is involved in ligand binding. Cell 79, 659-667. Damle, N. K., Klussman, K., Leytze, G., Aruffo, A., Linsley, P. S., and Ledbetter, J. A. (1993a). Costimulation with integrin ligands intercellular adhesion molecule-1 or vascular cell adhesion molecule-1 augments activation-induced death of antigen specific CD4+ Tlymphocytes.J. Immunol. 151, 2368-2379. Damle, N . K., Klussman, K., Leytze, G., and Linsley, P. S. (1993b). Proliferation of human T lymphocytes induced with superantigens is not dependent on costimulation by the CD28 counterreceptor B7.J Immunol. 150, 726-735. Danen, E. H. J., Aota, S., van Kraats, A. A., Yamada, K. M., Ruiter, D. J., and van Muijen, G. N. P. (1995). Requirement for the synergy site for cell adhesion to fibronectin depends on the activation state of integrin a5/31.J. Biol. Chem. 270, 21612-21618. Danilov, Y. N., and Juliano, R. L. (1989). Phorbol ester modulation of integrin-mediated cell adhesion: A postreceptor event. J. Cell Biol. 108, 1925-1933. Das, A. M., Williams, T. J., Lobb, R., and Nourshargh, S. (1995). Lung eosinophilia is dependent on IL-5 and the adhesion molecules CD18 and VLA-4, in a guinea-pig model. lmmunology 84, 41-46.

INTEGRINS IN THE IMMUNE SYSTEM

359

Davignon, D., Martz, E., Reynolds, T., Kurzinger, K., and Springer, T. A. (1981a). Monoclonal antibody to a novel lymphocyte function-associated antigen (LFA-1): Mechanism of blockade of T lymphocyte-mediated kdling and effects on other T and B lymphocyte functions. 1.Immunol. 127,590-595. Davignon, D., Martz, E., Reynolds, T., Kurzinger, K., and Springer, T. A. (1981b). Lymphocyte function-associated antigen 1 (LFA-1): A surface antigen distinct from Lyt-2,3 that participates in T lymphocyte-mediated killing. Proc. Natl. Acad. Sci. U.S.A. 78,4535-4539. de Fougerolles, A. R., Stacker, S. A., Schwarting, R., and Springer, T. A. (1991). Characterization of ICAM-2 and evidence for a third counter-receptor for LFA-1. 1. Exp. Med. 174, 253-267. de Fougerolles, A. R., Qin, X., and Springer, T. A. (1994). Characterization of the function of intercellular adhesion molecule (ICAM)-3 and comparison with JCAM-1 and ICAM2 in immune responses. I. Exp. Med. 179,619-629. de Fougerolles, A. R., and Springer, T. A. (1992). Intercellular adhesion molecule 3, a third adhesion counter-receptor for lymphocyte function-associated molecule 1on resting lymphocytes. 1.Exp. Med. 175, 185-190. Degitz, K., Lian-Jie, L., and Caughman, S. W. (1991). Cloning and characterization of the 5’-transcriptional regulatory region of the human intercellular adhesion molecule 1gene. 1.Biol. Chem. 266, 14024-14030. Diamond, M. S., Staunton, D. E., Marlin, S. D., and Springer, T. A. (1991). Binding of the integrin Mac-1 (CDllbICD18) to the third immunoglobulin-like domain of ICAM1 (CD54) and its regulation by glycosylation. Cell 65, 961-971. Diamond, M. S., Garcia-Aguilar, J.. Bickford, J. K., Corbi, A. L., and Springer, T. A. (1993). The I domain is a major recognition site on the leukocyte integrin Mac-1 (CDllbICD18) for four distinct adhesion ligands. J. Cell Biol. 120, 1031-1043. Diamond, M. S., and Springer, T. A. (1994). Dynamic regulation of integrins: How signals inside translate into adhesion outside. Cum. Biol. 4 , 506-517. Diaz-Gonzalez, F., Forsyth, J., Steiner, B., and Ginsberg, M. H. (1996). Trans-dominant inhibition of integrin function. MoZ. Biol. Cell 7 , 1939-1951. Diaz-Gonzalez, F., and Ginsberg, M. H. (1996). Cell adhesion molecules and endothelial cells in arthritis. In “Arthritis and Allied conditions” (W. J. Koopman, ed.), pp. 479-489. Williams & Wilkins, Baltimore, Maryland. Diaz-Gonzalez, F., and Sanchez-Madrid, F. (1998). Inhibition of leukocyte adhesion: An alternative mechanism of action for anti-inflammatory drugs. Zmmunol. Today 19, 169-172. Dickenson, S. K., Walsh, J. J., and Santoro, S. A. (1997). Contributions of the I and E F hand domains to the divalent cation-dependent collagen binding activity of the a 2 p l integrin. I. Biol. Chem. 272, 7661-7668. Dittel, B. N., McCarthy, J. B., Wayner, E. A., and LeBien, T. W. (1993). Regulation of human B-cell precursor adhesion to bone marrow stromal cells by cytokines that exert opposing effects on the expression of vascular cell adhesion molecule-1 (VCAM-1).Blood 81,2272-2282. Doussis-Anagnostopodou, I,, Kaklamanis, L., Cordell, J. L., Jones, M., Turley, H., Pulford, K., Simmons, D., Mason, D., and Gatter, K. (1993). ICAM-3 expression on endothelium in lymphoid malignancy. Am. 1.Pathol. 143, 1040-1043. Du, X., Gu, M., Weisel, J. W., Nagaswami, C., Bennett, J. S., Bowditch, R. D., and Ginsberg, M. H. (1993). Long range propagation of conformational changes in integrin a,&. J. B i d Chem. 268, 23087-23092. Dunon, D., and Imhof, B. A. (1993). Mechanisms of thymus homing. Blood 81, 1-8.

360

YOJI SHIMIZU d nl.

Dunon, D., and Imhof, B. A. (1996).T cell migration during ontogeny and T cell repertoire generation. Curr. Top. Microbiol. lmmunol. 212, 79-93. Dustin, M. L., and Springer, T. A. (1989).T-cell receptor cross-linkingtransiently stimulates adhesiveness through LFA-1. Nature 341, 619-624. Emsley, J., King, S. L., Bergelson, J. M., and Liddington, R. C. (1997). Crystal structure of the I domain from integrin a 2 p l . J. Biol. Chem. 272, 28512-28517. Engelhardt, B., Conley, F. K., Kilshaw, P. J., and Butcher, E. C. (1995). Lymphocytes infiltrating the CNS during inflammation display a distinctive phenotype and bind to VCAM-1 but not to MAdCAM-1. lnt. lmmunol. 7, 481-491. Engelhardt, B. (1998).The role of a4-integrin in T lymphocyte migration into the inflamed and noninflamed central nervous system. Cum Top. Microbiol. lmmunol. 231,51-64. Erle, D. J., Briskin, M. J., Butcher, E. C., Garcia-Pardo, A,, Lazarovits, A. I., and Tidswell, M. (1994).Expression and function of the MAdCAM-1receptor, integrin a4p7, on human leukocytes. /. lmmunol. 153,517-528. Fassler, R., and Meyer, M. (1995). Consequences of lack of @lintegrin gene expression in mice. Genes Dev. 9, 1896-1908. Faull, R. J., Kovach, N. L., Harlan, J. M., and Ginsberg, M. H. (1994). Stimulation of integrin-mediated adhesion of T lymphocytes and monocytes: Two mechanisms with divergent biological consequences. 1. Exp. Med. 179, 1307-1316. Fawcett, J., Holness, C. L., Needham, L. A., Turley, H., Gatter, K. C., Mason, D. Y., and Simmons, D. L. (1992). Molecular cloning of ICAM-3, a third ligand for LFA-1, constitutively expressed on resting leukocytes. Nature 360, 481-484. Fernandez, M. A,, Munoz-Fernandez, M . A., and Fresno, M. (1993). Involvement of pl integrins in the binding and entry of Typanosoma cruzi into human macrophages. Eur. J. lmmunol. 23,552-557. Fine, J. S., and Kruisbeek, A. M. (1991). The role of LFA-l/ICAM-1 interactions during murine T lymphocyte development. /. lmmunol. 147, 2852-2859. Fitzgerald, L. A., and Phillips, D. R. (1985). Calcium regulation of the platelet membrane glycoprotein IIb-IIIa complex. J. Biol. Chem. 260, 11366-11374. Fleming, J. C., Pahl, H. L., Gonzalez, D. A., Smith, T. F., and Tenen, D. G . (1993). Structural analysis of the C D l l b gene and phylogenetic analysis of the X a-integrin gene family demonstrate remarkable conservation of genomic X organization and suggest early diversification during evolution. J. lmmunol. 150, 480-490. Fornaro, M. and Languino, L. R. (1997). Alternatively spliced variants: A new view of the integrin cytoplasmic domain. Matrix Biol. 16, 185-193. Freedman, A. S., Rhynhart, K., Nojima, Y., Svahn, J., Eliseo, L., Benjamin, C. D., Morimoto, C., and Vivier, E. (1993). Stimulation of protein tyrosine phosphorylation in human B cells after ligation of the pl integrin VLA-4. J. lmmunol. 150, 1645-1652. Frelinger, A. L., 111, Lam, S. C., Plow, E. F., Smith, M. A., Loftus, J. C., and Ginsberg, M. H. (1988). Occupancy of an adhesive glycoprotein receptor modulates expression of an antigenic site involved in cell adhesion. J. Biol. Chem. 263, 12397-12402. Frelinger, A. L., 111, Cohen, I., Plow, E. F., Smith, M. A,, Roberts, J., Lam, S. C.-T., and Ginsberg, M. H. (1990). Selective inhibition of integrin function by antibodies specific for ligand-occupied receptor conformers. J. Biol. Chem. 265, 6346-6352. Friedrich, C., Cybulsky, M. I., and Gutierrez-Ramos, J. C. (1996). Vascular cell adhesion molecule-1 expression by hematopoiesis-supporting stromal cells is not essential for lymphoid or myeloid differentiation in oivo or in uitro. Eur. J. lmmunol. 26, 2773-2780. Frisch, S. M., and Ruoslahti, E. (1997). Integrins and anoikis. Curr. +in. CeZZ Bial. 9, 701-706.

I N T E G R I N S IN THE I M M U N E SYSTEM

361

Gallatin, W. M., Weissman, I. L., and Butcher, E. C. (1983).A cell-surface molecule involved in organ-specific homing of lymphocytes. Nature 304, 30-34. Georges-Labouesse, E., Messaddeq, N., Yehia, G., Cadalbert, L., Dierich, A., and Le Meur, M. (1996). Absence of integrin a 6 leads to epidermolysis bullosa and neonatal death in mice. Nature Genet. 13, 370-373. Giancotti, F. G. (1997). Integrin signaling: specificity and control of cell survival and cell cycle progression. Curr. Opin. Cell Biol. 9, 691-700. Ginsberg, M. H., Forsyth, J., Lightsey, A,, Chediak, J., and Plow, E. F. (1983). Reduced surface expression and binding of fibronectin by thrombin-stimulated thrombasthenic platelets. 1.Clin. Inuest. 71, 619-624. Ginsberg, M. H., Pierschbacher, M. D., Ruoslabti, E., Marguerie, G. A,, and Plow, E. F. (1985).Inhibition of fibronectin binding to platelets by proteolFc fragments and synthetic peptides which support fibroblast adhesion. J. Biol. Chem. 260, 3931-3936. Ginsberg, M. H., Lightsey, A,, Kunicki, T. J., Kaufman, A., Marguerie, G. A,, and Plow, E. F. (1986). Divalent cation regulation of the surface orientation of platelet membrane glycoprotein IIb: Correlation with fibrinogen binding function and definition of a novel variant of Glanzmann thrombasthenia. J. Clin. Inuest. 78, 1103-1111. Ginsberg, M. H., Loftus, J. C., Ryckwaert, J.-J., Pierschbacher, M. D., Pytela, R., Ruoslahti, E., and Plow, E. F. (1987). Immunochemical and amino-terminal sequence comparison of two cytoadhesins indicates they contain similar or identical p subunits and distinct a subunits. J Biol. Chem. 262, 5437-5440. Ginsberg, M. H., Du, X., and Plow, E. F. (1992a). Inside-out integrin signahg. Cum. Opin. Cell Biol. 4, 766-771. Ginsberg, M. H., O’Toole, T. E., Loftus, J. C., and Plow, E. F. (199213). Ligand binding to integrins: Dynamic regulation and common mechanisms. CSH Symp. @ant. B i d . 57,221-231. Girard, J. P., and Springer, T. A. (1995). High endothelial venules (HEVs): Specialized endothelium for lymphocyte migration. Immunol. Today 16, 449-457. Goodman, T., and Lefrancois, L. (1989). Intraepithelial lymphocytes. Anatomical site, not T cell receptor form, dictates phenotype and function. 1. Exp. Med. 170, 1569-1581. Greenspoon, N., Hershkoviz, R., Alon, R., Varon, D., Shenkman, B., Mam, G., Federman, S., Kapustina, G., and Lider, 0. (1993). Structural analysis of integrin recognition and the inhibition of integrin-mediated cell functions by novel nonpeptidic surrogates of the Arg-Gly-Asp sequence. Biochemistry 32, 1001-1008. Gresham, H. D., Goodwin, J. L., Allen, P. M., Anderson, D. C., and Brown, E. J. (1989). A novel member of the integrin receptor family mediates Arg-Cly-Asp-stimulatedneutrophil phagocytosis. /. Cell Bid. 108, 1935-1943. Grinnell, F. (1977). Cellular adhesiveness and extracellular substrata. Int. Reu. Cytol. 53, 65-144. Guan, J. L. (1997). Focal adhesion kinase in integrin signaling. Matrix Biol. 16, 195-200. Gulino, D., Boudignon, C., Zhang, L. Y., Concord, E., Rabiet, M. J., and Marguerie, G. (1992). Ca(2f)-binding properties of the platelet glycoprotein IIb ligand-interacting domain. J. Bwl. Chem. 267, 1001-1007. Gunn, M. D.,Tangemann, K., Tam, C., Cyster, J. G., Rosen, S. D., and Williams, L. T. (1998). A chemokine expressed in lymphoid high endothelial venules promotes the adhesion and chemotaxis of naive T lymphocytes. Proc. Natl. Acad. Sci. U.S.A. 95, 258-263. Hahne, M., Lenter, M., Jager, U., Isenmann, S., and Vestweber, D. (1993). VCAM-1 is not involved in LPAM-1 (a4ppla4p7) mediated binding of lymphoma cells to high endothelial venules of mucosa-associated lymph nodes. Eur. 1. Cell Biol. 61, 290-298.

362

YOJI SHIMIZU et al

Halstensen, T. S., Scott, H., and Brandtzaeg, P. (1989). Intraepithelial T cells of the TcR gammddelta+ CD8- and V Sl/J61+ phenotypes are increased in coeliac disease. Scud. 1.Immunol. 30,665-672. Hamm, H. E. (1998). The many faces of G protein signaling.]. Biol. Chem. 273,669-672. Hanninen, A., Taylor, C., Streeter, P. R., Stark, L. S., Sarte, J. M., and Shizuru, J. A. (1993). Vascular addressins are involved on islet vessels during insulitis in nonobese diabetic mice and are involved in lymphoid cell binding to islet endothelium. 1. Clin. Invest. 92,2509-2515. Hanninen, A., Salmi, M., Simell, O., and Jalkanen, S. (1996). Mucosa-associated (p7integrinhigh) lymphocytes accumulate early in the pancreas of NOD mice and show aberrant recirculation behavior. Diabetes 45, 1173-1180. Hantgan, R. R., Braaten, J. V., and Rocco, M. (1993). Dynamic light scattering studies of aIIbp3 solution conformation. Biochemistry 32,3935-3941. Hayashi, Y., Haimovich, B., Reszka, A., Boettiger, D., and Honvitz, A. (1990). Expression and function of chicken integrin /31 subunit and its cytoplasmic domain mutants in mouse NIH 3T3 cells. 1.Cell Biol. 110, 175-184. Hedrick, J. A., and Zlotnik, A. (1997). Identification and characterization of a novel /3 chemokine containing six conserved cysteines. 1.Immunol. 159, 1589-1593. Hemler, M. E. (1990). VLA proteins in the integrin family: Structures, functions, and their role on leukocytes. Annu. Rev. Immunol. 8, 365-400. Hemler, M., and Lobb, R. (1995).The leukocyte /3lintegrin. Cum. Opin. Hematol. 2,61-67. Hemler, M. E., Crouse, C., and Sonnenberg, A. (1989). Association of the V I A a 6 subunit with a novel protein. A possible alternative to the common VIA /3l subunit of certain cell lines. J. B i d Chem. 264,6529-6535. Hemler, M. E., Weitzman, J. B., Pasqualini, R., Kawaguchi,S., Kassner, P. D., and Berdichevsky, F. B. (1994). Structure, biochemical properties, and biological functions of integrin cytoplasmic domains. In “Integrins: The Biological Problem” (Y.Takada, ed.), pp. 1-36. CRC Press, Cleveland, Ohio. Hermiston, M. L., and Gordon, J. I. (1995). Inflammatory bowel disease and adenomas in mice expressing a dominant negative N-cadherin. Science 270, 1203-1207. Hernandez-Caselles, T., Rubio, G., Campanero, M. R., del Pozo, M. A., Muro, M., SanchezMadrid, F., and Aparicio, P. (1993). ICAM-3, the third LFA-1 counterreceptor, is a costimulatory molecule for both resting and activated T lymphocytes. Eur. J. Immunol. 23,2799-2806. Hesterberg, P. E., Winsor-Hines, D., Briskin, M. J., Soler-Ferran, D., M e d , C., Mackay, C. R., Newman, W., and Ringler, D. 1. (1996). Rapid resolution of chronic colitis in the cotton-top tamarin with an antibody t o a gut-homing integrin a4/37. Gastroenterology 111, 1373-1380. Higgins, J. M., Mandlebrot, D. A., Shaw, S . K., Russell, G. J., Murphy, E. A., Chen, Y.T., Nelson, W. J., Parker, C.M., and Brenner, M. B. (1998). Direct and regulated interaction of integrin aE/37 with E-cadherin. J. Cell Biol. 140, 197-210. Hirsch, E., Iglesias, A., Potocnik, A. J., Hartmann, U., and Fassler, R. (1996). Impaired migration but not differentiation of haematopoietic stem cells in the absence of& integrins. Nature 380, 171-175. Hogervorst, F., Kuikman, I., Von dem Borne, A. E. G., and Sonnenberg, A. (1990). Cloning and sequence analysis of /3-4 cDNA: An integrin subunit that contains a unique 118 kd cytoplasmic domain. EMBO J. 9, 765-770. Holzmann, B., McIntyre, B. W., and Weissman, I. L. (1989). Identification of a murine Peyer’s patch-specific lymphocyte homing receptor as an integrin molecule with an a chain homologous to human VLA-4a. Cell 56, 37-46.

INTEGRINS IN THE IMMUNE SYSTEM

363

Holzmann, B. and Weissman, I. L. (1989). Peyer's patch-specific lymphocyte homing receptors consist of a VLA-4-like a chain associated with either of two integrin p chains, one of which is novel. EMBO J. 8, 1735-1741. Horst, E., Meijer, C. J., Duijvestijn, A. M., Hartwig, N., Van der Harten, H. J., and Pals, S. T. (1990). The ontogeny of human lymphocyte recirculation: High endothelial cell antigen (HECA-452) and CD44 homing receptor expression in the development of the immune system. Eur. J. Immunol. 20,1483-1489. Horton, M. A., Dorey, E. L., Nesbitt, S. A., Samanen, J., Ah, F. E., Stadel, J. M., Nichols, A., Greig, R., and Helfrich, M. H. (1993). Modulation of vitronectin receptor-mediated osteoclast adhesion by Arg-Gly-Asp peptide analogs: A structure-function analysis. J. Bone Miner. Res. 8,239-247. Howe, A., Aplin, A. E., Alahari, S. K., and Juliano, R. L. (1998). Integrin signaling and cell growth control. Cum. Opin. Cell Biol. LO, 220-231. Hromas, R., Kim, C. H., Klemsz, M., Krathwohl, M., Fife, K., Cooper, S., Schnizlein-Bick, C., and Broxmeyer, H. E. (1997). Isolation and characterization of Exodus-2, a novel C-C chemokine with a unique 37-amino acid carboxyl-terminal extension. J. Immrnol. 159,2554-2558. Huang, C., Lu, C., and Springer, T. A. (1997). Folding of the conserved domain but not of flanking regions in the integrin /32 subunit requires association with the a subunit. Proc. Nutl. Acud. Sci. U.S.A. 94, 3156-3161. Huang, C., and Springer, T. A. (1997). Folding of the &propeller domain of the integrin aL subunit is independent of the I domain and dependent on the /32 subunit. Proc. Nutl. Acad. Sci. U.S.A. 94,3162-3167. Huang, X., Chen, A,, Agrez, M., and Sheppard, D. (1993). A point mutation in the integrin & subunit abolishes both binding to fibronectin and receptor localization to focal adhesion plaques. Mol. Biol. Cell 4, 283a (abstract). Hughes, A. L. (1992). Coevolution of the vertebrate integrin a-and /3-chain genes. Mol. Biol. Evol. 9, 216-234. Huleatt, J. W., and Lefrancois, L. (1995). Antigen-driven induction of C D l l c on intestinal intraepithelial lymphocytes and CD8' T cells in uiuo. J. Immunol. 154, 5684-5693. Huleatt, W., and Lefrancois, L. (1996).p2 integrins and ICAM-1 are involved in establishment of the intestinal mucosal T cell compartment. Immunity 5, 263-273. Humphries, M. J., Akiyama, S. K., Komoriya, A., Olden, K., and Yamada, K. M. (1986). Identification of an alternatively spliced site in human plasma fibronectin that mediates cell type-specific adhesion. J. Cell Biol. 103, 2637-2647. Hurley, R. W., McCarthy, J. B., and Verfaillie, C. M. (1995). Direct adhesion to bone marrow stroma via fibronectin receptors inhibits hematopoietic progenitor proliferation. J. Clin. Invest. 96, 511-519. Huttenlocher, A., Ginsberg, M. H., and Honvitz, A. F. (1996). Modulation of cell migration by integrin mediated cytoskeletal linkages and ligand binding affinity. J. Cell Bwl. 134, 1551-1562. Hwang, S. T., Singer, M. S., Giblin, P. A,, Yednock, T. A., Bacon, K. B., Simon, S. I., and Rosen, S. D. (1996). GlyCAM-1, a physiologic ligand for L-selectin, activates P2 integrins on naive peripheral lymphocytes. 1. Exp. Med. 184, 1343-1348. Hynes, R. 0. (1987). Integrins: A family of cell surface receptors. Cell 48, 549-554. Hynes, R. 0. (1992). Integrins: Versatility, modulation, and signaling in cell adhesion. Cell 69, 11-25. Hynes, R. O., and Wagner, D. D. (1996). Genetic manipulation of v a s d a r adhesion molecules in mice. J. Clin. fnuest. 98, 2193-2195. J,

364

YOJI SHIMIZU et al

Iademarco, M. F., McQuillan, J. J., Rosen, G. D., and Dean, D. C. (1992).Characterization of the promoter for vascular cell adhesion molecule-1 (VCAM-1).J. Biol. C h m . 267, 16323-16329. Iademarco, M. F., Barks, J. L., and Dean, D. C. (1995). Regulation ofvascular cell adhesion molecule-1 expression by IL-4 and TNF-a in cultured endothelial cells. J. Clin. Invest. 95, 264-271. Imhof, B. A., Weerasinghe, D., Brown, E. J., Lindberg, F. P., Hammel, P., Pi&, L., Dessing, M., and Gisler, R. (1997). Cross talk between a(v)/33 and a 4 P l integrins regulates lymphocyte migration on vascular cell adhesion molecule. Eur. J. Immunol. 27, 32423252. Irani, D. N., and Griffin, D. E. (1996). Regulation of lymphocyte homing into the brain during viral encephalitis at various stages of infection. J. Immunol. 156, 3850-3857. Irie, A,, Kamata, T., and Takada, Y. (1997). Multiple loop structures critical for ligand binding of the integrin a 4 subunit in the upper face of the &propeller mode 1. Proc. Natl. Acad. Sci. U.S.A. 94, 7198-7203. Isberg, R. R. and Leong, J. M. (1990). Multiple /3l chain integrins are receptors for invasin, a protein that promotes bacterial penetration into mammalian cells. Cell 60, 861-871. Isobe, M., Yagita, H., Ohmura, K., and Ihara, A. (1992). Specific acceptance of cardiac allograft after treatment with antibodies to ICAM-1 and LFA-1. Science 255,1125-1127. Jacobsen, K., Kravitz, J., Kincade, P. W., and Osmond, D. G. (1996). Adhesion receptors on bone marrow stromal cells: In vivo expression of vascular cell adhesion molecule-1 by reticular cells and sinusoidal endothelium in normal and gamma-irradiated mice. Blood 87,73-82. Jahbowski, A., Rosa, M. D., Bixler, S., Lobb, R., and Burkly, L. C. (1995). Vascular cell adhesion molecule (VCAM)-Ig fusion protein defines distinct affinity states of the very late antigen-4 (VLA-4) receptor. Cell Adhes. Commun. 3, 131-142. Jennings, L. K., and Phillips, D. R. (1982). Purification of glycoproteins IIb and I11 from human platelet plasma membranes and characterization of a calcium-dependent glycoprotein IIb-111 complex. J. Biol. Chem. 257, 10458-10466. Johnson, G. R., and Moore, M. A. (1975). Role of stem cell migration in initiation of mouse foetal liver haemopoiesis. Nature 258, 726-728. Kajiji, S., Tamura, R. N., and Quaranta, V. (1989). A novel integrin (a&) from human epithelial cells suggests a fourth family of integrin adhesion receptors. EMBO J. 8, 673-680. Kamata, T., Puzon, W., and Takada, Y. (1994). Identification of putative ligand binding sites within I domain of integrin a& (VLA-2, CD49b/CD29).]. Biol. Chem. 269,9659-9663. Kamata, T., Puzon, W., and Takada, Y. (1995). Identification of putative ligand-binding sites of the integrin a4/3l (VLA-4, CD49dKD29). Biochem. J. 305, 945-951. Kamata, T., Irie, A., Tohhira, M., and Takada, Y. (1996). Critical residues of integrin aIIb subunit for binding of aIIb/33 (glycoprotein IIb-IIIa) to fibrinogen and ligand-mimetic antibodies (PAC-1, OP-G2, and LJ-CPS). J. Biol. Chem. 271, 18610-18615. Kamata, T., and Takada, Y. (1994). Direct binding of collagen to the I domain of integrin a2Bl (VLA-2, CD49b/CD29) in a divalent cation-independent manner. J. Biol. Chem. 269, 26006-26010. Karecla, P. I., Bowden, S. J., Green, S. J., and Kilshaw, P. J. (1995). Recognition of Ecadherin on epithelial cells by the mucosal T cell integrin cy M290 07 (aEP7).Eur. J, Immunol. 25, 852-856. Karecla, P. I., Green, S. J., Bowden, S. J., Coadwell,J., and Kilshaw, P. J. (1996).Identification of a binding site for integrin aEP7 in the N-terminal domain of E-cadherin. J. Biol. Chem. 271,30909-30915.

INTEGRINS IN THE IMMUNE SYSTEM

365

Kedl, R. M., and Mescher, M. F. (1997). Migration and activation of antigen-specific CD8' T cells upon in uiuo stimulation with allogeneic tumor. J. Immunol. 159, 650-663. Kern, A., Briesewitz, R., Bank, I., and Marcantonio, E. E. (1994). The role of the I domain in ligand binding of the human integrin alp1.1.Biol. Chem. 269, 22811-22816. Kerst, J. M., Sanders, J. B., Slaper-Cortenbach, I. C., Doorakkers, M. C., Hooibrink, B., van Oers, R. H., von dem Borne, A. E., and van der Schoot, C. E. (1993). a 4 P l and a5Pl are differentially expressed during myelopoiesis and mediate the adherence of human CD34' cells to fibronectin in an activation-dependent way. Blood 81, 344-351. Keszthelyi, E., Karlik, S., Hyduk, S., Rice, G. P., Gordon, C., Yednock, T., and Homer, H. (1996). Evidence for a prolonged role of (Y 4 integrin throughout active experimental allergic encephalomyelitis. Neurology 47, 1053- 1059. Kilger, G., Needham, L. A., Nielsen, P. J., Clements, J., Vestweber, D., and Holzmann, B. (1995). Differential regulation of a, integrin-dependent binding to domains 1 and 4 of vascular cell adhesion molecule-1. ]. BioL Chem. 270,5979-5984. Kilshaw, P. J. (1993). Expression of the mucosal T cell integrin (Y M290 P7 by a major subpopulation of dendritic cells in mice. Eur. ]. Immunol. 23,3365-3368. Kilshaw, P. J., and Baker, K. C. (1988).A unique surface antigen on intraepithelial lymphocytes in the mouse. Immunol. Lett. 18, 149-154. Kilshaw, P. J., and Karecla, P. (1997). Structure and function of the mucosal T-cell integrin cuEP7. Biochem. Soc. Trans. 25,433-439. Kilshaw, P. J., and Murant, S. J. (1990).A new surface antigen on intraepithelial lymphocytes in the intestine. Eur. ]. Immunol. 20, 2201-2207. Kinashi, T., Escobedo, J. A,, Williams, L. T., Takatsu, K., and Springer, T. A. (1995). Receptor tyrosine kinase stimulates cell-matrix adhesion by phosphatidylinosito13 kinase and phospholipase C-gamma 1 pathways. Blood 86,2086-2090. King, S. L., Kamata, T., Cunningham, J. A,, Emsley, J., Liddington, R. C., Takada, Y., and Bergelson, J. M. (1997). Echovirus 1 interaction with the human vely late antigen-:! (integrin a2Pl) I domain. Identification of two independent virus contact sites distinct from the metal ion-dependent adhesion site. J. Bid. Chem. 272, 28518-28522. Kishirnoto, T. K., Hollander, N., Roberts, T. M., Anderson, D. C., and Springer, T. A. (1987). Heterogeneous mutations in the /3 subunit common to the LFA-1, Mac-1, and p150,95 glycoproteins cause leukocyte adhesion deficiency. Cell 50, 193-202. Kishimoto, T. K., Warnock, R. A,, Jutila, M. A,, Butcher, E. C., Lane, C., Anderson, D. C., and Smith, C. W. (1991).Antibodies against human neutrophil LECAM-1 (LAMl/Leu-8/DREG-56 antigen) and endothelial cell ELAM-1 inhibit a common CD18independent adhesion pathway in uitro. Blood 78, 805-811. Kitani, A,, Nakashima, N., Matsuda, T., Xu, B., Yu, S., Nakamura, T., and Matsuyama, T. (1996). T cells bound by vascular cell adhesion molecule-l/CDlO6 in synovial fluid in rheumatoid arthritis inhibitor role of soluble vascular cell adhesion molecule-1 in T-cell activation. ]. Immunol. 156, 2300-2308. Kitazawa, H., Muegge, K., Badolato, R., Wang, J. M., Fogler, W. E., Ferris, D. K., Lee, C. K., Candeias, S., Smith, M. R., Oppenheirn, J. J., and Durum, S. K. (1997). IL-7 activates a 4 P l integrin in murine thynocytes. /. Immunol. 159, 2259-2264. Klickstein, L. B., York, M. R., Fougerolles, A. R., and Springer, T. A. (1996). Lochation of the binding site on intercellular adhesion molecule-3 ( ICAM-3) for lymphocyte functionassociated antigen 1 (LFA-1).]. Bid. Chem. 271, 23920-23927. Koivunen, E., Cay, D. A,, and Ruoslahti, E. (1993). Selection of peptides binding of the a& integrin from phage &splay Library ]. Biol. Chem. 268,20205-20210. Kolanus, W., Nagel, W., SchiUer, B., Zeitlmann, L., Godar, S., Stockinger, H., and Seed, B. (1996).aLP2 integrin/LFA-1 binding to ICAM-1 induced hy cytohesin-1, a cytoplasmic regulatory molecule. Cell 86, 233-242.

366

YOJI SHIMIZU d al

Kolanus, W., and Seed, B. (1997).Integrins and inside-out signal transduction: Converging signals from PKC and PIPS. Cum w i n . Cell Biol. 9, 725-731. Komoriya, A,, Green, L. J., Mervic, M., Yamada, S. S., Yamada, K. M., and Humphries, M. J. (1991).The minimal essential sequence for a major cell type-specificadhesion site (CS1)within the alternativelyspliced Type I11 connecting segment domain of fibronectin is leucine-aspartic acid-valine. 1.Biol. Chem. 266, 15075-15079. Kondo, S., Kono, T., Brown, W. R., Pastore, S., McKenzie, R. C., and Sauder, D. N. (1994). Lymphocyte function-associatedantigen-1 is required for maximum elicitation of allergic contact dermatitis. Br. 1.Dermutol. 131, 354-359. Koopman, G., Keehnen, R. M. J., Lindhout, E., Newman, W., Shimizu, Y., VanSeventer, G. A., decroot, C., and Pals, S. T. (1994).Adhesion through the LFA-1 (CDllaICD18)pathways prevents apoptosis of ICAM-1 (CD54) and VLA-4(CD49d)-VCAM-l(CD106) germinal center B cells. 1.lmmunol. 152, 3760-3767. Korenaga, R., Ando, J., Kosaki, Isshiki, M., Takada, Y.,and Kamiya, A. (1997). Negative transcriptional regulation of VCAM-1 gene by fluid shear stress in murine endothelial cells. Am. J. Physiol. 273, C1506-C1515. Kraehenbuhl, J. P., and Neutra, M. R. (1992). Molecular and cellular basis of immune protection of mucosal surfaces. Physiol. Rev. 72, 853-879. Kretsinger, R. H. (1980).Structure and evolution of calcium-modulatedproteins. CRC Crit. Rev. Biochem. 8, 119-174. Kucik, D. F., Dustin, M. L., Miller, J. M., and Brown, E. J. (1996). Adhesion-activating phorbol ester increases the mobility of leukocyte integrin LFA-1 in cultured lymphocytes. 1. Clin. Invest. 97, 2139-2144. Kuijpers, K. C., Kuijpers, T. W., Zeijlemaker, W. P., Lucas, C. J., Van Lier, R. A., and Miedema, F. (1990). Analysis of the role of leukocyte function-associated antigen-1 in activation of human influenza virus-specific T cell clones. 1.lmmunol. 144,3281-3287. Kuijpers, T. W., Mul, E. P. J., Blom, M., Kovach, N. L., Gaeta, F. C. A., Tollefson, V., Elices, M. J., and Harlan, J. M. (1993). Freezing adhesion molecules in a state of highavidity binding blocks eosinophil migration.]. Exp. Med. 178,279-284. Kuijpers, T. W., Van Lier, R. A., Hamann, D., de Boer, M., Thung, L. Y.,Weening, R. S., Verhoeven, A. J., and Roos, D. (1997). Leukocyte adhesion deficiency type 1 (LAD-1)/ variant. A novel immunodeficiencysyndrome characterized by dysfunctional,82 integrins. 1.Clin. Invest. 100, 1725-1733. Kunicki, T. J., Pidard, D., Rosa, J., and Nurden, A. T. (1981). The Formation of Ca++dependent complexes of platelet membrane glycoproteins IIb and IIIa in solution as determined by crossed immunoelectrophoresis. Blood 58, 268-277. Kushnir, N., Liu, L., and MacPherson, G. G. (1998). Dendritic cells and resting B cells form clusters in uitro and in uivo: T cell independence, partial LFA-1 dependence, and regulation by cross-linkingsurface molecules.1.Immunol. 160, 1774-1781. Kwee, L., Baldwin, S., Stewart, C. S., Buck, C. A., and Labow, M. A. (1995). Defective development of embryonic and extraembryoniccirculatory systems in vascular cell adhesion molecule (VCAM-1) deficient mice. Development 121, 489-503. LaFlamme, S. E., Homan, S. M., Bodeau, A. L., and Mastrangelo, A. M. (1997). Integrin cytoplasmic domains as connectors to the cell's signal transduction apparatus. Matrix Biol. 16, 153-163. Lam, S. C., Plow, E. F., Smith, M. A., Andrieux, A., Ryckwaert, J.-J., Marguerie, G. A., and Ginsberg, M. H. (1987).Evidence that Arginyl-Glycyl-Aspartatepeptides and fibrinogen gamma chain peptides share a common binding site on platelets. J. Bwl. Chem. 262,947-950.

INTEGRINS IN THE IMMUNE SYSTEM

367

Lam, S. C., Plow, E. F., and Ginsberg, M. H. (1989). Platelet membrane glycoprotein IIb heacy chain forms a complex with glycoprotein IIa that binds Arg-Gly-Asp peptides. Blood 73, 1513-1518. Lam, S. C. T. (1992). Isolation and characterization of a chymotryptic fragment of platelet glycoprotein IIb-IIIa retaining arg-gly-asp binding activity. J. Biol. Chem. 267, 56495655. Lannes-Vieira, J., Chammas, R., Villa-Verde, D. M., Vannier-dos-Santos, M. A,, MelloCoelho, V., de Souza, S. J., Brentani, R. R., and Savino, w. (1993). Extracellular matrix components of the mouse thymic microenvironment. 111. Thymic epithelial cells express the VLA6 complex that is involved in laminin-mediated interactions with thymocytes. Int. Immunol. 5, 1421-1430. Larson, R. S., Corbi, A. L., Berman, L., and Springer, T. A. (1989). Primary structure of the leukocyte function-associated molecule-1 LY subunit: An integrin with an embedded domain defining a protein superfamily.J. Cell Biol. 108, 703-712. Lawrence, M. B., Berg, E. L., Butcher, E. C., and Springer, T. A. (1995). Rolling of lymphocytes and neutrophils on peripheral node addressin and subsequent arrest on ICAM-1 in shear flow. Eur. J. Immunol. 25, 1025-1031. Lawrence, M. B., and Springer, T. A. (1991). Leukocytes roll on selectin at physiologic flow rates: Distinction from and prerequisite for adhesion through integrins. Cell 65,859-873. Le Deist, F., Blanche, S., Keable, H., Gaud, C., Pham, H., Descamp-Latscha, B., Wahn, V., Griscelli, C., and Fischer, A. (1989). Successful HLA nonidentical bone marrow transplantation in three patients with the leukocyte adhesion deficiency. Blood 74, 512-516. Leahy, D. J., Hendrickson, W. A,, Aukhil, I., and Erickson, H. P. (1992). Structure of a fibronectin type I11 domain from tenascin phased by MAD analysisof the selenomethionyl protein. Science 258, 987-991. Leahy, D. J., Aukhil, I., and Erickson, H. P. (1996). 2.0 A crystal structure of a four-domain segment of human fibronectin encompassing the RGD loop and synergy region. CeZZ 84, 155-164. Leahy, D. J. (1997). Implications of atomic-resolution structures for cell adhesion. Annu. Rev. Cell Dev. Biol. 13, 363-393. Leca, G., Mansur, S. E., and Bensussan, A. (1995). Expression of VCAM-1 (CD106) by a subset of TCR-gd-bearing lymphocyte clones. Involvement of a metalloprotease in the specific hydrolytic release of the soluble isoform. J. Immunol. 154, 1069-1077. Lee, J., Rieu, P., Amaout, M. A., and Liddington, R. (1995). Crystal structure of the A domain from the a! subunit of integrin CR3 (CDllb/CDl8). Cell 80, 631-638. Lefrancois, L., and Goodman, T. (1989). In vivo modulation of cytolytic activity and Thy1expressionin TCR-gamma delta intraepithelial lymphocytes.Science 243,1716-1718. Lefrancois, L., and Olson, S. (1994). A novel pathway of thymus-directed T lymphocyte maturation. J. Immunol. 153,987-995. Leppert, D., Waubant, E., Galardy, R., Bunnett, N . W., and Hauser, S. L. (1995). T cell gelatinases mediate basement membrane transmigration in uitro.J. Immund. 154,43794389. Levesque, J. P., Leavesley, D. I., Niutta, S., Vadas, M., and Simmons, P. J. (1995). Cytokines increase human hemopoietic cell adhesiveness by activation of very late antigen (VLA)4 and VLA-5 integrins. 1.Exp. Med. 181, 1805-1815. Li, R., Nortamo, P., Kantor, C., Kovanen, P., Timonen, T., and Gahmberg, C. G. (1993). A leukocyte integrin binding peptide from intercellular adhesion molecule-2 stimulates T cell adhesion and natural killer cell activity. J . Biol. Chem. 268,21474-21477.

+

368

YOJI SHIMIZU et al

Li, R., Xie, Y., Kantor, C., Koistinen, V., Altieri, D. C., Nortamo, P., and Gahmberg, C. G. (1995). A peptide derived from the intercellular adhesion molecule-2 regulates the avidity of the leukocyte integrin CDllbICD18 and CDllc/CDl8.]. Cell B i d . 129,11431153. Liesveld, J. L., Winslow, J. M., Frediani, K. E., Ryan, D. H., and Abboud, C. N. (1993). Expression of integrins and examination of their adhesive function in normal and leukemic hematopoietic cells. Blood 81, 112-121. Lin, E. C. K., Ratnikov, B. I., Tsai, P. M., Carron, C. P., Myers, D. M., Barbas, C. F., and Smith, J. W. (1997a). Identification of a region in the integrin p3 subunit that confers ligand binding. 1.Biol. Chem. 272,23912-23920. Lin, E. C. K., Ratnikov, B. I., Tsai, P. M., Gonzalez, E. R., McDonald, S., Pelletier, A. J., and Smith, J. W. (199713). Evidence that the integrin p3 and p5 subunits contain a metal ion-dependent adhesion site-like motif but lack an I domain.]. Biol. Chem. 272,1423614243. Lin, T., Matsuzaki, G., Kenai, H., Nakamura, T., and Nomoto, K. (1993).Thymus influences the development of extrathymically derived intestinal intraepithelial lymphocytes. Eur. 1.Immunol. 23, 1968-1974. Lobb, R. R., Antognetti, G., Pepinsky, R. B., Burkly, L. C., Leone, D. R., and Whitty, A. (1995).A direct binding assay for the vascular cell adhesion molecule-1 (vcaml) interaction with a 4 integrins. Cell Adhes. Commun. 3, 385-397. Loftus, J. C., OToole, T. E., Plow, E. F., Glass, A,, Frelinger, A. L., 111, and Ginsberg, M. H. (1990). A p3integrin mutation abolishes ligand binding and alters divalent cationdependent conformation. Science 249,915-918. Loftus, J. C., Smith, J. W., and Ginsberg, M. H. (1994). Integrin-mediated cell adhesion: The extracellular face. 1.Biol. Chem. 269, 25235-25238. Loftus, J. C., Halloran, C. E., Ginsberg, M. H.. Feigen, L. P., Zablocki, J. A., and Smith, J. W. (1996).The amino-terminal one-third of a l l b defines the ligand recognition specificity of integrin all&'. I. Biol. Chem. 271,2033-2039. Lollo, B. A,, Chan, K. W., Hanson, E. M., May, V. T., and Brian, A. A. (1993). Direct evidence for two affinity states for lymphocyte function-associated antigen 1 on activated T cells. I. B i d . Chem. 268, 21693-21700. Lotz, M. M., Burdsal, C. A., Erickson, H. P., and McClay, D. R. (1989). Cell adhesion to fibronectin and tenascin: Quantitative measurements of initial binding and subsequent strengthening response. 1. Cell Biol. 109, 1795-1805. Lu, C., Oxvig, C., and Springer, T. A. (1998). The structure of the &propeller domain and C-terminal region of the integrin a M subunit. Dependence on p subunit association and prediction of domains. 1. B i d . Chem. 273, 15138-15147. Mackay, C. R., Marston, W. L., and Dudler, L. (1990). Naive and memory T cells show distinct pathways of lymphocyte recirculation. /. Exp. Med. 171,801-817. Mackay, C. R., Andrew, D. P., Briskin, M., Ringler, D. J., and Butcher, E. C. (1996). Phenotype, and migration properties of three major subsets of tissue homing T cells in sheep. Eur. I. lmmunol. 26, 2433-2439. Maeurer, M. J., Martin, D., Walter, W., Liu, K., Zitvogel, L., Halusczcak, K., Rabinowich, H., Duquesnoy, R., Storkus, W., and Lotze, M. T. (1996). Human intestinal V61+ lymphocytes recognize tumor cells of epithelial origin. 1.Exp. Med. 183, 1681-1696. Main, A. L., Harvey, T. S., Baron, M., Boyd, J., and Campbell, I. D. (1992). The threedimensional structure of the tenth type I11 module of fibronectin: An insight into RGDmediated interactions. Cell 71, 671-678. Manabe, A., Murti, K. G., Coustan-Smith, E., Kumagai, M., Behm, F. G., Raimondi, S. C., and Campana, D. (1994). Adhesion-dependent survival of normal and leukemic human B lymphoblasts on bone marrow stromal cells. BZood 83, 758-766.

INTEGRINS IN THE IMMUNE SYSTEM

369

Manie, S. N., Astier, A,, Wang, D., Phifer, J. S., Chen, J., Lazarovits, A. I., Morimoto, C., and Freedman, A. S. (1996). Stimulation of tyrosine phosphorylation after ligation of p7 and pl integrins on human B cells. Blood 87, 1855-1861. Manie, S. N., Astier, A., Haghayeghi, N., Canty, T., Druker, B. J., Hirai, H., and Freedman, A. S. (1997a). Regulation of integrin-mediated p130(Cas) tyrosine phosphorylation in human B cells. A role for p59(Fyn) and SHP2. J. Biol. Chem. 272, 15636-15641. Manie, S. N., Beck, A. R., Astier, A,, Law, S. F., Canty, T., Hirai, H., Druker, B. J., Avraham, H., Haghayeghi, N., Sattler, M., Salgia, R., Griffin, J. D., Golemis, E. A,, and Freedman, A. S. (1997b). Involvement of pl3O(Cas) and pl05(HEFl), a novel Cas-like docking protein, in a cytoskeleton-dependent signaling pathway initiated by ligation of integrin or antigen receptor on human B cells. J. Biol. Chem. 272, 4230-4236. Manie, S. N., Sattler, M., Astier, A,, Phifer, J. S., Canty, T., Morimoto, C., Druker, B. J., Salgia, R., Griffin, J. D., and Freedman, A. S. (1997~). Tyrosine phosphorylation of the product of the c-cbl protooncogene is [corrected] induced after integrin stimulation. Exp. Hemnutol. 25, 45-50. [Erratum, Exp. Hematol. 25(2), 177.1 Marcantonio, E. E., and David, F. S. (1997). Integrin receptor signaling: the propagation of an a-helix model. Matrix Biol. 16, 179-184. Marlin, S. D., and Springer, T. A. (1987).Purified intercellular adhesion molecule-1 (ICAM1)is a ligand for lymphocyte function-associated antigen 1 (LFA-1). Cell 51, 813-819. May, M. J., Entwistle, G., Humphries, M. J., and Ager, A. (1993).VCAM-1 is a CS1 peptideinhibitable adhesion molecule expressed by lymph node high endothelium. J. CeZZ Sci. 106, 109-119. McKay, B. S., Annis, D. S., Honda, S., Christie, D., and Kunich, T. J. (1996). Molecular requirements for assembly and function of a minimized human integrin aIIbp3. J. Biol. Chem. 271,30544-30547. Mebius, R. E., Skeeter, P. R., Michie, S., Butcher, E. C., and Weissman, I. L. (1996). A developmental switch in lymphocyte homing receptor and endothelial vascular addressin expression regulates lymphocyte homing and permits CD4+ CD3- cells to colonize lymph nodes. Proc. Nutl. Acud. Sci. U.S.A. 93, 11019-11024. Michishita, M., Videm, V., and Arnaout, M. A. (1993). A novel divalent cation-binding site in the A domain of the Pa integrin CR3 (CdllbKD18) is essential for ligand binding. Cell 72, 857-867. Miyake, K., Weissman, I. L., Greenberger, J. S., and Kincade, P. W. (1991). Evidence for a role of the integrin VLA-4 in lympho-hemopoiesis.J. Exp. Med. 173, 599-607. Miyamoto, S., Teramoto, H., Coso, 0. A,, Gutkind, J. S.,Burbelo, P. D., Akiyama, S. K., and Yamada, K. M. (1995). Integrin function: Molecular hierarchies of cytoskeletal and signaling molecules. J. CeZl Biol. 131, 791-805. Mobley, J. L., Reynolds, P. J., and Shimizu, Y. (1993). Regulatory mechanisms underlying T cell integrin receptor function. Semin. lmmunol. 5, 227-236. Mojcik, C. F., Salomon, D. R., Chang, A. C., and Shevach, E. M. (1995). Differential expression of integrins on human thymocyte subpopulations. Blood 86, 4206-4217. Morimoto, C., Iwata, S., and Tachibana, K. (1998). VM-4-mediated signaling. Cum. Top. Microbiol. Imnwnol. 231, 1-22. Mould, A. P., Askari, J. A,, Aota, S., Yamada, K. M., Irie, A,, Takada, Y., Mardon, H. J., and Humphries, M. J. (1997). Defining the topology of integrin a5pl-fibronectin interactions using inhibitory anti-a5 and anti-pl monoclonal antibodies. Evidence that the synergy sequence of fibronectin is recognized by the amino-terminal repeats of the a5 subunit. 1. Biol. Chem. 272, 17283-17292. Muchowksi, P. J., Zhang, L., Chang, E. R., Soule, H. R., Plow, E. F., and Moyle, M. (1994). Functional interaction between the integrin antagonist neutrophil inhibitory factor and the I domain of CDllbICD18. J. Biol. Chem. 269, 26419-26423.

370

YOJI SHIMIZU et of.

Nagel, T., Resnick, W. J,, Atkinson, W. J., Dewey, C. F., and Gimbrone, M. A. J, (1994). Shear stress selectively upregulates intercellularadhesion molecule-1expression in culture human vascular endothelial cells. 1.Clin. Invest. 94, 885-891. Nagel, W., Zeitlmann, L., Schilcher, P., Geiger, C., Kolanus, J., and Kolanus, W. (1998). Phosphoinositide 3-OH kinase activates the p2 integrin adhesion pathway and induces membrane recruitment of cytohesin-1. J. Biol. Chem. 273,14853-14861. Naik, U. P., and Parise, L. V. (1997).Structure and function of platelet aIIbp3. C u r . @in. H W t o l . 4, 317-322. Nakache, M., Berg, E. L., Streeter, P. R., and Butcher, E. C. (1989).The mucosal vascular addressin is a tissue-specific endothelial cell adhesion molecule for circulating lymphocytes. Nature 337, 179-181. Nakajima, S., Roswit, W. T., Look, D. C., and Holtzman, M. J. (1995). A hierarchy for integrin expression and adhesiveness among T cell subsets that is linked to TCR gene usage and emphasizes V delta 1+gamma delta T cell adherence and tissue retention. 1.Immunol. 155,1117-1131. Nathan, C., Srimal, S., Farber, C., Sanchez, E., Kabbash, L., Asch, A., Gailit, J., and Wright, S. D. (1989).Cytokine-induced respiratory burst of human neutrophils: Dependence on extracellular matrix proteins and CDlUCD18 integrins. J. Cell Biol. 109, 1341-1349. Needham, L. A., Van Dijk, S., Pigott, R., Edwards, R. M., Shepherd, M., Hemingway, I., Jack, L., and Clements, J. M. (1994). Activation dependent and independent VLA-4 binding sites on vascular cell adhesion molecule-1. Cell Adhes. Commun. 2, 87-99. Nermut, M. V., Green, N. M., Eason, P., Yamada, S. S., and Yamada, K. M. (1988).Electron microscopy and structural model of human fibronectin receptor, EMBO J. 7,4093-4099. Newham, P., Craig, S. E., Seddon, G. N., Schofield, N. R., Rees, A., Edwards, R. M., Jones, E. Y., and Humphries, M. J. (1997). a 4 integrin binding interfaces on VCAM-1 and MAdCAM-1. Integrin binding footprints identify accessory binding sites that play a role in integrin specificity. J. Biol. Chem. 272, 19429-19440. Newton, R. A., Thiel, M., and H o g , N. (1997). Signaling mechanisms and the activation of leukocyte integrins. 1.Leukoc. Biol. 61,422-426. Nickoloff, B. J., Mitra, R. S., Green, J., Zheng, X. G., Shimizu,Y.,Thompson, C., and Turka, L. A. (1993). Accessory cell function of keratinocytes for superantigens. Dependence on lymphocyte function-associated antigen-l/intercellular adhesion molecule-1 interaction. 1.lmmunol. 150,2148-2159. Nowlin, D. M., Gorcsan, F., Moscinski, M., Chiang, S. L., Lobl, T. J., and Cardarelli, P. M. (1993).A novel cyclic pentapeptide inhibits a4pl and a5pl integrin-mediated cell adhesion. J. Biol. Chem. 268, 20352-20359. ODoherty, U., Peng, M., Gezelter, S., Swiggard, W. J., Betjes, M., Bhardwaj, N., and Steinman, R. M. (1994).Human blood containstwo subsets of dendriticcells, one immunologically mature and the other immature. lmmunology 82,487-493. O’Toole, T. E. (1997). Integrin signaling: Building connections beyond the focal contact? Matrix Biol. 16, 165-171. OToole, T. E., Loftus, J. C., Plow, E. F., Glass, A., Harper, J. R., and Ginsberg, M. H. (1989). Efficient surface expression of platelet GPIIb-IIIa requires both subunits. Blood 74, 14-18. Obara, M., Kang, M. S., and Yamada, K. M. (1988). Site-directed mutagenesis of the cell-binding domain of human fibronectin: Separable, synergistic sites mediate adhesive function. Cell 53, 649-657. Odum, N., Hofmann, B., Morling, N., Platz, P., Ryder, L. P., Tvede, N., Geisler, C., and Svejgaard, A. (1988). Differences between primed allogeneic T-cell responses and the primary mixed leucocyte reaction. Primed T cells become independent of the blocking

INTEGRINS IN THE IMMUNE SYSTEM

371

effects of monoclonal antibodies against IL-1 P and the CD5, C D l l a (LFA-l), and C D l l c (p 150, 95) molecules. Scand. J. lmmunol. 27,405-411. Obh, M., and Takei, F. (1994).Interferon-y and phorbol myristate acetate-response elements involved in intercellular adhesion molecule-1 mRNA stabilization. 1. Biol. Chem. 269, 30117-30120. Ohh, M., and Takei, F. (1996). New insight into the regulation of ICAM-1 gene expression. Leukemia Lymphoma 20, 223-228. Ohtsuka, A., Ando, J., Korenaga, R., Kamiya, A., Toyama-Sorimachi, N., and Miyasaka, M. (1993). The effect of flow on the expression of vascular cell adhesion molecule-1 by cultured mouse endothelial cells. Biochem. Biophys. Res. Commun. 193,303-310. Osborn, L., Vassallo, C., and Benjamin, C. D. (1992). Activated endothelium binds lymphocytes through a novel binding site in the alternately spliced domain of vascular cell adhesion molecule-1. J. Exp. Med. 176, 99-107. Osborn, L., Vassallo, C., Browning, B. G., Tizard, R., Haskard, D. O., Benjamin, C. D., Dougas, I., and Kirchhausen, T. (1994). Arrangement of domains, and amino acid residues required for binding of vascular cell adhesion molecule-1 to its counter-receptor VLA4((r4/31).J. Cell Biol. 124, 601-608. Oxvig, C., and Springer, T. A. (1998). Experimental support for a P-propeller domain in integrin a-subunits and a calcium binding site on its lower surface. Proc. Natl. Acad. Sci. U.S.A. 95,4870-4875. Palecek, S. P., Schmidt, C. E., Lauffenburger, D. A,, and Horwitz, A. F. (1996). Integrin dynamics on the tail region of migrating fibroblasts.1.Cell Sci. 109, 941-952. Palecek, S. P., Loftus, J. C., Ginsberg, M. H., Honvitz, A. F., and Lauffenburger, D. A. (1997).Integrin-ligand binding properties govern cell migration speed through cellsubstratum adhesiveness. Nature 385, 537-540. Panettieri, R. A. J., Lazaar, A. L., Pure, E., and Albelda, S. M. (1995). Activation of CAMPdependent pathways in human ainvay smooth muscle cells inhibits TNF-a-induced ICAM1 and VCAM-1 expression and T-lymphocyte adhesion. J. Immunol. 154, 2358-2365. Papayannopoulou, T., Craddock, C., Nakamoto, B., Priestley, G. V., and Wolf, N. S. (1995). The VLAWCAM-1 adhesion pathway defines contrasting mechanisms of lodgement of transplanted murine hemopoietic progenitors between bone marrow and spleen. Proc. Natl. Acad. Sci. U.S.A. 92, 9647-9651. Papayannopoulou, T., and Nakamoto, B. (1993). Peripheralization of hemopoietic progenitors in primates treated with anti-VLA4 integrin. Proc. Natl. Acad. Sci. U.S.A.90,93749378. Parise, L. V., Helgerson, S. L., Steiner, B., Nannizzi, L., and Phillips, D. R. (1987). Synthetic peptides derived from fibrinogen and fibronectin change the conformation of purified platelet glycoprotein IIb-IIIa. J. Biol. Chem.262, 12597-12602. Parise, L. V., and Phillips, D. R. (1985). Platelet membrane glycoprotein IIb-IIIa complex incorporated into phospholipid vesicles. Preparation and morphology. 1. Biol. Chem. 260, 1750-1756. Parise, L. V., and Phillips, D. R. (1986). Fibronectin-binding properties of the purified platelet glycoprotein IIb-IIIa complex. J. Biol. Chem. 261, 14011-14017. Patey, N., Vazeux, R., Canioni, D., Potter, T., Gallatin, W. M., and Browse, N. (1996). Intercellular adhesion molecule-3 on endothelial cells. Expression in tumors but not in idammatory responses. Am. 1.Pathol. 148, 465-472. Perkins, S . J., Smith, K. F., Williams, S . C., Haris, P. I., Chapman, D., and Sim, R. B. (1994). The secondary structure of the von Willebrand factor w e A domain in factor B of human complement by Fourier transform infrared spectroscopy. Its occurrence in

372

YOJI SHIMIZU et ~l

collagen types VI,VII, X I , andXV, the integrins and other proteins by averaged structure predictions. J. Mol. Biol. 238, 104-119. Peschon, J. J., Morrissey, P. J., Grabstein, K. H., Ramsdell, F. J., Maraskovsky, E., Gliniak, B. C., Park, L. S., Ziegler, S. F., Williams, D. E., and Ware, C. B. (1994). Early lymphocyte expansion is severely impaired in interleukin 7 receptor-deficient mice. J. Exp. Med. 180, 1955-1960. Phillips, D. R., and Agin, P. P. (1977). Platelet plasma membrane glycoproteins. Evidence for the presence of non-equivalent disulfide bonds using non-reduced-reduced twodimensional gel electrophoresis. J. Biol. Chem. 252, 2121-2126. Pidard, D., Didly, D., Kunicki, T. J,, and Nurden, A. T. (1986). Temperature-dependent effects of EDTA on the membrane glycoprotein IIb-IIIa complex and platelet aggregability. Blood 67,604-611. Pierschbacher, M. D., and Ruoslahti, E. (1984a). Cell attachment activity of fibronectin can be duplicated by small synthetic fragments of the molecule. Nature 309, 30-33. Pierschbacher, M. D., and Ruoslahti, E. (198413). Variants of the cell recognition site of fibronectin that retain attachment-promoting activity. Proc. Natl. Acud. Sci. U.S.A. 81,5985-5988. Pigott, R., Dillon, L. P., Hemingway, I., and Gearing, A. J. H. (1992). Soluble forms of Eselectin, ICAM-l, and VCAM-1 are present in the supernatant of cytokine activated culture endothelial cells. Biochem. Biophys. Res. Commun. 187,584-589. Pober, J . S., Slowik, M. R., DeLuca, L. G., and Ritchies, A. J. (1993). Elevated cyclic AMP inhibits endothelial cell synthesis and expression of TNF-induced endothelial leukocyte adhesion molecule-1 and vascular cell adhesion molecule-1, but not intercellular adhesion molecule-1. J. Immunol. 150, 5114-5123. Postigo, A. A., Garcia-Vicuna, R., Dim-Gonzalez, F., Arroyo, A. G., DeLandazuri, M. O., Chi-Rosso, G., Lobb, R. R., Laffon, A,, and Sanchez-Madrid, F. (1992).Increased binding of synovial T lymphocytes from rheumatoid arthritis to endothelial-leukocyte adhesion molecule-1 (ELAM-1) and vascular cell adhesion molecule-l(VCAM-1). 1.Cltn. Invest. 89, 1445-1452. Puzon-McLaughlin, W., and Takada, Y. (1996). Critical residues for ligand binding in an I domain-like structure of the integrin pl subunit. J. Biol. Chem. 271, 20438-20443. Pytela, R., Pierschbacher, M. D., and Ruoslahti, E. (1985). Identification and isolation of a 140 kd cell surface glycoprotein with properties expected of a fibronectin receptor. Cell 40, 191-198. Qu, A., and Leahy, D. J. (1995). Crystal structure of the I-domain from the cdlldcdl8 (Ifa-1, 0192) integrin. Proc. Natl. Acad. Sci. U.S.A.92, 10277-10281. Randi, A. M., and Hogg, N. (1994). I domain of 0 2 integrin lymphocyte function-associated antigen-1 contains a binding site for ligand intercellular adhesion molecule-1. J. Bid. Chem. 269, 12395-12398. Relman, D., Tuomanen, E., Falkow, S . , Golenbock, D. T., Saukkonen, K., and Wright, S. D. (1990). Recognition of a bacterial adhesin by an integrin: Macrophage CR3 (aMP2, CDllbICD18) binds filamentous hemagglutinin of Bordetellu pertussis. Cell 61, 13751382. Roberts, K. and Kilshaw, P. J. (1993). The mucosal T cell integrin aM290 P7 recognizes a ligand on mucosal epithelial cell lines. Eur. J. Immunol. 23, 1630-1635. Roldan, E., Garcia-Pardo, A., and Brieva, J. A. (1992). VLA-4-fibronectin interaction is required for the terminal differentiation of human bone marrow cells capable of spontaneous and high rate immunoglobulin secretion. J. Exp. Med. 175, 1739-1747. Romanic, A. M., Graesser, D., Baron, J. L., Visintin, I., Janeway, C. A,, Jr,, and Madri, J. A. (1997). T cell adhesion to endothelial cells and extracellular matrix is modulated upon transendothelial cell migration. Lab. Invest. 76, 11-23.

INTEGRINS IN THE IMMUNE SYSTEM

373

Romanic, A. M., and Madri, J. A. (1994). The induction of 72-kD gelatinase in T cells upon adhesion to endothelial cells is VCAM-1 dependent. J. Biol. Chem. 125, 1165-1178. Rose, J. R., Williams, M. J., Rott, L. S., Butcher, E. C., and Greenberg, H. B. (1998). Expression of the mucosal homing receptor 4 3 7 correlates with the ability of CD8+ memory T cells to clear rotavirus infection. J. Virol. 72, 726-730. Ross, L., Hassman, F., and Molony, L. (1992). Inhibition of Molt-4-endothelid adherence by synthetic peptides from the sequence of ICAM-1.1. Bid. Chem. 267, 8537-8543. Rothlein, R., Mainolfi, E., Czajkowski, M., and Marlin, S. D. (1991). A form of circulating ICAM-1 in human serum. J. lmmunol. 147,3788-3793. Rott, L. S., Briskin, M. J., Andrew, D. P., Berg, E. L., and Butcher, E. C. (1996). A fundamental subdivision of circulating lymphocytes defined by adhesion to mucosal addressin cell adhesion molecule-1. Comparison with vascular cell adhesion molecule-1 and correlation with /37 integrins and memory differentiation. J. Immunol. 156, 3727-3736. Ruggeri, Z. M., Houghten, R. A., Russell, S. R., and Zimmerman, T. S. (1986). Inhibition of platelet function with synthetic peptides designed to be high-affinity antagonists of fibrinogen binding to platelets. Proc. Natl. Acad. Sci. U.S.A. 83, 5708-5712. Ruiz, P., Wiles, M. V., and Imhof, B. A. (1995).a 6 integrins participate in pro-T cell homing to the thymus. Eur. J. Immunol. 25,2034-2041. Ryan, D. H., Nuccie, B. L., and Abboud, C. N. (1992). Inhibition of human bone marrow lymphoidprogenitor colonies by antibodies to VLA integrins.]. Immunol. 149,3759-3764. Salmi, M. and Jalkanen, S. (1997).How do lymphocytes know where to go: Current concepts and enigmas of lymphocyte homing. Adv. lmmunol. 64, 139-218. Salomon, D. R., Mojcik, C. F., Chang, A. C., Wadsworth, S., Adams, D. H., Coligan, J. E., and Shevach, E. M. (1994). Constitutive activation of integrin a& defines a unique stage of human thymocyte development. J. Exp. Med. 179, 1573-1584. Salomon, D. R., Crisa, L., Mojcik, C. F., Ishii, J. K., Klier, G., and Shevach, E. M. (1997). Vascular cell adhesion molecule-1 is expressed by cortical thymic epithelial cells and mediates thymocyte adhesion. Implications for the function of a 4 p l (VLA4) integrin in T-cell development. Blood 89,2461-2471. Sanchez-Madrid, F., Krensky, A. M., Ware, C. F., Robbins, E., Strominger, J. L., Burakoff, S. J., and Springer, T. A. (1982). Three distinct antigens associated with human Tlymphocyte-mediated cytolysis: LFA-1, LFA-2, and LFA-3. Proc. Natl. Acad. Sci. U.S.A. 79, 7489-7493. Sanders, M. E., Makgoba, M. W., Sharrow, S. O., Stephany. D., Springer, T. A., Young, H. A., and Shaw, S. (1988). Human memory T lymphocytes express increased levels of three cell adhesion molecules (LFA-3, CD2, and LFA-1) and three other molecules (UCHL1, CDw29, and Pgp-1) and have enhanced IFN-gamma production. J. Immunol. 140, 1401-1407. Santoro, S. A. and Lawing, W. J., Jr. (1987). Competition for related but nonidentical binding sites on the glycoprotein IIb-IIIa complex by peptides derived from platelet adhesive proteins. Cell 48, 867-873. Sarnacki, S., Begue, B., Buc, H., Le Deist, F., and Cerf-Bensussan, N. (1992).Enhancement of CD3-induced activation of human intestinal intraepithelial lymphocytes by stimulation of the p7-containing integrin defined by HML-1 monoclonal antibody. Eur. J. Immunol. 22,2887-2892. Sasseville, V. G., Newman, W. A,, Lackner, A. A,, Smith, M. O., Lausen, N. C., Beall, D., and Ringler, D. J. (1992). Elevated vascular cell adhesion molecule-1 in AIDS encephalitis induced by simian immunodeficiency virus. Am. J. Pathol. 141, 1021-1030. Sastry, S. K. and Honvitz, A. F. (1993).Integrin cytoplasmic domains: mediators of cytoskeletal linkages and extra- and intracellular initiated transmembrane signaling. Cum. @in. Cell Bwl. 5, 819-831.

374

YOJI SHIMIZU et al.

Savino, W., Dardenne, M., and Carnaud, C. (1996).The conveyor belt model for intrathymic T-cell migration [letter]. Immunol. Today 17, 97-98. Sawada, M., Nagamine, J., Takeda, K., Utsumi, K., Kosugi, A., Tatsumi, Y., Hamaoka, T., Miyake, K., Nakajima, K., Watanabe, T., Sakakibara, S., and Fujiwara, H. (1992). Expression of VLA-4 on thymocytes. Maturation stage-associated transition and its correlation with their capacity to adhere to thymic stromal cells, J. Immunol. 149, 3517-3524. Scharffetter-Kochanek, K., Lu, H., Norman, K., van Nood, N., Munoz, F., Grabbe, S., McArthur, M., Lorenzo, I., Kaplan, S., Ley, K., Wayne Smith, C., Montgomery, C. A., Rich, S.,and Beaudet, A. L. (1998). Spontaneous skin ulceration and defective T cell function in CD18 null mice. J. Exp. Med. 188, 119-131. Schleimer, R. P., Sterbinsky, S. A., Kaiser, J., Bickel, C. A., Klunk, D. A., Tomioka, K., Newman, W., Luscinskas, F. W., Gimbrone, M. A. J., McIntyre, B. W., and Bochner, B. S. (1992). IL-4 induces adherence of human eosinophils and basophils but not neutrophils to endothelium. Association with expression of VCAM-1. J. Immunol. 148, 10861092. Schmits, R., Kundig, T. M., Baker, D. M., Shumaker, G., Simard, J. J. L., Duncan, G., Wakeham, A,, Shahinian, A., van der Heiden, A,, Bachmann, M. F., Ohashi, P. S., Mak, T. W., and Hickstein, D. D. (1996). LFA-1-deficient mice show normal CTL responses to virus but fail to reject immunogenic tumor. J. Exp. Med. 183, 1415-1426. Schriever, F., Korinth, D., Salahi, A,, Lefterova, P., Schmidt-Wolf, I. G., and Behr, S. I. (1997). Human T-lymphocytes bind germinal centers of human tonsils via integrin a41 VCAM-1 and LFA-l/ICAM-1,2. Eur. J. Immunol. 27, 35-39. Schwartz,M. A., Schaller, M. D., andGinsberg, M. H. (1995).Integrins: Emergingparadigms of signal transduction. Annu. Rev. Cell Deu. Biol. 11,549-599. Schweighoffer,T., Tanaka, Y., Tidswell, M., Erle, D. J., Horgan, K. J.. Luce, G. E., Lazarovits, A. I., Buck, D., and Shaw, S. (1993). Selective expression of integrin a4/37 on a subset of human CD4+ memory T cells with hallmarks of gut-tr0phism.J. Immunol. 151,717-729. Scollay, R., and Godfrey, D. I. (1995). Thymic emigration: Conveyor belts or lucky dips? Immunol. Today 16,268-273; discussion, 273-274. Shattil, S. J., Brass, L. F., Bennett, J. S., and Pandhi, P. (1985). Biochemical and functional consequences of dissociation of the platelet membrane glycoprotein IIb-IIIa complex. Blood 66,92-98. Shattil, S . J., Kashiwagi, H., and Pampori, N. (1998). Integrin signaling: The platelet paradigm. Blood 91,2645-2657. Shattil, S . J., and Ginsberg, M. H. (1997). Integrin signaling in vascular biology. J , Clin. Inuest. 100, 1-5. Shaw, S., Luce, G. E., Quinones, R., Gress, R. E., Springer, T. A,, and Sanders, M. E. (1986). Two antigen-independent adhesion pathways used by human cytotoxic T-cell clones. Nature 323, 262-264. Shaw, S. K., Cepek, K. L., Murphy, E. A., Russell, G. J., Brenner, M. B., and Parker, C. M. (1994). Molecular cloning of the human mucosal lymphocyte integrin aEsubunit. Unusual structure and restricted RNA distribution. J. Biol. Chem. 269, 6016-6025. Shelton-Inloes, B. B., Titani, K., and Sadler, J, E. (1986). cDNA sequences for human von Willebrand factor reveal five types of repeated domains and five possible protein sequence polymorphisms. Biochemistry 25,3164-3171. Shier, P., Otulakowski, G., Ngo, K., Panakos, J., Chourmouzis, E., Christjansen, L., Lau, C. Y., and Fung-Leung, W. P. (1996). Impaired immune responses toward alloantigens and tumor cells but normal thymic selection in mice deficient in the 0 2 integrin leukocyte function-associated antigen-1. J. Immunol. 157, 5375-5386.

INTEGRINS IN THE IMMUNE SYSTEM

375

Shimizu, Y., Seventer, G. A. V., Horgan, K. J., and Shaw, S. (1990a). Costimulation of proliferative responses of resting CD4+ T cells by the interaction of VLA4 and VLA-5 with fibronectin or VLA-6 with laminin. J. Immunol. 145, 59-67. Shimizu, Y., VanSeventer, G. A., Horgan, K. J., and Shaw, S. (1990b). Regulated expression and binding of three VLA (Bl) integrin receptors on T cells. Nature 345, 250-253. Shimizu, Y., Newman, W., Tanaka, Y., and Shaw, S. (1992). Lymphocyte interactions with endothelial cells. Immunol. Today 13, 106-112. Shimizu, Y., Mobley, J. L., Finkelstein, L. D., and Chan, A. S. H. (1995). A role for phosphatidylinositol3-kinasein the regulation of Pl integrin activity by the CD2 antigen. 1.Cell Biol. 131, 1867-1880. Shimizu, Y., and Hunt, S. W., 3rd (1996). Regulating integrin-mediated adhesion: one more function for PI 3-!&ase? Immunol. Today 17,565-573. S h y , J. Y., and Chien, S. (1997). Role of integrins in cellular responses to mechanical stress and adhesion. Curr. Opin. Cell Biol. 9, 707-713. Sikorski, M. A., Staunton, D. E., and Mier, J. W. (1996). L-Selectin crosslinking induces integrin-dependent adhesion: Evidence for a signaling pathway involving PTK but not PKC. Cell Adhes. Commun. 4,355-367. Sim, G. K. (1995). Intraepithelial lymphocytes and the immune system. Adu. Immunol. 58,297-343. Simmons, P. J., Masinovsky, B., Longenecker, B. M., Berenson, R., Torok-Storb, B., and Gallatin, W. M. (1992). Vascular cell adhesion molecule-1 expressed by bone marrow stromal cells mediates the binding of hematopoietic progenitor cells. Blood 80,388-395. Sligh, J. E., Ballantyne, C. M., Rich, S. S., Hawkins, H. K., Smith, C. W., Bradley, A., and Beaudet, A. L. (1993).Inflammatory and immune responses are impaired in mice deficient in intracellular adhesion molecule 1. Proc. Natl. Acad. Sci. U.S.A. 90, 8529-8533. Smith, J. W. (1997). AUostery and proteolysis: Two novel modes of regulating integrin function. Matrix Biol. 16, 173-178. Smith, J. W., and Cheresh, D. A. (1988). The Arg-Gly-Asp binding domain of the vitronectin receptor. Photoaffinity cross-linking implicates amino acid residues 61-203 of the P subunit. I. Biol. Chem. 263, 18726-18731. Smith, J. W., and Cheresh, D. A. (1990). Integrin (a&)-ligand interaction. Identification of a heterodimeric RGD binding site on the vitronectin receptor.]. Biol. Chem. 265,21682172. Soilu-Hanninen, M., Roytta, M., Salmi, A,, and Salonen, R. (1997). Therapy with antibody against leukocyte integrin VLA-4 (CD49d) is effective and safe in virus-facilitated experimental allergic encephalomyelitis.J. Neuroimmunol. 72,95-105. Springer, T. A. (1990). Adhesion receptors of the immune system. Nature 346,425-434. Springer, T. A. (1994).Traffic signals for lymphocyte recirculation and leukocyte emigration: The multistep paradigm. Cell 76,301-314. Springer, T. A. (1995). Traffic signals on endothelium for lymphocyte recirculation and leukocyte emigration. Annu. Rev. Physiol. 57, 827-872. Springer, T. A. (1997). Folding of the N-terminal, ligand-binding region of integrin asubunits into a P-propeller domain. Proc. Natl. Acad. Sci. U.S.A. 94, 65-72. St-Pierre, Y., Hugo, P., Legault, D., Tremblay, P., and Potworowski, E. F. (1996). Modulation of integrin-mediated intercellular adhesion during the interaction of thymocytes with stromal cells expressing VLA-4 and LFA-1 ligands. Eur. J. Immunol. 26, 2050-2055. Staatz, W. D., Fok, K. F., Zutter, M. M., Adams, S. P., Rodriguez, B. A., and Santoro, S. A. (1991). Identification of a tetrapeptide recognition sequence for the (u-2 P-1 integrin in collagen. J. Biol. Chem. 266, 7363-7367.

376

YOJI SHIMIZU et a1

Stanley, P., Bates, P. A,, Harvey, J,, Bennett, R. I., and Hogg, N. (1994). Integrin LFA-1 a contains an ICAM-1 binding site in domains V and VI. EMBO J. 13, 1790-1798. Starling, G. C., McLellan, A. D., Egner, W., Sorge, R. V., Fawcett, J., Simmons, D. L., and Hart, D. N. (1995). Intercellular adhesion molecule-3 is the predominant co-stimulatory ligand for leukocyte function antigen-1 on human blood dendritic cells. Eur. J. Immunol. 25, 2528. Staunton, D. E., Marlin, S. D., Stratowa, C., Dustin, M. L., and Springer, T. A. (1988). Primary structure of ICAM-1 demonstrates interaction between members of the immunoglobulin and integrin supergene families. Cell 52, 925-933. Staunton, D. E., Dustin, M. L., Erickson, H. P., and Springer, T. A. (1990).The arrangement of immunoglobulin-likedomains of ICAM-1 and the binding sites of LFA-1 and rhinovirus. Cell 61, 243-254. Steeber, D. A., Engel, P., Miller, A. S., Sheetz, M. P., and Tedder, T. F. (1997). Ligation of L-selectin through conserved regions within the lectin domain activates signal transduction pathways and integrin function in human, mouse, and rat leukocytes. J. Immunol. 159, 952-963. Steffen, P. J., Breier, G., Butcher, E. C., Schulz, M., and Engelhardt, B. (1996). ICAM-1, VCAM-1, and MAdCAM-1 are expressed on choroid plexus epithelium but no endothelium and mediates binding of lymphocytes in vitro. Am. J. Puthol. 148, 1819-1838. Steiner, B., Parise, L. V., Leung, B., and Phillips, D. R. (1991). Ca2+-dependentstructural transitions of the platelet glycoprotein IIb-IIIa comp1ex.J. Biol. Chem. 266,14986-14991. Stephens, L. E., Sutherland, A. E., Klimanskaya, I. V., Andrieux, A., Meneses, J., Pedersen, R. A., and Damsky, C. H. (1995). Deletion of pl integrins in mice results in inner cell mass failure and periimplantation lethality. Genes Dev. 9, 1883-1893. Suzuki, S . , and Naitoh, Y. (1990). Amino acid sequence of a novel integrin p4 subunit and primary expression of the mRNA in epithelial cells. EMBO J. 9, 757-763. Swanborg, R. H. (1995). Experimental autoimmune encephalomyelitis in rodents as a model for human demyelinating disease. Clin. Immunol. Immunopathol. 77, 4-13. Szabo, M. C., Butcher, E. C., and McEvoy, L. M. (1997). Specialization of mucosal follicular dendritic cells revealed by mucosal addressin-celladhesion molecule-1disp1ay.J. Immunol. 158,5584-5588. Takada, Y., Ylanne, J., Mandelman, D., Puzon, W., and Ginsberg, M. H. (1992). A point mutation of integrin p1subunit blocks binding of a& to fibronectin and invasin but not recruitment to adhesion plaques. J. Cell Biol. 119, 913-921. Takada, Y., and Hemler, M. E. (1989).The primary structure of the VLA-2kollagen receptor a 2 subunit (platelet GPla): Homology to other integrins and the presence of a possible collagen-binding domain. J. Cell Biol. 109, 397-407. Takagi, J., Kamata, T., Meredith, J., Puzon-McLaughlin,W., andTakada, Y. (1997).Changing ligand specificities of cwpl and cwp3 integrins by swapping a short diverse sequence of the p subunit. J. BioZ. Chem. 272, 19794-19800. Tamura, R. N., Rozzo, C., Starr, L., Chambers, J., Reichardt, L. F., Cooper, H. M., and Quaranta, V. (1990). Epithelial integrin a6P4: Complete primary structure of a 6 and variant forms of p4.J. Cell Biol. 111, 1593-1604. Tan, K., Casasnovas, J. M., Liu, J., Briskin, M. J., Springer, T. A., and Wang, J. (1998). The structure of immunoglobulin superfamily domains 1 and 2 of MAdCAM-1 reveals novel features important for integrin recognition. Structure 6, 793-801. Tanaka, Y.,Adams, D. H., Hubscher, S., Hirano, H., Siebenlist, U., and Shaw, S. (1993). T-cell adhesion induced by proteoglycan-immobilizedcytokine MIP-1 p [see comments]. Nature 361, 79-82.

INTEGRINS IN THE IMMUNE SYSTEM

377

Taub, R., Could, R. J., Garsky, V. M., Ciccarone, T. M., Hoxie, J., Friedman, P. A,, and Shattil, S. J. (1989).A monoclonal antibody against the platelet fibrinogen receptor contains a sequence that mimics a receptor recognition domain in fibrinogen. J. Biol. Chem. 264,259-265. Tedder, T. F., Steeber, D. A,, and Pizcueta, P. (1995). L-Selectin-deficient mice have impaired leukocyte recruitment into inflammatory sites. J. Exp. Med. 181, 2259-2264. Thomhill, M. H., Wellicome, S. M., Mahiouz, D. L., Lanchbury, J. S., Kyan, A. U., and Haskard, D. 0. (1991).Tumor necrosis factor combines with IL-4 of IFN-7 to selectively enhance endothelial cell adhesiveness for T cells. J. Immunol. 146, 592-598. Ticchioni, M., Deckert, M., Bernard, G., Calandra, D., Breittmeyer, J. P., Imbert, V., Peyron, J. F., and Bernard, A. (1995). Comitogenic effects of very late activation antigens on CD3-stimulated human thymocytes. Involvement of various tyrosine kinase pathways. J. Immunol. 154,1207-1215. Titani, K., Kumar, S., Takio, K., Ericsson, L. H., Wade, R. D., Ashida, K., Walsh, K. A,, Chopek, M. W., Sadler, J. E., and Fujikawa, K. (1986). Amino acid sequence of human von Willebrand factor. Biochemistry 25, 3171-3184. Tomiyama, Y., Brojer, E., Ruggeri, Z. M., Shattil, S. J., Smiltneck, J., Gorski, J., Kumar, A,, Kieber-Emmons, T., and Kunicki, T. J. (1992). A molecular model of RGD ligands: Antibody D gene segments that direct specificity for the integrin aIIbP3. J. Biol. Chem. 267, 18085-18092. Tozer, E. C., Liddington, R. C., Sutcliffe, M. J., Smeeton, A. H., and Loftus, J. C. (1996). Ligand binding to integrin is dependent on a MIDAS-like domain in the P3subunit. J. Biol. Chem. 271, 21978-21984. Udagwa, T., Woodside, D. G., and McIntyre, B. W. (1996). a 4 P l (CD49aCD29) integrin costimulation of human T cells enhances transcription factor and cytokine induction in abscence of altered sensitivity to anti-CD3 stimulation. J. Immunol. 157, 1965- 1972. Utsumi, K., Sawada, M., Narumiya, S., Nagamine, J., Sakata, T., Iwagami, S., Kita, Y., Teraoka, H., Hirano, H., Ogata, M., Hamaoka, T., and Fujiwara, H. (1991). Adhesion of immature thymocytes to thymic stromal cells through fibronectin molecules and its signL6cance for the induction of thymocyte differentiation. Proc. Natl. Acud. Sci. U.S.A. 88,5685-5689. van Kooyk, Y., Weder, P., Heije, K., and Figdor, C. G. (1994). Extracellular Ca" modulates leukocyte function-associated antigen-1 cell surface distribution on T lymphocytes and consequently affects cell adhesion. J. Cell Biol. 124, 1061-1070. van Kooyk, Y., Binnerts, M. E., Edwards, C. P., Champe, M., Berman, P. W., Figdor, C. G., and Bodary, S. C. (1996). Critical amino acids in the lymphocyte function-associated antigen-1 I domain mediate intercellular adhesion molecule 3 binding and immune function. J. Exp. Med. 183, 1247-1252. van Kooyk, Y., and Figdor, C. G. (1997). Signalling and adhesive properties of the integrin leucocyte function-associated antigen 1 (LFA-1). Biochem. SOC. Trans. 25,515-520. van Seventer, G. A,, Newman, W., Shimizu, Y., Nutman, T. B., Tanaka, Y., Horgan, K. J., Gopal, T. V., Ennis, E., O'Sullivan, D., and Grey, H. (1991a). Analysis of T cell stimulation by superantigen plus major histocompatibility complex class I1 molecules or by CD3 monoclonal antibody: Costimulation by purified adhesion ligands VCAM-1, ICAM-1, but not ELAM-1.1. Exp. Med. 174, 901-913. van Seventer, G. A,, Newman, W., Shimizu, Y., Nutman, T. B., Tanaka, Y., Horgdn, K. J., Gopal, T. V., Ennis, E., OSullivan, D., Grey, H., et al. (1991b). Analysis of T cell stimulation by superantigen plus major histocompatibilitycomplex class I1 molecules or by CD3 monoclonal antibody: Costimulation by purified adhesion ligands VCAM-1, ICAM-1, but not ELAM-1. J. Exp. Med. 174, 901-913.

378

YOJl SHIMIZU et al.

Verfaillie, C. M., McCarthy, J. B., and McGlave, P. B. (1992). Mechanisms underlying abnormal trafficking of malignant progenitors in chronic myelogenous leukemia. Decreased adhesion to stroma and fibronectin but increased adhesion to the basement membrane components laminin and collagen type IV. 1.Clin. Invest. 90, 1232-1241. Verfaillie, C. M., Hurley, R., Lundell, B. I., Zhao, C., and Bhatia, R. (1997). Integrinmediated regulation of hematopoiesis: Do BCWABL-induced defects in integrin function underlie the abnormal circulation and proliferation of CML progenitors? Ada Haematol. 97,40-52. Venveij, C. L., Diergaarde, P. J., Hart, M., and Pannekoek, H. (1986). Full-length von Willebrand factor (vWF) cDNA encodes a highly repetitive protein considerably larger than the mature vWF subunit. EMBO]. 5, 1838-1847. Viney, J. L., Jones, S., C h u , H. H., Lagrimas, B., Renz, M. E., Presta, L. G., Jackson, D., Hillan, K. J., Lew, S., and Fong, S. (1996). Mucosal addressin cell adhesion molecule-1. A structural and functional analysis demarcates the integrin binding motif. 1.Immunol. 157,2488-2497. von Andrian, U. H., Chambers, J. D., McEvoy, L. M., Bargatze, R. F., Arfors, K. E., and Butcher, E. C. (1991). Two-step model of leukocyte-endothelial cell interaction in inflammation distinct roles for LECAM-1 and the leukocyte p2 integrins in oiuo. Proc. Natl. Acad. Sci. U.S.A.88, 7538-7542. von Freeden-Jeffry, U., Vieira, P., Lucian, L. A., McNeil, T., Burdach, S. E., and Murray, R. (1995). Lymphopenia in interleukin (1L)-7 gene-deleted mice identifies IL-7 as a nonredundant cytokine.1.Exp. Med. 181,1519-1526. Vonderheide, R. H., Tedder, T. F., Springer, T. A., and Staunton, D. E. (1994). Residues within a conserved amino acid motif of domains 1 and 4 of VCAM-1 are required for binding to VLA-4.1. Cell Biol. 125, 215-222. Wagner, N., Lohler, J., Kunkel, E. J., Ley, K., Leung, E., Krissansen, G., Rajewsky, K., and Muller, W. (1996). Critical role for p7 integrins in formation of the gut-associated lymphoid tissue. Nature 382, 366-370. Wang, J., Pepinsky, R. B., Stehle, T., Liu, J., Karpusas, M., Browning, B., and Osborn, L. (1995). The crystal structure of an N-terminal two-domain fragment of vascular cell adhesion molecule 1(VCAM-1);a cyclic peptide based on domain 1C-D loop can inhibit VCAM-l-a4 integrin interaction. Proc. Natl. Acad. Sci. U.S.A.92, 5714-5718. Wang, R., Furihata, K., McFarland, J. G., Friedman, K., Aster, R. H., and Newman, P. J. (1992). An amino acid polymorphism within the RGD binding domain of platelet membrane glycoprotein IIIa is responsible for the formation of the pen-&pen-b alloantigen system. 1.Clin. Invest. 90, 2038-2043. Wamock, R. A., Askari, S., Butcher, E. C., and von Andrian, U. H. (1998). Molecular mechanisms of lymphocyte homing to peripheral lymph nodes.]. Exp. Med. 187,205-216. Washington, R., Burton, J., Todd, R. F., 3rd, Newman, W., Dragovic, L., and Dore-Duffy, P. (1994). Expression of immunologically relevant endothelial cell activation antigens on isolated central nervous system microvessels from patients with multiple sclerosis. Ann. Neurol. 35, 89-97. Wayner, E. A,, and Kovach, N. L. (1992).Activation-dependent recognitionby hematopoietic cells of the LDV sequence in the V region of fibronectin. 1.Cell Biol. 116,489-497. Weber, C., d o n , R., Moser, B., and Springer, T. A. (1996a). Sequential regulation of a& and a& integrin avidityby CC chemokines in monocytes: Implications for transendothelial chemotaxis. J. Cell Blol. 134, 1063-1073. Weber, C., Katayama, J., and Springer, T. A. (1996b). Differential regulation of pl and p2 integrin avidity by chemoattractants in eosinophils. Proc. Natl. Acad. Sci. U.S.A. 93, 10939-10944.

INTEGRINS IN THE IMMUNE SYSTEM

379

Weber, C., Lu, C. F., Casasnovas, J. M., and Springer, T. A. (1997). Role of aLp2 integrin avidity in transendothelial chemotaxis of mononuclear cells.]. Immunol. 159,3968-3975. Wei, J., Shaw, L. M., and Mercurio, A. M. (1997). Integrin signaling in leukocytes: Lessons from the a 6 p l integrin. 1.Leukoc. Biol. 61, 397-407. Weisel, J. W., Nagaswami, C., Vilaire, G., and Bennett, J. S. (1992). Examination of the platelet membrane glycoprotein IIb-IIIa complex and its interaction with fibrinogen and other ligands by electron microscopy. 1.Biol. Chem. 267, 16637-16643. Welsh, C. T., Rose, J. W., Hill, K. E., and Townsend, J. J. (1993).Augmentation of adoptively transferred experimental allergic encephalomyelitis by administration of a monoclonal antibody specific for LFA-1 a.1.Neuroimmunol. 43, 161-167. Williams, D. A,, Rios, M., Stephens, C., and Patel, V. P. (1991). Fibronectin and VLA-4 in haematopoietic stem cell-microenvironment interaction. Nature 352, 438-441. Williams, M. B., and Butcher, E. C. (1997). Homing of naive and memory T lymphocyte subsets to Peyer’s patches, lymph nodes, and spleen. J. ImmunoZ. 159, 1746-1752. Williams, M. J., Hughes, P. E., O’Toole, T. E., and Ginsberg, M. H. (1994). The inner world of cell adhesion: Integrin cytoplasmic domains. Trends Cell Biol. 4, 109-112. Wippler, J., Kouns, W. C., Schlaeger, E. J., Kuhn, H., Hadvary, P., and Steiner, B. (1994).The integrin all&, platelet glycoprotein IIb-IIIa, can form a functionally active heterodimer complexwithout the cysteine-rich repeats ofthe p3subunit.J. Biol. Chem. 269,8754-8761. Xiao, J., Messinger, Y., Jin, J., Myers, D. E., Bolen, J, B., and Uckun, F. M. (1996). Signal transduction through the pl integrin family surface adhesion molecules vla-4 and vla-5 of human B-cell precursors activates cd19 receptor-associated protein-tyrosine kinases. 1.Biol. Chem. 271, 7659-7664. Xu, J., Grewal, I. S., Geba, G. P., and Flavell, R. A. (1996).Impaired primary T cell responses in L-selectin-deficient mice. 1.Exp. Med. 183, 589-598. Yamada, K. M., and Geiger, B. (1997). Molecular interactions in cell adhesion complexes. Cum Opin. Cell Biol. 9, 76-85. Yaps, A. S., Brieher, W. M., and Gumbiner, B. M. (1997). Molecular and functional analysis of cadherin-based adherens junction. Annu. Reo. CeZZ Dev. B6Z. 13, 119-146. Yauch, R. L., Felsenfeld, D. P., Kraeft, S. K., Chen, L. B., Sheetz, M. P., and Hemler, M. E. (1997). Mutational evidence for control of cell adhesion through integrin diffusion/ clustering, independent of ligand binding. J. Exp. Med. 186, 1347-1355. Yednock, T. A,, Cannon, C., Fritz, L. C., Sanchez-Madrid, F., Steinman, L., and Karin, N. (1992). Prevention of experimental autoimmune encephalomyelitis by antibodies against a 4 p l integrin. Nature 356, 63-66. Ylanne, J., Chen, Y., O’Toole, T. E., Loftus, J, C., Takada, Y., and Ginsberg, M. H. (1993). Distinct functions of integrin a and p subunit cytoplasmic domains in cell spreading and formation of focal adhesions. J. Cell Biol. 122, 223-233. Yoshida, R., Nagira, M., Kitaura, M., Imagawa, N., Imai, T., and Yoshie, 0.(1998). Secondary lymphoid-tissue chemokine is a functional ligand for the CC chemokine receptor CCR7. J. Biol. Chem. 273, 7118-7122. Yurochko, A. D., Liu, D. Y., Eierman, D., and Haskill, S. (1992). Integrins as a primary signal transduction molecule regulating monocyte immediate-early gene induction. Proc. Nutl. h a d . Sci. U.S.A. 89, 9034-9038. Zauli, G., Bassini, A., Vitale, M., Gibellini, D., Celeghini, C., Caramelli, E., Pierpaok S., Guidotti, L., and Capitani, s, (1997). Thrombopoietin enhances the aIIbp3-dependent adhesion of megakaryocytic cells to fibrinogen or fibronectin through PI 3 kinase. Blood 89,883-895. Zell, T., Hunt, S. W., 3rd, Mobley, J. L., Finkelstein, L. D., and Shimizu, Y. (1996). CD28mediated up-regulation of pl-integrin adhesion involves phosphatidylinositol 3-kinase. 1.Immunol. 156, 883-886.

380

YOJI SHIMIZU et ol.

Zell, T., Warden, C. S., Chan, A. S., Cook, M. E., Dell, C. L., Hunt, S. W., 3rd, and Shimizu, Y. (1998). Regulation of /3l-integrin-mediated cell adhesion by the cbl adaptor protein. Cum Biol. 8, 814-822. Zhang, L., and Plow, E. F. (1996).Overlapping, but not identical, sites are involved in the recognition of C3bi, neutrophil inhibitory factor, and adhesive ligands by the aM/32 integrin. 1.Bid. Chen. 271, 18211-18216. Zhang, L., and Plow, E. F. (1997). Identification and reconstruction of the binding site within aMP2 for a specific and high affinity ligand, NIF.]. Bid. Chem. 272,17558-17564. Zhou, M., and Brown, E. J. (1993). Leukocyte response integrin and integrin-associated protein act as a signal transduction unit in generation of a phagocyte respiratory burst. J. Exp. Med. 178, 1165-1174. This article was accepted for publication on August 28, 1998.

INDEX

A Agammaglobulinemia,X-linked, 159-160 Allergies, 293 aEP7, 346-348 a4p7, 220-223 inflammation blocldng, 242-243 lymph node development, 343 lymphocyte homing, 220-222, 233-234 aLp2, 349-350 Antigens capture, 262 CD1,257-258 CLA, 265 intestinal, 237-240 presenting cells, 264 processing, 264-265 AP-1, 23 Apoptosis dendritic cells, 263-264 GTPases role, 47-49 Asthma, 293-294 Ataxia-telangiectasia gene, 180-181 lymphoid defects characterization, 179 dissecting, 181-185 tumorigenesis characterization, 179 dissecting, 181- 185 ATM, 180-181

B Bacteria, 294 B cells -dendritic cell interaction

B cell differentiation, 289-291 B cell proliferation, 289 dialogues, 288-291 follicular DC, 286-287 general, 285 germinal center DC, 287-288 development, 339-340 memory, 237-240 receptors, ITIM inhibition, 155-156 p7, 225-227 Blood mononuclear cells, 277 Btk, 160

C E-Cadherin, 333-334 Calcium BCR-triggered, 156-157 Btk effects, 160 Ca", dendritic cell, 273-274 ITAM effects, 150 SHIP effects, 159 Cathepsin D, 272 Cbl, 12-14 CD43, 226-227 CD44, 226-227 CD1 antigens, 257-258 Cdc-42, 33-34 CD40/CD40L, 269-270 Cdc42sp, 36-37 Cell migration dendritic, 265-268 diapedesis, 338 C3 exotoxin, 33-34 Chemolo'nes, lymphocyte trafficking, 227-228 Chlamydia truchomutis, 214 Colitis, ulcerative, 213 381

382

INDEX

Crohn's disease, 213 C-type lectin receptor, 263 Cutaneous lymphocyte antigen, 265 Cytomegalovirus, 296-297 Cytoskeleton reorganization, 27-34

D Delayed-type hypersensitivity, 349-350 Dendritic cells adhesion molecules, 265 antigen capture, 262 antigen presentation, 264 apoptosis, 263-264 -B cell interaction B cell differentiation, 289-291 B cell proliferation, 289 dialogues, 288 fokcular DC, 286-287 general, 285 germinal center DC, 287-288 Ca", 273-274 clinical studies allergies, 293 asthma, 293-294 bacteria, 294 parasites, 294-295 psoriasis, 291 retroviruses, 298-300 rheumatoid arthritis, 291 transplantation, 291-293 tumors general, 300-302 human, 303-305 mouse, 302-303 viruses, 295-298 cytomegdovirus, 296-297 herpesvirus, 296 influenza, 297 measles, 297-298 costimulatory molecules, 268-269 C-type lectin receptors, 263 development, 262 lymphoid pathway, 277-278 stages, 259-260 theories, current, 278-279 enzymes, 272-273 FCEreceptors, 262 Fcy receptors, 262

FLT-3 ligmd, 278 generation blood mononuclear cells, 277 hematopoietic progenitor cells, 274-277 murine cell lines, 274 humoral responses in uitm, 289 in uiuo, 288-289 identification, 255 immune response, 255 macropinocytosis, 262 mannose receptor, 263 maturation IL-10 inhibition, 280-281 stimulators, 279-280 MHC class I loading, 264-265 migration control, 266-268 patterns, 265-266 morphology, 257 NK phenotype, 273 physiology, 255-257 signaling CD40/CD40L, 269-271 FadFasL, 272 OX4O/OX401, 271 RANK-UTRANCE/ODF, 271-272 RANKA'RANCE-Wosteoprotein, 271-272 TNFmNF-R, 269 -T cell interaction cytokine role, 282 in uiuo association, 281-282 T cell priming, 282-283 tolerance central, 283-284 peripheral, 284 Dinitrofluorobenzene, 282-283 DNFB, see Dinitrofluorobenzene

E ECM proteins, see Extracellular matrix proteins Encephalomyelitis, experimental autoimmune, 348-353 Endothelial cells HEV, levels, 213-214 -lymphocyte interaction p7 role, 225-227

383

INDEX

characterization, 211 IEL role, 225-227 LFA-1 role, 225 E-selection role, 225-227 L-selection role, 224-225 P-selection role, 225-227 sialomucin role, 226-227 ERKs, Extracellular-regulated kinases Exotoxin, C3, 33-34 Experimental autoimmune encephalomyelitis, 348-353 Extracellular matrix proteins, 27-28 Extracellular-regulated kinases IRR activation, 3 NFAT activation, 34-35 T cell activation, 16-17

F Fas, 47-49 Fas/FasL, 272 FCEreceptors, 262 Fcy receptors characterization, 154-158 dendritic cells, 262 Fetal liver kinase 2, 278-279 FLT-3, 278 Focal adhesion kinase, 66-68 Fyn, 117-119

G GAPS,see GTPase-activation proteins

Gastrointestinal tract inflammation a4P7 blocking, 234-237 lymphocyte recruitment role, 242-243 Grb2, 37-38 Grb2iSo.s Ra9 activation, 8-11 T cell activation, 11-12 GTPase-activation proteins function, 2 Ras activation, 14-15 GTPases Fas apoptosis, 47-49 HIV infection, 59-63 Rab family, 52-53 Ras family CD28 signaling, 49-51

cytokine receptor signaling, 25 function, 1-2 IRR activation, 3-5 Cbl role, 12-14 Grb2/Sos role, 8-11 PTK role, 5-8 SLP-76 role, 12-14 Vav role, 12-14 IRR signaling effectors, 15-18 negative factors, 23-25 transcription factors, 18-23 lymphocyte development, 64-66 regulation, 1-2 T cell anergy, 53-59 regulation, 68-69 Rho family CD28 signaling, 51-52 cell adhesion, 41-43 chemoattractant responses, 44-46 hematopoietic cells regulation, 38-39 Vav integration, 34-35 integrin adhesion, 41-43 IRR signaling activation, 39-41 characterization, 25-26 leukocytes cytoskeleton, 27-34 lymphocyte development, 66-68 p38 pathway, 68 SEKl pathway, 68 selections, 43-44 signal control, 34-35 Wiskott-Aldrich syndrome, 63-64

H Hematopoietic cells Ras signals, 34-35 Rho signals regulation, 38-39 Vav integration, 34-35 Hematopoietic progenitor cells, 274-277 Hematopoietic stem cells, 338-339 Herpes simplex virus, 295-296 Human immunodeficiency virus dendritic cells, 298-300 G protein role, 59-63 Hypersensitivity, delayed-type, 349-350

384

INDEX

I ICAMs, see Intercellular adhesion molecules Immuglobulins IgA, 290 IgM, 289-290 Immune receptor tyrosine-based inhibitory motifs characterization, 149-150 families, 150, 152, 154 inhibitory signal, FcyRII-mediated, 154- 158 KIR receptor inhibition, 161-163 mouse models, 161-165 Immune recognition receptors Ras activation, 3-5 Cbl role, 12-14 Grb2/Sos role, 8-11 PTK role, 5-8 Shc role, 8-11 SLP-76 role, 12-14 Vav role, 12-14 signaling Ras effectors, 15-18 negative factors, 23-25 transcription factors, 18-23 Rho characterization, 25-26 effectors, 39-41 signal transduction, 2-3 Immune system integrin ligands function, 330-331,337-343 localization, 333-334 signaling, 335-337 structure, 331-333 Inflammation a4p7 blocking, 242-243 integrin-mediated, 350-353 lymphocyte-mediated clinical study, 350-351 recruitment role, 234-237 Inff uenza virus, 297 Inositol 5-polphosphate phosphatase SH2-containing FcyRII inhibition, 156 mechanism, 158-161 mouse models, 163-165 phosphoinositol hydrolyzation, 150 properties, 24-25

Integrin adhesion receptors ~~4p7,220-223 autoimmune blocking, 242-243 gastrointestinal tract inflammation, 242-243 inflammation blocking, 242-243 lymph node development, 343 lymphocyte homing, 220-222, 233-234 lymphoid malignancies, 241 vaccine development, 241-242 p7,225-227 LFA-1 cascading, 232-233 characterization, 225 signaling, 335-337 Integrins B cell development, 339-340 characterization, 325-326 aEP7, 346-348 hematopoietic organ seeding, 338-339 inflammatory disease, 350-351 aLp2, 349-350 ligand binding sites, 326-330 function, 330-331 inflammatory response, 351-353 locahzation, 333-334 mucosal immunity, 334-335 structure, 331-333 lymphocyte recirculation, 344-348 signaling, 335-337 trafficking control, 337-338 IntercelIular adhesion molecules dendritic cells, 265 inflammatory response, 351-353 integrin ligand domains, 333-334 Rho function, 33-34 Interleukin-2, 4-5, 289-290 Interleukin-10, DC, 280-281 Intestinal antigens, 237-240 Intraepithelial lymphocytes, 346-347 y-Irradiation, 181 IRRs, see Immune recognition receptors ITIMs, see Immune receptor tyrosine-based inhibitory motifs

J JNK cascade, 36-37

INDEX

-c-Jun, 21 IRR signaling, 39-41

Killer cell inhibitory receptors, 161-163

L LAT, see Linker activation of T cells Lck HIV, 61-62 pre-TCR sensor, 117-119 Lectin receptor, C-type, 263 Leishmania spp., 294-295 Leukocytes, see also specijic leukocyte cytoskeletons, 27-34 Rab function, 52-53 trafficking, 337-338 LFA-1 cascading, 232-233 characterization, 225 dependent adhesion, 33-34 signaling, 335-337 Ligands, integrin function, 330-331 inflammatory response, 351-353 localization, 333-334 mucosal immunity, 334-335 structure, 331-333 Linker activation of T cells, 12 Liver, fetal, 338-339 LPA, see Lysophosphatidic acid Lymphocytes, see also specijic lymphocyte blood-borne, recruitment chemotactic factors, 227-229 integrin-triggering, 227-229 required steps, 211-212 development Ras role, 64-66 Rho role, 66-68 -endothelium interaction P7 role, 225-227 characterization, 211 IEL role, 225-227 LFA-1 role, 225 E-selection, 225-227 L-selection role, 224-225 P-selection, 225-227 sidornucin role, 226-227

385

growth signals, 34-35 homing 014017 preactivation, 233-234 014p7 characterization, 220-223 segregation, 237-241 adhesion cascade naive effectors, 232-233 nonintestinal tissue, 234 MAdCAM-1, 213-220 memory effectors, 231-232 naive effectors, 231-232 T cells in uiuo, 229-230 inflammatory disease, 350-351 mixed reactions, 268 trafficking alterations, 234-237 characterization, 209 muscularis patterns, 237 serosa patterns, 237 studies, 210-211 theory, 210 Lyn/FceRI ratio, 121 Lysophosphatidic acid, 33

Macropinocytosis, 262 MAdCAm-1,see Mucosal vascular addressin Major histocompatibility complex deiidritic cell loading, 264-265 maturation, 279-280 Mannose receptor, 263 MAPKAP kinase-2, 39-41 Measles virus, 297-298 MEK-1, 118 MEKK-1, 21-22, 39-41 Memory cells, 237-241 Metal ion-dependent adhesion site, 329 MIDAS, see Metal ion-dependent adhesion site MIPBP, 228 MKK7,39-41 Mucosal vascular addressin autoimmune blocking, 242-243 aEP7 integrins, 346-348 inflammation blocking, 242-243 integrin ligands, 334-335

INDEX

lymph node development, 343 lymphocyte-endotheliuminteraction, 213-220,233-234 Mutations adaptive responses, 138-140 redundancy, 138-140

N NADPH oxidase chemoattractant responses response, 46-47 Rho signal regulation, 38-39 Natural killer cells CD4 role, 108 DC phenotype, 273 as lymphoid DCs, 257 Rho signal regulation, 38-39 NFAT, Nuclear factor of activated T cells NKR-P1, 273 Nuclear factor of activated T cells characterization, 5 growth signals, 34-35 Ras signaling, 19-20, 23

0 OX4O/OX4OL, 271

P P38 -JNK, 39-41 -SEKI, 68 Parasites, 294-295 Phagocytosis, 34 Phospholipase D, 47 Phosphotyrosine phophatases, SH2containing, 24 Precursor cells, 337-338 Pre-T cell receptors delic exclusion, 127-128 assembly, 116-117 CD3 components, 114-115 cell survival, 131-138 function, 131 aPyS lineage role, 124-127

sensor characterization, 119, 121-122 downstream effectors, 117-119 TCR-@locus role, 127-128 Programmed cell death, see Apoptosis

R Rab, 52-53 Rac-1, 23 Raf-1 kinase characterization, 16 function, 22 RAG-1, 118 RAG-2, 182 RANK-L/TRANCE/ODF, 271-272 Rapl, 5 Ras CD28 signaling, 49-51 cytokine receptor signaling, 25 function, 1-2 IRR activation, 3-5 Cbl role, 12-14 GrbZ/Sos role, 8-11 PTK role, 5-8 Shc role, 8-11 SLP-76 role, 12-14 IRR signaling effectors, 15-18 negative factors, 23-25 transcription factors, 18-23 lymphocyte development, 64-66 regulation, 1-2 T cell anergy, 53-59 Rel-B, 280 Retroviruses, 298-300 Rheumatoid arthritis, 291 Rho CD28 signaling, 51-52 cell adhesion, 41-43 chemoattractant responses, 44-46 cytoskeleton reorganization, 27-34 hematopoietic cells regulation, 38-39 Vuu integration, 34-35

INDEX

integrin adhesion, 41-43 IRR signaling activation, 39-41 characterization, 25-26 p38 pathway, 68 SEKl pathway, 68 selections, 43-44 RHO family, 34-35

Scdl, 36-37 SDF-1, 227-228 SEKl, 68 E-Selection, 225-227 L-Selection, 224-225, 232-233 P-Selection, 225-227 Serum response element, 22 Serum response factor, 22 SHIP, see Inositol 5-polyphosphate phosphatase Sialomucins, 226-227 Signal transduction cytokine receptors, 25 dendritic cells CD40/CD40L, 269-271 FadFasL, 272 OX4O/OX401,271 RANK-L/TRANCE/ODF, 271-272

RANK/TRANCE-Wosteoprotein, 271-272 TNF/TNF-R, 269 FqRII, 154-158 integrin, 335-337 IRR characterization, 2-3 Ras effectors, 15-18 Rho effectors, 39-41 transcription factors, 18-23 SLCIGCkine, 228 SLP-76, 12-14 SRE, see Serum response element SRF, see Serum response factor Stem cells, hematopoietic, 338-339 Syk, 118, 122

T T cell receptors, see a h Pre-T cell receptors CD3 contribution, 116-117

CD3P chain components, 113-114 CD3.9 chain components, 112-113 CD3y chain components, 114 cell suMval, 131-138 FceRIy components, 115-116 ap chain sensors, 128-131 p chain allelic exclusion, 127-128 components, 109, 111 y6 chain, 124-127 mutations, 138-140 pTa chain components, 111-112 maturation independent of, 122-124 T cells activation linker, 11-12 anergy Rapl-induced, 5 Ras induced, 53-59 CD28 signaling Ras role, 49-51 Rho role, 51-52 -dendritic cell interaction cytokine role, 282 T cell priming, 282-283 in uiuo association, 281-282 development CD3 function, 103, 105-106 integrin function, 340-343 mouse model, 106-109 ERK activation, 16-17 memory, recirculation, 344345 naive, recirculation, 344 Raf-1 activation, 16-17 shape regulation, 43-44 Thymocytes development, 341342 a-Tubulin, 31-32 Tumorigenesis, AT characterization, 179 dissecting, 181-185 Tumor necrosis factors dendritic cells generation, 269-277 maturation, 279-280 signahng, 269-272

387

388

INDEX

U Ulcerative colitis, 213

W Wiskott-Aldrich syndrome, 63-64

V Vascular cell adhesion molecules inflammatory response, 351-353 integrin ligand domains, 331-332, 334 lyniphocyte homing, 220-221 Vav G protein regulation, 68-69 Ras activation, 12-14 Rho signals, 34-35 V(D)J recombination, 182-185 Viruses, 295-298

X X-linked agammaglobulinemia, 159-160

ZAP-70 -CD3 complex, 57 characterization, 118 recruitment, 121-122

CONTENTS OF RECENT VOLUMES

Volume 68

Immunological Treatment of Autoimmune Diseases J. R. KALDEN,F. C. BREEDVELD, AND G. R. BURMESTER H. BURKHARDT,

Posttranscriptional Regulation of mRNAs Important in T Cell Function JAMES S. MALTER

INDEX

Molecular and Cellular Mechanisms of T Lymphocyte Apoptosis JOSEF M. PENNINGER AND GUIDO KROEMER

Volume 69 Molecular and Cellular Events in Early Thymocyte Development A N D HANS HANS-REIMER RODEWALD JORG FEHLINC

Prenylation of Ras GTPase Superfamily Proteins and Their Function in Immunobiology ROBERTB. LOBELL

Regulation of Immunoglobulin Light Chain Isotype Expression FREDERICK W. ALT AND JAMES R. GORMAN

Generation and TAP-Mediated Transport of Peptides for Major Histocompatibility Complex Class I Molecules FRANK MOMBURG AND GONTHER J. HAMMERLING

Role of Immunoreceptor Tyrosine-Based Activation Motif in Signal Transduction from Antigen and Fc Receptors NOAHISAKOV

Adoptive Tumor Immunity Mediated by Lymphocytes Bearing Modified AntigenSpecific Receptors THOMAS BROCKER AND KLAUS KARJAIAINEN Membrane Molecules as Differentiation Antigens of Murine Macrophages ANDREWJ. M c K A N~D SIAMON ~ ~ ~ GORDON Major Histocompatibility Complex-Directed Susceptibility to Rheumatoid Arthritis GEHALD T. NEPOM

Atypical Serine Proteases of the Complement System G~RAR J. D ARLAUD,JOHN E. VOLANAKIS, NICOLEM. THIELENS, STHANAM V. L. NARAYANA, V~RONIQUE Ross], AND YUANYUANXU ~

Accessibility Control of V(D)J Recombination: Lessons from Gene Targeting M. HEMPEL, ISABELLE LEDUC, WILLIAM NOELLEMATHIEU, RAJKAMALTRIPATHI, AND PIERRE FERRIER 389

390

CONTENTS OF RECENT VOLUMES

Phylogenetic Emergence and Molecular Interactions between the Immune System Evolution of the Immunoglobulin Family and Gene Therapy Vectors: Bidirectional JOHNJ. MARCHALONIS, SAMUEL F. Regulation of Response and Expression SCHLUTER, RALPH M. BERNSTEIN, JONATHAN S. BROMBERG, LISADEBRUYNE, SHANXIANG SHEN,AND ALLENB. AND LIHUIQIN EDMUNDSON Major Histocompatibility Complex Genes Influence Individual Odors and Mating Preferences DUSTINPENNAND WAYNE Pons Olfactory Receptor Gene Regulation ANDREWCHESS

INDEX

Current Insights into the “Antiphospholipid Syndrome: Clinical, Immunological, and Molecular Aspects DAVIDA. KANDIAH,ANDREJSALI, YONGHUASHENG,EDWARD J. VICTORIA, DAVIDM. MARQUIS, STEPHEN M. COUTTS, AND STEVEN A. KRILIS

INDEX

Volume 71 Volume 70 Biology of the Interleukin-2 Receptor BRADH. NELSONAND DENNIS M. WILLERFORD Interleukin-12: A Cytokine at the Interface of Inflammation and Immunity GIORGIO TRINCHIERI Recent Progress on the Regulation of Apoptosis by Bcl-2 Family Members ANDYJ. MINN, RACHELE. SWAIN, AVERIL MA. AND CFLUGB. THOMPSON Interleukin-18: A Novel Cytokine That Augments Both Innate and Acquired Immunity HARUKI OKAMURA, HIROKO TSUTSUI, SHIN-ICHIRO KASHIWAMURA, TOMOHIRO YOSHIMOTO, AND KENJI NAKANISHI

a&S Lineage Commitment in the Thymus of Normal and Genetically Manipulated Mice HANSJORGFEHLING, SUSAN GILFILLAN, AND RHODRI CEREDIG Immunoregulatory Functions of yS T Cells WILLIBORN,CAROL CADY, JESSICA JONESCARSON, AKIKOMUKASA, MICHAEL LAHN, AND REBECCAO’BRIEN STATs as Mediators of Cytokine-Induced Responses TIMOTHY HOEYAND MICHAEL J. GRUSBY CD95(APO-UFas)-Mediated Apoptosis: Live and Let Die PETERH. KRAMMER A CXC Chemokine SDF-UPBSF: A Ligand for a HIV Coreceptor, CXCR4 TAKASHI NAGASAWA, KAZUNOBU TACHIBANA, AND KENJIKAWABATA

T Lymphocyte Tolerance: From Thymic CD4’ T-cell Induction and Effector Deletion to Peripheral Control Mechanisms Functions: A Comparison of Immunity BRIGITTA STOCKINGER against Soluble Antigens and Viral Infections ANNETTEOXENIUS, ROLFM. Confrontation between Intracellular Bacteria ZINKERNAGEL. AND HANSHENGARTNER and the Immune System ULRICHE. SCHAIBLE, HELENL. COLLINS, Current Views in Intracellular Transport: AND STEFAN H. E. KAUFMANN Insights from Studies in Immunology VICTORW. Hsu AND PETERJ. PETERS INDEX

This Page Intentionally Left Blank