Advances in Cancer Research, Volume 84

Advances in CANCER RESEARCH Volume 84 This Page Intentionally Left Blank Advances in CANCER RESEARCH Volume 84 E...

Author: George Klein | George F. Vande Woude | George F. Vande Woude

32 downloads 1165 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Advances in

CANCER RESEARCH Volume 84

This Page Intentionally Left Blank

Advances in

CANCER RESEARCH Volume 84

Edited by

George F. Vande Woude Van Andel Research Institute Grand Rapids, Michigan

George Klein Microbiology and Tumor Biology Center Karolinska Institute Stockholm, Sweden

San Diego

San Francisco New York Boston London Sydney Tokyo

∞ This book is printed on acid-free paper. C 2002 by ACADEMIC PRESS Copyright

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-2002 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. 0065-230X/2002 $35.00 Explicit permission from Academic Press is not required to reproduce a maximum of two ﬁgures or tables from an Academic Press chapter in another scientiﬁc or research publication provided that the material has not been credited to another source and that full credit to the Academic Press chapter is given.

Academic Press An Elsevier Science Imprint. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com

Academic Press 32 Jamestown Road, London NW1 7BY, UK http://www.academicpress.com International Standard Book Number: 0-12-006684-X PRINTED IN THE UNITED STATES OF AMERICA 02 03 04 05 06 07 EB 9 8 7 6 5 4

3

2

1

Contents

Contributors to Volume 84 ix

Gene Expression in Inherited Breast Cancer ˚ Borg ´ Jeffrey M. Trent, and Ake Ingrid A. Hedenfalk, Markus Ringner, I. II. III. IV. V. VI. VII. VIII.

Introduction 2 Epidemiology of Familial Breast Cancer 3 The BRCA1 and BRCA2 Genes 3 Tumor Progression and the Role of Estrogen in Breast Cancer Development 10 Characteristics of Hereditary Breast Cancers 11 Other Causes of Breast Cancer 13 Gene Expression Analysis 16 Concluding Remarks 26 References 27

Multiparameter Analyses of Cell Cycle Regulatory Proteins in Human Breast Cancer: A Key to Deﬁnition of Separate Pathways in Tumorigenesis ¨ Goran Landberg I. II. III. IV. V.

Breast Cancer 36 Cell Cycle Aberrations 37 Deregulation of Some Important G1 /S Regulatory Proteins in Breast Cancer 39 Multiparameter Analyses of G1 /S Regulatory Proteins in Breast Cancer 44 Future Perspectives 48 References 50

Rho GTPases in Transformation and Metastasis Aron B. Jaffe and Alan Hall I. Introduction 57 II. Rho GTPases and Transformation 60 III. Rho GTPases and Metastasis 70

v

vi

Contents

IV. Conclusions and Future Directions 75 References 75

The myc Oncogene: MarvelouslY Complex Sara K. Oster, Cynthia S. W. Ho, Erinn L. Soucie, and Linda Z. Penn I. II. III. IV. V. VI.

Introduction 82 Myc and Cancer 82 Biological Activities of Myc 92 Functions of Myc as a Regulator of Transcription 111 Identiﬁcation of Myc-Interacting Proteins 126 Overall Perspectives 140 References 141

Genetic Requirements for the Episomal Maintenance of Oncogenic Herpesvirus Genomes Christopher M. Collins and Peter G. Medveczky I. II. III. IV.

Introduction 156 Latent Episomal Genomes of Oncogenic Herpesviruses 157 The Role of EBNA-1 and OriP in EBV Episomal Replication 157 The Role of LANA1 and Terminal Repeats in Episomal Replication of KSHV and HVS 162 V. Conclusions 170 References 171

Treatment of Epstein–Barr Virus-Associated Malignancies with Speciﬁc T Cells Stephen Gottschalk, Helen E. Heslop, and Cliona M. Rooney I. II. III. IV. V. VI. VII.

Introduction 176 EBV-Associated Lymphoproliferative Disease 179 Hodgkin’s Disease 187 Nasopharyngeal Carcinoma 190 Burkitt’s Lymphoma 191 EBV-Associated Non-Hodgkin’s Lymphoma in HIV Patients 191 Conclusions 192 References 194

Role of Glycogen Synthase Kinase-3 in Cancer: Regulation by Wnts and Other Signaling Pathways Armen S. Manoukian and James R. Woodgett I. Origins of GSK-3 204 II. GSK-3 Functions: Clues From Other Species 205

Contents

III. IV. V. VI.

vii

Mammalian Wnt Signaling and Cancer 214 Regulation of GSK-3 by Other Signaling Pathways 218 Other Targets of GSK-3 Involved in Cancer 221 GSK-3 and Cancer 223 References 224

Chronic Immune Activation and Inﬂammation in the Pathogenesis of AIDS and Cancer Angus G. Dalgleish and Kenneth J. O’Byrne I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV.

The Pathogenesis of HIV-1 Infection 232 SIV and Simian AIDS 233 Chronic Immune Activation 233 Immunological Imbalance in HIV Infection 236 Cancer and the Immune Response 237 Kaposi’s Sarcoma (KS) and HIV Infection as a Paradigm for Inﬂammatory Cancer Development 240 The Interrelationship between the Immune Response and Angiogenesis 243 Inﬂammation and Apoptosis 246 The Biology of Oncogenic Viruses: Impact on the Immune Response, Angiogenesis, and Apoptosis 249 Features of Tumors Lacking an Obvious Premalignant Inﬂammatory Process 253 Environment Created by Noninﬂammatory Factors That Predispose to Malignant Disease 254 Fractal Mathematics, Carcinogenesis, and the Progression of Malignant Disease 256 Implications for Chemoprevention and Future Treatment of Malignant Disease 257 Conclusions 258 References 259

Molecular Biology of Hodgkin’s Lymphoma ¨ Ralf Kuppers I. II. III. IV. V. VI. VII.

Introduction 278 Pathology 278 Origin of Hodgkin and Reed–Sternberg Cells 279 Pathogenesis 288 Phenotype and Gene Expression Patterns of HRS Cells 293 HRS Cells in Their Microenvironment 298 Concluding Remarks 301 References 304

Index 313

This Page Intentionally Left Blank

Contributors

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

˚ Ake Borg, Departments of Oncology, Lund University, SE-22100 Lund, Sweden (1) Christopher M. Collins, Department of Medical Microbiology and Immunology, H. Lee Mofﬁtt Cancer Center, University of South Florida, Tampa, Florida 33612-4799 (155) A. J. Dalgleish, Department of Oncology, St. George’s Hospital Medical School, London, SW17 0RE, United Kingdom (231) Stephen Gottschalk, Center for Cell and Gene Therapy, Department of Pediatrics, Baylor College of Medicine, Houston, Texas 77030 (175) Alan Hall, MRC Laboratory for Molecular Cell Biology and Cell Biology Unit, CRC Oncogene and Signal Transduction Group, University College London, London WC1E 6BT, United Kingdom (57) Ingrid A. Hedenfalk, Cancer Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland 20892; Department of Oncology, Lund University, SE-22100 Lund, Sweden (1) Helen E. Heslop, Center for Cell and Gene Therapy, Department of Pediatrics and Department of Medicine, Baylor College of Medicine, Houston, Texas 77030 (175) Cynthia S. W. Ho, Division of Cellular and Molecular Biology, Ontario Cancer Institute, Princess Margaret Hospital, Department of Medical Biophysics, University of Toronto, Toronto, Ontario M5G 2M9, Canada (81) Aron B. Jaffe, MRC Laboratory for Molecular Cell Biology and Cell Biology Unit, CRC Oncogene and Signal Transduction Group, University College London, London WC1E 6BT, United Kingdom (57) ¨ Ralf Kuppers, Institute for Genetics and Department of Internal Medicine I, University of Cologne, D-50931 Cologne, Germany (277) ¨ Goran Landberg , Division of Pathology, Department of Laboratory Medi¨ cine, Lund University, Malmo¨ University Hospital, SE-205 02 Malmo, Sweden (35) Armen S. Manoukian, Divisions of Experimental Therapeutics and Cellular and Molecular Biology, Ontario Cancer Institute, University of Toronto, Toronto, Ontario M5G 2M9, Canada (203) ix

x

Contributors

Peter G. Medveczky, Department of Medical Microbiology and Immunology, H. Lee Mofﬁtt Cancer Center, University of South Florida, Tampa, Florida 33612-4799 (155) Ken O’Byrne, University Department of Oncology, Leicester Royal Inﬁrmary, Leicester LE1 5WW, United Kingdom (231) Sara K. Oster, Division of Cellular and Molecular Biology, Ontario Cancer Institute, Princess Margaret Hospital, Department of Medical Biophysics, University of Toronto, Toronto, Ontario M5G 2M9, Canada (81) Linda Z. Penn, Division of Cellular and Molecular Biology, Ontario Cancer Institute, Princess Margaret Hospital, Department of Medical Biophysics, University of Toronto, Toronto, Ontario M5G 2M9, Canada (81) Markus Ringn´er, Cancer Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland 20892 (1) Cliona M. Rooney, Center for Cell and Gene Therapy, Department of Pediatrics and Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, Texas 77030 (175) Erinn L. Soucie, Division of Cellular and Molecular Biology, Ontario Cancer Institute, Princess Margaret Hospital, Department of Medical Biophysics, University of Toronto, Toronto, Ontario M5G 2M9, Canada (81) Jeffrey Trent, Cancer Genetics Branch, National Human Genome Research Institute, National Institutes of Health, Bethesda, Maryland 20892 (1) James R. Woodgett, Divisions of Experimental Therapeutics and Cellular and Molecular Biology, Ontario Cancer Institute, University of Toronto, Toronto, Ontario M5G 2M9, Canada (203)

Gene Expression in Inherited Breast Cancer Ingrid A. Hedenfalk,*,† Markus Ringner,* ´ ˚ Jeffrey M. Trent,* and Ake Borg† *Cancer Genetics Branch National Human Genome Research Institute National Institutes of Health Bethesda, Maryland 20892 † Department of Oncology Lund University SE-221 85 Lund, Sweden

I. Introduction II. Epidemiology of Familial Breast Cancer III. The BRCA1 and BRCA2 Genes A. Gene Structures and Mutation Spectra B. Normal Regulation and Expression of BRCA1 and BRCA2 C. Functions of the BRCA1 and BRCA2 Genes IV. Tumor Progression and the Role of Estrogen in Breast Cancer Development V. Characteristics of Hereditary Breast Cancers A. Pathology and Histology of Hereditary Breast Cancer B. Somatic Genetic Changes in Hereditary Breast Cancer VI. Other Causes of Breast Cancer A. Syndromes with Increased Incidence of Breast Cancer B. Other Genes Conferring Breast Cancer Susceptibility C. Low-Penetrance Susceptibility Genes D. The Search for BRCA3 VII. Gene Expression Analysis A. Microarray Technique B. Data Analysis C. Gene Expression Proﬁles in BRCA1/2 Breast Cancers D. Discovering New Classes within the BRCAx Breast Cancers E. Copy Number and Gene Expression Analysis VIII. Concluding Remarks References

Large proportions of hereditary breast cancers are due to mutations in the two breast cancer susceptibility genes BRCA1 and BRCA2. Considerable effort has gone into studying the function(s) of these tumor suppressor genes, both in attempts to better understand why individuals with these inherited mutations acquire breast (and ovarian) cancer and

Advances in CANCER RESEARCH 0065-230X/02 $35.00

1

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

2

Ingrid A. Hedenfalk et al.

to potentially develop better treatment strategies. The advent of tools such as cDNA microarrays has enabled researchers to study global gene expression patterns in, for example, primary tumors, thus providing more comprehensive overviews of tumor development and progression. Our recent study (Hedenfalk et al., 2001) strongly supports the principle that genomic approaches to classiﬁcation of hereditary breast cancers are possible, and that further studies will likely identify the most signiﬁcant genes that discriminate between subgroups and may inﬂuence prognosis and treatment. A large number of hereditary breast cancer cases cannot be accounted for by mutations in these two genes and are believed to be due to as yet unidentiﬁed breast cancer predisposition genes (BRCAx). Subclassiﬁcation of these non-BRCA1/2 breast cancers using cDNA microarray-based gene expression proﬁling, followed by linkage analysis and/or investigation of genomic alterations, may help in the recognition of novel breast cancer predisposition loci. To summarize, gene expression-based analysis of hereditary breast cancer can potentially be used for classiﬁcation purposes, as well as to expand upon our knowledge of differences between different forms of hereditary breast cancer. Initial studies indicate that a patient’s genotype does in fact leave an identiﬁable trace on her/his cancer’s gene expression proﬁle. C 2002 Academic Press.

I. INTRODUCTION Breast cancer is one of the most common malignancies affecting women in the Western world today, the lifetime risk being approximately 10% (Casey, 1997). Breast cancer is both genetically and histopathologically heterogeneous, and the mechanism(s) underlying breast cancer development remains largely unknown. Approximately 5–10% of breast cancers are of hereditary origin, and two major breast cancer susceptibility genes have been identiﬁed to date, BRCA1 and BRCA2. These two genes were initially proposed to be responsible for the majority of inherited breast cancer (Easton, 1999; Miki et al., 1994; Wooster et al., 1995), but more recent population-based studies suggest that they account for a far smaller portion of familial breast cancer, with considerable variation between different populations (Szabo and King, 1997). Presumably, additional BRCA genes with high-penetrance alleles may exist (Kainu et al., 2000), but familial breast cancer may also be due to low-penetrance or recessively acting susceptibility alleles (Cui et al., 2001). Large-scale gene expression studies using microarrays have come to play an important role in our effort to better understand diseases such as cancer. First, microarrays can be used to subclassify tumors into homogeneous entities based on gene expression proﬁles. Second, genome-wide expression data can help us further characterize the biology of these “new” subgroups. Finally, microarray experiments can aid in the search for new therapeutic targets and in the identiﬁcation of novel diagnostic markers.

Gene Expression in Inherited Breast Cancer

3

II. EPIDEMIOLOGY OF FAMILIAL BREAST CANCER Germline mutations in BRCA1 have been identiﬁed in 15–45% of women with a strong family history of breast cancer and in 60–80% of women with a family history of both breast and ovarian cancer (Couch et al., 1997; Narod et al., 1995b; Peto et al., 1999). The lifetime risk for breast cancer is 60–80% in females carrying a BRCA1 mutation, although penetrance estimates vary depending on the study population (Easton et al., 1993; Struewing et al., 1996). Lifetime ovarian cancer risks are approximately 20–40% (Easton et al., 1995; Struewing et al., 1997) in BRCA1 mutation carriers, and, to a much lesser extent, males have an increased risk of prostate cancer (Ford et al., 1994). A correlation between early onset of disease and bilateral breast cancer, and family history has been shown. The lifetime breast cancer risk for BRCA2 mutation carriers is estimated to be 60–85%, and the lifetime ovarian cancer risk is approximately 10–20% (Easton et al., 1997; Ford et al., 1998). Male BRCA2 mutation carriers are also at increased risk of developing breast cancer, with a lifetime risk of 6% (Breast Cancer Linkage Consortium (BCLC), 1999). Moreover, BRCA2 mutations may also be associated with an increased risk for prostate, pancreas, colon, gall bladder, bile duct, and stomach cancers, as well as malignant melanoma (BCLC, 1999).

III. THE BRCA1 AND BRCA2 GENES In 1990 Hall and colleagues reported linkage of early-onset breast cancer families to chromosome 17q12 (Hall et al., 1990). In 1994, BRCA1 was cloned by Miki and colleagues, as they constructed a transcriptional map of a 600 kb region at 17q12, ﬁnding mutations that segregated with 17q-linked susceptibility for breast and ovarian cancer (Miki et al., 1994). The search for breast cancer susceptibility genes continued, as it was realized that only approximately 45% of families with multiple cases of early-onset breast cancer showed evidence of linkage to BRCA1. In 1994 Wooster and colleagues performed genetic linkage analysis on families with multiple cases of early-onset breast cancer, but without evidence of linkage to BRCA1 (Wooster et al., 1994). Cosegregation of disease with chromosome 13q markers was found, and in 1995 BRCA2 was identiﬁed (Wooster et al., 1995). The BRCA1 and BRCA2 genes are thought to account for the majority of breast and ovarian cancer families (Narod et al., 1995a,b). Nonetheless, despite the considerable variation in the contribution to breast cancer from BRCA1 and BRCA2 in different populations, it remains evident that additional breast cancer susceptibility genes are still to be identiﬁed.

4

Ingrid A. Hedenfalk et al.

A. Gene Structures and Mutation Spectra BRCA1 is a large gene spread over 80 kb of genomic DNA composed of 22 coding exons that are transcribed into a 7.8-kb mRNA encoding a protein containing 1863 amino acids (Miki et al., 1994; Smith et al., 1996) (Fig. 1; see color insert). The approximate molecular mass of the BRCA1 protein is 220 kDa. The BRCA1 gene bears no homology with other genes, with the exception of a RING ﬁnger motif at the amino-terminal end. In other proteins, such a motif has been shown to interact with nucleic acids and to form protein complexes, suggesting a role for BRCA1 in transcription. In addition, there is a nuclear localization sequence (NLS) in exon 11, and a conserved acidic carboxy terminus, the BRCT (BRCA1 carboxyl-terminal) domain. To date, more than 600 different mutations in the BRCA1 gene have been reported (Breast Cancer Information Core: http://www.nhgri.nih.gov/Intramural research/Lab transfer/Bic/). The majority of these are frameshift or nonsense mutations located throughout the gene and typically result in premature translation termination; in the most subtle form merely the last 11 residues of the protein are lost (Breast Cancer Information Core, see above). Many tumor-associated point mutations in BRCA1 are found in conserved domains such as the RING ﬁnger and the BRCT domain (Irminger-Finger et al., 1999). However, the majority of missense variants reported cannot readily be distinguished as either disease-associated mutations or benign polymorphisms, posing a very relevant problem in genetic counseling. Several BRCA1 founder mutations have been identiﬁed; the two most common are 185delAG and 5382insC, which account for approximately 10% of all the mutations seen in BRCA1 (Couch and Weber, 1996). Nevertheless, mutations span the whole BRCA1 gene, a large proportion of which appear in exon 11, which constitutes 60% of the gene. Because of this fact and the size of the gene, mutation screening is both time-consuming and laborious. Traditionally, mutation screening entails the use of the protein truncation test (PTT), single-strand conformational polymorphism (SSCP) analysis, and direct sequencing of the coding region to pinpoint the mutation. One shortcoming of this traditional approach is that large rearrangements may not be detected, and it has been suggested that as many as 30% of mutations in the BRCA1 gene are undetected by standard mutation detection methods (Unger et al., 2000). Consequently, a proportion of the families that initially test negative for BRCA1 and BRCA2 mutations may do so because of cryptic mutations in these genes that are not detectable by conventional PCR-based methods. BRCA2, like BRCA1, is a large gene, consisting of 27 exons that encode a transcript of approximately 12 kb, contained within 70 kb of genomic sequence (Fig. 1). The BRCA2 protein consists of 3418 amino acids, with an estimated molecular mass of 384 kDa (Wooster et al., 1995). Also in common

Gene Expression in Inherited Breast Cancer

5

with BRCA1, BRCA2 shows no homology to other known proteins and contains no previously deﬁned functional domains. There are eight copies of a 30- to 80-amino-acid repeat (BRC repeats) that are present within exon 11 of the gene (Bignell et al., 1997). BRCA2 mutations span the whole coding region of the gene, and most of these mutations cause premature protein truncation, leading to loss of protein function (Tavtigian et al., 1996). To date, more than 250 mutations have been found (Breast Cancer Information Core, see above). No mutation hotspots have been identiﬁed so far.

B. Normal Regulation and Expression of BRCA1 and BRCA2 Both BRCA1 and BRCA2 are ubiquitously expressed, with the highest levels found in thymus and testis (Miki et al., 1994). BRCA1 and BRCA2 are required for proliferation in early embryogenesis and are up-regulated with the proliferation of breast epithelial cells during puberty and pregnancy (Rajan et al., 1997). Estrogen levels are high in both breast and ovarian tissue during these phases, suggesting that estrogen might stimulate this expression. In normal cells BRCA1 and BRCA2 are nuclear proteins (Bertwistle et al., 1997; Scully et al., 1996). The expression of BRCA1 and BRCA2 increases in late G1 phase of the cell cycle (Gudas et al., 1996; Wang et al., 1997). In mitotic cells, BRCA1, BRCA2, and RAD51 interact and colocalize in a punctate pattern in the nucleus during the S phase of the cell cycle (Chen et al., 1998; Scully et al., 1997). BRCA1 function is regulated by phosphorylation; it is hyperphosphorylated during late G1 and S phase, and dephosphorylated in M phase (Ruffner et al., 1999). Id4 (inhibitor of DNA binding 4) has been shown to negatively regulate BRCA1 (Beger et al., 2001). Overexpression of Id4 and concomitant reduction of BRCA1 expression are associated with anchorage-independent growth. Interestingly, estrogen reduces Id4 expression, hence increasing the expression of BRCA1. Conversely, estrogen receptor (ER) negative cells may overexpress Id4, with consequent reduction in BRCA1 expression. In addition, NF-κB has been shown to up-regulate the expression of BRCA2 by binding the BRCA2 promoter (Wu et al., 2000). Studies of mammalian cells deﬁcient in BRCA1 have suggested that it is involved in DNA double-strand break repair, transcription-coupled repair, and cell cycle control, all of which are important for maintaining genomic stability (reviewed by Deng and Scott, 2000).

C. Functions of the BRCA1 and BRCA2 Genes Both BRCA1 and BRCA2 encode large, multifunctional proteins, and both function as tumor suppressor genes. BRCA1 and BRCA2 proteins are

6

Ingrid A. Hedenfalk et al.

thought to be involved in two main fundamental cellular processes—DNA damage repair and transcriptional regulation (reviewed by Monteiro, 2000; Scully and Livingston, 2000; Wang et al., 2000; Welcsh et al., 2000; Zheng et al., 2000a). In addition, chromatin remodeling functions have been attributed to both BRCA1 and BRCA2 (reviewed by Irminger-Finger et al., 1999; Welcsh et al., 2000). A schematic overview of the roles of BRCA1 and BRCA2 is provided in Fig. 2 (see color insert).

1. TRANSCRIPTIONAL REGULATION AND CHROMATIN REMODELING BRCA1 contains several functional domains that interact directly or indirectly with a variety of molecules, including tumor suppressors, oncogenes, DNA damage repair proteins, cell cycle regulators, transcriptional activators, and repressors. In support of a role for BRCA1 in transcription, the C-terminal domain of BRCA1 (BRCT) interacts with RNA polymerase II subunits hRPB2 and hRPB10α, as well as with components of the RNA polymerase holoenzyme, including RNA helicase A, CBP/p300, and the BRG1 subunit of SWI/SNF (Bochar et al., 2000; Irminger-Finger et al., 1999; Pao et al., 2000). More N-terminal domains of BRCA1 interact with sequence speciﬁc transcription factors, such as p53, c-myc, STAT1, ERα, cAMP-dependent transcription factor-1 (ATF1), and a zinc ﬁnger/KRABdomain protein, ZBRK1 (Fan et al., 2001; Houvras et al., 2000; Monteiro, 2000; Scully and Livingston, 2000; Welcsh et al., 2000; Zheng et al., 2000b). Thus, BRCA1 may serve as a coactivator and bridging factor to RNA polymerase II holoenzyme, thereby altering the expression of target genes, such as p21 and GADD45, involved in cell cycle arrest. Via its BRCT domains, BRCA1 also binds to transcriptional repressor proteins CtIP/CtBP, pRB, and the histone deacetylases RbAp46, RbAp48, HDAC1, and HDAC2 (Chen et al., 2001; Deng and Brodie, 2000; Yarden and Brody, 1999; Yu and Baer, 2000). Taken together, these data suggest that BRCA1 may have either a positive or a negative modulator effect on transcription, depending on the context. Target genes with a ZBRK1 motif, potentially being repressed by BRCA1, include Bax, TopoIIa, and TIMP-1/2 (Zheng et al., 2000b). An additional putative transcriptional activation domain, AD1, containing a coiled-coil motif, has been mapped to a region close to the BRCT domains of BRCA1 (Hu et al., 2000). In vitro transactivation assays suggest a role for an N-terminal region of BRCA2 in transcriptional regulation, and another N-terminal region of BRCA2 interacts with P/CAF (Scully and Livingston, 2000; Welcsh et al., 2000). BRCA2 may regulate transcription through recruitment of the histone-acetyltransferase activity of the P/CAF coactivator. Cancer-predisposing mutations in BRCA1 and BRCA2 that abolish

Gene Expression in Inherited Breast Cancer

7

transcriptional activation also prevent chromatin remodeling, presumably with a direct effect on DNA replication and repair processes.

2. DNA REPAIR BRCA1 binds to BRCA2, p53, RAD51, and many other proteins involved in cell cycling and DNA damage response (Scully et al., 1997; Scully and Livingston, 2000). BRCA1 becomes phosphorylated at critical serine/ threonine residues by the ATM (ataxia telangiectasia mutated) and ATMrelated kinase ATR proteins in response to DNA damage (Cortez et al., 1999). BRCA1 phosphorylation in response to double-strand breaks induced by ionizing radiation (IR) may also be controlled by ATM via CHK2 (checkpoint kinase 2) (Lee et al., 2000). ATM also phosphorylates CtIP, which dissociates from BRCA1, allowing activation of GADD45 to take place (Li et al., 2000; Scully and Livingston, 2000; Welcsh et al., 2000). The activity of ATR is ATM-independent and is also induced by DNA damage caused by UV light and hydroxyurea, suggesting that ATM- and ATRinduced BRCA1 activation is triggered by different types of DNA damage (Scully and Livingston, 2000). In undamaged cells BRCA1 and BRCA2 colocalize with RAD51 in nuclear foci during the S and G2 phases of the cell cycle. The interaction between RAD51 and BRCA2 is directly mediated by six of the eight central BRC repeats of BRCA2, whereas the RAD51–BRCA1 interaction may be more indirect and possibly is mediated by BRCA2. RAD51 has a known role in double-strand break (DSB) repair in both pro- and eukaryotic cells by promoting joint molecule formation and strand exchange between homologous duplex DNA. RAD51 also forms the synaptonemal complexes during meiotic homologous recombination (HR). BRCA1 and BRCA2 have been shown to participate in both of these processes (Monteiro, 2000; Scully and Livingston, 2000; Welcsh et al., 2000), indicating involvement in recombination-mediated repair of double-stranded breaks and the maintenance of chromosome integrity (Chen et al., 1999). Upon DNA damage hyperphosphorylated BRCA1 relocalizes, together with RAD51, BRCA2, and additional components, to sites of DNA synthesis (as shown by PCNA staining), and presumably at stalled replication forks. BRCA1 is also part of the RAD50–MRE11–NBS1/p95 complex, an essential component of recombination-mediated repair of DNA double-strand breaks (Zhong et al., 1999). NBS1/p95 is phosphorylated by ATM in response to IR, and the RAD50-MRE11-NBS1/p95 complex is responsible for the end processing of DNA-DSBs that precedes both nonhomologous end joining (NHEJ) and HR. It is possible that BRCA1 may couple the RAD50– MRE11–NBS1/p95 associated end processing and Rad51-induced strand exchange during HR. It has been shown that BRCA1 binds directly to DNA,

8

Ingrid A. Hedenfalk et al.

thereby inhibiting the nucleolytic activity of the RAD50–MRE11–NBS1/p95 complex, an enzyme implicated in numerous aspects of double-strand break repair (Paull et al., 2001). BRCA1 may also function as a coordinator of an even larger BRCA1-associated genome surveillance complex (BASC), including additional tumor suppressor and DNA repair proteins such as MSH2, MSH6, MLH1, as well as the Bloom syndrome gene (BS), RecQ type DNA helicase (Monteiro, 2000; Scully and Livingston, 2000; Welcsh et al., 2000). BRCA1 has also been found to interact with one of the Fanconi anemia genes, FANCD2 (Garcia-Higuera et al., 2001). BRCA1 may play an important general role in the maintenance of genome integrity during DNA synthesis, acting directly downstream of DNA damage sensors and checkpoint genes in coordinating the assembly of DNA repair complexes. These repair activities may in large be HR-related, and sister chromatid recombination constitutes one process to execute DNA repair at persistent single-strand DNA tracts and stalled replication forks (Scully et al., 2000). RAD51 is a key protein in HR, and BRCA2 may serve as a scaffold for regulation of RAD51-induced nucleoprotein ﬁlaments, as well as for its nuclear localization (Davies et al., 2001). An additional role for BRCA1 in transcription-coupled repair (TCR), speciﬁcally of oxidative DNA damage, has also been suggested (Gowen et al., 1998). TCR requires an active RNA polymerase II, which is compatible with the interaction between BRCA1 BRCT domains and several components of the core and holoenzyme (Monteiro, 2000; Scully and Livingston, 2000; Welcsh et al., 2000). Removal of oxidative DNA damage requires excision repair proteins CSA and CSB of the Cockayne syndrome, the XPG product of the xeroderma pigmentosum syndrome (XP), and the mismatch repair protein MSH2 (Gowen et al., 1998). BARD1, a RING ﬁnger and BRCT domain protein that interacts with the RING ﬁnger of BRCA1, is also present in the BRCA1–BRCA2–RAD51 nuclear complex (Monteiro, 2000; Scully and Livingston, 2000; Welcsh et al., 2000). The BRCA1–BARD1 interaction is mediated by regions adjacent to the RING ﬁngers, thus forming a RING ﬁnger heterodimer with ubiquitine activity (Hashizume et al., 2001). A disease-associated BRCA1 mutation, C61G, located in the RING ﬁnger abolishes the polyubiquitination ability of BRCA1–BARD1. The substrates targeted for degradation are still unknown, but could be nuclear proteins involved in DNA damage repair and/or transcription (Hashizume et al., 2001). BAP1, another BRCA1-associated protein that binds to the RING ﬁnger, encodes a ubiquitin C-terminal hydrolase or thiol protease that catalyzes proteolytic processing of ubiquitin (Scully and Livingston, 2000; Welcsh et al., 2000). Moreover, BARD1 interacts with the mRNA polyadenylation factor CstF, which may reﬂect a link between repression of nuclear mRNA processing, DNA repair, and tumor progression (Kleiman and Manley, 2001).

Gene Expression in Inherited Breast Cancer

9

3. CELL CYCLE CHECKPOINT FUNCTIONS AND CENTROSOME REGULATION As mentioned previously, BRCA1 and BRCA2 are expressed late in the G1 phase of the cell cycle, possibly by estrogen-dependent stimulation (Monteiro, 2000; Scully and Livingston, 2000; Welcsh et al., 2000). Cells without functional BRCA1 do not arrest at the G2/M checkpoint after DNA damage and are deﬁcient in TCR (Gowen et al., 1998; Larson et al., 1997). Mouse embryonic ﬁbroblasts (MEFs) with a homozygous deletion of BRCA1 exon 11 display normal (p53-induced) G1 arrest upon IR, but are defective in G2/M arrest, resulting in extensive chromosomal abnormalities. These cells also contain multiple centrosomes, leading to unequal chromosome segregation at mitosis, and aneuploidy. BRCA1, as well as other tumor suppressor proteins such as p53 and pRb, localizes to centrosomes, possibly via γ -tubulin, during mitosis (Hsu and White, 1998; Xu et al., 1999b), suggesting a role for BRCA1 in centrosome ampliﬁcation and the G2/M checkpoint (Scully and Livingston, 2000; Welcsh et al., 2000). Centrosome hyperampliﬁcation is frequently seen in advanced stages of breast cancer (Carroll et al., 1999). In fact, it has been suggested that BRCA1 regulates the transition between G2 and M phases through regulation of cdc25 kinase activity (Yarden et al., 2001b). In addition, it has been suggested that BRCA1 plays a role in the regulation of apoptosis (Harkin et al., 1999; Shao et al., 1996). Mice have been created with several different homozygous Brca1 or Brca2 mutations, and null mutations result in embryonic lethality for both genes (see, e.g., Gowen et al., 1996, and Sharan et al., 1997). Of interest, elimination of one p53 allele completely rescues this embryonic lethality in Brca1 null mice and restores normal mammary gland development (Xu et al., 2001). These recent ﬁndings may provide a link to understanding the mechanism for BRCA1associated breast carcinogenesis. BRCA2 has been found to interact with the DNA binding protein BRCA2associated factor 35 (BRAF35) in close association with condensed chromatin (Marmorstein et al., 2001). A role for the BRCA2–BRAF35 complex in resolving and packaging of entangled chromatin ﬁbers or maintenance of chromosome integrity throughout segregation at mitosis has been suggested. In addition, a role for the complex in DNA repair and/or recombination is possible (Marmorstein et al., 2001). BRCA2 has been shown to interact with and become phosphorylated by the mitotic checkpoint protein hBUBR1 in cells with microtubuli disruption, and BRCA2 or hBUBR1 deﬁciency could result in genomic instability (Futamura et al., 2000). Finally, an interaction has also been noted between BRCA2 and DSS1, a conserved and largely uncharacterized protein of importance for proper cell cycle completion in yeast (Marston et al., 1999).

10

Ingrid A. Hedenfalk et al.

IV. TUMOR PROGRESSION AND THE ROLE OF ESTROGEN IN BREAST CANCER DEVELOPMENT Estrogen receptor levels are low in the normal breast, but vary from woman to woman, and high levels have been directly associated with an increased risk of breast cancer (Khan et al., 1994). There are two types of estrogen receptors, α and β, and the α receptor has a higher afﬁnity for estrogen than β. The relative expression of ERα to ERβ is higher in invasive tumors than in normal breast tissue (Leygue et al., 1998), suggesting that the balance between these receptors is important for the relative risk of breast carcinogenesis. The level of expression of ERα is widely used as a marker of hormone responsiveness and efﬁcacy of treatment with antiestrogenic drugs, such as tamoxifen. Levels of ER expression in tumors show an age-dependent variation; tumors from young women are often ER negative, whereas tumors from older women and male breast cancer patients often express ER (Loman et al., 1998). It was recognized early on that tumors from BRCA1 mutation carriers were often ER negative, even when the carriers were compared with age- and stage-matched controls (Johannsson et al., 1997), whereas tumors from BRCA2 mutation carriers are often ER positive (Loman et al., 1998). Different models have been proposed to account for the differences in hormone receptor status of breast cancers (Parl, 2000). One suggests that all breast cancers are ER positive initially and gradually evolve into hormone independence, resulting in loss of ER expression. An alternative model suggests that the ER status is a basic characteristic of each tumor, not a marker of tumor progression and dedifferentiation, and possibly has its origin in speciﬁc histogenesis. Alternatively, breast tumors could be of polyclonal origin and ER positive and negative cells may coexist transiently before clonal outgrowth. Moreover, it has been shown that ER positive and negative breast tumors display distinct gene expression proﬁles, and can be readily separated even after exclusion of ER-related genes, supporting the notion that the differences between ER positive and negative breast tumors reﬂect not only differences in hormone responsiveness, but also possibly differences in histogenesis (Gruvberger et al., 2001). It has been suggested that the resting adult breast epithelium is organized into myoepithelial (basal) cells and luminal epithelial cells and that the latter become multilayered upon hormonal stimulation (Osborne, 1996). Although myoepithelial cells do not express ER, the luminal component comprises both ER positive and ER negative cells. Cytokeratins 7, 8, 18, and 19 are almost exclusively expressed by luminal cells, whereas cytokeratins 5, 13, 14, and 17 are predominantly expressed by myoepithelial cells (Ronnov-Jessen et al., 1996). It is conceivable that the breast epithelium includes several cell types of distinctive or successive lineage and that these can act as progenitors of

Gene Expression in Inherited Breast Cancer

11

different types of breast cancer. As a consequence, the differences in phenotype between BRCA1 and BRCA2 breast cancers could be related to differences in stem cell origin in terms of hormone status or genetic repertoire. BRCA1 and BRCA2 are likely to play an important role in the regulation of growth and differentiation of the mammary gland. Mice with a conditional disruption of Brca1 in breast epithelial cells display abnormal mammary morphogenesis, with smaller glands and ducts and endbuds that fail to branch out, most likely due to DNA-damage-induced growth arrest (Xu et al., 1999a). A major factor promoting development of breast cancer seems to be estrogen stimulation of mammary epithelia. Moreover, BRCA1 may also directly regulate proliferation of epithelial cells in the breast by modulating estrogen-dependent transcriptional pathways (Fan et al., 1999). Wild-type BRCA1 can suppress the ligand-dependent transcriptional activity (AF-2) of ERα (Fan et al., 2001), and mutations in BRCA1 can result in the loss of this ability, contributing to tumorigenesis in individuals with BRCA1 mutations. Perou et al. (2000) have suggested that breast cancers can be divided into four distinct subclasses: ER positive/luminal like, ER negative/basal-like, ER negative/Erb-B2 ampliﬁed, and ER negative/luminal-like (Erb-B2 negative, keratin 8 positive). In our investigation of gene expression proﬁles in hereditary breast cancers, we found that the BRCA1 tumors included in our study showed low expression of keratin 8, Erb-B2, and ER (Hedenfalk et al., 2001). This does not, however, necessarily imply a basal cell origin for BRCA1 breast tumors, as phenotypic changes may occur during tumor evolution. Indeed, it has been demonstrated that luminal cells can give rise to myoepithelial cells (Pechoux et al., 1999). Moreover, we found that a majority of BRCA1 tumors stained positively for the luminal marker MUC-1 on tissue microarrays (see Olopade et al., 2001, for a discussion). Consequently, the distinct phenotypes of BRCA1 and BRCA2 tumors could be related to different stem cell origins or responsiveness to hormonal stimuli.

V. CHARACTERISTICS OF HEREDITARY BREAST CANCERS A. Pathology and Histology of Hereditary Breast Cancer There is emerging evidence that BRCA1- and BRCA2-associated breast cancers have distinct histopathological features. A correlation between BRCA1 and high mitotic count, continuous pushing margins, lymphocyte inﬁltration, and medullary carcinoma has been shown (BCLC, 1997; Lakhani

12

Ingrid A. Hedenfalk et al.

et al., 1998). BRCA2-related breast cancers have also been associated with speciﬁc tumor types, although the BRCA2 phenotype may be more heterogeneous. In the review performed by the BCLC, both BRCA1 and BRCA2 breast cancers were associated with a high histological grade, but BRCA2 tumors had a high grade only because of decreased tubule formation, showing no difference from sporadic breast cancers in mitotic count or pleomorphism (BCLC, 1997). As mentioned previously, differences in steroid receptor levels between BRCA1 and BRCA2 breast cancers have been found in several studies. BRCA1 breast cancers have been found to most often be ER and progesterone receptor (PR) negative (Johannsson et al., 1997; Karp et al., 1997; Osin et al., 1998; Verhoog et al., 1998, 1999). In contrast, BRCA2 breast cancers, although more heterogeneous in steroid receptor levels, more often tend to be ER and PR positive (Osin et al., 1998; Verhoog et al., 1999). A study of the histological features of cancers in families not attributable to mutations in BRCA1 or BRCA2 indicated that these breast cancers differed histologically from both BRCA1 and BRCA2 breast cancers (Lakhani et al., 2000). These non-BRCA1/2 breast tumors were generally of lower grade and showed less nuclear pleomorphism and lower mitotic activity than BRCA1 and BRCA2 breast cancers. The study also suggested that non-BRCA1/2 breast cancers differ from nonfamilial breast cancers (Lakhani et al., 2000). The apparently more aggressive appearance of BRCA1 tumors could be related to their intrinsic chromosomal instability, defective DNA repair, and dysfunctional centrosome regulation. The overall high grade of BRCA2 tumors is mainly attributable to the low degree of tubule formation, suggesting a state of dedifferentiation in these tumors. The group of non-BRCA1/2 cancers clearly has a less aggressive appearance than both BRCA1 and BRCA2 tumors, implying that the underlying susceptibility genes are of a different type than BRCA1 and BRCA2.

B. Somatic Genetic Changes in Hereditary Breast Cancer The high degree of aneuploidy associated with BRCA1 and, to a lesser extent, BRCA2 tumors is compatible with a role in maintaining genomic stability. Comparative genomic hybridization (CGH) (Kallioniemi et al., 1992) has been used to characterize genomic alterations in these tumor types, and a high frequency of copy number alterations has been shown in BRCA1 and BRCA2 breast cancers, compared to sporadic cases (Tirkkonen et al., 1997). Moreover, distinct proﬁles of copy number gains and losses have been found for both BRCA1 and BRCA2 breast cancers, suggesting that progression

Gene Expression in Inherited Breast Cancer

13

of breast cancer traverses down distinct pathways in the different types of hereditary breast cancer (Kainu et al., 2000; Wistuba et al., 2000). In the study by Tirkkonen et al., loss of 5q, 4q, 4p, 2q, and 12q was found to be signiﬁcantly higher in BRCA1 tumors than in sporadic tumors. In BRCA2 tumors, loss of 13q and 6q, as well as gain of 17q23 and 20q13, was signiﬁcantly more common (Tirkkonen et al., 1997).

VI. OTHER CAUSES OF BREAST CANCER A. Syndromes with Increased Incidence of Breast Cancer Breast cancer is part of the disease spectrum in a number of multicancer syndromes of known genetic origin, such as the Li–Fraumeni syndrome, Li–Fraumeni-like syndrome, Cowden’s disease, and Peutz–Jeghers syndrome where affected individuals inherit mutations in p53, hCHK2, PTEN, and STK11/LKB1, respectively (Bell et al., 1999; Boardman et al., 1998; Liaw et al., 1997; Malkin et al., 1990). In addition, breast cancer occurs in some women who are affected with Muir–Torre syndrome and who have been found to harbor mutations in the DNA repair genes MSH2 and MLH1 (Lairmore and Norton, 1997). These syndromes however, are rare in the population and account for only a small portion of hereditary breast cancer cases. Nonetheless, this suggests that additional syndromes with other combinations of malignancies may exist.

B. Other Genes Conferring Breast Cancer Susceptibility Complete characterization of the components in the BRCA1 and BRCA2 signaling pathways is likely to uncover novel breast-cancer-predisposing genes. Breast cancer may be part of the AT (-like) syndromes in which ATM, NBS1, or MRE11 mutations have been reported (Li et al., 2000; Stewart et al., 1999; Swift et al., 1991). In addition, epidemiological studies of AT families suggest that heterozygote AT carriers may have an increased risk for developing breast cancer, although this observation remains controversial (Athma et al., 1996; Easton, 1994). Suspected disease-associated variants of BARD1 have been seen in patients with breast, ovarian, and uterine cancer (Thai et al., 1998), and somatic CBP/p300 mutations have been reported in breast cancer patients (Gayther et al., 2000). A single-nucleotide polymorphism (SNP) in the 5 UTR of RAD51 has been shown to modify the penetrance of BRCA2 mutations (Levy-Lahad et al., 2001). In addition,

14

Ingrid A. Hedenfalk et al.

the location of BRAF35 and BRG1 on chromosome 19p coincides with a commonly deleted region in ovarian cancer, and RAD50 is located in a region that is often deleted in BRCA1-associated breast cancers. A study has reported the interaction of BRCA1 with a novel protein, termed BACH1, a function that was found to be important for DNA doublestrand break repair (Cantor et al., 2001). In addition, mutations in the BACH1 gene were found in two early-onset breast cancer patients, suggesting that BACH1 might be a novel target for germline breast-cancer-inducing mutations. Moreover, a recent study reported the association between an SNP in the prohibitin gene and breast cancer in women with a ﬁrst-degree relative with the disease (Jupe et al., 2001).

C. Low-Penetrance Susceptibility Genes Low-penetrance susceptibility genes, or “modiﬁer genes,” are deﬁned as polymorphic genes with speciﬁc alleles that are associated with an altered risk for disease susceptibility. Mutations in these low-penetrance genes might be relatively common in the general population; therefore, although each variant may be associated with only a small increased risk for breast cancer in an individual, the risk in the population as a whole might be high. Based on the apparent variability in breast cancer risk in carriers of BRCA1 and BRCA2 mutations, and the notion that genes that affect breast cancer risk in the general population may also presumably affect breast cancer risk in BRCA1 and BRCA2 mutation carriers, a number of studies have evaluated variants in candidate genes looking for modiﬁers of penetrance. Modiﬁer genes mediate a low or moderate increase in breast cancer risk. A number of candidate genetic variants have been identiﬁed in association studies to be associated with breast cancer risk. These include genes involved in steroid metabolism pathways, genes involved in metabolism of exogenous carcinogens, DNA repair pathways, and immunomodulatory pathways (Dunning et al., 1999). These variant alleles are, however, only associated with risks of approximately 1.5-fold and are predicted to account for only a few percent of breast cancer incidence. A number of polymorphisms in CYP1A1, a gene that encodes aryl hydrocarbon hydroxylase, which catalyzes the conversion of estradiol to hydroxylated estrogen, have been investigated in relation to their association with breast cancer risk (Crofts et al., 1994). Alterations in the activity of CYP1A1 could lead to changes in the levels of estrogen, which could affect breast cancer risk. Other members of the cytochrome P-450 family that have been investigated in association with breast cancer risk include CYP2D6, CYP2E1, and CYP1 (Rebbeck, 1999). The glutathione S-transferases (GSTs, e.g., GSTM1, GSTT1) constitute a family of genes that encode for enzymes that catalyze the conjunction of

Gene Expression in Inherited Breast Cancer

15

reactive chemical intermediates to soluble glutathione conjugates to facilitate clearance. Since these enzymes metabolize environmental carcinogens, there has been interest in determining if the inability to metabolize exogenous chemicals by way of GSTs may increase breast cancer risk, but this remains to be resolved. The N-acetyltransferase genes NAT1 and NAT2 are also important in the metabolism of carcinogens. Polymorphisms in these genes are associated with an altered rate of metabolism of carcinogens, such that variant alleles result in a slow acetylator phenotype. Findings by Rebbeck et al. (1997) suggest that BRCA1 mutation carriers who smoke are at an increased risk for breast cancer if they also are slow acetylators, possibly due to an association between smoking and altered steroid hormone metabolism. An association between age at onset of breast cancer in BRCA1 mutation carriers and length of the polymorphic androgen receptor (AR) CAG repeat has been suggested (Rebbeck et al., 1999). AR alleles containing longer CAG repeats are associated with a decreased ability to activate androgenresponsive genes. Together with the ﬁnding that BRCA1 interacts physically with and is a coactivator of the AR promoter (Park et al., 2000), this provides evidence that allelic variation in AR may affect breast cancer penetrance in BRCA1 mutation carriers. Similarly, a correlation between glutamine repeat lengths in the AIB1 gene and breast cancer risk in women with BRCA1 mutations has been suggested, but no evidence for such a correlation was found in a large study of breast cancer patients (Haiman et al., 2000). It remains possible, however, that the AIB1 genotype may be involved in breast cancer risk in individuals highly predisposed to breast cancer. Other allelic variants that have been associated with an increased breast cancer risk include ATM (mutated in ataxia telangiectasia), ERCC (excision-repair cross-complementation) family members, BRCA1-associated RING domain-1 (BARD1), and the 17β-hydroxysteroid dehydrogenase 1 (HSD-17B1) (Feigelson et al., 2001; Rebbeck, 1999).

D. The Search for BRCA3 Several genomic regions have been suggested as candidate loci for additional breast cancer susceptibility genes, but they remain to be conﬁrmed in other studies. Chromosome 8p has been proposed to harbor a breast cancer predisposition gene (Seitz et al., 1997), although no evidence of linkage was found in a subsequent study, suggesting that if a breast cancer susceptibility exists at this locus it accounts for only a very small proportion of familial breast cancer (Rahman et al., 2000). Kainu et al. (2000) found a high frequency of deletions at 13q21, more distal than the location of BRCA2 and Rb, in non-BRCA1/2 inherited tumors, suggesting the presence of an additional tumor suppressor gene. However, evidence against this locus has been based

16

Ingrid A. Hedenfalk et al.

on a collection of 119 families (Thompson et al., 2001), emphasizing genetic heterogeneity and the need to cluster families prior to linkage analysis.

VII. GENE EXPRESSION ANALYSIS Until recently, the approach to understanding the molecular basis of complex diseases such as cancer was to study the behavior of individual genes one at a time. The recent development of two powerful tools, microarrays and serial analysis of gene expression (SAGE), to determine the expression patterns of thousands of genes simultaneously enables scientists to study overall gene expression patterns, thereby revealing global gene expression proﬁles. There are two types of microarrays, those containing arrayed cDNA clones of approximately 500–2000 kb in length (Schena et al., 1995) and oligonucleotide arrays where the arrayed probes consist of 20- to 80-mer oligonucleotide fragments (Lockhart et al., 1996). The principle of SAGE (Velculescu et al., 1995) is based on the generation of a library of short oligonucleotide fragments (SAGE tags, typically 9–10 bp) and concatenation of these sequence tags, followed by serial sequencing of multiple tags within a clone. One difﬁculty in the study of BRCA1- and BRCA2-induced breast cancer is the limited availability of in vitro model systems. Only one established breast cancer cell line with a disease-causing BRCA1 mutation has been established to date, HCC1937 (Tomlinson et al., 1998), and no breast cancer cell lines with BRCA2 mutations have been described. The HCC1937 cell line has been used to study BRCA1-dependent global gene expression patterns upon DNA damage, with results that support the role of BRCA1 in chromatin remodeling and DNA repair and replication (Yarden et al., 2001a). Because of the differences in histopathology, genomic alterations, and steroid receptor levels, as well as in function of the genes, it seems likely that BRCA1 and BRCA2 breast cancers might display identiﬁable differences at the gene expression level. To address this possibility, we have analyzed gene expression levels in primary breast tumors from BRCA1 and BRCA2 mutation carriers, as well as sporadic tumors using cDNA microarrays containing approximately 6500 cDNA clones (Hedenfalk et al., 2001).

A. Microarray Technique The use of cDNA microarrays to study gene expression patterns in cancer was ﬁrst described by DeRisi et al. (1996). cDNA microarrays offer a systematic method for performing very extensive gene expression proﬁling

Gene Expression in Inherited Breast Cancer

17

within single cancer specimens. The technology is based on robotic spotting of thousands of cDNA probes onto glass microscope slides in a high-density manner (Fig. 3; see color insert). Fluorescently labeled tumor cDNA is hybridized onto the array together with a differentially labeled reference cDNA. The ratio of the two ﬂuorescence intensities at each spot indicates the relative expression of that gene within the tumor and can be compared with the expression ratios for all other tumors analyzed. cDNA microarray analysis is a high-throughput technology and makes possible molecular classiﬁcation of, for example, cancers. Moreover, this technology will provide researchers with new possibilities for identifying prognostic markers as well as targets for treatment. The advent of microarray technology has made possible genome-wide expression proﬁling of diseases such as cancer. It has been proposed that a distinct cancer taxonomy can be identiﬁed by thus analyzing global gene expression patterns, and, to date, classiﬁcation of rhabdomyosarcoma (Khan et al., 1998), sporadic breast cancer (Perou et al., 1999), leukemia (Golub et al., 1999), lymphoma (Alizadeh et al., 2000), and melanoma (Bittner et al., 2000), as well as hereditary breast cancer (Hedenfalk et al., 2001) and, recently, different forms of childhood cancer (Khan et al., 2001), has been performed.

B. Data Analysis The unprecedented quantity of data on gene expression patterns that is generated by microarray experiments has led to a need for extensive computational tools for analyzing the results. A number of available computational tools have been used to analyze gene expression patterns (see, e.g., Quackenbush, 2001, and Brazma and Vilo, 2000, for reviews). It is no overstatement to say that the methods used to analyze gene expression patterns can have an inﬂuence on the interpretation of the results. It should be clear that the choice of an appropriate algorithm is dependent on the biological question explored and therefore is an integral part of the experimental design. Here we will limit ourselves to the discussion of algorithms primarily in the context of gene expression analysis of cancer. The methods can be separated into two groups: unsupervised and supervised. In unsupervised methods the gene expression patterns are grouped solely according to the expression data. If one has some previous information or prejudice about which samples or genes that are expected to group together, this information can be utilized in a supervised method. Many of the algorithms used to analyze microarray data are based on the pairwise comparison of expression patterns of either genes or samples. This is addressed by mathematically deﬁning a distance between genes or samples

18

Ingrid A. Hedenfalk et al.

in “expression space.” Clustering algorithms group samples or genes based on their separation in expression space, as given by the distance metric. It should be noted that different choices of distance metric will place different objects in different clusters (Quackenbush, 2001). One of the most commonly used tools for displaying large data sets generated from microarray experiments is two-dimensional hierarchical clustering (Eisen et al., 1998). It is an agglomerative method in which expression proﬁles are joined to form groups, which are further joined until completion, and a single hierarchical tree is formed. The groups are joined based on their distance in expression space, such that close samples are joined early in the process, whereas more dissimilar samples are added to more distant nodes in the tree. Genes are grouped independent of samples into a separate tree. One advantage with hierarchical clustering is that the results are simply visualized as a tree. In addition, once the genes and samples are sorted in their respective trees, a color matrix is commonly used to display the variation in gene expression across all samples and genes, allowing visual inspection of gene expression patterns. Hierarchical clustering has, for example, been used to investigate molecular proﬁles of human breast tumors (Perou et al., 2000), to classify diffuse large B-cell lymphoma (Alizadeh et al., 2000), and to distinguish colon adenocarcinomas from adenomas and paired normal colon tissues (Notterman et al., 2001). Another unsupervised approach used to ﬁnd groups is self organizing maps (SOMs) (Kohonen, 2001), a neural-network-based divisive clustering approach in which samples are partitioned into a predeﬁned number of clusters. Compared to other methods that group samples into a set of clusters, such as k-means clustering (Bishop, 1995), SOMs have the additional feature that the clusters are presented in a geometric conﬁguration, typically a two-dimensional grid. The two-dimensional grid is ordered so that clusters containing samples that are similar to one another are located close on the grid. In this way, one achieves not only a partitioning of the samples but also an indication of relationships between the clusters. SOMs have been used to cluster both genes and samples. Tamayo et al. (1999) used SOMs to cluster genes in experiments designed to investigate hematopoietic differentiation, and Golub et al. (1999) used SOMs to classify leukemias. Since the number of genes measured is very large, one cannot visualize samples in expression space directly. One way to reduce the dimensionality of the samples to allow them to be shown in a lower dimensional space, which can be more easily visualized, is multidimensional scaling (MDS) analysis. In MDS the samples are placed in a lower dimensional space with the objective of preserving the distance between samples in expression space as well as possible. Standard MDS is unsupervised and the resulting low-dimension representation is dependent only on the distances between samples. MDS has been utilized to categorize rhabdomyosarcoma (Khan et al., 1998). To be

Gene Expression in Inherited Breast Cancer

19

able to make use of previous knowledge about groups of samples, MDS has been extended to a supervised method, in which genes are weighted according to their contribution to separating the samples into the clusters expected from previous information (Bittner et al., 2000). Based on this weighted gene analysis, the genes are ranked, and the MDS can be redone using a smaller number of top-ranked genes. This supervised MDS analysis has been applied to melanoma (Bittner et al., 2000), as well as to the investigation of gene expression proﬁles in BRCA1 and BRCA2 tumors (Hedenfalk et al., 2001). Another approach to visualizing samples or genes is principal component analysis (PCA) (Joliffe, 1986), which is a mathematical technique that reduces the effective dimension of gene expression space without signiﬁcant loss of information and ﬁnds the view of the data in a lower dimensional space which best separates the data. There are many supervised approaches to investigating, gene by gene, which genes individually are good discriminators for a given grouping of samples. A standard t test has been used to identify genes that discriminate BRCA1 from BRCA2 tumors (Hedenfalk et al., 2001). To identify genes that discriminate acute myeloid leukemia from acute lymphoblastic leukemia, Golub et al. (1999) used a signal-to-noise ratio that was designed to ﬁnd genes that on average were expressed differently in the two groups, but also had a small variation of expression within each group. Another approach is to ﬁnd good discriminator genes using the total number of misclassiﬁcations (TNoM) score, which, based on a threshold expression value, measures the number of misclassiﬁed samples for each gene (Ben-Dor et al., 2000). The TNoM score has been used to analyze melanoma (Bittner et al., 2000) and to distinguish BRCA1 from BRCA2 tumors (Hedenfalk et al., 2001). The t test, signal-to-noise ratio, and TNoM score have all been used to build classiﬁers to classify samples. The idea is to extract good discriminator genes for a given grouping of a set of samples and then use these genes to classify additional samples. These methods are all used to rank genes, but how many (if any) of the top-ranked genes are signiﬁcantly associated with distinguishing the given cancer groups remains to be evaluated. The common approach to this aim is random permutation tests. In these tests, one randomly permutes which group each sample belongs to, and for each random grouping of samples the discriminatory power for each gene is calculated. This way one can assess whether the discriminatory power of a gene is signiﬁcantly associated with the original grouping of samples one is interested in. An appealing feature of the TNoM score is that the signiﬁcance of a score can be assessed analytically and the random permutations do not have to be explicitly performed. Another class of supervised methods, which includes supervised artiﬁcial neural networks (ANNs) (Bishop, 1995) and support vector machines (SVMs) (Cristianini and Shawe-Taylor, 2000), is based on ﬁnding a model,

20

Ingrid A. Hedenfalk et al.

deﬁned by a set of parameters, which maps a sample from expression space to a given phenotype. In contrast to methods that evaluate each gene independently, these methods can potentially ﬁnd more complex patterns of gene expression that are associated with the phenotypes of interest. A socalled training set of samples is used to calibrate the models. The models are presented with the correct classiﬁcation of the training samples and this information is used to tune the parameters of the models to optimally classify the training samples. The calibrated models are subsequently used to classify additional test samples. A cross-validation procedure, in which the training set is split into two groups, with one used for calibration and one for validation, is generally used to evaluate the performance of the models. SVMs have been used to predict functional roles for uncharacterized yeast open reading frames (Brown et al., 2000), as well as to classify cancer tissue samples (Furey et al., 2000). In addition, ANNs have been used for the diagnostic classiﬁcation of cancer samples into four groups of childhood cancer (Khan et al., 2001) and to investigate the gene expression patterns associated with ER status in sporadic breast tumors (Gruvberger et al., 2001). Even though these methods can be used as “black boxes,” which classify tumor samples based on their gene expression patterns, it is important to note that one can evaluate which features or genes were important for the classiﬁcation. For ANNs a method for ranking the genes based on their importance to the classiﬁcation has been developed (Khan et al., 2001). The supervised approaches to classifying human disease states using patterns of gene expression are very promising and they can potentially have great impact on the classiﬁcation of cancer. However, the advantage of supervised methods, that is, that one can make use of previous knowledge about classes of disease states, obviously restricts their applicability to investigations where one has previous knowledge. In particular, they cannot be directly applied to ﬁnding new classes of cancer. Of note is that once new classes of cancer have been suggested by unsupervised class discovery methods they can be veriﬁed using supervised classiﬁcation schemes. The ﬁeld of class discovery based on gene expression patterns is still in an early stage and great activity is directed to developing methods for this application. Hierarchical clustering has been used to discover two molecularly distinct types of diffuse large B-cell lymphoma in which the patients in the two subgroups showed signiﬁcant differences in overall survival (Alizadeh et al., 2000), and to categorize breast cancer into new subtypes (Perou et al., 2000). MDS weighted gene analysis has been used together with hierarchical clustering and a nonhierarchical clustering algorithm to separate cutaneous malignant melanoma into two classes that differ in their invasive properties (Bittner et al., 2000). SOMs were used to show that the already known categorization of lymphoma into acute myeloid and acute lymphoblastic lymphoma could be discovered solely from gene expression patterns (Golub

Gene Expression in Inherited Breast Cancer

21

et al., 1999). Another approach is to score candidate partitions of samples and to ﬁnd the partition with the largest score. Part of the problem is to ﬁnd a good scoring function, which should be based on biological criteria. One such method is designed to ﬁnd the partition of samples that has an overabundance of genes separating the suggested groups (Ben-Dor et al., 2001). In other words, the discovered classes are those that have the largest number of discriminatory genes separating them. Regardless of which method is used to discover classes, it may be fruitful to initially obtain an estimate of the number of classes to expect from the gene expression data. For example, from the view presented by PCA the number of classes to expect can be estimated. In addition, PCA can be a powerful technique when used in conjunction with another classiﬁcation technique such as SOMs (Quackenbush, 2001). It can also be used together with supervised methods such as ANNs (Khan et al., 2001). An interesting application of class discovery methods is to ﬁnd new classes within the group of non-BRCA1/2 hereditary breast cancers, as it can be suspected that they comprise a heterogeneous entity. In this context, it may be beneﬁcial to use hybrid methods in which some previous knowledge, such as information about families or populations, is used to restrict possible new classes.

C. Gene Expression Proﬁles in BRCA1/2 Breast Cancers As mentioned previously, there are pathological as well as genomic and functional differences between different types of hereditary breast cancer that might suggest differences in gene expression patterns. Although a BRCA1derived breast cancer displays certain histopathological characteristics that may aid in its characterization as a BRCA1 tumor, these tumors do not constitute an entirely uniform group. Moreover, BRCA2 breast cancers make up a considerably more heterogeneous group. Extended knowledge of the defect(s) causing the development of breast cancer may greatly improve both treatment schemes and intervention strategies for the affected individuals. Indeed, using a microarray of approximately 6500 cDNA clones, we have shown that it was possible to subclassify BRCA1 and BRCA2 breast cancer, as well as to separate them from sporadic cases (Hedenfalk et al., 2001). Further investigation of the genes that were found to distinguish BRCA1 from BRCA2 breast cancers suggested the involvement of distinct pathways of pathogenesis in these breast tumors. Moreover, the ﬁnding of a BRCA1like tumor within the sporadic group warranted further analysis, revealing the presence of hypermethylation of the BRCA1 promoter region, causing down-regulation of BRCA1 gene expression in this tumor (Hedenfalk et al.,

22

Ingrid A. Hedenfalk et al.

2001). This implies that a somatic down-regulation of BRCA1 expression can give rise to a phenotype similar to that of germline mutations and emphasizes the signiﬁcance of BRCA1 deﬁciency in tumor development. Such silencing of the BRCA1 gene has been shown in a small fraction of sporadic breast cancers displaying a BRCA1-like phenotype (Esteller et al., 2000), and this is especially interesting in light of the fact that BRCA1 is rarely, if ever, found to be altered by somatic mutations in sporadic cases of breast cancer. This ﬁnding illustrates the power and sensitivity of gene expression proﬁling of cancers. Interestingly, it has been suggested that methylation of the BRCA1 promoter can in some instances constitute the “second hit” in tumor development (Knudson, 1971) when loss of BRCA1 is not due to loss of heterozygosity (LOH) (M. Esteller, personal communication). Because of the differential expression of steroid hormone receptors between BRCA1 and BRCA2 breast cancers, it is likely that a certain degree of the differences in gene expression levels is attributable to this fact. However, as mentioned in our study, this does not fully account for the observed differences, since separation of BRCA1 and BRCA2 tumors was possible even after removal of ER- and PR-related genes from the analysis. Moreover, the sporadic tumors included in the study did not cluster with the BRCA1 and BRCA2 tumors based on the differences in hormone receptor expression, with the exception of the ER negative sporadic case displaying hypermethylation of BRCA1 clustering with the hormone receptor negative BRCA1 tumors. Aberrant methylation of BRCA1 has been shown to be associated with ER negativity (Catteau et al., 1999); however, this tumor clustered with the BRCA1 mutated cases even upon removal of ER-related genes from the analysis. This suggests that a substantial portion of the gene expression proﬁles of BRCA1 and BRCA2 breast cancers are due to the underlying mutations in these genes. The tumor from a male BRCA2 mutation carrier displayed a gene expression proﬁle similar to that of the other BRCA2 tumors in the study, but was, upon class prediction using a small number of differentially expressed genes, classiﬁed as a non-BRCA2 tumor, suggesting slightly different properties in breast tumors arising in males and females carrying BRCA2 mutations (Hedenfalk et al., 2001). Similarly, two BRCA2 tumors with the most N-terminal mutation (i.e., causing truncation of all the RAD51 binding BRC repeats) were also misclassiﬁed in the BRCA2 positive or negative classiﬁcation (Hedenfalk et al., 2001), implying that genotype–phenotype correlations may exist. Accordingly, it has been suggested that frameshift and nonsense mutations occurring within the ovarian cancer cluster region (OCCR) of BRCA2, which largely coincides with the location of the BRC repeats and potentially results in truncating proteins lacking one or more RAD51 binding sites, are associated with a lower risk of breast cancer and higher risk of ovarian cancer than truncating mutations in the N- or C-terminal part of

Gene Expression in Inherited Breast Cancer

23

the gene (Gayther et al., 1997; Thompson and Easton, 2001). A similar correlation with disease phenotype may exist for mutations in BRCA1, where truncating mutations positioned 3 of the large exon 11 have been associated with a lower risk of ovarian cancer than mutations occurring in the 5 part of the gene (Gayther et al., 1995), and where certain amino acid substitutions in the BRCT domains could behave distinctly (Vallon-Christersson et al., 2001). One explanation for such a genotype–phenotype correlation could be that the mutant protein products retain varying degrees of wildtype BRCA1 function depending on the presence of speciﬁc domains. For instance, domains in the N-terminal part of the gene, 5 of the supposed change point, might supply important BRCA1 function in ovarian cells but not in breast epithelium, rendering these individuals more susceptible to ovarian cancer than to breast cancer. A possible means of elucidating such a genotype–phenotype correlation is to employ cDNA microarray analysis to investigate the relationship between gene expression proﬁles and the locations of mutations within the gene, the hypothesis being that the mutation site will be reﬂected in speciﬁc and unique gene expression proﬁles. Finally, one sporadic tumor was classiﬁed as BRCA2 mutation positive in the BRCA2 classiﬁcation (Hedenfalk et al., 2001). One might speculate that the BRCA2-like gene expression proﬁle of this tumor is due to epigenetic silencing, as was the case with the sporadic tumor classiﬁed as BRCA1 mutation positive. However, there has been no evidence of aberrant methylation of CpG islands within the BRCA2 promoter region in breast cancer tissues (Collins et al., 1997). The analysis of genes separating BRCA1 from BRCA2 tumors revealed a number of genes with high expression in BRCA1-associated tumors compared to BRCA2-associated tumors (Hedenfalk et al., 2001). Many of these genes are known to be induced by p53 in response to DNA damage (e.g., MSH2, MSH6, GADD34). It is, however, known that p53 is mutated in a majority of BRCA1-associated breast cancers, leading to the possibility of a p53-independent activation of DNA damage response pathways in BRCA1deﬁcient tumors. Moreover, in a previous study we found that the oncogene MYB was ampliﬁed and overexpressed in a subset of BRCA1 breast carcinomas (Kauraniemi et al., 2000), suggesting a role for this cell cycle regulator and transcription factor in the progression of some BRCA1 tumors. That study showed how identiﬁcation of genomic alterations by, for examples CGH, followed by targeted studies (e.g., with microarrays) to pinpoint the putative target gene(s) within the amplicon, can be a useful approach to recognizing potentially signiﬁcant genetic alterations in breast carcinogenesis. These studies have indicated that genomic approaches to classiﬁcation of hereditary breast cancers are possible and that further studies will likely pinpoint the most signiﬁcant genes to differentiate subgroups and inﬂuence

24

Ingrid A. Hedenfalk et al.

prognosis and treatment. To further validate and extend cDNA microarray ﬁndings, one can use tissue microarrays containing large numbers of parafﬁnembedded tissue specimens (see Kallioniemi et al., 2001, for a review). This technology allows analysis of DNA, RNA, and protein across hundreds of tissue specimens in parallel and has successfully been applied to, for example verifying genes of signiﬁcance in renal carcinomas and breast cancer (Moch ¨ et al., 1999; Barlund et al., 2000). In addition, it appears possible that more extended studies in this ﬁeld may lead to the development of a “classiﬁcation chip” containing a small number of highly differentially expressed genes that could be used in the clinical setting to rapidly screen for mutations.

D. Discovering New Classes within the BRCAx Breast Cancers cDNA microarray-based gene expression proﬁling provides a powerful tool for the elucidation of differences in tumor phenotypes and may also be used for the discovery of novel subgroups within the group of non-BRCA1/2 hereditary breast tumors. It is likely that the most efﬁcient approach to identifying additional breast cancer predisposition genes is the combination of large-scale gene expression proﬁling with positional linkage information. As mentioned previously, the non-BRCA1/2 subgroup of breast cancers appears to be part of a histologically heterogeneous group, indicating the presence of multiple underlying alterations. This fact, in addition to the possible presence of sporadic cases in these families and population heterogeneity, limits the power of traditional linkage analysis in the search for new breast cancer predisposition genes. We are therefore currently employing the strategy of initial gene expression proﬁling followed by the use of various statistical methods in an effort to subclassify these families into genetically homogeneous entities. This approach is then followed by CGH analysis to identify common regions of deletion within each subgroup and genome-wide scans to pinpoint regions of linkage. Preliminary results from genome-wide expression proﬁling of breast tumors that have tested negative for mutations in BRCA1 and BRCA2 show that the gene expression proﬁles of these tumors are distinct from those of BRCA1 and BRCA2 tumors (data not shown). This is consistent with the overall phenotype of these tumors being different from that of BRCA1 and BRCA2 tumors (Lakhani et al., 2000). Various class discovery tools are applied to these data sets to subclassify these breast tumors, and the emerging pattern has revealed that certain tumors tend to cluster together, suggesting the possibility of a common underlying genetic defect in these individuals. Tumors from related individuals tend to cluster together, implying, again, that the cause for familial clustering is genetic and not a chance clustering of breast cancer in these families (Fig. 4).

Fig. 4 Multidimensional scaling plot of BRCAx breast tumors. Individual tumors from non-BRCA1/2 mutation carriers are plotted in three-dimensional space to reveal the relatedness between tumors on a gene expression level. Filled circles in the pedigrees represent individuals affected with breast cancer. The numbers below indicate age at diagnosis. Tumors from closely related individuals (pedigrees A and B) appear close to each other in this “gene expression space,” indicating similarities in global gene expression patterns. One tumor from a family previously shown to be linked to chromosome 13q21 (Kainu et al., 2000) is shown (pedigree C), and it is believed that other tumors with similar gene expression proﬁles also may be linked to this region.

26

Ingrid A. Hedenfalk et al.

E. Copy Number and Gene Expression Analysis cDNA microarrays can also be used to analyze genomic DNA instead of RNA expression levels. As initially described by Pollack et al. (1999), CGH can be performed in an array format using the same cDNA microarrays. Copy number data can be compared with expression data to deﬁne candidate genes associated with either gain or loss of chromosomal regions. Moreover, targeted microarrays covering particular regions of interest can be constructed, as described by Monni et al. (2001), who constructed a chromosome 17q-speciﬁc microarray to analyze both copy number alterations and gene expression proﬁles within this commonly altered region in human breast cancer. One could envision the subclassiﬁcation of non-BRCA1/2 breast tumors into genetically homogeneous entities based on gene expression proﬁling and similarities in genomic alterations, followed by analysis using targeted microarrays with complete coverage of the region(s) of interest to identify the putative cancer causing gene(s). Unfortunately, the analysis of genomic alterations is currently limited to the detection of DNA ampliﬁcation because of the insufﬁcient degree of sensitivity of cDNA microarrays for detecting deletions. Moreover, based on the assumption that the underlying cause of cancer in these individuals is loss of a tumor suppressor gene, this approach is hampered by the fact that a mutation in a putative tumor suppressor gene does not necessarily result in decreased expression of the corresponding mRNA. In some instances, however, premature termination codons have been shown to initiate degradation of mutant mRNA transcripts by a mechanism termed nonsense-mediated messenger RNA decay (NMD) (Frischmeyer and Dietz, 1999). A recent report describes a strategy for identifying genes harboring nonsense mutations based on pharmacologically inhibiting the NMD pathway, resulting in stabilization of nonsense transcripts, thereby rendering detection of mutations in such genes possible (Noensie and Dietz, 2001).

VIII. CONCLUDING REMARKS Considerable effort has focused on elucidation of the function(s) of BRCA1 and BRCA2 over the past two decades: from a biological point of view to increase our understanding of these tumor suppressor proteins, and from a clinical point of view in the hope of improved treatment and prophylaxis. Two important questions remain to be answered: Why the tissue speciﬁcity of tumor development in BRCA1 and BRCA2 mutation carriers, and why are sporadic mutations in these genes so rare? A potential explanation relates to the hormonal environment of breast and ovarian cells, and possibly to the

Gene Expression in Inherited Breast Cancer

27

tissue-speciﬁc expression of as yet unidentiﬁed genes. The development of novel tools to address these questions should shed further light on the roles of BRCA1 and BRCA2 in normal development and in tumorigenesis. There have been many advances in the area of merging genomic research with the study of hereditary cancer, including breast cancer, and we should be prepared for surprises. Because of the power of cDNA microarray analysis, the future is likely to bring substantial changes to the molecular and pathological classiﬁcation of tumors. Moreover, large-scale expression analysis will likely become increasingly useful in the search for novel therapeutic targets, as well as in the establishment of new prognostic markers for disease.

ACKNOWLEDGMENTS We thank D. Leja for assistance with scientiﬁc illustrations. M.R. was supported by a postdoctoral fellowship from the Swedish Research Council.

REFERENCES Alizadeh, A. A., Eisen, M. B., Davis, R. E., Ma, C., Lossos, I. S., Rosenwald, A., Boldrick, J. C., Sabet, H., Tran, T., Yu, X., Powell, J. I., Yang, L., Marti, G. E., Moore, T., Hudson, J., Jr., Lu, L., Lewis, D. B., Tibshirani, R., Sherlock, G., Chan, W. C., Greiner, T. C., Weisenburger, D. D., Armitage, J. O., Warnke, R., Levy, R., Wilson, W., Grever, M. R., Byrd, J. C., Botstein, D., Brown, P. O., and Staudt, L. M. (2000). Nature 403, 503–511. Athma, P., Rappaport, R., and Swift, M. (1996). Cancer Genet. Cytogenet. 92, 130–134. ¨ Barlund, M., Forozan, F., Kononen, J., Bubendorf, L., Chen, Y., Bittner, M. L., Torhorst, J., Haas, P., Bucher, C., Sauter, G., Kallioniemi, O. P., and Kallioniemi, A. (2000). J. Natl. Cancer Inst. 92, 1252–1259. Beger, C., Pierce, L. N., Kruger, M., Marcusson, E. G., Robbins, J. M., Welcsh, P., Welch, P. J., Welte, K., King, M. C., Barber, J. R., and Wong–Staal, F. (2001). Proc. Natl. Acad. Sci. USA 98, 130–135. Bell, D. W., Varley, J. M., Szydlo, T. E., Kang, D. H., Wahrer, D. C., Shannon, K. E., Lubratovich, M., Verselis, S. J., Isselbacher, K. J., Fraumeni, J. F., Birch, J. M., Li, F. P., Garber, J. E., and Haber, D. A. (1999). Science 286, 2528–2531. Ben-Dor, A., Bruhn, L., Friedman, N., Nachman, I., Schummer, M., and Yakhini, Z. (2000). J. Comput. Biol. 7, 559–583. Ben-Dor, A., Friedman, N., and Yakhini, Z. (2001). In “Proceedings of the Fifth Annual International Conference on Computational Molecular Biology (RECOMB).” ACM Press. Bertwistle, D., Swift, S., Marston, N. J., Jackson, L. E., Crossland, S., Crompton, M. R., Marshall, C. J., and Ashworth, A. (1997). Cancer Res. 57, 5485–5488. Bignell, G., Micklem, G., Stratton, M. R., Ashworth, A., and Wooster, R. (1997). Hum. Mol. Genet. 6, 53–58. Bishop, C (1995). “Neural Networks for Pattern Recognition.” Oxford University Press, Oxford.

28

Ingrid A. Hedenfalk et al.

Bittner, M., Meltzer, P., Chen, Y., Jiang, Y., Seftor, E., Hendrix, M., Radmacher, M., Simon, R., Yakhini, Z., Ben-Dor, A., Sampas, N., Dougherty, E., Wang, E., Marincola, F., Gooden, C., Lueders, J., Glatfelter, A., Pollock, P., Carpten, J., Gillanders, E., Leja, D., Dietrich, K., Beaudry, C., Berens, M., Alberts, D., Sondak, V., Hayward, N., and Trent, J. (2000). Nature 406, 536–540. Boardman, L. A., Thibodeau, S. N., Schaid, D. J., Lindor, N. M., McDonnell, S. K., Burgart, L. J., Ahlquist, D. A., Podratz, K. C., Pittelkow, M., and Hartmann, L. C. (1998). Ann. Intern. Med. 128, 896–899. Bochar, D. A., Wang, L., Beniya, H., Kinev, A., Xue, Y., Lane, W. S., Wang, W., Kashanchi, F., and Shiekhattar, R. (2000). Cell 102, 257–265. Brazma, A., and Vilo, J. (2000). FEBS Lett. 480, 17–24. Breast Cancer Linkage Consortium (1997). Lancet 349, 1505–1510. Breast Cancer Linkage Consortium (1999). J. Natl. Cancer Inst. 91, 1310–1316. Brown, M. P., Grundy, W. N., Lin, D., Cristianini, N., Sugnet, C. W., Furey, T. S., Ares, M., Jr., and Haussler, D. (2000). Proc. Natl. Acad. Sci. USA 97, 262–267. Cantor, S. B., Bell, D. W., Ganesan, S., Kass, E. M., Drapkin, R., Grossman, S., Wahrer, D. C., Sgroi, D. C., Lane, W. S., Haber, D. A., and Livingston, D. M. (2001). Cell 105, 149–160. Carroll, P. E., Okuda, M., Horn, H. F., Biddinger, P., Stambrook, P. J., Gleich, L. L., Li, Y. Q., Tarapore, P., and Fukasawa, K. (1999). Oncogene 18, 1935–1944. Casey, G. (1997). Curr. Opin. Oncol. 9, 88–93. Catteau, A., Harris, W. H., Xu, C. F., and Solomon, E. (1999). Oncogene 18, 1957–1965. Chen, G. C., Guan, L. S., Yu, J. H., Li, G. C., Choi Kim, H. R., and Wang, Z. Y. (2001). Biochem. Biophys. Res. Commun. 284, 507–514. Chen, J., Silver, D. P., Walpita, D., Cantor, S. B., Gazdar, A. F., Tomlinson, G., Couch, F. J., Weber, B. L., Ashley, T., Livingston, D. M., and Scully, R. (1998). Mol. Cell. 2, 317–328. Chen, J. J., Silver, D., Cantor, S., Livingston, D. M., and Scully, R. (1999). Cancer Res. 59, 1752–1756s. Collins, N., Wooster, R., and Stratton, M. R. (1997). Br. J. Cancer 76, 1150–1156. Cortez, D., Wang, Y., Qin, J., and Elledge, S. J. (1999). Science 286, 1162–1166. Couch, F. J., and Weber, B. L. (1996). Hum. Mutat. 8, 8–18. Couch, F. J., DeShano, M. L., Blackwood, M. A., Calzone, K., Stopfer, J., Campeau, L., Ganguly, A., Rebbeck, T., and Weber, B. L. (1997). N. Engl. J. Med. 336, 1409–1415. Cristianini, N., and Shawe-Taylor, J. (2000). “An Introduction to Support Vector Machines.” Cambridge University Press, Cambridge, U.K. Crofts, F., Taioli, E., Trachman, J., Cosma, G. N., Currie, D., Toniolo, P., and Garte, S. J. (1994). Carcinogenesis 15, 2961–2963. Cui, J., Antoniou, A. C., Dite, G. S., Southey, M. C., Venter, D. J., Easton, D. F., Giles, G. G., McCredie, M. R., and Hopper, J. L. (2001). Am. J. Hum. Genet. 68, 420–431. Davies, A. A., Masson, J. Y., McIlwraith, M. J., Stasiak, A. Z., Stasiak, A., Venkitaraman, A. R., and West, S. C. (2001). Mol. Cell 7, 273–282. Deng, C. X., and Brodie, S. G. (2000). Bioessays 22, 728–737. Deng, C. X., and Scott, F. (2000). Oncogene 19, 1059–1064. DeRisi, J., Penland, L., Brown, P. O., Bittner, M. L., Meltzer, P. S., Ray, M., Chen, Y., Su, Y. A., and Trent, J. M. (1996). Nat. Genet. 14, 457–460. Dunning, A. M., Healey, C. S., Pharoah, P. D., Teare, M. D., Ponder, B. A., and Easton, D. F. (1999). Cancer. Epidemiol. Biomarkers. Prev. 8, 843–854. Easton, D. F. (1994). Int. J. Radiat. Biol. 66, S177–S182. Easton, D. F. (1999). Breast Cancer Res. 1, 14–17. Easton, D. F., Bishop, D. T., Ford, D., and Crockford, G. P. (1993). Am. J. Hum. Genet. 52, 678–701. Easton, D. F., Ford, D., and Bishop, D. T. (1995). Am. J. Hum. Genet. 56, 265–271.

Gene Expression in Inherited Breast Cancer

29

Easton, D. F., Steele, L., Fields, P., Ormiston, W., Averill, D., Daly, P. A., McManus, R., Neuhausen, S. L., Ford, D., Wooster, R., Cannon-Albright, L. A., Stratton, M. R., and Goldgar, D. E. (1997). Am. J. Hum. Genet. 61, 120–128. Eisen, M. B., Spellman, P. T., Brown, P. O., and Botstein, D. (1998). Proc. Natl. Acad. Sci. USA 95, 14863–14868. Esteller, M., Silva, J. M., Dominguez, G., Bonilla, F., Matias-Guiu, X., Lerma, E., Bussaglia, E., Prat, J., Harkes, I. C., Repasky, E. A., Gabrielson, E., Schutte, M., Baylin, S. B., and Herman, J. G. (2000). J. Natl. Cancer Inst. 92, 564–569. Fan, S., Ma, Y. X., Wang, C., Yuan, R. Q., Meng, Q., Wang, J. A., Erdos, M., Goldberg, I. D., Webb, P., Kushner, P. J., Pestell, R. G., and Rosen, E. M. (2001). Oncogene 20, 77–87. Fan, S., Wang, J., Yuan, R., Ma, Y., Meng, Q., Erdos, M. R., Pestell, R. G., Yuan, F., Auborn, K. J., Goldberg, I. D., and Rosen, E. M. (1999). Science 284, 1354–1356. Feigelson, H. S., McKean-Cowdin, R., Coetzee, G. A., Stram, D. O., Kolonel, L. N., and Henderson, B. E. (2001). Cancer Res. 61, 785–789. Ford, D., Easton, D. F., Bishop, D. T., Narod, S. A., and Goldgar, D. E. (1994). Lancet 343, 692–695. Ford, D., Easton, D. F., Stratton, M., Narod, S., Goldgar, D., Devilee, P., Bishop, D. T., Weber, B., Lenoir, G., Chang-Claude, J., Sobol, H., Teare, M. D., Struewing, J., Arason, A., Scherneck, S., Peto, J., Rebbeck, T. R., Tonin, P., Neuhausen, S., Barkardottir, R., Eyfjord, J., Lynch, H., Ponder, B. A., Gayther, S. A., Birch, J. M., Lindblom, A., Stoppa-Lyonnet, D., Bignon, Y., ˚ Hamann, U., Haites, N., Scott, R. J., Maugard, C. M., Vasen, H., Seitz, S., CannonBorg, A., Albright, L. A., Schoﬁeld, A., and Zelada-Hedman, M., and the Breast Cancer Linkage Consortium. (1998). Am. J. Hum. Genet. 62, 676–689. Frischmeyer, P. A., and Dietz, H. C. (1999). Hum. Mol. Genet. 8, 1893–1900. Furey, T. S., Cristianini, N., Duffy, N., Bednarski, D. W., Schummer, M., and Haussler, D. (2000). Bioinformatics 16, 906–914. Futamura, M., Arakawa, H., Matsuda, K., Katagiri, T., Saji, S., Miki, Y., and Nakamura, Y. (2000). Cancer Res. 60, 1531–1535. Garcia-Higuera, I., Taniguchi, T., Ganesan, S., Meyn, M. S., Timmers, C., Hejna, J., Grompe, M., and D’Andrea, A. D. (2001). Mol. Cell 7, 249–262. Gayther, S. A., Batley, S. J., Linger, L., Bannister, A., Thorpe, K., Chin, S. F., Daigo, Y., Russell, P., Wilson, A., Sowter, H. M., Delhanty, J. D., Ponder, B. A., Kouzarides, T., and Caldas, C. (2000). Nat. Genet. 24, 300–303. Gayther, S. A., Mangion, J., Russell, P., Seal, S., Barfoot, R., Ponder, B. A., Stratton, M. R., and Easton, D. (1997). Nat. Genet. 15, 103–105. Gayther, S. A., Warren, W., Mazoyer, S., Russell, P. A., Harrington, P. A., Chiano, M., Seal, S., Hamoudi, R., van Rensburg, E. J., Dunning, A. M., Love, R., Evans, G., Easton, D., Clayton, D., Stratton, M. R., and Ponder, B. A. J. (1995). Nat. Genet. 11, 428–433. Golub, T. R., Slonim, D. K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J. P., Coller, H., Loh, M. L., Downing, J. R., Caligiuri, M. A., Bloomﬁeld, C. D., and Lander, E. S. (1999). Science 286, 531–537. Gowen, L. C., Avrutskaya, A. V., Latour, A. M., Koller, B. H., and Leadon, S. A. (1998). Science 281, 1009–1012. Gowen, L. C., Johnson, B. L., Latour, A. M., Sulik, K. K., and Koller, B. H. (1996). Nat. Genet. 12, 191–194. ˚ Ferno, ¨ M., Peterson, Gruvberger, S., Ringn´er, M., Chen, Y., Panavally, S., Saal, L. H., Borg, A., C., and Meltzer, P. S. (2001). Cancer Res. 61, 5979–5984. Gudas, J. M., Li, T., Nguyen, H., Jensen, D., Rauscher III, F. J., and Cowan, K. H. (1996). Cell Growth Differ. 7, 717–723. Haiman, C. A., Hankinson, S. E., Spiegelman, D., Colditz, G. A., Willett, W. C., Speizer, F. E., Brown, M., and Hunter, D. J. (2000). Breast Cancer Res. 2, 378–385.

30

Ingrid A. Hedenfalk et al.

Hall, J. M., Lee, M. K., Newman, B., Morrow, J. E., Anderson, L. A., Huey, B., and King, M. C. (1990). Science 250, 1684–1689. Harkin, D. P., Bean, J. M., Miklos, D., Song, Y. H., Truong, V. B., Englert, C., Christians, F. C., Ellisen, L. W., Maheswaran, S., Oliner, J. D., and Haber, D. A. (1999). Cell 97, 575–586. Hashizume, R., Fukuda, M., Maeda, I., Nishikawa, H., Oyake, D., Yabuki, Y., Ogata, H., and Ohta, T. (2001). J. Biol. Chem. 276, 14537–14540. Hedenfalk, I., Duggan, D., Chen, Y., Radmacher, M., Bittner, M., Simon, R., Meltzer, P., ˚ and Trent, J. (2001). Gusterson, B., Esteller, M., Kallioniemi, O. P., Wilfond, B., Borg, A., N. Engl. J. Med. 344, 539–548. Houvras, Y., Benezra, M., Zhang, H., Manfredi, J. J., Weber, B. L., and Licht, J. D. (2000). J. Biol. Chem. 275, 36230–36237. Hsu, L.-C., and White, R. L. (1998). Proc. Natl. Acad. Sci. USA 95, 12983–12988. Hu, Y. F., Miyake, T., Ye, Q., and Li, R. (2000). J. Biol. Chem. 275, 40910–40915. Irminger-Finger, I., Siegel, B. D., and Leung, W. C. (1999). Biol. Chem. 380, 117–128. ˚ Barkardottir, R. B., Egilsson, V., and Johannsson, O. T., Idvall, I., Anderson, C., Borg, A., Olsson, H. (1997). Eur. J. Cancer 33, 362–371. Joliffe, I. T. (1986). “Principal Component Analysis.” Springer Verlag, New York. Jupe, E. R., Badgett, A. A., Neas, B. R., Craft, M. A., Mitchell, D. S., Resta, R., Mulvihill, J. J., Aston, C. E., and Thompson, L. F. (2001). Lancet 357, 1588–1589. Kainu, T., Juo, S. H., Desper, R., Schaffer, A. A., Gillanders, E., Rozenblum, E., Freas-Lutz, D., Weaver, D., Stephan, D., Bailey-Wilson, J., Kallioniemi, O. P., Tirkkonen, M., Syrjakoski, K., Kuukasjarvi, T., Koivisto, P., Karhu, R., Holli, K., Arason, A., Johannesdottir, G., Bergthorsson, J. T., Johannsdottir, H., Egilsson, V., Barkardottir, R. B., Johannsson, O., ˚ Vehmanen, Haraldsson, K., Sandberg, T., Holmberg, E., Gronberg, H., Olsson, H., Borg, A., P., Eerola, H., Heikkila, P., Pyrhonen, S., and Nevanlinna, H. (2000). Proc. Natl. Acad. Sci. USA 97, 9603–9608. Kallioniemi, A., Kallioniemi, O. P., Sudar, D., Rutovitz, D., Gray, J. W., Waldman, F., and Pinkel, D. (1992). Science 258, 818–821. Kallioniemi, O. P., Wagner, U., Kononen, J., and Sauter, G. (2001). Hum. Mol. Genet. 10, 657–662. Karp, S. E., Tonin, P. N., Begin, L. R., Martinez, J. J., Zhang, J. C., Pollak, M. N., and Foulkes, W. D. (1997). Cancer 80, 435–441. Kauraniemi, P., Hedenfalk, I., Persson, K., Duggan, D. J., Tanner, M., Johannsson, O., Olsson, ˚ (2000). Cancer Res. 60, 5323–5328. H., Trent, J. M., Isola, J., and Borg, A. Khan, J., Simon, R., Bittner, M., Chen, Y., Leighton, S. B., Pohida, T., Smith, P. D., Jiang, Y., Gooden, G. C., Trent, J. M., and Meltzer, P. S. (1998). Cancer. Res. 58, 5009–5013. Khan, J., Wei, J. S., Ringn´er, M., Saal, L. H., Ladanyi, M., Westermann, F., Berthold, F., Schwab, M., Antonescu, C. R., Peterson, C., and Meltzer, P. S. (2001). Nat. Med. 7, 673–679. Khan, S. A., Rogers, M. A., Obando, J. A., and Tamsen, A. (1994). Cancer. Res. 54, 993–997. Kleiman, F. E., and Manley, J. L. (2001). Cell 104, 743–753. Knudson, A. G., Jr. (1971). Proc. Natl. Acad. Sci. USA 68, 820–823. Kohonen, T. (2001). “Self-Organizing Maps.” Springer Verlag, New York. Lairmore, T. C., and Norton, J. A. (1997). Am. J. Surg 173, 37–41. Lakhani, S. R., Gusterson, B. A., Jacquemier, J., Sloane, J. P., Anderson, T. J., van de Vijver, M. J., Venter, D., Freeman, A., Antoniou, A., McGuffog, L., Smyth, E., Steel, C. M., Haites, N., Scott, R. J., Goldgar, D., Neuhausen, S., Daly, P. A., Ormiston, W., McManus, R., Scherneck, S., Ponder, B. A., Futreal, P. A., Peto, J., Stoppa-Lyonnet, D., Bignon, Y. J., and Stratton, M. R. (2000). Clin. Cancer Res. 6, 782–789. Lakhani, S. R., Jacquemier, J., Sloane, J. P., Gusterson, B. A., Anderson, T. J., van de Vijver, M. J., Farid, L. M., Venter, D., Antoniou, A., Storfer-Isser, A., Smyth, E., Steel, C. M., Haites, N., Scott, R. J., Goldgar, D., Neuhausen, S., Daly, P. A., Ormiston, W., McManus, R., Scherneck, S., Ponder, B. A. J., Ford, D., Peto, J., Stoppa-Lyonnet, D., Bignon, Y.-J.,

Gene Expression in Inherited Breast Cancer

31

Struewing, J. P., Spurr, N. K., Bishop, D. T., Klijn, J. G. M., Devilee, P., Cornelisse, C. J., Lasset, C., Lenoir, G., Barkardottir, R. B., Egilsson, V., Hamann, U., Chang-Claude, J., Sobol, H., Weber, B., Stratton, M. R., and Easton, D. F. (1998). J. Natl. Cancer Inst. 90, 1138–1145. Larson, J. S., Tonkinson, J. L., and Lai, M. T. (1997). Cancer. Res. 57, 3351–3355. Lee, J.-S., Collins, K. M., Brown, A. L., Lee, C.-H., and Chung, J. H. (2000). Nature 404, 201–204. Levy-Lahad, E., Lahad, A., Eisenberg, S., Dagan, E., Paperna, T., Kasinetz, L., Catane, R., Kaufman, B., Beller, U., Renbaum, P., and Gershoni-Baruch, R. (2001). Proc. Natl. Acad. Sci. USA 98, 3232–3236. Leygue, E., Dotzlaw, H., Watson, P. H., and Murphy, L. C. (1998). Cancer. Res. 58, 3197–3201. Li, S., Ting, N. S., Zheng, L., Chen, P. L., Ziv, Y., Shiloh, Y., Lee, E. Y., and Lee, W. H. (2000). Nature 406, 210–215. Liaw, D., Marsh, D. J., Li, J., Dahia, P. L., Wang, S. I., Zheng, Z., Bose, S., Call, K. M., Tsou, H. C., Peacocke, M., Eng, C., and Parsons, R. (1997). Nat. Genet. 16, 64–67. Lockhart, D. J., Dong, H., Byrne, M. C., Follettie, M. T., Gallo, M. V., Chee, M. S., Mittmann, M., Wang, C., Kobayashi, M., Horton, H., and Brown, E. L. (1996). Nat. Biotechnol. 14, 1675–1680. ˚ Ferno, ¨ M., and Olsson, H. (1998). Cancer Loman, N., Johannsson, O., Bendahl, P. O., Borg, A., 83, 310–319. Malkin, D., Li, F. P., Strong, L. C., Fraumeni, J. F., Jr., Nelson, C. E., Kim, D. H., Kassel, J., Gryka, M. A., Bischoff, F. Z., Tainsky, M. A., and Friend, S. H. (1990). Science 250, 1233–1238. Marmorstein, L. Y., Kinev, A. V., Chan, G. K., Bochar, D. A., Beniya, H., Epstein, J. A., Yen, T. J., and Shiekhattar, R. (2001). Cell 104, 247–257. Marston, N. J., Richards, W. J., Hughes, D., Bertwistle, D., Marshall, C. J., and Ashworth, A. (1999). Mol. Cell. Biol. 19, 4633–4642. Miki, Y., Swensen, J., Shattuck-Eidens, D., Futreal, P. A., Harshman, K., Tavtigian, S., Liu, Q., Cochran, C., Bennett, L. M., Ding, W., Bell, R., Rosenthal, J., Hussey, C., Tran, T., McClure, M., Frye, C., Hattier, T., Phelps, R., Haugen-Strano, A., Katcher, J., Yakumo, K., Gholalmi, Z., Shaffer, D., Stone, S., Bayer, S., Wray, C., Bogden, R., Dayananth, P., Ward, J., Tonin, P., Narod, S., Bristow, P. K., Norris, F. J., Helvering, L., Morrison, P., Rostek, P., Lai, M., Barrett, J. C., Lewis, C., Neuhausen, S., Cannon-Albright, L., Goldgar, D., Wiseman, R., Kamb, A., and Skolnick, M. H. (1994). Science 266, 66–71. Moch, H., Schraml, P., Bubendorf, L., Mirlacher, M., Kononen, J., Gasser, T., Mihatsch, M. J., Kallioniemi, O. P., and Sauter, G. (1999). Am. J. Pathol. 154, 981–986. ¨ Monni, O., Barlund, M., Mousses, S., Kononen, J., Sauter, G., Heiskanen, M., Paavola, P., Avela, K., Chen, Y., Bittner, M. L., and Kallioniemi, A. (2001). Proc. Natl. Acad. Sci. USA 98, 5711–5716. Monteiro, A. N. (2000). Trends Biochem. Sci. 25, 469–474. Narod, S., Ford, D., Devilee, P., Barkardottir, R. B., Eyfjord, J., Lenoir, G., Serova, O., Easton, D., and Goldgar, D. (1995a). Am. J. Hum. Genet. 57, 957–958. Narod, S. A., Ford, D., Devilee, P., Barkardottir, R. B., Lynch, H. T., Smith, S. A., Ponder, B. A., Weber, B. L., Garber, J. E., Birch, J. M., Cornelis, R. S., Kelsell, D. P., Spurr, N. K., Smyth, E., Haites, N., Sobol, H., Bignon, Y. J., Chang-Claude, J., Hamann, U., Lindblom, A., Borg, ˚ Piver, M. S., Gallion, H. H., Struewing, J. P., Whittemore, A., Tonin, P., Goldgar, D. E., A., and Easton, D. F., and the Breast Cancer Linkage Consortium. (1995b). Am. J. Hum. Genet. 56, 254–264. Noensie, E. N., and Dietz, H. C. (2001). Nat. Biotechnol. 19, 434–439. Notterman, D. A., Alon, U., Sierk, A. J., and Levine, A. J. (2001). Cancer Res. 61, 3124–3130. Olopade, O. I., Grushko, T., Sudbo, J., Reith, A., Lingjaerde, O. C., Hedenfalk, I., Simon, R., and Trent, J. (2001). N. Engl. J. Med. 344, 2028–2029.

32

Ingrid A. Hedenfalk et al.

Osborne, M. P. (1996). “Breast Development and Anatomy.” Lippincott-Raven, Philadelphia. Osin, P., Crook, T., Powles, T., Peto, J., and Gusterson, B. (1998). Lancet 351, 1487. Pao, G. M., Janknecht, R., Ruffner, H., Hunter, T., and Verma, I. M. (2000). Proc. Natl. Acad. Sci. USA 97, 1020–1025. Park, J. J., Irvine, R. A., Buchanan, G., Koh, S. S., Park, J. M., Tilley, W. D., Stallcup, M. R., Press, M. F., and Coetzee, G. A. (2000). Cancer Res. 60, 5946–5949. Parl, F. F. (2000). In “Estrogens, Estrogen Receptor and Breast Cancer.” IOS Press, Amsterdam. Paull, T. T., Cortez, D., Bowers, B., Elledge, S. J., and Gellert, M. (2001). Proc. Natl. Acad. Sci. USA 98, 6086–6091. Pechoux, C., Gudjonsson, T., Ronnov-Jessen, L., Bissell, M. J., and Petersen, O. W. (1999). Dev. Biol. 206, 88–99. Perou, C. M., Jeffrey, S. S., van de Rijn, M., Rees, C. A., Eisen, M. B., Ross, D. T., Pergamenschikov, A., Williams, C. F., Zhu, S. X., Lee, J. C. F., Lashkari, D., Shalon, D., Brown, P. O., and Botstein, D. (1999). Proc. Natl. Acad. Sci. USA 96, 9212–9217. Perou, C. M., Sorlie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., Rees, C. A., Pollack, J. R., Ross, D. T., Johnsen, H., Akslen, L. A., Fluge, O., Pergamenschikov, A., Williams, C., Zhu, S. X., Lonning, P. E., Borresen-Dale, A. L., Brown, P. O., and Botstein, D. (2000). Nature 406, 747–752. Peto, J., Collins, N., Barfoot, R., Seal, S., Warren, W., Rahman, N., Easton, D. F., Evans, C., Deacon, J., and Stratton, M. R. (1999). J. Natl. Cancer Inst. 91, 943–949. Pollack, J. R., Perou, C. M., Alizadeh, A. A., Eisen, M. B., Pergamenschikov, A., Williams, C. F., Jeffrey, S. S., Botstein, D., and Brown, P. O. (1999). Nat. Genet. 23, 41–46. Quackenbush, J. (2001). Nat. Rev. Genet. 2, 418–427. Rahman, N., Teare, M. D., Seal, S., Renard, H., Mangion, J., Cour, C., Thompson, D., Shugart, Y., Eccles, D., Devilee, P., Meijers, H., Nathanson, K. L., Neuhausen, S. L., Weber, B., ChangClaude, J., Easton, D. F., Goldgar, D., and Stratton, M. R. (2000). Oncogene 19, 4170–4173. Rajan, J. V., Marquis, S. T., Gardner, H. P., and Chodosh, L. A. (1997). Dev. Biol. 184, 385–401. Rebbeck, T. R. (1999). Cancer 86, 2493–2501. Rebbeck, T. R., Kantoff, P. W., Krithivas, K., Neuhausen, S., Blackwood, M. A., Godwin, A. K., Daly, M. B., Narod, S. A., Garber, J. E., Lynch, H. T., Weber, B. L., and Brown, M. (1999). Am. J. Hum. Genet. 64, 1371–1377. Rebbeck, T. R., Rogatko, A., and Viana, M. A. (1997). Genet. Test. 1, 157–164. Ronnov-Jessen, L., Petersen, O. W., and Bissell, M. J. (1996). Physiol. Rev. 76, 69–125. Ruffner, H., Jiang, W., Craig, A. G., Hunter, T., and Verma, I. M. (1999). Mol. Cell Biol. 19, 4843–4854. Schena, M., Shalon, D., Davis, R. W., and Brown, P. O. (1995). Science 270, 467–470. Scully, R., and Livingston, D. M. (2000). Nature 408, 429–432. Scully, R., Chen, J., Plug, A., Xiao, Y., Weaver, D., Feunteun, J., Ashley, T., and Livingston, D. M. (1997). Cell 88, 265–275. Scully, R., Ganesan, S., Brown, M., De Caprio, J. A., Cannistra, S. A., Feunteun, J., Schnitt, S., and Livingston, D. M. (1996). Science 272, 123–126. Scully, R., Puget, N., and Vlasakova, K. (2000). Oncogene 19, 6176–6183. Seitz, S., Rohde, K., Bender, E., Nothnagel, A., Kolble, K., Schlag, P. M., and Scherneck, S. (1997). Oncogene 14, 741–743. Shao, N., Chai, Y. L., Shyam, E., Reddy, P., and Rao, V. N. (1996). Oncogene 13, 1–7. Sharan, S. K., Morimatsu, M., Albrecht, U., Lim, D. S., Regel, E., Dinh, C., Sands, A., Eichele, G., Hasty, P., and Bradley, A. (1997). Nature 386, 804–810. Smith, T. M., Lee, M. K., Szabo, C. I., Jerome, N., McEuen, M., Taylor, M., Hood, L., and King, M. C. (1996). Genome Res. 6, 1029–1049. Stewart, G. S., Maser, R. S., Stankovic, T., Bressan, D. A., Kaplan, M. I., Jaspers, N. G., Raams, A., Byrd, P. J., Petrini, J. H., and Taylor, A. M. (1999). Cell 99, 577–587.

Gene Expression in Inherited Breast Cancer

33

Struewing, J. P., Hartge, P., Wacholder, S., Baker, S. M., Berlin, M., McAdams, M., Timmerman, M. M., Brody, L. C., and Tucker, M. A. (1997). N. Engl. J. Med. 336, 1401–1408. Struewing, J. P., Tarone, R. E., Brody, L. C., Li, F. P., and Boice, J. D., Jr. (1996). Lancet 347, 1493. Swift, M., Morrell, D., Massey, R. B., and Chase, C. L. (1991). N. Engl. J. Med. 325, 1831– 1836. Szabo, C. I., and King, M. C. (1997). Am. J. Hum. Genet. 60, 1013–1020. Tamayo, P., Slonim, D., Mesirov, J., Zhu, Q., Kitareewan, S., Dmitrovsky, E., Lander, E. S., and Golub, T. R. (1999). Proc. Natl. Acad. Sci. USA 96, 2907–2912. Tavtigian, S. V., Simard, J., Rommens, J., Couch, F., Shattuck-Eidens, D., Neuhausen, S., Merajver, S., Thorlacius, S., Ofﬁt, K., Stoppa-Lyonnet, D., Belanger, C., Bell, R., Berry, S., Bogden, R., Chen, Q., Davis, T., Dumont, M., Frye, C., Hattier, T., Jammulapati, S., Janecki, T., Jiang, P., Kehrer, R., Leblanc, J. F., Mitchell, J. T., McArthur-Morrison, J., Nguyen, K., Peng, Y., Samson, C., Schroeder, M., Snyder, S. C., Steele, L., Stringfellow, M., Stroup, C., Swedlund, B., Swensen, J., Teng, D., Thomas, A., Tran, T., Tran, T., Tranchant, M., Weaver-Feldhaus, J., Wong, A. K. C., Shizuya, H., Eyfjord, J. E., CannonAlbright, L., Labrie, F., Skolnick, M. H., Weber, B., Kamb, A., and Goldgar, D. E. (1996). Nat. Genet. 12, 333–337. Thai, T. H., Du, F., Tsan, J. T., Jin, Y., Phung, A., Spillman, M. A., Massa, H. F., Muller, C. Y., Ashfaq, R., Mathis, J. M., Miller, D. S., Trask, B. J., Baer, R., and Bowcock, A. M. (1998). Hum. Mol. Genet. 7, 195–202. Thompson, D., and Easton, D. (2001). Am. J. Hum. Genet. 68, 410–419. Thompson, D., Mangion, J., Szabo, C., van Eijk, R., Odefrey, F., Seal, S., Barfoot, R., Teare, D., Weber, B., Devilee, P., Easton, D., Goldgar, D., and Stratton, M., and the International BRCA3 Linkage Consortium. (2001). Proc. Am. Assoc. Cancer. Res. 42 (Abstr. No. LB-53) Tirkkonen, M., Johannsson, O., Agnarsson, B. A., Olsson, H., Ingvarsson, S., Karhu, R., Tanner, ˚ and Kallioniemi, O.-P. (1997). Cancer. Res. 57, M., Isola, J., Barkardottir, R. B., Borg, A., 1222–1227. Tomlinson, G. E., Chen, T. T.-L., Stastny, V. A., Virmani, A. K., Spillman, M. A., Tonk, V., Blum, J. L., Schneider, N. R., Wistuba, I. I., Shay, J. W., Minna, J. D., and Gazdar, A. F. (1998). Cancer. Res. 58, 3237–3242. Unger, M. A., Nathanson, K. L., Calzone, K., Antin-Ozerkis, D., Shih, H. A., Martin, A. M., Lenoir, G. M., Mazoyer, S., and Weber, B. L. (2000). Am. J. Hum. Genet. 67, 841–850. Vallon-Christersson, J., Cayanan, C., Haraldsson, K., Loman, N., Bergthorsson, J. T., Brondum˚ and Monteiro, Nielsen, K., Gerdes, A. M., Moller, P., Kristoffersson, U., Olsson, H., Borg, A., A. N. (2001). Hum. Mol. Genet. 10, 353–360. Velculescu, V. E., Zhang, L., Vogelstein, B., and Kinzler, K. W. (1995). Science 270, 484–487. Verhoog, L. C., Brekelmans, C. T., Seynaeve, C., Dahmen, G., van Geel, A. N., Bartels, C. C., Tilanus-Linthorst, M. M., Wagner, A., Devilee, P., Halley, D. J., van den Ouweland, A. M., Meijers-Heijboer, E. J., and Klijn, J. G. (1999). J. Clin. Oncol. 17, 3396–3402. Verhoog, L. C., Brekelmans, C. T., Seynaeve, C., van den Bosch, L. M., Dahmen, G., van Geel, A. N., Tilanus-Linthorst, M. M., Bartels, C. C., Wagner, A., van den Ouweland, A., Devilee, P., Meijers-Heijboer, E. J., and Klijn, J. G. (1998). Lancet 351, 316–321. Wang, Q., Zhang, H., Fishel, R., and Greene, M. I. (2000). Oncogene 19, 6152–6158. Wang, S. C., Lin, S. J., Su, L. K., and Hung, M. C. (1997). Biochem. Biophys. Res. Commun. 234, 247–251. Welcsh, P. L., Owens, K. N., and King, M.-C. (2000). Trends. Genet. 16, 69–74. Wistuba, II, Tomlinson, G. E., Behrens, C., Virmani, A., Geradts, J., Blum, J. L., Minna, J. D., and Gazdar, A. F. (2000). Gene. Chromosome Cancer 28, 359–369. Wooster, R., Bignell, G., Lancaster, J., Swift, S., Seal, S., Mangion, J., Collins, N., Gregory, S., Gumbs, C., Micklem, G., Barfoot, R., Hamoudi, R., Patel, S., Rice, C., Biggs, P., Hashim, Y.,

34

Ingrid A. Hedenfalk et al.

Smith, A., Connor, F., Arason, A., Gudmundsson, J., Ficenec, D., Kelsell, D., Ford, D., Tonin, P., Bishop, D. T., Spurr, N. K., Ponder, B. A. J., Eeles, R., Peto, J., Devilee, P., Cornelisse, C., Lynch, H., Narod, S., Lenoir, G., Egilsson, V., Barkardottir, R. B., Easton, F. E., Bentley, D. R., Futreal, P. A., Ashworth, A., and Stratton, M. R. (1995). Nature 378, 789–792. Wooster, R., Neuhausen, S. L., Mangion, J., Quirk, Y., Ford, D., Collins, N., Nguyen, K., Seal, S., Tran, T., Averill, D., Fields, P., Marshall, G., Narod, S., Lenoir, G. M., Lynch, H., Feunteun, J., Devilee, P., Cornelisse, C. J., Menko, F. H., Daly, P. A., Ormiston, W., McManus, R., Pye, C., Lewis, C. M., Cannon-Albright, L. A., Peto, J., Ponder, B. A. J., Skolnick, M. H., Easton, D. F., Goldgar, D. E., and Stratton, M. R. (1994). Science 265, 2088–2090. Wu, K., Jiang, S. W., Thangaraju, M., Wu, G., and Couch, F. J. (2000). J. Biol. Chem. 275, 35548–35556. Xu, X., Qiao, W., Linke, S. P., Cao, L., Li, W. M., Furth, P. A., Harris, C. C., and Deng, C. X. (2001). Nat. Genet. 28, 266–271. Xu, X., Wagner, K.-U., Larson, D., Weaver, Z., Li, C., Ried, T., Hennighausen, L., WynshawBoris, A., and Deng, C.-X. (1999a). Nat. Genet. 22, 37–43. Xu, X., Weaver, Z., Linke, S. P., Li, C., Gotay, J., Wang, X. W., Harris, C. C., Ried, T., and Deng, C. X. (1999b). Mol. Cell. 3, 389–395. Yarden, R. I., and Brody, L. C. (1999). Proc. Natl. Acad. Sci. USA 96, 4983–4988. Yarden, R. I., Duggan, D. J., Sgagias, M. K., Cowan, K. H., and Brody, L. C. (2001a). Proc. Am. Assoc. Cancer Res. 42 (Abstr. No. 2296) Yarden, R. I., Pardo-Reoyo, S., Sgagias, M. K., Cowan, K. H., and Brody, L. C. (2001b). Proc. Am. Assoc. Cancer. Res. 42 (Abstr. No. 721) Yu, X., and Baer, R. (2000). J. Biol. Chem. 275, 18541–18549. Zheng, L., Li, S., Boyer, T. G., and Lee, W. H. (2000a). Oncogene 19, 6159–6175. Zheng, L., Pan, H., Li, S., Flesken-Nikitin, A., Chen, L. P., Boyer, G. T., and Lee, H. W. (2000b). Mol. Cell 6, 757–768. Zhong, Q., Chen, C.-F., Li, S., Chen, Y., Wang, C.-C., Xiao, J., Chen, P.-L., Sharp, Z. D., and Lee, W.-H. (1999). Science 285, 747–750.

Multiparameter Analyses of Cell Cycle Regulatory Proteins in Human Breast Cancer: A Key to Deﬁnition of Separate Pathways in Tumorigenesis Goran ¨ Landberg Division of Pathology Department of Laboratory Medicine Lund University Malmo¨ University Hospital ¨ Sweden S-205 02 Malmo,

I. Breast Cancer II. Cell Cycle Aberrations III. Deregulation of Some Important G1 /S Regulatory Proteins in Breast Cancer A. Cyclin D1 B. Cyclin E C. CDK Inhibitors IV. Multiparameter Analyses of G1 /S Regulatory Proteins in Breast Cancer V. Future Perspectives References

Breast cancer is one of the most common cancer forms affecting many women. The disease nevertheless has widely varying behavior and therefore patient outcome, and an important undertaking is to deﬁne and understand the molecular mechanisms behind these actions. Defects in the G1/S transition in the cell cycle affect both tumor proliferation and the ﬁdelity of check points responsible for chromosomal integrity and DNA damage response and has lately been shown to represent one of a rather limited set of key aberrations in the transformation process. Many cell cycle regulatory proteins are either oncogenes or suppressor genes or are closely associated to the transformation process. The types of aberrations in the G1/S transition seem to be different in various cancers but are nevertheless often linked to clinical behaviors. In this review the role of multiparameter analyses of cell cycle regulatory proteins in breast cancer will be outlined with special attention to pattern analyses as well as the deﬁnition of two contrasting pathways in tumorigenesis deﬁned by either cyclin D1 or cyclin E overexpression. C 2002 Academic Press.

Advances in CANCER RESEARCH 0065-230X/02 $35.00

35

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

36

¨ Goran Landberg

I. BREAST CANCER Breast cancer is a heterogeneous disease presenting in various forms with large variations in morphologic appearance, metastatic potential, response to various treatments, and eventual patient survival. There are certain features such as tumor size, differentiation grade, estrogen/progesterone receptor expression, proliferation rate, and node positivity that today are used to predict tumor aggressiveness, but it is clear that we need better and more objective parameters to envisage tumor behavior. Ideally, a set of key gene products would be deﬁned and analyzed in each tumor, and different aberrations in these gene products would form patterns that were associated with speciﬁc tumor behaviors and importantly with potential response to various treatment strategies. These analyses of key gene products in the tumor could be performed on the DNA, RNA, or protein level, and there are array-based methods that readily can be used to screen large amounts of genes for aberrations either on the DNA level or regarding RNA expression (Snijders et al., 2000). For proteins, 2D gels can be used for a less selective screen with the advantage that protein modiﬁcations also can be detected, whereas antibodybased detection is ideal but limited because of shortcomings in the screening setting (Vercoutter-Edouart et al., 2001). It is, however, clear that many gene products involved in tumor progression are regulated by posttranslational mechanisms, and in order to detect these alterations, the analyses have to be performed on the protein level. An alternative would be if the array-based RNA expression pattern obtained with analyses of a large proportion of all transcribed genes would completely match the information of protein modiﬁcations and regulation of protein stability. Because of the complexity of deﬁning patterns in the RNA-based analysis of thousands of gene products, it is nevertheless a more appealing approach to use a set of deﬁned gene products and perform these analyses on the protein level. DNA-based analysis suffers from the drawback that not only must genetic mutations and translocations be considered, but also epigenetic phenomena such as hypermethylation of CpG islands. Perou and co-workers have reported a cDNA-based technology to delineate “molecular portraits” of breast cancer samples (Perou et al., 2000). This represents an important contribution in the attempt to deﬁne expression patterns that are formed when analyzing large amounts of gene products. Using different statistical approaches, several patterns and clusters were revealed as the “proliferation cluster,” including growth-related gene products or the interferon cluster, the estrogen receptor cluster, and the HER2/neu cluster. It was also obvious that cytokeratins were coexpressed with sets of other gene products, including cell cycle regulatory genes forming either a luminal- or basal-cell-like expression pattern. The cytokeratin expression

Cell Cycle and Breast Cancer

37

mirrors the normal expression of cytokeratins in luminal or basal cells in nontransformed ductuli, and it is intriguing to speculate whether the origin of the transformed cell will decide what genetic pattern the tumor will obtain during the transformation process. An alternative could be that the basal or luminal expression pattern is a characteristic of a differentiation pathway that tumors follow. It is nevertheless obvious that it will not be an easy task to deﬁne the important information in the overwhelming variety of data. This is also illustrated by the fact that tumors analyzed before and after cytostatic treatments were always more alike compared to other treated tumors, suggesting that there will probably be a great variety of tumor patterns (Perou et al., 2000). Nontransformed cells that are included in array analyses or 2D gel analyses also contribute to the complexity of the data analyses as shown by gene products clusters in the tumors that were similar to cell lines from different origins (Perou et al., 2000). The importance of the crosstalk between stromal and tumor cells cannot be overestimated (Elenbaas and Weinberg, 2001), but preferably this information should be separated and tumor cells and nontransformed cells should be analyzed individually. This could be achieved by microdissection or sorting of different cell types before analysis, but it also highlights the fact that the impending analysis of sets of proteins or gene products in tumors have to be based on section analyses so each cell type could readily be evaluated separately.

II. CELL CYCLE ABERRATIONS Malignant transformation is characterized by the acquisition of multiple genetic abnormalities leading to activation of oncogenes and/or inactivation of suppressor genes (Elenbaas et al., 2001). Many important regulatory mechanisms are targeted in the transformation process even though some key processes or check points probably need to be deregulated in order to obtain a fully neoplastic phenotype characterized by invasive behavior and ﬁnally metastatic capacity. Tumor progression is a central phenomenon responsible for the sequential gain of genetic defects in the transformation process, and a prerequisite for this event is genetic instability and/or aberrations in DNA repair mechanisms (Hanahan and Weiberg, 2000). A further requirement is an augmented and unlimited net growth and lifespan through increased proliferation and/or diminished cell death and activation of telomerase (DePinho, 2000). Aberrations in the G1/S control in the cell cycle affect many of the listed features such as proliferation control and chromosomal stability, as well as certain DNA-repair mechanisms, highlighting how crucial these defects probably are in the transformation process (Sherr, 2000). Deregulation of the cell cycle machinery is also a common ﬁnding in human

38

¨ Goran Landberg

tumors and has even been proposed to be obligatory in tumor development (Lukas et al., 1995a). The G1/S transition especially seems to be targeted in the transformation processes, and proteins controlling the G1/S phase transition are often aberrantly expressed in malignancies (Sherr, 2000). Families of highly conserved proteins consisting of cyclins, cyclin-dependent kinases (CDKs), and activating and inhibitory proteins are involved in the control of the G1/S transition (Adams, 2001). In principal, cyclins D and E sequentially activate CDK4/6 and CDK2, triggering phosphorylation of key substrates such as the retinoblastoma protein (pRb) and release of E2F, thereby initiating DNA replication and passage through the restriction point (Muller and Helin, 2000). Cyclin Ds preferentially activate CDK4/6, whereas cyclin Es mainly activate CDK2 (Adams, 2001). CDKs also need to be phosphorylated by CAK (CDK-activating kinases) before the CDK complex can be fully functional (Wu et al., 2001). There are two families of CDK inhibitors affecting the activity of the kinase complexes and therefore contributing to proper control of the G1/S transition. The INK family of proteins (p15, p16, p18, and p19) consists of speciﬁc CDK inhibitors mainly affecting the cyclin D–CDK4/CDK6 complexes (Rocco and Sidransky, 2001). The other class of inhibitors, the CIP/KIP family (including p21, p27 and p57), has a less selective inhibitory effect on many CDK complexes but with its main activity during G1 and S phase (Philipp-Staheli et al., 2001). Interestingly, both p21 and p27 also have CDK stabilizing functions and seem to be needed for fully active CDK complexes (Dotto, 2000). The shuttling of CDK inhibitors between different CDKs is further an important regulatory mechanism that can affect the activity of CDKs; for example, can p27 be sequestered by CDK4 in favor of an active CDK2 complex (Blain et al., 1997). The cellular localization of CDK inhibitors also inﬂuences the inhibitory activity, and p27, for example, can be impounded in the cytoplasma, as also has been observed in several malignancies (Orend et al., 1998). In summary, the formation of various active CDK complexes that are needed for cells to pass the G1/S transition and S-phase represents a highly complex orchestration cascade of activating and inhibitory events offering a multitude of possibilities for aberrations in malignancies. The molecular mechanisms behind aberrations in G1/S regulatory proteins are numerous and include common activating and inactivating mechanism such as mutations, deletions, gene ampliﬁcations, and translocations as well as epigenetic phenomena such as silencing of promoters by hypermethylation of CpG island (Steeg and Zhou, 1998; Donnellan and Chetty, 1999; Rocco and Sidransky, 2001; Zheng and Lee, 2001). Besides disturbances in the RNA stability as have been described for cyclin D1 (Dufourny et al., 2000), speciﬁc alterations in the ubiquitin-proteosome pathway leading to aberrations in the protein degradation of several cell cycle regulatory proteins have been

Cell Cycle and Breast Cancer

39

revealed (Nakayama et al., 2000) and represent an intriguing mechanism to deregulate the G1/S check point as will be elaborated later. The retinoblastoma protein has a central function in the cell cycle as well as in many other mechanisms such as differentiation, chromatin remodeling, and maintenance of genome integrity (Zheng and Lee, 2001), but despite being the most important substrate for CDKs in the G1/S control, it is clear that there must exist other substrates for CDKs. Overexpression of cyclin E and c-myc can both induce S-phase in serum-starved Rb inactivated cells indicating that other CDK dependent and independent regulatory pathways are present downstream of pRb (Santoni-Rugiu et al., 2000). The Rb related pocket proteins p107 and p130 also inﬂuence cell cycle control, and especially p130 has speciﬁc inhibitory functions on cyclin E/CDK2 complexes that inﬂuence, for example, the estrogen response in breast cancer (Classon and Dyson, 2001). The presence of p107 and p130, however, cannot fully clarify how Rb negative cells can enter S-phase after CDK2 activation, and other potentially important substrates are clearly present and need to be revealed. A key CDK substrate downstream of pRb will, however, be an important G1/S regulatory molecule that ideally can be targeted in novel treatment schedules.

III. DEREGULATION OF SOME IMPORTANT G1 /S REGULATORY PROTEINS IN BREAST CANCER A. Cyclin D1 Cyclin D1 is the most extensively studied cyclin D in the cyclin D family consisting of cyclin D1–3. The protein is induced by estrogen and growth factors and can act as a cellular sensor for growth-related factors (Bernards, 1999; Zhou et al., 2000; Planas-Silva et al., 2001). Transgenic mice overexpressing cyclin D1 in a tissue-speciﬁc system developed mammary hyperplasia and adenocarcinoma indicating that cyclin D1 is one of several genetic alterations necessary to induce breast cancer (Wang et al., 1994; Hutchinson and Muller, 2000). The relevance for cyclin D1 in breast cancer is further emphasized by the ﬁnding that cyclin D1 knockout mice show a marked defect in breast epithelium development during pregnancy (Fantl et al., 1995; Sicinski et al., 1995). Cyclin D1 has both CDK-dependent and -independent functions of importance for breast cancer (Bernards, 1999). Besides activating CDK4/6 and CDK2 in the G1/S transition, cyclin D1 can activate the estrogen receptor independent of both estrogens as well as CDK-binding to cyclin D1 (Zwijsen et al., 1997; Prall et al., 1998; Zwijsen et al., 1998; Lamb

40

¨ Goran Landberg

et al., 2000). Cyclin D1 reacts not only with ER but also with several members of related steroid receptor coactivators (SRCs) (McMahon et al., 1999) and brieﬂy, strong synergy between cyclin D1 and estrogen in activation of ER has been observed (Zwijsen et al., 1997). Cyclin D1 can further bind to the transcription factor and Myb-like protein DMP1 and counteract the growth inhibitory effect of DMP1 in a CDK-independent manner (Hirai and Sherr, 1996; Inoue and Sherr, 1998). Gene ampliﬁcation of the 11q13, including the cyclin D1 gene, is found in 15–20% of patients with breast cancer and has been associated with unfavorable prognosis (Borg et al., 1991; Schuuring et al., 1992). Studies analyzing protein levels have shown somewhat contradictory results and its prognostic value has been disputed, but in a large study analyzing 345 primary breast cancer samples, cyclin D1 overexpression was clearly associated with improved survival for up to 20 years (Barnes, 1997). The signiﬁcance of protein overexpression of cyclin D1 in breast cancer has been evaluated in many studies using both immunohistochemistry and Western blotting, and up to 50% of all breast cancer seems to overexpress the protein (Barnes, 1997; Nielsen et al., 1997; Wong et al., 2001). Cyclin D1 overexpression is also indisputably associated with estrogen receptor positivity in breast cancer (Nielsen et al., 1997). The association to proliferation is more ambiguous, and some studies have reported a lower proliferation or at least an unaltered proliferation in cyclin-D1-overexpressing tumors (Nielsen et al., 1999; Shoker et al., 2001; Wong et al., 2001). Recent reports have also shown that cyclin D1 is one of the targets of β-catenin in breast cancer cells linking the Wnt pathway to deregulation of cyclin D1 and the pRb pathway (Lin et al., 2000).

B. Cyclin E Cyclin E is one of the main activating proteins besides cyclin D1 in the G1/S transition (Reed, 1997). Interestingly, cyclin E is both a regulator and a target of the E2F transcription family, and E2F can bind to the cyclin E promoter, forming an autoregulatory loop between cyclins, Rb phosphorylation, release of E2F, and cyclin E transcription as discussed in more detail later (Donnellan and Chetty, 1999). Cyclin E is normally rapidly degraded after it has fulﬁlled its function, with maximal expression during the G1/S transition and decreasing levels during S-phase (Ohtosubo et al., 1995). Despite the fact that cyclin E is overexpressed in many tumors, it has not been fully clariﬁed whether cyclin E is a protooncogene or whether the high expression observed in tumors is secondary to other important aberrations such as pRb inactivation (Keyomarsi et al., 1995; Gray-Bablin et al., 1996; Nielsen et al., 1996; Steeg and Zhou, 1998; Donnellan and Chetty, 1999). Gene ampliﬁcation involving the cyclin E gene has nevertheless been observed in

Cell Cycle and Breast Cancer

41

1.3% of breast carcinomas and 12.5% of ovarian carcinomas, indicating that overexpression of cyclin E could represent a primary genetic event in the transformation process (Courjal et al., 1996). Also, transgenic mice with targeted overexpression of cyclin E in the mammary glands during pregnancy and lactation develop mammary carcinomas, although with rather low frequency and long latency, indicating that additional genetic events besides cyclin E overexpression are necessary to trigger malignancy (Bortner and Rosenberg, 1997; Hutchinson and Muller, 2000). Besides gene ampliﬁcation, altered cyclin E degradation through the ubiquitin-proteasome pathway might induce overexpression or unscheduled expression of the protein in malignancies. Mice lacking Skp2, an F-box protein and substrate recognition component of the Skp1-cullin-F-box protein (SCF) ubiquitin ligase, showed increased cyclin E and p27 protein as well as loss of cyclin E periodicity, but nonetheless without increased risk for cancer development (Nakayama et al., 2000). We earlier analyzed cyclin E by immunohistochemistry as well as by Western blotting in more than 80 primary breast cancer samples demonstrating large variations in the intensity of the cyclin E staining in each tumor with totally negative tumor cells up to extremely intense nuclear cyclin E staining, including tumors with extremely high protein quantities determined by Western blotting (Nielsen et al., 1996). This supports that cyclin E is cell-cycle-speciﬁcally expressed with an apparently intact degradation machinery in most breast cancer samples, including the cases with extreme protein contents, suggesting that other mechanisms are also involved in the deregulation of the protein. Variant forms of cyclin E mRNA and protein have also been observed in several malignancies, including breast cancer (Keyomarsi et al., 1995). Some of these forms seem to arise from posttranslational actions of proteases, whereas others are splice variants (Harwell et al., 2000). A shorter splice variant of cyclin E further activates CDK2 more robustly than full-length cyclin E, suggesting that the variant forms often detected in malignancies could contribute to increased CDK2 activation (Porter and Keyomarsi, 2000). In addition, cyclin E overexpression in different cell lines can induce genomic instability and increased percentage of polyploid cells (Spruck et al., 1999). This links cyclin E overexpression to central processes in tumor progression as inaccurate duplication and segregation of chromosomes as discussed further below. We and others have shown that cyclin E overexpression is associated with an increased risk of death for breast cancer patients as well as lack of estrogen receptor (Dutta et al., 1995; Porter et al., 1997; Nielsen et al., 1997; Donnellan et al., 2001). Similar results were obtained for a lymph-node negative subset of patients (Porter et al., 1997). In multiple regression analysis, nevertheless, cyclin E is often not an independent prognostic factor (Nielsen et al., 1997), which could be explained by the close associations between high cyclin E expression and other risk factors.

42

¨ Goran Landberg

C. CDK Inhibitors The most important CDK inhibitors in the breast cancer transformation and progression processes are p27, p21, and p16. These inhibitory molecules affect various central mechanisms in the control and response of growth factor signals, cell cycle and DNA synthesis, stress response, and differentiation (Adams, 2001). The multitude of functions that CDK inhibitors are involved in highlights the importance of the deregulation of these molecules in malignancies. Both p27 and p16 are commonly down-regulated in breast cancer by different mechanisms, whereas genetic alterations in p21 seem to be rare (Dotto, 2000; Rocco and Sidransky, 2001; Philipp-Staheli et al., 2001). The protein content of p21, though, varies in breast cancer, probably reﬂecting different regulatory events affecting p21. The CDK inhibitor p27 is closely associated with cyclin E, and besides restraining cyclin E–cdk2 activity the protein can also serve as a substrate for the cyclin E–cdk2 complex triggering degradation of p27 by ubiquitinmediated proteolysis (Philipp-Staheli et al., 2001). This mechanism provides a negative regulatory feedback loop that makes the G1/S transition irreversible (Sheaff et al., 1997; Vlach et al., 1997; Nguyen et al., 1999). As expected of a CDK inhibitor involved in the start and arrest of the cell cycle, the expression of p27 differs during the cell cycle, with high p27 protein levels in resting G0 cells and rapidly declining levels in cells entering the active cell cycle (Hengst and Reed, 1996; Sherr, 1996). p27 is affected by both intrinsic and extrinsic factors, such as TGF-β, cell–cell contact, or elevated cyclic AMP levels causing increased expression of the inhibitor with subsequent arrest in G1 or cell cycle exit (Polyak et al., 1994; Kato et al., 1994; Fornaro et al., 1999). Apart from cell cycle regulatory functions, there have been reports suggesting that p27 is involved in the regulation of apoptosis (Katayose et al., 1997). Conﬂicting results have been reported regarding the relation between p27 expression and proliferation in tumors, indicating that deregulation of p27 in malignancies might not affect the net proliferation ´ signiﬁcantly (Sanchez-Beato et al., 1997; Erlanson and Landberg, 2001). We have showed that p27 status in breast cancer affects the cyclin-E-associated kinase activity, which subsequently is strongly linked to proliferation, suggesting that it is necessary to include the status of cyclins, CDKs, and CDK inhibitors before any relations to proliferation can be fully detailed (Lod´en et al., 1999). p27 can probably function as a haploinsufﬁcient tumor suppressor, and it has been hypothesized that the molecule can function as a rheostat rather than as an on/off switch (Philipp-Staheli et al., 2001). Low p27 protein content in a fraction of breast cancer samples has been observed in several studies, but inactivation of p27 by mutations seems to be rare in ˜ breast cancer as well as in other malignancies (Ponce-Castaneda et al., 1995;

Cell Cycle and Breast Cancer

43

Catzavelos et al., 1997; Porter et al., 1997; Lod´en et al., 1999; PhilippStaheli et al., 2001). Alternative mechanisms that might explain the low protein content in breast cancer could be aberrant DNA methylation, posttranscriptional and translational defects, or posttranslational inactivation (Loda et al., 1997; Qian et al., 1998; Masciullo et al., 1999; Millard et al., 2000). In summary, most studies detailing the prognostic importance of p27 protein content in breast cancer have shown that low p27 was associated with an adverse outcome (Porter et al., 1997; Tan et al., 1997; Chiarle et al., 2001). There are, nevertheless, some studies demonstrating that p27 protein content was not associated with survival, indicating that the composition of the studied patient cohort might affect the results (Barbareschi et al., 2000). p21 is a multifunctional protein acting both as an inhibitor of CDKs and as a positive modulator of cyclin/CDK complex formation and nuclear localization (Harper et al., 1995; LaBaer et al., 1997). Besides cell cycle functions, the protein is also involved in p53-mediated G1/S arrest (Dotto, 2000), modulation of apoptosis (Harvey et al., 1998), and differentiation processes (DiCunto et al., 1998). The regulation of estrogen-dependent growth is of interest in breast cancer because of the opportunity to use antiestrogen treatment for patients with tumors expressing the estrogen receptor. Reports have linked p21 as well as p27 to the growth-stimulatory effect of estrogens, and the CDK inhibitors are probably two of the main regulators mediating the therapeutic effect of antiestrogens in breast cancer (Planas-Silva and Weinberg, 1997; Cariou et al., 2000; Swarbrick et al., 2000; Foster et al., 2001). The protein content of p21 varies greatly in breast cancer and is further strongly linked to both p53 status and the cyclin-E-associated kinase activity (Lod´en et al., 1999), all together supporting the idea that p21 proteins most likely have many important regulatory functions in breast cancer cells. The CDK-inhibitor p16 fulﬁlls all criteria to be deﬁned as a suppressor gene (Haber and Harlow, 1997), but there is some confusion around p16’s exact role as a tumor suppressor due to disparities between human molecular genetic data and the cancer-prone phenotype of INK4a/ARF locus mouse knockout models (Rocco and Sidransky, 2001). The INK4a/ARF locus also encodes an alternate gene product with tumor suppressor gene functions called p14ARF affecting p53 stability by interaction with MDM2 (Quelle et al., 1997; Bates et al., 1998; Sherr, 2000). The presence of two gene products with tumor suppressor functions in the same gene locus commonly targeted in malignancies clearly complicates the valuation of aberrations in each gene product with regard to affect of the transformation process, and further studies are needed to exactly pinpoint the separate functions (Sharpless and DePinho, 1999). The tumor suppressor function for p16 is nevertheless believed to be that p16 blocks CDK4/6, leading to formation of inhibitory CDK4/6–p16 complexes and loss of stimulatory CDK4/6–cyclin

44

¨ Goran Landberg

D1 complexes. This results in degradation of free cyclin D1 by a ubiquitindependent proteasome pathway and accumulation of p27 that can inhibit cyclin E-A/CDK2 dependent phosphorylation of pRb (Diehl et al., 1997; Zhang et al., 1999). The end result of a p16 increase is an Rb-dependent G1/S block by prevention of E2F-dependent transcription (Lukas et al., 1995b). p16 can be inactivated by homozygous deletions, mutations, or hypermethylation of CpG islands in the promoter area as also observed in breast cancer in rather low but substantial fractions of the tumors (Brenner and Aldaz, 1995; Cairns et al., 1995; Hui et al., 2000; Nielsen et al., 2001). Down-regulation of p16 protein in breast cancer has also been observed in several studies, supporting the existence and importance of p16 deregulation in breast cancer (Geradts and Wilson, 1996; Nielsen et al., 2001; Rocco and Sidransky, 2001).

IV. MULTIPARAMETER ANALYSES OF G1 /S REGULATORY PROTEINS IN BREAST CANCER Because of the complexity of the G1/S transition and the multitude of regulators as well as defects in tumors, we and others have performed multiparameter analyses of several G1/S regulatory proteins in breast cancer in an attempt to deﬁne patterns of aberrantly expressed proteins (Nielsen et al., 1997, 1999; Gerardts and Ingraham, 2000; Milde-Langosch et al., 2000; Weng et al., 2001; Wong et al., 2001). It is obvious that there exists an intimate interplay among the different cyclins, CDK inhibitors, and pRb, and the deregulation of one or several regulators will probably affect the balance between the different activating and inhibitory molecules. Since many of the G1/S regulatory proteins affect the same pathway, called the pRb pathway (including p16, cyclin D1-3, CDK4, and pRb), it has also been argued that there might only exist one aberration in the same pathway, as, for example, supported by the inverse association between p16 and pRb protein content observed in several malignancies (Guan et al., 1994; Lukas et al., 1995b). A more complete understanding of the presence and interplay between aberrations mainly affecting the G1/S transition can therefore only be achieved by multiparameter analyses of sets of key proteins and is, despite the difﬁculties associated with this type of analysis, a favorable approach to use. We have earlier reported that one or several aberrations of G1/S regulatory proteins in breast cancer are common and observed in approximately 90% of all tumors (Nielsen et al., 1999). In the study tumors were divided according to observed combinations of aberrations in the protein content of p16, cyclin E and D1, and pRb as determined by Western blotting and p27 characterized by immunohistochemistry as summarized in Fig. 1. Isolated low p16

Cell Cycle and Breast Cancer

45

Fig. 1 Division of 100 breast cancer samples according to observed combinations of aberrations in cyclin E, D1, p16, p27 and pRb protein content determined by Western blotting or immunohistochemistry (Nielsen et al., 1999). “No defect” indicates absence of protein aberrations in the above-indicated G1/S regulatory proteins.

protein was the most common alteration (38.6%), followed by high cyclin D1 with low p16 (12.5%), isolated high cyclin E (12.5%), high cyclin E in combination with low p27 (6.8%), isolated high cyclin D1 (5.7%), and a combination of low p16 and p27 (4.5%). All pRb-inactivated tumors also had high p16 and high cyclin E protein content but varied concerning p27 (4.5% normal and 2.3% low p27) (Nielsen et al., 1997, 1999). Even though only 100 tumors were characterized, the results clearly indicate that aberrations in G1/S regulatory proteins are very common in breast cancer and will probably be observed in all tumors if additional regulatory proteins are analyzed. This supports the idea that deregulation of the G1/S transition is an obligatory event in the transformation process. The different subsets of G1/S defects indicated in Fig. 1 also had different proliferative capacities determined by Ki-67, and the highest proliferation was observed in pRbinactivated tumors, followed by tumors with combination of high cyclin E protein, whereas the lowest proliferation was observed in the different group of tumors with high cyclin D1 protein content (Nielsen et al., 1999). Also,

46

¨ Goran Landberg

Fig. 2 Scatter plot showing the relation between cyclin D1 and cyclin E protein in a material of 120 primary breast cancer samples. Cyclin D1 and E protein contents were determined by Western blotting and expressed as relative ratios to a cell line standard. The estrogen receptor status (ER) for the tumors is further indicated in the ﬁgure (Nielsen et al., 1997).

patient survival differed between the groups, with worse prognoses for patients with pRb-inactivated or high cyclin E tumors and best prognoses for patients with high cyclin D1 tumors, approximately mimicking the proliferative differences (Nielsen et al., 1999). In the multiparameter analyses of G1/S regulatory protein in breast cancer, we early observed a striking association between cyclin D1 and cyclin E protein forming a boomerang-shaped curve with either high cyclin D1 and low cyclin E or vice versa (Nielsen et al., 1997). As illustrated in Fig. 2, the estrogen receptor status also perfectly matched the cyclin E and D1 pattern, and all cyclin E high/cyclin D1 low tumors (denoted cyclin Ehigh) were also ER negative, whereas ER positivity was associated with high cyclin D1/low cyclin E tumors (denoted cyclin D1high). Rb-inactivated tumors deﬁned by absent nuclear reactivity in tumor cells, LOH of four different polymorphisms in the Rb gene, aberrant quote between Rb-phosphorylation/proliferation, and high p16 protein content (Nielsen et al., 1997) were also predominantly observed among cyclin Ehigh tumors. This suggests that cyclin Ehigh tumors distinctly differ from the cyclin D1high tumors, and we have therefore continued to delineate molecular differences. The p53 protein acts as a guardian of the genome and if DNA damage is sensed p53 can induce apoptosis or a cell cycle block with subsequent DNA repair. Mutations of the p53 gene have been reported in various malignancies and approximately 15% of all breast cancers are p53 inactivated because of mutations (Landberg and Roos, 1997). Increased expression of p53 protein is often used as an indicator of p53 gene mutations but can also represent an abnormal accumulation of an inactive p53 protein without gene mutations. Inactivation of p53 through insertions, deletions, and nonsense

Cell Cycle and Breast Cancer

47

point mutations as well as overexpression of p53 protein is also overrepresented in tumors with high cyclin E protein (Lindahl et al., submitted; Lod´en et al., submitted), suggesting that p53 inactivation is strongly linked to cyclin Ehigh tumors. The observed mutation pattern in breast cancer cells overexpressing cyclin E, though, differed markedly from that in tumors with low cyclin E, suggesting that cyclin E overexpression is associated with a speciﬁc type of inactivation of the p53, which is in line with data from Spruck and co-workers (Spruck et al., 1999). Besides inactivation of pRb and p53 in cyclin Ehigh tumors the potential suppressor gene candidate IGFBP-rP1 is also down-regulated in a large fraction of the cyclin Ehigh breast cancer samples (Landberg et al., 2001). A similar pattern has been observed for the CDK-inhibitor p27, and low p27 protein determined by immunohistochemistry was commonly observed in cyclin Ehigh tumors (Lod´en et al., submitted). In contrast, low p16 protein was mainly observed in low-cyclin tumors (Landberg et al., 1997; Lod´en et al., submitted), altogether indicating that four out of ﬁve tumor suppressor gene products are inactivated in the cyclin Ehigh tumor group and clearly linking cyclin E to alterations in many suppressor genes. Cyclin E overexpression has further been associated with tumor aneuploidy (Nielsen et al., 1996; Lindahl et al., submitted). These results and ﬁndings suggesting that cyclin E overexpression can cause genetic instability (Spruck et al., 1999) support a model where cyclin E overexpression represents a key event in the transformation process linking cell cycle aberrations with chromosomal instability and frequent inactivation of several tumor suppressor genes. The pRb function is central in the control of the cell cycle and as shown earlier pRb is inactivated in a fraction of cyclin Ehigh tumors (Nielsen et al., 1997). In the above-described tumor material we have also characterized in vivo phosphorylation of pRb using Western blotting and calculated the quote between the slower migrating phosphorylated pRb to the faster migrating unphosphorylated pRb representing the degree of pRb phosphorylation (Nielsen et al., 1997). The cyclin-E-dependent kinase activity in the same tumor material further mirrored the protein content of activating cyclin E and inhibitory functions for p21 and p27, and interestingly the cyclin-Edependent kinase activity was strongly associated with pRb phosphorylation. Besides supporting that pRb is an in vivo substrate for cyclin E, this validates the separate kinase measurements, Western blotting of cyclin E, and immunohistochemistry of p21 and p27. Surprisingly, pRb phosphorylation did not correlate with proliferation in the tumor material, whereas the cyclin-E-dependent kinase activity was strongly associated with proliferation. This suggests that there must exist other cyclin-E-associated events and substrates besides pRb that are rate-limiting for S-phase entrance. However, because of the marked differences between cyclin Ehigh and cyclin D1high tumors, these groups have been analyzed separately regarding associations

48

¨ Goran Landberg

between cyclin status, Rb-phosphorylation, and proliferation. Interestingly, in cyclin D1high tumors there was a signiﬁcant link between Rb phosphorylation and proliferation, suggesting that these tumors have an intact pRb pathway in contrast to cyclin Ehigh tumors either lacking pRb or probably overruling its function (Lod´en et al., submitted). It is obvious that phosphorylation of the 16 potential cyclin/CDK phosphorylation sites on pRb (Lin et al., 1991) is not controlled by a single cyclin/ CDK complex, and instead it appears that pRb is sequentially inactivated by multiple cyclin/CDK complexes that each target distinct functions. The ﬁrst step in the inactivation of pRb is probably phosphorylation of the C terminus by cyclin D/CDK4, inducing a disruption of the pRb/HDAC complex and resulting in a partial release of repressed E2F target genes, including cyclin E (Zheng and Lee, 2001). This facilitates formation of cyclin E/CDK2 complexes that now fully can phosphorylate pRb after a conformational change of pRb induced by the cyclin D/CDK4 phosphorylation. The combined cyclin-D- and cyclin-E-triggered phosphorylation of pRb disrupts both the A and B domain interfaces of the pRb pocket region, resulting in a release of free E2F followed by initiation of the S-phase (Zheng and Lee, 2001). The scenario just deﬁned seems to be present in cyclin D1high tumors with proliferation induction with corresponding cyclin D1 and E increase, suggesting that these tumors have an intact cyclin-D1- and cyclin-E-associated phosphorylation response of pRb. Cyclin Ehigh tumors, however, proliferate independently of pRb, targeting downstream or parallel pathways to pRb, nevertheless resulting in a high proliferation as observed in the tumor material (Nielsen et al., 1997).

V. FUTURE PERSPECTIVES As highlighted in this article, deregulation of the G1/S transition is most likely a major event in the transformation process. Also obvious is that when the normally tightly controlled check point is deranged through one or several defects, the aberration in regulatory gene products forms patterns linking certain types of aberrations. This could either be due to mutually inactivating mechanisms that are responsible for the separate defects or be due to secondary compensating mechanisms triggering aberrantly expressed protein levels after the primary damage to a regulator. It could also be that tumors follow different pathways in the transformation process and that there is a selective pressure toward certain predisposed aberrations, including various G1/S regulators. The clear difference between cyclin Ehigh tumors and the cyclin D1high tumors regarding type of genetic inactivation of suppressor genes, proliferative capacity, and dependence of pRb as summarized in

Cell Cycle and Breast Cancer

49

Fig. 3 A summary of some important molecular and functional differences between breast cancer divided according to cyclin E and D1 protein content.

Fig. 3 suggests that they represent two major pathways in the transformation process leading to breast cancer. In favor of this is also the difference in patient age between the groups, indicating that some hereditary breast cancers might be associated with cyclin E overexpression. Future studies must delineate whether the two pathways and further subdivisions of breast cancer can give predictive information regarding response to various therapies. Many analyses these days are array based, and to further explore patterns of protein aberrations in breast cancer and other tumors both in G1/S regulatory gene products and in other transformation-associated events, tissue arrays will be a valuable tool as exempliﬁed in Fig. 4 (see color insert). It is therefore an important task to continue the characterization of multiple gene products and use pattern analyses to deﬁne clusters of tumors. Eventually, this could lead to the ultimate tumor description, including a gene-productbased subgrouping that will give information on the transformation pathway and involved gene products, as well as tumor aggressiveness and expected response to different therapies. The described subdivision according to cyclin D1 and cyclin E overexpression is clearly a promising attempt to subdivide

50

¨ Goran Landberg

breast cancer according to mutual genetic aberrations. This will, it is hoped; be a starting point for further subdivisions that will result in better understanding of the complex transformation process and the different pathways that lead to breast cancer.

ACKNOWLEDGMENTS The present study was supported by grants from the Swedish Cancer Society, Malmo¨ University Hospital funds, and Lund University funds. The author thanks all co-workers in the lab and especially Martin Lod´en and Niels Hilmer Nielsen.

REFERENCES Adams, P. D. (2001). Regulation of the retinoblastoma tumor suppressor protein by cyclin/cdks. Biochem. Biophys. Acta 1471, M123–M133. Barbareschi, M., van Tinteren, H., Mauri, F. A., Veronese, S., Peterse, H., Maisonneuve, P., Caffo, O., Scaioli, M., Doglioni, C., Galligioni, E., Dalla Palma, P., and Michalides, R. (2000). P27(kip1) expression in breast carcinomas: an immunohistochemical study on 512 patients with long-term follow-up. Int. J. Cancer 20, 236–241. Barnes, D. (1997). Cyclin D1 in mammary carcinoma. J. Pathol. 181, 267–269. Bates, S., Phillips, A. C., Clark, P. A., Stott, F., Peters, G., Ludwig, R. L., and Vousden, K. H. (1998). P14ARF links tumour suppressors RB and p53. Nature 395, 124–125. Bernards, R. (1999). CDK-independent activities of D type cyclins. Biochem. Biophys. Acta 1424, M17–M22. Blain, S. W., Montalvo, E., and Massague, J. (1997). Differential interaction of the cyclindependent kinase (Cdk) inhibitor p27Kip1 with cyclin A-Cdk2 and cyclin D2-Cdk4. J. Biol. Chem. 272, 25863–25872. ˚ Sigurdsson, H., Clark, G. M., Ferno, M., Fuqua, S. A. W., Olsson, H., Killander, D., Borg, A., and McGurie, W. L. (1991). Association of INT2/HST1 co-ampliﬁcation in primary breast cancer with hormone-dependent phenotype and poor prognosis. Br. J. Cancer 63, 136–142. Bortner, D., and Rosenberg, M. (1997). Induction of mammary gland hyperplasia and carcinomas in transgenic mice expressing human cyclin E. Mol. Cell Biol. 17, 453–459. Brenner, A. J., and Aldaz, C. M. (1995). Chromosomal 9p allelic loss and p16/CDKN2 in breast cancer and evidence of p16 inactivation in immortal breast epithelial cells. Cancer Res. 55, 2892–2895. Cairns, P., Polasci, T. J., Eby, Y., Tokino, K., Califano, J., Merlo, A., Mao, L., Herath, J., Jenkins, R., Westra, W., Rutter, J. L., Buckler, A., Gabrielson, E., Tockman, M., Cho, K. R., Hedrick, L., Bova, G. S., Isaacs, W., Koch, W., Schwab, D., and Sidransky, D. (1995). Frequency of homozygous deletion at p16/CDKN2 in primary human tumours. Nat. Genet. 11, 210–212. Cariou, S., Donovan, J. C. H., Flanagan, W. M., Milic, A., Bhattacharya, N., and Slingerland, J. M. (2000). Down-regulation of p21WAF1/CIP1 or p27KIP1 abrogates antiestrogenmediated cell cycle arrest in human breast cancer cells. Proc. Natl. Acad. Sci. USA 97, 9042–9046.

Cell Cycle and Breast Cancer

51

Catzavelos, C., Bhattacharya, N., Ung, Y. C., Wilson, J. A., Roncari, L., Sandhu, C., Shaw, P., Yeger, H., Morava-Protzner, I., Kapusta, L., Franssen, E., Pritchard, K. I., and Slingerland, J. M. (1997). Decreased levels of the cell-cycle inhibitor p27Kip1 protein: Prognostic implications in primary breast cancer. Nat. Med. 3, 227–230. Chiarle, R., Pagano, M., and Inghirami, G. (2001). The cyclin dependent kinase inhibitor p27 and its prognostic role in breast cancer. Breast Cancer Res. 3, 91–94. Classon, M., and Dyson, N. (2001). P107 and p130: Versatile proteins with interesting pockets. Exp. Cell Res. 264, 135–147. Courjal, F., Louason, G., Speiser, P., Katsaros, D., Zeillinger, R., and Theillet, C. (1996). Cyclin gene ampliﬁcation and overexpression in breast and ovarian cancers: evidence for the selection of cyclin D1 in breast and cyclin E in ovarian tumors. Int. J. Cancer 69, 247–253. DePinho, R. A. (2000). The age of cancer. Nature 9, 248–254. Diehl, J. A., Zindy, F., and Sherr, C. J. (1997). Inhibition of cyclin D1 phosphorylation on threnine-286 prevents its rapid degradation via the ubiquitin-proteasome pathway. Genes Dev. 11, 53–61. Donnellan, R., and Chetty, R. (1999). Cyclin E in human cancers. FASEB J. 13, 773–780. Donnellan, R., Kleinschmidt, I., and Chetty, R. (2001). Cyclin E immunoexpression in breast ductal carcinoma: pathologic correlations and prognostic implications. Hum. Pathol. 32, 89–94. Dotto, G. P. (2000). p21WAF1/Ciip1: More than a break to the cell cycle? Biochem. Biophys. Acta 1471, M43–M56. Dufourny, B. D., van Teeffelen, H. A. A. M., Hamelers, I. H. L., Sussenbach, J. S., and Steenbergh, P. H. (2000). Stabilization of cyclin D1 mRNA via the phosphatidylinositol 3-kinase pathway in MCF-7 human breast cancer cells. J. Endocrinol. 166, 329–338. Dutta, A., Chandra, R., Leiter, L. M., and Lester, S. (1995). Cyclins as markers of tumor proliferation: immunocytochemical studies in breast cancer. Proc. Natl. Acad. Sci. USA 92, 386–5390. Elenbaas, B., and Weinberg, R. A. (2001). Heterotypic signaling between tumor cells and ﬁbroblasts in carcinoma formation. Exp. Cell Res. 264, 169–184. Elenbaas, B., Spirio, L., Koerner, F., Fleming, M. D., Zimonjic, D. B., Donaher, J. L., Popescu, N. C., Hahn, W. C., and Weinberg, R. A. (2001). Human breast cancer cells generated by oncogenic transformation of primary mammary epithelial cells. Genes Dev. 15, 50–65. Erlanson, M., and Landberg, G. (2001). Prognostic implications of p27 and cyclin E protein contents in malignant lymphomas. Leukemia Lymphoma 40, 461–470. Fantl, V., Stamp, G., Andrews, A., Rosewell, I., and Dickson, C. (1995). Mice lacking cyclin D1 are small and show defects in eye and mammary gland development. Genes Dev. 9, 2364–2372. Fornaro, M., Tallini, G., Zheng, D. Q., Flanagan, W. M., Manzotti, M., and Languino, L. R. (1999). P27(kip1) acts as a downstream effector of and is coexpressed with the beta1C integrin in prostatic adenocarcinoma. J. Clin. Invest. 103, 321–329. Foster, J. S., Henley, D. C., Bukosky, A., Seth, P., and Wimalasena, J. (2001). Multifaceted regulation of cell cycle progression by estrogen: regulation of cdk inhibitors and Cdc25 independent of cyclin D1-cdk4 function. Mol. Cell Biol. 21, 794–810. Geradts, J., and Ingraham, C. D. (2000). Abnormal expression of cell cycle regulatory proteins in ductal and lobular carcinomas of the breast. Mod. Pathol. 13, 945–953. Geradts, J., and Wilson, P. A. (1996). High frequency of aberrant p16 (INK4A) expression in human breast cancer. Am. J. Pathol. 149, 15–20. Gray-Bablin, J., Zalvide, J., Fox, M. P., Knickerbocker, C. J., DeCaprio, J. A., and Keyomarsi, K. (1996). Cyclin E, a reductant cyclin in breast cancer. Proc. Natl. Acad. Sci. USA 93, 15215–15220.

52

¨ Goran Landberg

Guan, K. L., Jenkins, C. W., Li, Y., Nichols, M. A., Wu, X., O’Keefe, C. L., Matera, A. G., and Xiong, Y. (1994). Growth suppression by p18, a p16INK4/MTS1- and p14INK4B/MTS2related CDK6 inhibitor, correlates with wild-type pRb function. Genes Dev. 8, 2939– 2952. Haber, D., and Harlow, E. (1997). Tumour-suppressor genes: Evolving deﬁnitions in the genomic age. Nat. Genet. 16, 320–322. Hanahan, D., and Weinberg, R. A. (2000). The hallmarks of cancer. Cell 7, 57–70. Harper, J. W., Elledge, S. J., Keyomarsi, K., Dynlacht, B., Tsai, L. H., Dobrowolski, S., Bai, C., Connell-Crowley, L., Swindell, E., Fox, M. P., and Wei, N. (1995). Inhibition of cyclindependent kinases by p21. Mol. Biol. Cell 6, 387–400. Harwell, R. M., Porter, D. C., Danes, C., and Keyomarsi, K. (2000). Processing of cyclin E differs between normal and tumor breast cells. Cancer Res. 15, 481–489. Harvey, K. J., Blomquist, J. F., Ucker, D. S., and Kirschner, M. W. (1998). Commitment and effector phases of the physiological cell death pathway elucidated with respect to Bcl-2 caspase, and cyclin-dependent kinase activities. Mol. Cell Biol. 18, 2912–2922. Hengst, L., and Reed, S. I. (1996). Translational control of p27Kip1 accumulation during the cell cycle. Science 271, 1861–1864. Hirai, H., and Sherr, C. J. (1996). Interaction of D-type cyclins with a novel myb-like transcription factor, DMP1. Mol. Cell Biol. 16, 6457–6467. Hui, R. H., Macmillan, R. D., Kenny, F. S., Musgrove, E. A., Blamey, R. W., Nicholson, R. I., Robertsoon, J. F. R., and Sutherland, R. L. (2000). INK4a gene expression and methylation in primary breast cancer: overexpression of p16INK4a messenger RNA is a marker of poor prognosis. Clin. Cancer Res. 6, 2777–2787. Hutchinson, J. N., and Muller, W. (2000). Transgenic mouse models of human breast cancer. Oncogene 19, 6130–6137. Inoue, K., and Sherr, C. J. (1998). Gene expression and cell cycle arrest mediated by transcription factor DMP1 is antagonized by D-type cyclins through a cyclin-dependent-kinaseindependent mechanism. Mol. Cell Biol. 18, 1590–1600. Katayose, Y., Kim, M., Rakkar, A. N., Li, Z., Cowan, K. H., and Seth, P. (1997). Promoting apoptosis: a novel activity associated with the cyclin-dependent kinase inhibitor p27. Cancer Res. 57, 5441–5445. Kato, J. Y., Matsuoka, M., Polyak, K., Massague, J., and Sherr, C. J. (1994). Cyclic AMPinduced G1 phase arrest mediated by an inhibitor (p27Kip1) of cyclin-dependent kinase 4 activation. Cell 79, 487–496. Keyomarsi, K., Conte, D. Jr., Toyofukum, W., and Fox, M. P. (1995). Deregulation of cyclin E in breast cancer. Oncogene 11, 941–950. LaBaer, J., Garrett, M. D., Stevenson, L. F., Slingerland, J. M., Sandhu, C., Chou, H. S., Fattaey, A., and Harlow, E. (1997). New functional activities for the p21 family of CDK inhibitors. Genes Dev. 11, 847–862. Lamb, J., Lahda, M. H., Mcmahon, C., Sutherland, R. L., and Ewen, M. E. (2000). Regulation of the functional interaction between cyclin D1 and the estrogen receptor. Mol. Cell Biol. 20, 8667–8675. Landberg, G., and Roos, G. (1997). The cell cycle in breast cancer. APMIS 105, 575–589. Landberg, G., Nielsen, N. H., Nilsson, P., Emdin, S. O., Cajander, J., and Roos, G. (1997). Telomerase activity is associated with cell cycle deregulation in human breast cancer. Cancer Res. 57, 549–554. Landberg, G., Ostlund, H., Nielsen, N. H., Roos, G., Emdin, S., Burger, A. M., and Seth, A. (2001). Downregulation of the potential suppressor gene IGFBP-rP1 in human breast cancer is associated with inactivation of the retinoblastoma protein, cyclin E overexpression and increased proliferation in estrogen receptor negative tumors. Oncogene 14, 3497– 3505.

Cell Cycle and Breast Cancer

53

Lin, B. T., Gruenwald, S., Morla, A. O., W. H., Lee, W. H., and Wang, J. Y. (1991). Retinoblastoma cancer suppressor gene product is a substrated of the cell cycle regulator cdc2 kinase. EMBO J. 10, 857–864. Lin, S. Y., Xia, W., Wang, J. C., Kwong, K. Y., Spohn, B., Wen, Y., Pestell, R. G., and Hung, M. C. (2000). B-catenin, a novel prognostic marker for breast cancer: its roles in cyclin D1 expression and cancer progression. Proc. Natl. Acad. Sci. USA 97, 4262–4266. Loda, M., Cukor, B., Tam, S. W., Lavin, P., Fiorentino, M., Draetta, G. F., Jessup, J. M., and Pagano, M. (1997). Increased proteasome-dependent defradation of the cyclin-dependent kinase inhibitor p27 in aggressive colorectal carcinomas. Nat. Med. 3, 231–234. Loden, ´ M., Nielsen, N. H., Roos, G., Emdin, S. O., and Landberg, G. (1999). Cyclin E dependent kinase activity in human breast cancer in relation to cyclin E, p27 and p21 expression and retinoblastoma protein phosphorylation. Oncogene 18, 2557–2566. Lukas, J., Aagard, L., Strauss, M., and Bartek, J. (1995a). Oncogenic aberrations of p16/ INK4/CDKN2 and cyclin D1 cooperate to deregulate G1 control. Cancer Res. 55, 4818– 4823. Lukas, J., Parry, D., Aagaard, L., Mann, D. J., Bartkova, J., Strauss, M., Peters, G., and Bartek, J. (1995b). Retinoblastoma-protein-dependent cell-cycle inhibition by the tumour suppressor p16. Nature 375, 503–506. Masciullo, V., Sgambato, A., Pacilio, C., Pucci, B., Ferrandina, G., Palazzo, J., Carbone, A., and Cittadini, A. R. (1999). Frequent loss of expression of the cyclin-dependent kinase inhibitor p27 in epithelial ovarian cancer. Cancer Res. 59, 3790–3794. McMahon, C., Suthiphongchai, T, DiRenzo, J., and Ewen, M. E. (1999). P/CAF associates with cyclin D1 and potentiates its activation of the estrogen receptor. Proc. Natl. Acad. Sci. USA 96, 5382–5387. ¨ Milde-Langosch, K., Bambergerm, A. M., Methner, C., Rieck, G., and d Loning, T. (2000). Expression of cell cycle-regulatory proteins Rb, p16/MTS1, p27/KIP1, p21/WAF1, cyclin D1 and cyclin E in breast cancer: correlations with expression of activating protein-1 family members. Int. J. Cancer 87, 468–472. Millard, S. S., Vidal, A., Markus, M., and Koff, A. (2000). A U-rich element in the 5untranlated ´ region is necessary for the translation of p27 mRNA. Mol. Cell Biol. 20, 5947–5959. Muller, H., and Helin, K. (2000). The E2F transcription factors: key regulators of cell proliferation. Biochem. Biophys. Acta 1470, M1–M12. Nakayama, K., Nagahama, H., Minamishima, Y. A., Matsumoto, M, Nakamichi, I., Kitagawa, K., Shirane, M., Tsunematsu, R., Tsukiyama, T., Ishida, N., Kitagawa, M., Nakayama, K. I., and Hatakeyama, S. (2000). Targeted disruption of Skp2 results in accumulation of cyclin E and p27Kip1, polyploidy and centrosome overduplication. EMBO J. 19, 2069–2081. Nguyen, H., Gitig, D. M., and Koff, A. (1999). Cell-free degradation of p27(kip1), a G1 cyclindependent kinase inhibitor, is dependent on CDK2 activity and the proteasome. Mol. Cell Biol. 19, 1190–1201. ¨ C., Emdin, S. O., and Landberg, G. (1996). Cyclin E overexpression, Nielsen, N. H., Arnerlov, a negative prognostic factor in breast cancer with strong correlation to oestrogen receptor status. British J. Cancer 74, 874–880. Nielsen, N. H., Emdin, S. O., Cajander, J., and Landberg, G. (1997). Deregulation of cyclin E and D1 in breast cancer is associated with inactivation of the retinoblastoma protein. Oncogene 14, 295–304. Nielsen, N. H., Lod´en, M., Cajander, J., Emdin, S. O., and Landberg, G. (1999). G1-S transition defects occur in most breast cancers and predict outcome. Breast Cancer Res. Treatm. 56, 105–112. Nielsen, N. H., Roos, G., Emdin, S. O., and Landberg, G. (2001). Methylation of the p16Ink4a tumor suppressor gene 5. CpG island in breast cancer. Cancer Lett. 163, 59–69.

54

¨ Goran Landberg

Ohtsubo, M., Theodoras, A. M., Schumacher, J., Roberts, J. M., and Pagano, M. (1995). Human cyclin E, a nuclear protein essential for the G1-to-S phase transition. Mol. Cell Biol. 15, 2612–2624. Orend, G., Hunter, T., and Ruoslahti, E. (1998). Cytoplasmic displacement of cyclin E-cdk2 inhibitors p21Cip1 and p27Kip1 in anchorage-independent cells. Oncogene 16, 2575–2583. Perou, C. M., Sorile, T., Eisen, M. B., van der Rijn, M., Jeffrey, S. S., Rees, C. A., Pollack, J. R., Ross, D. T., Johnsen, H., Akslen, L. A., Fluge, O., Pergamenschikov, A., Williams, C., Zhu, S. X., Lonning, P. E., Borresen-Dale, A- L., Brown, P. O., and Botstein, D. (2000). Molecular portraits of human breast tumours. Nature 406, 747–752. Philipp-Staheli, J., Payne, S. R., and Kemp, C. J. (2001). P27KIP1: Regulation and function of a haploinsufﬁcient tumor suppressor and its misregulation in cancer. Exp. Cell. Res. 264, 148–168. Planas-Silva, M. D., and Weinberg, R. A. (1997). Estrogen-dependent cyclin E-cdk2 activation through p21 redistribution. Mol. Cell Biol. 17, 4059–4069. Planas-Silva, M. D., Shang, Y., Donaher, J. L., Brown, M., and Weinberg, R. A. (2001). A1B1 enhances estrogen-dependent induction of cyclin D1 expression. Cancer Res. 61, 3858–3862. Polyak, K., Lee, M. H., Erdjument-Bromage, H., Koff, A., Roberts, J. M., Tempst, P., and Massague, J. (1994). Cloning of p27Kip1, a cyclin-dependent kinase inhibitor and a potential mediator of extracellular antimitogenic signals. Cell 78, 59–66. ˜ Ponce-Castaneda, M. V., Lee, M. H., Latres, E., Polyak, K., Lacombe, L., Montgomery, K., Mathew, S., Krauter, K., Sheinfeld, J., Massague, J., and Cordon-Cardo, C. (1995). p27Kip1: chromosomal mapping to 12p12-12p13.1 and absence of mutations in human tumors. Cancer Res. 55, 1211–1214. Porter, D. C., and Keyomarsi, K. (2000). Novel splice variants of cyclin E with altered substrate speciﬁcity. Nucleic Acids Res. 28, E101. Porter, P. L., Malone, K. E., Heagerty, P. J., Alexander, G. M., Gatti, L. A., Firpo, E.J., Daling, J. R., and Roberts, J. M. (1997). Expression of cell-cycle regulators p27kip1 and cyclin E, alone and in combination, correlate with survival in young breast cancer patients. Nature Med. 3, 222–225. Prall, O. W. J., Rogan, E. M., Musgrove, E. A., Watts, C. K. W. W., and Sutherland, R. L. (1998). C-Myc or cyclin D1 mimics estrogen effect on cyclin E-cdk2 activation and cell cycle reentry. Mol. Cell Biol. 18, 4499–4508. Qian, X., Jin, L., Kulig, E., and Lloyd, R. V. (1998). DNA methylation regulates p27kip1 expression in rodent pituitary cell lines. Am. J. Pathol. 153, 1475–1482. Quelle, D. E., Cheng, M., Ashmun, R. A., and Sherr, C. J. (1997). Cancer associated mutations at the INK4a locus cancel cell cycle arrest by p16INK4a but not by the alternative reading frame protein p19ARF. Proc. Natl. Acad. Sci. USA 94, 669–673. Reed, S. I. (1997). Control of the G1/S transition. Cancer Surv. 29, 7–23. Rocco, W. R., and Sidransky, D. (2001). P16(MTS-1/CDKN2/INK4a) in cancer progression. Exp. Cell Res. 264, 42–55. Sanchez-Beato, ´ M., Saez, A. I., Martinez-Montero, J. C., Sol-Mateo, M., Sanchez-Verde, L., Villuendas, R., Troncone, G., and Piris, M. A. (1997). Cyclin-dependent kinase inhibitor p27KIP1 in lymphoid tissue: p27KIP1 expression is inversely proportional to the proliferative index. Am. J. Pathol. 151, 151–160. Santoni-Rugiu, E., Falck, J., Mailand, N., Bartek, J., and Lukas, J. (2000). Involvement of Myc activity in a G(1)/S-promoting mechanism parallell to the pRb/E2F pathway. Mol. Cell Biol. 20, 3497–3509. Schuuring, E., Verhoeven, E., Tinteren, H., Peterse, J. L., Nunnink, B., Thunnissen, F. B. J. M., Devilee, P., Cornelisse, C. J., Vijer, M. J., Mooi, W. J., and Michalides, R. J. A. M. (1992). Ampliﬁcation of genes within chromosome 11q13 region is indicative of poor prognosis in patients with operable breast cancer. Cancer Res. 52, 5229–5234.

Cell Cycle and Breast Cancer

55

Sharpless, N. E., and DePinho, R. A. (1999). The INK4A/ARF locus and its two gene products. Curr. Opin. Genet. Dev. 9, 22–30. Sheaff, R. J., Groudine, M., Gordon, M., Roberts, J. M., and Clurman, B. E. (1997). Cyclin E-cdk2 is a regulator of p27Kip1. Genes Dev. 11, 1464–1478. Sherr, C. J. (1996). Cancer cell cycles. Science 274, 672–1677. Sherr, C. J. (2000). The Pezcoller Lecture: Cancer cell cycles revisited. Cancer Res. 60, 3689– 3695. Shoker, B. S., Jarvis, C., Davies, M. P., Iqbal, M., Sibson, D. R., and Sloane, J. P. (2001). Immunodetectable cyclin D(1) is associated with oestrogen receptor but not Ki67 in normal, cancerous and precancerous breast lesions. Br. J. Cancer 20, 1064–1069. Sicinski, P., Donaher, J. L., Parker, S. B., Li, T., Fazeli, A., Gardner, H., Haslam, S. Z., Bronson, R. T., Elledge, S. J., and Weinberg, R. A. (1995). Cyclin D1 provides a link between development and oncogenesis in the retina and breast. Cell 82, 621–630. Snijders, A. M., Meijer, G. A., Brakenhoff, R. H., van den Brule, A. J. C., and van Diest, P. J. (2000). Microarray techniques in pathology: tool or toy? J. Clin. Pathol: Mol. Pathol. 53, 289–294. Spruck, C. H., Won, K. A., and Reed, S. (1999). Deregulated cyclin E induces chromosomal instability. Nature 401, 297–300. Spruck et al. (2000). p. 16. Steeg, S. P., and Zhou, Q. (1998). Cyclins and breast cancer. Breast Cancer Res. Treatm. 52, 17–28. Swarbrick, A., Lee, C. S. L., Sutherland, R. L., and Musgrove, E. A. (2000). Cooperation of p27Kip1 and p18INK4c in progestin-mediated cell cycle arrest in T47D breast cancer cells. Mol. Cell Biol. 20, 2581–2591. Tan, P., Cady, B., Wanner, M., Worland, P., Cukor, B., Magi-Galluzzi, C., Lavin, P., Draetta, G., Pagano, M., and Loda, M. (1997). The cell cycle inhibitor p27 is an independent prognostic marker in small (T1a,b) invasive breast carcinomas. Cancer Res. 57, 1259–1263. Vlach, J., Hennecke, S., and Amati, B. (1997). Phosphorylation-dependent degradation of the cyclin-dependent kinase inhibitor p27. EMBO J. 16, 5334–5344. Wang, T. C., Cardiff, R. D., Zukerberg, L., Lees, E., Arnold, A., and Schmidt, E. V. (1994). Mammary hyperplasia and carcinoma in MMTV-cyclin D1 transgenic mice. Nature 369, 669–671. Weng, L. P., Brown, J. L., and Eng, C. (2001). PTEN coordinates G(1) arrest by down-regulating cyclin D1 via its protein phosphatase activity and up-regulating p27 via its lipid phosphatase activity in a breast cancer model. Hum. Mol. Genet. 15, 599–604. Wong, S. C. C., Chan, J. K. C., Lee, K. C., and Hsiao, W. L. W. (2001). Differential expression of p16/p21/p27 and cyclin D1/D3, and their relationships to cell proliferation, apoptosis, and tumor progression in invasive ductal carcinoma of the breast. J. Pathol. 194, 35–42. Wu, L., Chen, P., Shum, C. H., Chen, C., Barsky, L. W., Weinberg, K. I., Jong, A., and Triche, T. J. (2001). MAT1-modulated CAK activity regulates cell cycle G(1) exit. Mol. Cell Biol. 21, 260–270. Zhang, H. S., Postigo, A. A., and Dean, D. C. (1999). Active transcriptional repression by the Rb-E2F complex mediates G1 arrest triggered by p16INK4a, TGFbeta, and contact inhibition. Cell 97, 53–61. Zheng, L., and Lee, W. H. (2001). The retinoblastoma gene: A prototypic and multifunctional tumor suppressor. Exp. Cell Res. 264, 2–18. Zhou, Q, Wulkuhle, J., Ouatas, T., Fukushima, P., Steler-Stevenson, M., Miller, F. R., and Steeg, P. S. (2000). Cyclin D1 overexpression in a model of human breast premalignancy: preferential stimulation of anchorage-independent but not anchorage-dependent growth is associated with increased cdk2 activity. Breast Cancer Res. Treatm. 59, 27–39.

56

¨ Goran Landberg

Zwijsen, R. M. L., Wientjens, E., Klompmaker, R., van der Sman, J., Bernards, R., and Michalides, R. J. A. M. (1997). CDK-independent activation of estrogen receptor by cyclin D1. Cell 88, 405–415. Zwijsen, R. M., Buckle, R. S., Hijmans, E. M., Loomans, C. J., and Bernards, R. (1998). Ligandindependent recruitment of steroid receptor coactivators to estrogen receptor by cyclin D1. Genes Dev. 12, 3488–3498.

Rho GTPases in Transformation and Metastasis Aron B. Jaffe and Alan Hall MRC Laboratory for Molecular Cell Biology and Cell Biology Unit CRC Oncogene and Signal Transduction Group University College London London WC1E 6BT, U.K.

I. Introduction II. Rho GTPases and Transformation A. GEFs and GAPs B. Rho GTPases III. Rho GTPases and Metastasis A. Tiam-1 B. Rho IV. Conclusions and Future Directions References

During the development and progression of human cancer, cells undergo numerous changes in morphology, proliferation, and transcriptional proﬁle. Over the past couple of decades there have been intense efforts to understand the molecular mechanisms involved, and members of the Ras superfamily of small GTPases have emerged as important players. Mutated versions of the Ras genes were ﬁrst identiﬁed in human cancers some 20 years ago, but more recently, the Rho branch of the family has been receiving increased attention. In addition to the experimental evidence implicating Rho GTPase signaling in promoting malignant transformation, genetic analysis of human cancers has now revealed a few examples of direct alterations in the genes encoding regulators of Rho GTPases. In this review, we discuss the evidence implicating Rho GTPases in transformation and metastasis, as well as the progress made toward identifying their biochemical mechanism of action. C 2002 Academic Press.

I. INTRODUCTION Rho GTPases constitute a distinct family of the Ras superfamily of small GTPases. There are 24 predicted mammalian Rho GTPases (Schultz et al., 1998), 18 of which have been described, that share between 50 and 90% amino acid sequence homology: RhoA (Yeramian et al., 1987), RhoB, RhoC (Chardin et al., 1988), RhoD (Shimizu et al., 1997), RhoE/Rnd3 (Foster et al., Advances in CANCER RESEARCH 0065-230X/02 $35.00

57

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

58

Aron B. Jaffe and Alan Hall

1996), Rnd1, Rnd2 (Nobes et al., 1998), RhoG (Vincent et al., 1992), Rac1, Rac2 (Didsbury et al., 1989), Rac3 (Haataja et al., 1997), Cdc42 (Shinjo et al., 1990), TC10 (Drivas et al., 1990), RhoH/TTF (Dallery et al., 1995), Chp (Aronheim et al., 1998), TCL (Vignal et al., 2000), Rif (Ellis and Mellor, 2000), and Wrch-1 (Tao et al., 2001). The most intensively studied members are Rho, Rac, and Cdc42. Like all GTPases, they cycle between an active, GTP-bound state and an inactive, GDP-bound state. The balance between these two forms is regulated by three groups of factors (Van Aelst and D’Souza-Schorey, 1997) (Fig. 1). Guanine nucleotide exchange factors (GEFs) catalyze exchange of GDP for GTP, leading to activation in response Stimulus

GTP

GDP GEF

GDI

GDP

GTP GTPase

GTPase

effector

GAP Pi

GDI Downstream Signaling

Fig. 1 GTPase “life” cycle. Cycling between the active, GTP-bound, and the inactive, GDPbound forms is regulated by guanine nucleotide exchange factors (GEFs), GTPase activating proteins (GAPs), and guanine nucleotide dissociation inhibitors (GDIs). GTP-bound GTPase interacts with different target proteins (effectors), which mediate various cellular effects, such as actin cytoskeletal and transcriptional changes, and cell cycle progression.

Rho GTPases and Cancer

59

to various signals. GTPase-activating proteins (GAPs) increase the intrinsic GTPase activity, resulting in inactivation of the protein. Finally, guanine nucleotide dissociation inhibitors (GDIs) have the ability to block the cycling between the GDP- and GTP-bound forms by preventing the exchange of GDP for GTP, as well as inhibiting the intrinsic and GAP-stimulated GTPase activity. In the active, GTP-bound state, Rho GTPases interact with an ever-growing number of protein targets, termed effectors, to mediate numerous downstream cellular effects (Bishop and Hall, 2000). Their best-characterized role is in regulating actin cytoskeletal dynamics, and microinjection studies with constitutively active or dominant negative forms of Rho, Rac, and Cdc42 have revealed that Rho regulates the formation of stress ﬁbers, Rac controls the formation of lamellipodia or membrane rufﬂes, and Cdc42 governs the formation of ﬁlopodia (Kozma et al., 1995; Nobes and Hall, 1995; Ridley and Hall, 1992; Ridley et al., 1992). The regulation of the actin cytoskeleton by Rho GTPases has been implicated in promoting a variety of cellular processes such as changes in morphology, motility, and adhesion (Hall, 1998). Further studies have revealed signiﬁcant crosstalk between small GTPases; Ras and Cdc42 can each activate Rac, whereas Rac can either activate or inhibit Rho, depending on the cellular context (Nobes and Hall, 1995; Ridley et al., 1992; Sander et al., 1999). The extent to which GTPase-controlled pathways act alone or in combination during various biological processes is currently an intense area of study. Rho GTPases have been reported to have additional cellular activities that appear to be distinct from their effects on the actin cytoskeleton. In particular, Rac and Cdc42 can activate two MAP kinase pathways, JNK and p38, and all three Rho GTPases have been reported to activate the transcription factors SRF and NFκB (Van Aelst and D’Souza-Schorey, 1997). This suggests that these GTPases can regulate gene expression. In addition, several investigators have shown that Rho, Rac, and Cdc42 are required for G1 progression during the cell cycle (Olson et al., 1995), though whether this is due to their effects on gene transcription is not clear. Genetic analyses in Drosophila melanogaster and Caenorhabditis elegans, as well as molecular and cell biological studies using mammalian cells, have revealed that many biological processes, such as cell migration and cell differentiation, require changes both in the actin cytoskeleton and in gene expression. The ability of Rho GTPases to regulate the organization of the actin cytoskeleton coordinately with changes in gene transcription appears to be a key feature of their widespread role during development and in the adult. Alterations in cellular morphology, differentiation state, and cell motility are seen during the progression of a variety of human diseases, particularly cancer. During the process of tumorigenesis, a cell undergoes numerous morphological and behavioral changes, including increased proliferation, loss

60

Aron B. Jaffe and Alan Hall

of contact inhibition and serum dependence for growth, and the ability to grow in an anchorage-independent manner (Hanahan and Weinberg, 2000). In addition, a hallmark of many tumors is their ability to metastasize and invade neighboring tissues. This is thought to be a multistep process in which tumor cells sustain numerous alterations in their cytoskeletal organization and gene expression pattern, leading to changes in their adhesive properties, motility, and morphology. In this review, we will discuss the evidence implicating the Rho family of small GTPases in transformation and metastasis.

II. RHO GTPASES AND TRANSFORMATION The small GTPase Ras was the ﬁrst human oncogene identiﬁed (Santos et al., 1982; Hall et al., 1983), and it is now known that a Ras gene is mutated in approximately 30% of human tumors (Bos, 1989). Activated forms of the Ras protein (oncogenic Ras) have the ability to transform cells in culture and to promote tumor formation in mice (Greig et al., 1985; Stacey and Kung, 1984). Early studies on oncogenic Ras found that overexpression in ﬁbroblasts resulted in membrane rufﬂing (Bar-Sagi and Feramisco, 1986), and this was later shown to be due to activation of the endogenous Rac GTPase (Ridley et al., 1992). Since then, there has been a great deal of interest in whether Rho GTPases, like their Ras cousin, can contribute to the transformed phenotype of tumor cells. Two general lines of evidence supporting a role for Rho GTPases in promoting cellular transformation have come from the analysis of (A) regulators of Rho GTPases (GEFs and GAPs) and (B) mutant forms (activated and dominant-negative) of the Rho GTPases themselves.

A. GEFs and GAPs There are more than 50 predicted GEFs for the Rho GTPase family in the human genome (Venter et al., 2001), each characterized by a Dbl-homology (DH) domain immediately adjacent to a pleckstrin-homology (PH) domain. Many of these proteins were initially identiﬁed experimentally by their ability to induce foci when overexpressed in NIH3T3 ﬁbroblasts (Table I), and only later found to have exchange factor activity toward Rho GTPases (Glaven et al., 1996; Hart et al., 1991; Zheng et al., 1995). Transformation by some Rho GEFs has been shown to (a) induce cellular phenotypes that closely resemble those induced by expressing constitutively activated forms of Rho GTPases (Khosravi-Far et al., 1994), and (b) be inhibited by dominantnegative forms of Rho GTPases (Zheng et al., 1995). This suggests that

61

Rho GTPases and Cancer

Table I Oncogenic Guanine Nucleotide Exchange Factors GEF

GTPase speciﬁcity

Reference

Db1 Dbs/Ost Ect2 Fgd1 Lbc Lfc Lsc NET1 Tiam-1 Tim Vav

Rho, Cdc42 Rho, Cdc42 Rho, Rac, Cdc42 Cdc42 Rho Rho Rho Rho Rac ND Rho, Rac, Cdc42

Hart et al., 1991, 1994 Horii et al., 1994; Whitehead et al., 1995, 1999 Miki et al., 1993; Tatsumoto et al., 1999 Whitehead et al., 1998; Zheng et al., 1996 Zheng et al., 1995 Glaven et al., 1996 Glaven et al., 1996 Alberts and Treisman, 1998; Chan et al., 1996 Michiels et al., 1995; van Leeuwen et al., 1995 Chan et al., 1994 Katzav et al., 1989; Olson et al., 1996

the mechanism by which the Rho GEFs mediate transformation is via deregulated activation of endogenous Rho GTPases. It has also been shown that transformation by other oncogenes is mediated, at least in part, by Rho GEF activation (Fukuhara et al., 2001). Dominantnegative forms of Rho and Rac, for example, inhibit the focus-forming ability of the G-protein-coupled receptors Mas and G2A (Zohn et al., 1998, 2000). G2A-induced transformation is mediated by activation of the heterotrimeric G protein subunit Gα 13, which in turn interacts with a regulator of G protein signaling (RGS) domain in the Rho GEF Lsc (Zohn et al., 2000). Direct evidence for a role for Rho GEFs in human cancer is still rare, and there have only been two reports of tumor-associated mutations (Engers et al., 2000; Kourlas et al., 2000) (Table II). Tiam-1, a Rac-speciﬁc GEF, was found to have a point mutation in the PH domain of one of the two alleles, resulting in an alanine → glycine substitution at amino acid 441, in around 10% (4 of 35) of the renal cell carcinoma (RCC) samples (primary tumors and tumor cell lines) examined (Engers et al., 2000). Expression of this mutant in ﬁbroblasts induced focus formation to a similar degree as constitutively active (i.e., N-terminally deleted) Tiam-1, whereas wildtype Tiam-1 expression had no focus-forming activity, suggesting that the A441G mutation may act in a dominant gain-of-function manner in vivo. Surprisingly, 3 of 5 RCC cell lines had decreased expression of Tiam-1, including 2 that contained the A441G mutation. The signiﬁcance of this in light of the gain-of-function activity of this mutation is unclear. Another Rho GEF, leukemia-associated Rho guanine nucleotide exchange factor (LARG), has been isolated as a fusion partner of the mixed-lineage leukemia (MLL) gene in a patient with primary acute myeloid leukemia (AML) (Kourlas et al., 2000). This fusion protein contained the N-terminal portion of MLL in frame with the C-terminal 80% of LARG, which includes the DH and

Table II Tumor-associated Alterations in Components of Rho GTPase Signaling Pathways Gene

GTPase speciﬁcity

Tumor type

deleted in liver cancer 1 (DLC1)

Rho

Hepatocellular carcinoma (HCC)

p190-A

Rho

Gliomas/astrocytomas

GTPase regulator associated with the focal adhesion kinase pp125FAK (GRAF) Leukemia-associated Rho guanine nucleotide exchange factor (LARG)

Rho

Myelodysplastic syndrome/acute myeloid leukemia

Rho

Acute myeloid leukemia

Tiam1

Rac

Renal cell carcinoma (RCC)

Rho RhoA RhoC RhoH/TTF

Colon, breast, lung Testicular germ cell tumor Inﬂammatory breast cancer Pancreatic adenocarcinoma non-Hodgkin’s lymphoma (NHL) multiple myeloma

Rac1

Colorectal

Rac3

Breast

Aberration(s) One copy deleted in 7/16 primary HCCs and 10/11 HCC cell lines, and not expressed in 4/14 HCC cell lines Within a region that is deleted/rearranged 3/13 patients with one allele deleted have loss-of-function mutations in the other allele 5 end of the mixed-lineage leukemia (MLL) gene fused to the 3 end of LARG Decreased expression in 3/5 RCC cell lines; A441G substitution in 4/35 primary RCC tumors and tumor cell lines Increased protein levels Increased mRNA levels Increased mRNA levels Rearranged in 4/4 NHL patients and 1 NHL cell line (VAL) with +(3; 4) translocations; rearranged in 1 patient with multiple myeloma with +(4; 14)(p13; q32) translocation Increased mRNA levels of a splice variant resulting in a 19aa in-frame insertion Increased levels of GTP-bound Rac3 in highly proliferative breast cancer cell lines and 1/3 breast tumor samples

Reference Yuan et al., 1998; Homma and Emori, 1995 Tikoo et al., 2000; Ridley et al., 1993 Borkhardt et al., 2000

Kourlas et al., 2000

Engers et al., 2000; Michiels et al., 1995

Fritz et al., 1999 Kamai et al., 2001 van Golen et al., 1999; Suwa et al., 1998 Dallery et al., 1995; Preudhomme et al., 2000

Jordan et al., 1999

Mira et al., 2000

Rho GTPases and Cancer

63

PH domains, a domain with homology to the Rho GEF Lsc, and a nuclear localization signal (Kourlas et al., 2000). Overexpression of full-length or an N-terminally truncated form of LARG, corresponding to the region found fused to MLL, cooperated with Raf to transform ﬁbroblasts in culture, but neither induced transformation alone (Reuther et al., 2001). This is a surprising result, since LARG has GEF activity toward Rho (Reuther et al., 2001) and might therefore be expected to induce foci in a similar manner other Rho-speciﬁc GEFs such as Lbc. It remains to be established whether the MLL–LARG fusion protein is sufﬁcient to induce transformation in myeloid cells and how frequently the locus is affected in human cancer. Deregulated activation of Rho GTPases could also be achieved by a loss-offunction mutation, or decreased expression, of a GAP (Fig. 1). Studies have found deletions or mutations in Rho GAPs in different cancers (Borkhardt et al., 2000; Tikoo et al., 2000; Yuan et al., 1998). One of these, the human ortholog of the chicken GRAF (GTPase regulator associated with the focal adhesion kinase pp125FAK) gene, was isolated as a MLL fusion partner from a patient with juvenile myelomonocytic leukemia (Borkhardt et al., 2000). The fusion protein consisted of the N-terminal portion of MLL fused to the C-terminal region of GRAF, which lacks the GAP domain. Interestingly, human GRAF maps to chromosomal region 5q31, which is commonly deleted in myelodysplastic syndromes (MDSs) and AMLs (Van den Berghe and Michaux, 1997). Sequence analysis revealed that 3 of 13 patients in whom one allele of GRAF was deleted had mutations in the other GRAF allele that either inactivate or remove the GAP domain (Borkhardt et al., 2000). These results suggest that GRAF is a tumor suppressor gene that may contribute to the development of some hematological tumors. Two other GAPs, p190-A and deleted in liver cancer (DLC-1), are also localized to regions that are deleted in various tumor types (see Table II) (Tikoo et al., 2000; Yuan et al., 1998); however, their contribution to tumorigenesis remains unclear, since loss of heterozygosity at either locus has not been described. These in vivo observations of tumor-associated mutations suggest that gain-of-function alterations in GEFs, or loss-of-function alterations in GAPs, may play an important role in human cancers. With more than 50 GEFs and 60 GAPs in the human genome (Venter et al., 2001), it is too early to tell how widespread their role might be.

B. Rho GTPases Expression of constitutively activated forms of Rho, Rac, or Cdc42 in ﬁbroblasts results in low-level focus formation (approximately 100-fold less frequent than with activated Ras). These foci are morphologically distinct

64

Aron B. Jaffe and Alan Hall

Fig. 2 Rho- and Ras-induced focus formation. (A) Rho-induced focus in NIH 3T3 ﬁbroblasts. Rho foci appear compact with highly refractile cells on the periphery. (B) Ras-induced focus in NIH 3T3 ﬁbroblasts. Shown is approximately 1/3 of a Ras focus, which displays a swirling and highly spread morphology.

from the swirling and spread-out foci induced by activated Ras (Khosravi-Far et al., 1995; Perona et al., 1993; Qiu et al., 1995a, 1995b, 1997; Self et al., 1993), and instead are highly compact, resembling foci induced by overexpression of members of the Rho GEF family of oncogenes (Khosravi-Far et al., 1994; Lin et al., 1997) (Fig. 2). Interestingly, transformation by members of the Rho GEF family is much more efﬁcient than transformation by constitutively active Rho GTPases. For example, while the focus formation activity of each constitutively active Rho GTPase alone is extremely weak, requiring high levels of expression (Khosravi-Far et al., 1995; Qiu et al., 1995a, 1995b), Rho GEFs possess high levels of focus-forming activity even when expressed at low levels (Chan

65

Rho GTPases and Cancer

Table III Comparison of Transformation Properties of Rho GTPase Alleles Enhanced Growth in growth in soft agar low serum

Enhanced saturation density Tumorigenic

Allele

Foci

Ras (V12 or L61)

++++

++++

++++

++++

++++

Rho (V14 or L63)

+/−

+/−

−

++

++++

Rho (L30) Rac (V12 or L61) Rac (L28) Cdc42 (V12) Cdc42 (L28) Db1

+++ +/− + ND + ++++

+ + + ++++ ++++ +++

++ ++ ++ − ++ ++++

+++ ++ +++ − +++ ++

++++ ++++ ++++ ++++ ++++ ++++

References Khosravi-Far et al., 1995, 1996; Qiu et al., 1997 Khosravi-Far et al., 1995; Perona et al., 1993; Qiu et al., 1995; Self et al., 1993 Lin et al., 1999 Qiu et al., 1995 Lin et al., 1999 Qiu et al., 1997 Lin et al., 1999 Lin et al., 1999

et al., 1996; Lin et al., 1999). The reasons for this difference are not entirely clear. One possibility is that Dbl family members have other functions in addition to their role as GEFs for Rho GTPases, but to date, no other activity has been described. An alternative possibility is that efﬁcient transformation by Rho GTPases requires continuous cycling between the GTP- and GDPbound states. To address this possibility, mutant versions of Rho (F30L), Rac (F28L), and Cdc42 (F28L) having an increased basal nucleotide exchange rate have been constructed (Lin et al., 1997, 1999). These “fast-cycling” mutants do appear to have a more potent transforming activity as judged by various in vitro assays (see Table III), suggesting that for at least some of the characteristics associated with transformation, a complete cycle of GTP binding and hydrolysis is required. Several groups have reported that cells overexpressing constitutively active Rho or Rac, or a “fast-cycling” mutant of Cdc42, display increased growth rates, grow to a higher cell density, and divide in low serum (Khosravi-Far et al., 1995; Lin et al., 1997; Perona et al., 1993; Qiu et al., 1995a). However, in each case, these cellular phenotypes are much weaker than those induced by overexpression of oncogenic Ras. This suggests that additional signaling pathways are activated by Ras and are required to produce a fully transformed phenotype. Interestingly, an activated version of Ras that is unable to bind to Raf induces foci that closely resemble those induced by activated forms of Rho GTPases (Khosravi-Far et al., 1996). Since it is well known that Rho GTPases do not, at least directly, activate the Raf/ERK MAP kinase

66

Aron B. Jaffe and Alan Hall

pathway, it is likely that this is a key difference between transformation by Ras and Rho GTPases. Although constitutively active forms of Rho, Rac, and Cdc42 have a weak transforming activity when expressed alone, they each greatly enhance the focus formation induced by Ras (Khosravi-Far et al., 1995; Qiu et al., 1995a, 1995b; Roux et al., 1997) or Raf (Khosravi-Far et al., 1995; Qiu et al., 1995a, 1995b). Interestingly, the morphological changes induced by coexpression of activated Rho or Rac and activated Raf are somewhat different from those seen when the factors are expressed alone (Khosravi-Far et al., 1995; Qiu et al., 1995b), supporting the idea that the contribution of Rho GTPases to transformation is mediated, at least in part, by signaling events that are different from those induced by Ras. Additional data implicating Rho GTPases in cellular transformation have come from work showing that coexpression of dominant-negative forms of Rho, Rac, or Cdc42 impairs the focus-forming activity and morphological changes induced by oncogenic Ras (Khosravi-Far et al., 1995; Qiu et al., 1995a, 1995b, 1997). Furthermore, transfection of dominant-negative Rho or Cdc42 inhibits the transformed phenotype of cells stably transfected with activated Ras (Khosravi-Far et al., 1995; Qiu et al., 1995b, 1997), and farnesyltransferase inhibitors, a class of cancer therapeutics (Prendergast and Oliff, 2000), appear to suppress Ras transformation, at least in part, by inhibiting the activity of RhoB (Lebowitz et al., 1995). These results suggest that efﬁcient transformation by Ras is dependent on activation of Rho GTPase pathways. Several reports suggest that Rho, Rac, and Cdc42 might each make different contributions to Ras-mediated transformation. Closer examination of the morphological changes induced by stable expression of constitutively active and “fast-cycling” forms of each Rho GTPase indicated that each member produced a distinct phenotype (Table III) (Khosravi-Far et al., 1995; Lin et al., 1997, 1999; Qiu et al., 1997), consistent with earlier observations from microinjection experiments (Nobes and Hall, 1995; Ridley and Hall, 1992; Ridley et al., 1992). In addition, dominant-negative Rho or Cdc42 was much more efﬁcient at reverting the morphological phenotype of Ras-transformed cells than was dominant-negative Rac (Khosravi-Far et al., 1995; Qiu et al., 1997), a rather surprising observation given that Ras can act as a potent activator of Rac (Ridley et al., 1992). Only dominant-negative Rho and Cdc42 can inhibit transformation by activated Raf (Qiu et al., 1995a, 1995b, 1997). It may be that Ras signals to Rho and Cdc42 via the Raf/MAPK pathway, whereas the signaling to Rac is Raf-independent. The importance of signaling pathways connecting Ras and Rho GTPases has not been explored in human cancers. Expression of combinations of activated forms of Rho, Rac, and Cdc42 produces a much higher focus-forming activity than expression of each

Rho GTPases and Cancer

67

individually, suggesting that they may activate different pathways contributing to transformation (Roux et al., 1997). In addition to the changes in cell morphology observed by expression of constitutively active Rho, Rac, or Cdc42, some studies found that each could induce colony formation in soft agar and tumors in nude mice (albeit to a far lesser degree than constitutively active Ras) (Khosravi-Far et al., 1995; Lin et al., 1997; Perona et al., 1993; Qiu et al., 1995a); however, other studies failed to observe the same effects for overexpressing activated Rho (Qiu et al., 1995a, 1995b; Self et al., 1993). Moreover, a direct comparison among “fast-cycling” mutants of Rho, Rac, and Cdc42 found that the ability of Rho and Rac to induce colony formation in soft agar was severalfold less than that of Cdc42, whereas the ability of Rac and Cdc42 to induce focus formation was far less than that of Rho (Lin et al., 1999). Finally, dominant-negative Rac, but not Cdc42, dramatically inhibited the growth of Ras-transformed cells in low serum (Qiu et al., 1997); however, other studies suggest that inhibiting Cdc42 does impair Rasdependent growth in low serum, albeit to a lesser degree than inhibition of Rho or Rac (Sahai et al., 2001). These rather confusing, and sometimes conﬂicting, observations presumably reﬂect differences in cell types and assay conditions used by the different groups.

1. BIOCHEMICAL CONTRIBUTIONS OF RHO TO TRANSFORMATION To identify the signaling pathways activated by Rho that contribute to cell transformation, numerous mutational studies have been undertaken. It appears that loop 6 (amino acids 76–92) and the insert region (amino acids 123–137) of Rho are required for transformation (Zong et al., 1999), since amino acid substitutions in loop 6 (D87V/D90A) or deletion of the insert domain, in the context of a constitutively active form of Rho, impaired its ability to cooperate with active Raf to induce transformation. Interestingly, deleting the insert domain of Rho did not disrupt its interaction with the serine/threonine (Ser/Thr) kinase p160 Rho kinase, an effector protein that mediates stress ﬁber formation (Amano et al., 1997). This suggests that the ability of Rho to interact with p160 Rho kinase and to promote stress ﬁber formation is not sufﬁcient for Rho-mediated transformation. Indeed, the contribution that Rho makes to the transformed phenotype may be less through its effects on the actin cytoskeleton and more through its effects on transcription and cell cycle progression, since transformation of ﬁbroblasts with oncogenic Ras is accompanied by a decrease in stress ﬁbers (Dartsch et al., 1994). Further analysis of Ras-transformed Swiss3T3 ﬁbroblasts revealed that although stress ﬁbers are decreased, cells contain elevated levels of GTP-bound, active Rho after Ras transformation (Sahai et al., 2001). Inhibiting MEK, but not PI3-K, activation restored stress ﬁbers in these cells

68

Aron B. Jaffe and Alan Hall

without altering the levels of active Rho. This suggests that components of the MAP kinase pathway block the signaling pathways downstream of Rho that lead to stress ﬁber assembly, perhaps at the level of p160 Rho kinase (Sahai et al., 2001). The down-regulation of p160 Rho kinase by Ras signaling may be a way to differentially promote particular downstream effects of Rho, such as proliferation or transcriptional regulation. It is worth noting that the presence of stress ﬁbers is not incompatible with at least some aspects of cellular transformation, since cells transformed by the Rho GEFs Lbc, Dbl, and Vav maintain normal stress ﬁber levels (Khosravi-Far et al., 1994; Zheng et al., 1995), and others have concluded that p160 Rho kinase activity is required for Rho-mediated transformation (Sahai et al., 1999). The relevance of the role of Rho-mediated cytoskeletal changes in transformation remains unclear. Rho has been shown to be required for the progression through the G1 phase of the cell cycle (Olson et al., 1995). Furthermore, in the absence of Rho activity, constitutively active Ras induces high levels of the cyclindependent-kinase inhibitor p21Waf1/Cip1, inhibiting cell cycle progression. Expression of active Rho overcomes this cell cycle block by reducing the levels of p21Waf1/Cip1. This effect appears to be at the level of transcription, since activation of Rho causes a repression of transcription from the p21Waf1/Cip1 promoter (Olson et al., 1998); however, the precise signaling pathway that mediates Rho regulation of p21Waf1/Cip1 levels remains to be elucidated.

2. BIOCHEMICAL CONTRIBUTIONS OF RAC TO TRANSFORMATION The transforming activity of Rac also appears to be distinct from its function in regulating the actin cytoskeleton. One study identiﬁed two amino acid substitutions (E31V and F37L) that impaired or completely blocked lamellipodia formation but left the ability to transform ﬁbroblasts, as judged by focus formation and colony formation in soft agar in cooperation with activated Raf (Westwick et al., 1997). Conversely, two other mutations (N26D and N43D), which impaired the ability of activated Rac to cooperate with Raf to transform cells were still able to induce lamellipodia as efﬁciently as activated Rac alone. These data suggest that the primary role of Rac in transformation may not involve regulating the actin cytoskeleton. Another study reported that constitutively activated Rac lacking the insert region lost its ability to stimulate DNA synthesis in cooperation with activated Raf (Joneson and Bar-Sagi, 1998). This Rac mutant was still able to signal to the actin cytoskeleton and to mediate at least some transcriptional responses. Interestingly, the insert region appears to be required for the production of reactive oxygen species (ROS), a Rac-mediated effect originally

Rho GTPases and Cancer

69

identiﬁed in phagocytic cells (Abo et al., 1991; Diekmann et al., 1994), but later reported in many other cell types (Sundaresan et al., 1996). Blocking ROS production inhibits Rac-induced stimulation of DNA synthesis in ﬁbroblasts (Joneson and Bar-Sagi, 1998). However, Karnoub and colleagues (Karnoub et al., 2001) have come to an opposite conclusion, namely that the insert region was completely dispensable for Rac-induced mitogenesis and cell transformation, but was required for membrane rufﬂing. The reason for the discrepancies between these two studies is not clear. The possibility that Rac contributes to transformation in human cancers is supported by another study that found high levels of the active, GTPbound form of the Rac family member Rac3, in rapidly proliferating breast cancer cell lines (Mira et al., 2000). In addition, Mira and colleagues also found increased active Rac3 in one out of three metastatic breast cancer tissue samples, although whether this represents a feature of 33% of all metastatic tumors remains to be shown. No genetic alterations were found in the Rac3 coding sequence, suggesting that the increased activity could be caused by increased GEF activity or decreased GAP activity. The proliferative effect of activated Rac3 may be mediated by p21-activated kinase (PAK), which is an effector for Rac and Cdc42, since increased levels of active PAK1 and PAK2 were found in cell lines with high Rac3-GTP levels. Moreover, dominant-negative Rac3, as well as various inhibitory fragments of PAK, blocked S-phase entry in one of the highly proliferative cell lines (Mira et al., 2000). Using a completely different cell type, immortalized ﬁbroblasts, others have reported that the interaction between Rac1 and PAK1 is neither necessary nor sufﬁcient for Rac1 to cooperate with activated Raf in transformation (Westwick et al., 1997).

3. CONTRIBUTION OF Cdc42 TO TRANSFORMATION Although studies have shown that Cdc42 is both necessary and sufﬁcient for progression through the G1 phase of the cell cycle of Swiss 3T3 ﬁbroblasts (Olson et al., 1995), others have reported that cells expressing constitutively active Cdc42 do not display enhanced growth in low serum and do not grow to a higher saturation density in normal serum, unlike cells expressing activated Rac (Qiu et al., 1997). However, expression of the “fast-cycling” Cdc42 mutant does result in increased serum-independent growth and higher saturation density (Lin et al., 1997), and dominant-negative Cdc42 inhibits proliferation induced by activated Ras, although apparently not as well as dominant-negative Rac or inhibition of Rho with C3 toxin (Sahai et al., 2001). In addition, a “fast-cycling” Cdc42 mutant promoted growth in soft agar severalfold more efﬁciently than “fast-cycling” Rac or Rho, suggesting that the primary role of Cdc42 in transformation might be in promoting anchorage-independent growth (Lin et al., 1999).

70

Aron B. Jaffe and Alan Hall

The insert region of Cdc42 has been reported to be essential for serum and anchorage-independent growth and increased saturation density, in the context of a “fast-cycling” Cdc42 (Wu et al., 1998). In addition, the insert region is completely dispensable for the actin cytoskeletal and at least some transcriptional effects (Wu et al., 1998), suggesting that these Cdc42 signaling pathways are not sufﬁcient for transformation. Binding studies using the “fast-cycling” mutant with or without the insert region have revealed an unexpected interaction with the γ -subunit of the coatomer complex (γ COP) that appears to be necessary and sufﬁcient for Cdc42-mediated transformation (Wu et al., 2000). The coatomer complex is involved in intracellular trafﬁcking; overexpression of the “fast-cycling” Cdc42, but not a mutant that abolished binding to γ COP, accelerated intracellular trafﬁcking from the ER to the Golgi, as judged by a time course of processing of the vesicular stomatitis virus glycoprotein (VSV-G). This interaction and its apparent requirement in Cdc42-mediated transformation suggest a novel pathway by which Cdc42 acts to promote tumorigenesis.

III. RHO GTPASES AND METASTASIS The process of tumor metastasis can be viewed as a series of steps that ultimately lead to the formation of additional tumors at sites distal to the original tumor. These steps can be generally viewed as (1) invasion of, and passage through, the extracellular matrix (ECM), (2) entry into the circulatory system (intravasation), (3) travel through blood/lymphoid vessels, (4) exit from the circulatory system (extravasation), and (5) colonization (see Fig. 3). A number of in vitro and in vivo models have been established that attempt to recapitulate one or more of the steps involved in metastasis. The ability of cells to invade and pass through the ECM (step 1) can be determined using a modiﬁed Boyden chamber, in which cells are placed on a porous transwell ﬁlter that has been coated with one or more ECM components and are then examined for their ability to move to the opposite side of the ﬁlter. Alternatively, cells can be plated as a monolayer on the surface of a three-dimensional collagen matrix and monitored for their ability to pass into the gel (Banyard et al., 2000). Both of these assays require three crucial steps associated with invasion: detachment from other tumor cells, degradation of ECM, and movement through the ECM. A variation of these two in vitro assays has been used to investigate the invasive ability of nonadherent T lymphoma cells, which are plated on a monolayer of adherent ﬁbroblasts and examined for their ability to inﬁltrate the monolayer (Evers et al., 2000). Two mouse models are typically used to assess the metastatic ability of tumor cells (del Peso et al., 1997). The “experimental metastasis assay”

Fig. 3 Tumor metastasis. Metastatic cells (1) invade and pass through the basement membrane and extracellular matrix (invasion), (2) enter the lymphatic or venous circulatory system (intravasation), (3) travel through the circulatory system, (4) invade the vascular basement membrane and the extracellular matrix (extravasation), and (5) reattach and proliferate at a new site (colonization).

72

Aron B. Jaffe and Alan Hall

assesses the ability of tumor cells to carry out steps 3–5 (Fig. 3) associated with metastasis, since the cells are directly injected into the circulatory system via the tail vein. A “spontaneous metastasis assay” requires the tumor cells to perform all ﬁve steps associated with metastasis, since the tumor cells are injected subcutaneously into the footpad of the mouse. In both the experimental and the spontaneous metastasis assays, metastatic potential is scored as the latency and frequency of new tumor formation at distal sites, such as the lung. Since Rho GTPases have a role in many cellular processes that accompany tumor metastasis, such as migration, alterations in adhesion, and morphological changes, it has been proposed that changes in their activity are involved in speciﬁcally promoting invasion and metastasis.

A. Tiam-1 The T lymphoma invasion and metastasis (Tiam-1) gene was identiﬁed using a viral insertional mutagenesis screen in which noninvasive T lymphoma cells were infected with a virus and then plated on a monolayer of ﬁbroblasts. Cells that were able to invade the monolayer were isolated and, after several rounds of this selection process, analyzed for the sites of proviral insertions. Several independently infected lines contained proviral insertions in the Tiam-1 locus, which encodes a 1591aa protein with similarity to GEFs for Rho GTPases. Overexpression of various truncated forms of Tiam-1 protein conferred an invasive phenotype in T-lymphoma cells, and an apparently spontaneously invasive clone contained greatly elevated levels of full-length Tiam-1 protein, suggesting that the level of Tiam-1 is a key event in the progression to tumor invasion (Habets et al., 1994). Further studies indicated that Tiam-1 could act as a GEF, primarily for Rac, and transient overexpression of Tiam-1 in NIH3T3 cells induced extensive rufﬂing, similar to the phenotype seen by overexpression of constitutively active Rac (Michiels et al., 1995). In addition, the invasiveness conferred by overexpression of Tiam-1 in T-lymphoma cells is phenocopied by overexpression of constitutively active Rac, suggesting that Tiam-1-mediated invasion is via activation of Rac (Michiels et al., 1995). The ability of Tiam-1/Rac signaling to promote invasion of adherent cell types, such as ﬁbroblasts and epithelial cells, appears to be more complex and dependent on the ECM on which the cells are attached. Overexpression of an N-terminally truncated (i.e., constitutively activated) Tiam-1 inhibits the migration of NIH3T3 cells on ﬁbronectin (Sander et al., 1999) and of Madin Darby canine kidney (MDCK) epithelial cells on both ﬁbronectin and laminin, but increases MDCK motility on collagen (Sander et al., 1998). Tiam-1 activation of Rac appears to be dependent on PI3-kinase activity, since treatment of cells with the PI3-kinase inhibitor wortmannin blocked

Rho GTPases and Cancer

73

the increased levels of GTP-bound Rac seen when Tiam-1 is overexpressed (Sander et al., 1998). It is likely that the role of PI3-kinase is to promote binding of Tiam-1 to the plasma membrane by generating the phospholipid PIP3 (Michiels et al., 1997; Stam et al., 1997), since this is essential for Racdependent membrane rufﬂing and JNK activation (Michiels et al., 1997). Since many tumors have deletions in PTEN (Simpson and Parsons, 2001), the gene encoding PIP3 phosphatase, it is possible that Tiam-1 activity could be increased in these cells. Whether deregulated activation of Tiam-1 is important in promoting tumor invasion in human cancers remains to be seen; an analysis of renal cell carcinoma cell lines, using a modiﬁed Boyden chamber coated with a mixture of ECM components, found that Tiam-1 expression levels were inversely correlated with invasive ability in four out of ﬁve cell lines examined (Engers et al., 2000). Other studies suggest that Tiam-1 can interact with the hyaluronic acid receptor, CD44, which is sometimes up-regulated during tumor metastasis (Iida and Bourguignon, 1995), to mediate breast tumor cell migration (Bourguignon et al., 2000b). In addition, Tiam-1 has been shown to physically interact with the cytoskeletal protein ankyrin, and this interaction promotes Rac activation as well as breast tumor cell invasion and migration (Bourguignon et al., 2000a). Whether these interactions play a role in the invasive properties of other tumor cell types remains to be seen. In addition to the evidence indicating that deregulated Tiam-1 activity could promote tumor invasion, another study implicates Tiam-1 in tumor metastasis. This study found that the tumor metastasis suppressor nm23H1 associates with Tiam-1 in vivo, and this association leads to a suppression of the ability of Tiam-1 to activate Rac, thereby inhibiting Rac-induced morphological and transcriptional changes (Otsuki et al., 2001). This suggests that inhibiting Tiam-1 activity may block the progression of tumors to a metastatic state.

B. Rho The involvement of Rho in tumor metastasis and invasion has been demonstrated using both in vitro models for invasion and in vivo metastasis models (Clark et al., 2000; del Peso et al., 1997; Yoshioka et al., 1998). Overexpression of wild-type or constitutively active forms of Rho can induce an invasive behavior in noninvasive rat hepatoma cells in vitro (Yoshioka et al., 1998) and promote a metastatic behavior in both NIH3T3 ﬁbroblasts and poorly metastatic melanoma cells in vivo (Clark et al., 2000; del Peso et al., 1997), whereas inhibition of Rho can impair invasion of highly invasive rat hepatoma cells in vitro as well as highly metastatic melanoma cells in vivo (Clark et al., 2000; Yoshioka et al., 1998).

74

Aron B. Jaffe and Alan Hall

One way in which Rho might contribute to the invasive phenotype of tumor cells is through the actomyosin system. This idea is supported by the fact that overexpression of constitutively active forms of Rho or its downstream effector p160 Rho kinase can stimulate the phosphorylation of myosin light chain (MLC) (Itoh et al., 1999; Kureishi et al., 1997; Yoshioka et al., 1998). Rho-induced phosphorylation of MLC can be blocked by chemical inhibitors of MLC kinase (Yoshioka et al., 1998) or p160 Rho kinase (Itoh et al., 1999), and the levels of phosphorylated MLC appear to correlate with the in vitro and in vivo invasiveness of tumor cells (Itoh et al., 1999; Yoshioka et al., 1998). The precise signaling cascade leading from Rho activation to MLC phosphorylation and how this cascade contributes to metastasis remain to be elucidated. Additional evidence implicating Rho in metastasis comes from a number of studies indicating that Rho is overexpressed in a variety of tumors and that the levels correlate with tumor stage (Fritz et al., 1999; Kamai et al., 2001; Suwa et al., 1998; van Golen et al., 1999) (Table II). Interestingly, RhoC is overexpressed in more tumor types than RhoA. In addition, another study has found that increased levels of RhoC are selected for during the metastatic progression of tumor cells in an experimental model system (Clark et al., 2000). This system utilized a poorly metastatic melanoma cell line that, when injected into the tail vein of mice, formed few pulmonary metastases. Highly invasive derivatives were selected by repeated isolation of the metastases and reinjection into mice. After several rounds, the gene expression proﬁle of the highly metastatic cells was compared with that of the parental cell line. RhoC was found to be overexpressed in three independently isolated, highly metastatic derivatives. The same screen was also performed with a poorly metastatic mouse melanoma cell line, and RhoC was also found to be overexpressed in the resulting highly metastatic cells. Similar to RhoA, overexpression of RhoC is sufﬁcient to promote an invasive behavior in vitro of noninvasive melanoma and human mammary epithelial cells (Clark et al., 2000; van Golen et al., 2000), as well as a metastatic behavior of poorly metastatic melanoma cells (Clark et al., 2000). It is unclear why RhoC rather than RhoA overexpression is associated with these metastatic cells, since the two GTPases are 94% identical (i.e., only 11 amino acid differences; Ihara et al., 1998). Interestingly, almost all of these differences are located near the C-terminus, close to the A5 α helix (Ihara et al., 1998). The A5 helix is thought to act as one of the regions in small GTPases that determine the speciﬁcity of effector interactions, as it is involved in mediating contact between RhoA and one of its effectors PKN/PRK1 (Maesaki et al., 1999) as well as Cdc42 and two of its effectors WASP and ACK (Abdul-Manan et al., 1999; Mott et al., 1999). The amino acid differences between RhoA and RhoC, however, are not the residues within the A5 helix that directly contact PKN/PRK1 (Maesaki et al., 1999),

Rho GTPases and Cancer

75

though they may inﬂuence the relative interactions of RhoA and RhoC with other effector proteins. It is worth noting that RhoC has been shown, at least in one study, to be a better motogen than RhoA when expressed in melanoma cells (Clark et al., 2000), suggesting a functional difference between the two in this system.

IV. CONCLUSIONS AND FUTURE DIRECTIONS There are now several lines of experimental evidence that implicate Rho GTPases in multiple aspects of cancer. In addition to the many in vitro studies indicating that activation of Rho GTPases is necessary and sufﬁcient to induce transformation as well as tumor invasion, a few genetic alterations have now been found in GEFs and GAPs that regulate Rho, Rac, and Cdc42 in human tumors. Genetic alterations in the GTPases themselves have not been described, but several studies have found that one or more of the Rho GTPases are overexpressed in different tumors. Taken together, these studies provide compelling evidence that members of the Rho GTPase family play an integral role in tumor progression, and the challenge remains to understand their precise contribution to this process. Advances such as the completion of the human genome (Lander et al., 2001; Venter et al., 2001) and the adaptation of RNA interference to mammalian cell culture (Elbashir et al., 2001) should provide some of the tools necessary to dissect the pathways that Rho, Rac, and Cdc42 utilize to contribute to tumorigenesis, invasion, and metastasis.

ACKNOWLEDGMENTS We thank Annette Self for providing the pictures for Fig. 1. A.B.J. is supported by a postdoctoral fellowship from the International Agency for Research on Cancer (IARC). A.H. is supported by a programme grant from the Cancer Research Campaign (UK).

REFERENCES Abdul-Manan, N., Aghazadeh, B., Liu, G. A., Majumdar, A., Ouerfelli, O., Siminovitch, K. A., and Rosen, M. K. (1999). Nature 399, 379–383. Abo, A., Pick, E., Hall, A., Totty, N., Teahan, C. G., and Segal, A. W. (1991). Nature 353, 668–670. Alberts, A. S., and Treisman, R. (1998). EMBO J. 17, 4075–4085.

76

Aron B. Jaffe and Alan Hall

Amano, M., Chihara, K., Kimura, K., Fukata, Y., Nakamura, N., Matsuura, Y., and Kaibuchi, K. (1997). Science 275, 1308–1311. Aronheim, A., Broder, Y. C., Cohen, A., Fritsch, A., Belisle, B., and Abo, A. (1998). Curr. Biol. 8, 1125–1128. Banyard, J., Anand-Apte, B., Symons, M., and Zetter, B. R. (2000). Oncogene 19, 580–591. Bar-Sagi, D., and Feramisco, J. R. (1986). Science 233, 1061–1068. Bishop, A. L., and Hall, A. (2000). Biochem. J. 348 Pt 2, 241–255. Borkhardt, A., Bojesen, S., Haas, O. A., Fuchs, U., Bartelheimer, D., Loncarevic, I. F., Bohle, R. M., Harbott, J., Repp, R., Jaeger, U., Viehmann, S., Henn, T., Korth, P., Scharr, D., and Lampert, F. (2000). Proc. Natl. Acad. Sci. USA 97, 9168–9173. Bos, J. L. (1989). Cancer Res. 49, 4682–4689. Bourguignon, L. Y., Zhu, H., Shao, L., and Chen, Y. W. (2000a). J. Cell Biol. 150, 177–191. Bourguignon, L. Y., Zhu, H., Shao, L., and Chen, Y. W. (2000b). J. Biol. Chem. 275, 1829–1838. Chan, A. M., McGovern, E. S., Catalano, G., Fleming, T. P., and Miki, T. (1994). Oncogene 9, 1057–1063. Chan, A. M., Takai, S., Yamada, K., and Miki, T. (1996). Oncogene 12, 1259–1266. Chardin, P., Madaule, P., and Tavitian, A. (1988). Nucleic Acids Res. 16, 2717. Clark, E. A., Golub, T. R., Lander, E. S., and Hynes, R. O. (2000). Nature 406, 532–555. Dallery, E., Galiegue-Zouitina, S., Collyn-d’Hooghe, M., Quief, S., Denis, C., Hildebrand, M. P., Lantoine, D., Deweindt, C., Tilly, H., Bastard, C., and Kerckaert, J. P. (1995). Oncogene 10, 2171–2178. Dartsch, P. C., Ritter, M., Haussinger, D., and Lang, F. (1994). Eur. J. Cell Biol. 63, 316–325. del Peso, L., Hernandez-Alcoceba, R., Embade, N., Carnero, A., Esteve, P., Paje, C., and Lacal, J. C. (1997). Oncogene 15, 3047–3057. Didsbury, J., Weber, R. F., Bokoch, G. M., Evans, T., and Snyderman, R. (1989). J. Biol. Chem. 264, 16378–16382. Diekmann, D., Abo, A., Johnston, C., Segal, A. W., and Hall, A. (1994). Science 265, 531–533. Drivas, G. T., Shih, A., Coutavas, E., Rush, M. G., and D’Eustachio, P. (1990). Mol. Cell Biol. 10, 1793–1798. Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K., and Tuschl, T. (2001). Nature 411, 494–498. Ellis, S., and Mellor, H. (2000). Curr. Biol. 10, 1387–1390. Engers, R., Zwaka, T. P., Gohr, L., Weber, A., Gerharz, C. D., and Gabbert, H. E. (2000). Int. J. Cancer 88, 369–376. Evers, E. E., van der Kammen, R. A., ten Klooster, J. P., and Collard, J. G. (2000). Methods Enzymol. 325, 403–415. Foster, R., Hu, K. Q., Lu, Y., Nolan, K. M., Thissen, J., and Settleman, J. (1996). Mol. Cell Biol. 16, 2689–2699. Fritz, G., Just, I., and Kaina, B. (1999). Int. J. Cancer 81, 682–687. Fukuhara, S., Chikumi, H., and Gutkind, J. S. (2001). Oncogene 20, 1661–1668. Glaven, J. A., Whitehead, I. P., Nomanbhoy, T., Kay, R., and Cerione, R. A. (1996). J. Biol. Chem. 271, 27374–27381. Greig, R. G., Koestler, T. P., Trainer, D. L., Corwin, S. P., Miles, L., Kline, T., Sweet, R., Yokoyama, S., and Poste, G. (1985). Proc. Natl. Acad. Sci. USA 82, 3698–3701. Haataja, L., Groffen, J., and Heisterkamp, N. (1997). J. Biol. Chem. 272, 20384–20388. Habets, G. G., Scholtes, E. H., Zuydgeest, D., van der Kammen, R. A., Stam, J. C., Berns, A., and Collard, J. G. (1994). Cell 77, 537–549. Hall, A. (1998). Science 279, 509–514. Hall, A., Marshall, C. J., Spurr, N. K., and Weiss, R. A. (1983). Nature 303, 396–400. Hanahan, D., and Weinberg, R. A. (2000). Cell 100, 57–70. Hart, M. J., Eva, A., Evans, T., Aaronson, S. A., and Cerione, R. A. (1991). Nature 354, 311–314.

Rho GTPases and Cancer

77

Hart, M. J., Eva, A., Zangrilli, D., Aaronson, S. A., Evans, T., Cerione, R. A., and Zheng, Y. (1994). J. Biol. Chem. 269, 62–65. Homma and Emori (1995). Table II Horii, Y., Beeler, J. F., Sakaguchi, K., Tachibana, M., and Miki, T. (1994). EMBO J. 13, 4776– 4786. Ihara, K., Muraguchi, S., Kato, M., Shimizu, T., Shirakawa, M., Kuroda, S., Kaibuchi, K., and Hakoshima, T. (1998). J. Biol. Chem. 273, 9656–9666. Iida, N., and Bourguignon, L. Y. (1995). J. Cell Physiol. 162, 127–133. Itoh, K., Yoshioka, K., Akedo, H., Uehata, M., Ishizaki, T., and Narumiya, S. (1999). Nat. Med. 5, 221–225. Joneson, T., and Bar-Sagi, D. (1998). J. Biol. Chem. 273, 17991–17994. Jordan, P., Brazao, R., Boavida, M. G., Gespach, C., and Chastre, E. (1999). Oncogene 18, 6835–6839. Kamai, T., Arai, K., Tsujii, T., Honda, M., and Yoshida, K. (2001). BJU Int. 87, 227–231. Karnoub, A. E., Der, C. J., and Campbell, S. L. (2001). Mol. Cell. Biol. 21, 2847–2857. Katzav, S., Martin-Zanca, D., and Barbacid, M. (1989). EMBO J. 8, 2283–2290. Khosravi-Far, R., Chrzanowska-Wodnicka, M., Solski, P. A., Eva, A., Burridge, K., and Der, C. J. (1994). Mol. Cell. Biol. 14, 6848–6857. Khosravi-Far, R., Solski, P. A., Clark, G. J., Kinch, M. S., and Der, C. J. (1995). Mol. Cell Biol. 15, 6443–6453. Khosravi-Far, R., White, M. A., Westwick, J. K., Solski, P. A., Chrzanowska-Wodnicka, M., Van Aelst, L., Wigler, M. H., and Der, C. J. (1996). Mol. Cell Biol. 16, 3923–3933. Kourlas, P. J., Strout, M. P., Becknell, B., Veronese, M. L., Croce, C. M., Theil, K. S., Krahe, R., Ruutu, T., Knuutila, S., Bloomﬁeld, C. D., and Caligiuri, M. A. (2000). Proc. Natl. Acad. Sci. USA 97, 2145–2150. Kozma, R., Ahmed, S., Best, A., and Lim, L. (1995). Mol. Cell Biol. 15, 1942–1952. Kureishi, Y., Kobayashi, S., Amano, M., Kimura, K., Kanaide, H., Nakano, T., Kaibuchi, K., and Ito, M. (1997). J. Biol. Chem. 272, 12257–12260. Lander, E. S., Linton, L. M., Birren, B., Nusbaum, C., Zody, M. C., Baldwin, J., Devon, K., Dewar, K., Doyle, M., FitzHugh, W., Funke, R., Gage, D., Harris, K., Heaford, A., Howland, J., Kann, L., Lehoczky, J., LeVine, R., McEwan, P., McKernan, K., Meldrim, J., Mesirov, J. P., Miranda, C., Morris, W., Naylor, J., Raymond, C., Rosetti, M., Santos, R., Sheridan, A., Sougnez, C., Stange-Thomann, N., Stojanovic, N., Subramanian, A, Wyman, D., Rogers, J., Sulston, J., Ainscough, R., Beck, S., Bentley, D., Burton, J., Clee, C., Carter, N., Coulson, A., Deadman, R., Deloukas, P., Dunham, A., Dunham, I., Durbin, R., French, L., Grafham, D., Gregory, S., Hubbard, T., Humphray, S., Hunt, A., Jones, M., Lloyd, C., McMurray, A., Matthews, L., Mercer, S., Milne, S., Mullikin, J. C., Mungall, A., Plumb, R., Ross, M., Shownkeen, R., Sims, S., Waterston, R. H., Wilson, R. K., Hillier, L. W., McPherson, J. D., Marra, M. A., Mardis, E. R., Fulton, L. A., Chinwalla, A. T., Pepin, K. H., Gish, W. R., Chissoe, S. L., Wendl, M. C., Delehaunty, K. D., Miner, T. L., Delehaunty, A., Kramer, J. B., Cook, L. L., Fulton, R. S., Johnson, D. L, Minx, P. J., Clifton, S. W., Hawkins, T., Branscomb, E., Predki, P., Richardson, P., Wenning, S., Slezak, T., Doggett, N., Cheng, J. F., Olsen, A., Lucas, S., Elkin, C., Uberbacher, E., Frazier, M., Gibbs, R. A., Muzny, D. M., Scherer, S. E., Bouck, J. B., Sodergren, E. J., Worley, K. C., Rives, C. M., Gorrell, J. H., Metzker, M. L., Naylor, S. L., Kucherlapati, R. S., Nelson, D. L., Weinstock, G. M., Sakaki, Y., Fujiyama, A., Hattori, M., Yada, T., Toyoda, A., Itoh, T., Kawagoe, C., Watanabe, H., Totoki, Y., Taylor, T., Weissenbach, J., Heilig, R., Saurin, W., Artiguenave, F., Brottier, P., Bruls, T., Pelletier, E., Robert, C., Wincker, P., Smith, D. R., Doucette-Stamm, L., Rubenﬁeld, M., Weinstock, K., Lee, H. M., Dubois, J., Rosenthal, A., Platzer, M., Nyakatura, G., Taudien, S., Rump, A., Yang, H., Yu, J., Wang, J., Huang, G., Gu, J., Hood, L., Rowen, L., Madan, A., Qin, S., Davis, R. W., Federspiel, N. A., Abola, A. P., Proctor, M. J., Myers, R. M., Schmutz, J., Dickson, M., Grimwood, J., Cox, D. R., Olson, M. V., Kaul, R., Shimizu, N.,

78

Aron B. Jaffe and Alan Hall

Kawasaki, K., Minoshima, S., Evans, G. A., Athanasiou, M., Schultz, R., Roe, B. A., Chen, F., Pan, H., Ramser, J., Lehrach, H., Reinhardt, R., McCombie, W. R., de la Bastide, M., Dedhia, N., Blocker, H., Hornischer, K., Nordsiek, G., Agarwala, R., Aravind, L., Bailey, J. A., Bateman, A., Batzoglou, S., Birney, E., Bork, P., Brown, D. G., Burge, C. B., Cerutti, L., Chen, H. C., Church, D., Clamp, M., Copley, R. R., Doerks, T., Eddy, S. R., Eichler, E. E., Furey, T. S., Galagan, J., Gilbert, J. G., Harmon, C., Hayashizaki, Y., Haussler, D., Hermjakob, H., Hokamp, K., Jang, W., Johnson, L. S., Jones, T. A., Kasif, S., Kaspryzk, A., Kennedy, S., Kent, W. J., Kitts, P., Koonin, E. V., Korf, I., Kulp, D., Lancet, D., Lowe, T. M., McLysaght, A., Mikkelsen, T., Moran, J. V., Mulder, N., Pollara, V. J., Ponting, C. P., Schuler, G., Schultz, J., Slater, G., Smit, A. F., Stupka, E., Szustakowki, J., Thierry-Mieg, D., Thierry-Mieg, J., Wagner, L., Wallis, J., Wheeler, R., Williams, A., Wolf, Y. I., Wolfe, K. H., Yang, S. P., Yeh, R. F., Collins, F., Guyer, M. S., Peterson, J., Felsenfeld, A., Wetterstrand, K. A., Patrinos, A., and Morgan, M. J. (2001). Nature 409, 860–921. Lebowitz, P. F., Davide, J. P., and Prendergast, G. C. (1995). Mol. Cell Biol. 15, 6613–6622. Lin, R., Bagrodia, S., Cerione, R., and Manor, D. (1997). Curr. Biol. 7, 794–797. Lin, R., Cerione, R. A., and Manor, D. (1999). J. Biol. Chem. 274, 23633–23641. Maesaki, R., Ihara, K., Shimizu, T., Kuroda, S., Kaibuchi, K., and Hakoshima, T. (1999). Mol. Cell 4, 793–803. Michiels, F., Habets, G. G., Stam, J. C., van der Kammen, R. A., and Collard, J. G. (1995). Nature 375, 338–340. Michiels, F., Stam, J. C., Hordijk, P. L., van der Kammen, R. A., Ruuls-Van Stalle, L., Feltkamp, C. A., and Collard, J. G. (1997). J. Cell Biol. 137, 387–398. Miki, T., Smith, C. L., Long, J. E., Eva, A., and Fleming, T. P. (1993). Nature 362, 462–465. Mira, J. P., Benard, V., Groffen, J., Sanders, L. C., and Knaus, U. G. (2000). Proc. Natl. Acad. Sci. USA 97, 185–189. Mott, H. R., Owen, D., Nietlispach, D., Lowe, P. N., Manser, E., Lim, L., and Laue, E. D. (1999). Nature 399, 384–388. Nobes, C. D., and Hall, A. (1995). Cell 81, 53–62. Nobes, C. D., Lauritzen, I., Mattei, M. G., Paris, S., Hall, A., and Chardin, P. (1998). J. Cell Biol. 141, 187–197. Olson, M. F., Ashworth, A., and Hall, A. (1995). Science 269, 1270–1272. Olson, M. F., Pasteris, N. G., Gorski, J. L., and Hall, A. (1996). Curr. Biol. 6, 1628–1633. Olson, M. F., Paterson, H. F., and Marshall, C. J. (1998). Nature 394, 295–299. Otsuki, Y., Tanaka, M., Yoshii, S., Kawazoe, N., Nakaya, K., and Sugimura, H. (2001). Proc. Natl. Acad. Sci. USA 27, 27. Perona, R., Esteve, P., Jimenez, B., Ballestero, R. P., Ramon y Cajal, S., and Lacal, J. C. (1993). Oncogene 8, 1285–1292. Prendergast, G. C., and Oliff, A. (2000). Semin. Cancer Biol. 10, 443–452. Preudhomme, C., Roumier, C., Hildebrand, M. P., Dallery-Prudhomme, E., Lantoine, D., Lai, J. L., Daudignon, A., Adenis, C., Bauters, F., Fenaux, P., Kerckaert, J. P., and Gali`equeZouitina, S. (2000). Oncogene 19, 2023–2032. Qiu, R. G., Abo, A., McCormick, F., and Symons, M. (1997). Mol. Cell Biol. 17, 3449–3458. Qiu, R. G., Chen, J., Kirn, D., McCormick, F., and Symons, M. (1995a). Nature 374, 457–459. Qiu, R. G., Chen, J., McCormick, F., and Symons, M. (1995b). Proc. Natl. Acad. Sci. USA 92, 11781–11785. Reuther, G. W., Lambert, Q. T., Booden, M. A., Wennerberg, K., Becknell, B., Marcucci, G., Sondek, J., Caligiuri, M. A., and Der, C. J. (2001). J. Biol. Chem. 23, 23. Ridley, A. J., and Hall, A. (1992). Cell 70, 389–399. Ridley, A. J., Paterson, H. F., Johnston, C. L., Diekmann, D., and Hall, A. (1992). Cell 70, 401–410. Ridley et al. (1993). Table II

Rho GTPases and Cancer

79

Roux, P., Gauthier-Rouviere, C., Doucet-Brutin, S., and Fort, P. (1997). Curr. Biol. 7, 629–637. Sahai, E., Ishizaki, T., Narumiya, S., and Treisman, R. (1999). Curr. Biol. 9, 136–145. Sahai, E., Olson, M. F., and Marshall, C. J. (2001). EMBO J. 20, 755–766. Sander, E. E., ten Klooster, J. P., van Delft, S., van der Kammen, R. A., and Collard, J. G. (1999). J. Cell Biol. 147, 1009–1022. Sander, E. E., van Delft, S., ten Klooster, J. P., Reid, T., van der Kammen, R. A., Michiels, F., and Collard, J. G. (1998). J. Cell Biol. 143, 1385–1398. Santos, E., Tronick, S. R., Aaronson, S. A., Pulciani, S., and Barbacid, M. (1982). Nature 298, 343–347. Schultz, J., Milpetz, F., Bork, P., and Ponting, C. P. (1998). Proc. Natl. Acad. Sci. USA 95, 5857–5864. Self, A. J., Paterson, H. F., and Hall, A. (1993). Oncogene 8, 655–661. Shimizu, F., Watanabe, T. K., Okuno, S., Omori, Y., Fujiwara, T., Takahashi, E., and Nakamura, Y. (1997). Biochim. Biophys. Acta 1351, 13–16. Shinjo, K., Koland, J. G., Hart, M. J., Narasimhan, V., Johnson, D. I., Evans, T., and Cerione, R. A. (1990). Proc. Natl. Acad. Sci. USA 87, 9853–9857. Simpson, L., and Parsons, R. (2001). Exp. Cell Res. 264, 29–41. Stacey, D. W., and Kung, H. F. (1984). Nature 310, 508–511. Stam, J. C., Sander, E. E., Michiels, F., van Leeuwen, F. N., Kain, H. E., van der Kammen, R. A., and Collard, J. G. (1997). J. Biol. Chem. 272, 28447–28454. Sundaresan, M., Yu, Z. X., Ferrans, V. J., Sulciner, D. J., Gutkind, J. S., Irani, K., GoldschmidtClermont, P. J., and Finkel, T. (1996). Biochem. J. 318, 379–382. Suwa, H., Ohshio, G., Imamura, T., Watanabe, G., Arii, S., Imamura, M., Narumiya, S., Hiai, H., and Fukumoto, M. (1998). Br. J. Cancer 77, 147–152. Tao, W., Pennica, D., Xu, L., Kalejta, R. F., and Levine, A. J. (2001). Genes Dev. 15, 1796–1807. Tatsumoto, T., Xie, X., Blumenthal, R., Okamoto, I., and Miki, T. (1999). J. Cell Biol. 147, 921–928. Tikoo, A., Czekay, S., Viars, C., White, S., Heath, J. K., Arden, K., and Maruta, H. (2000). Gene 257, 23–31. Van Aelst, L., and D’Souza-Schorey, C. (1997). Genes Dev. 11, 2295–2322. Van den Berghe, H., and Michaux, L. (1997). Cancer Genet. Cytogenet. 94, 1–7. van Golen, K. L., Davies, S., Wu, Z. F., Wang, Y., Bucana, C. D., Root, H., Chandrasekharappa, S., Strawderman, M., Ethier, S. P., and Merajver, S. D. (1999). Clin. Cancer Res. 5, 2511– 2519. van Golen, K. L., Wu, Z. F., Qiao, X. T., Bao, L. W., and Merajver, S. D. (2000). Cancer Res. 60, 5832–5838. van Leeuwen, F. N., van der Kammen, R. A., Habets, G. G., and Collard, J. G. (1995). Oncogene 11, 2215–2221. Venter, J. C., Adams, M. D., Myers, E. W., Li, P. W., Mural, R. J., Sutton, G. G., Smith, H. O., Yandell, M., Evans, C. A., Holt, R. A., Gocayne, J. D., Amanatides, P., Ballew, R. M., Huson, D. H., Wortman, J. R., Zhang, Q., Kodira, C. D., Zheng, X. H., Chen, L., Skupski, M., Subramanian, G., Thomas, P. D., Zhang, J., Gabor Miklos, G. L., Nelson, C., Broder, S., Clark, A. G., Nadeau, J., McKusick, V. A., Zinder, N., Levine, A. J., Roberts, R. J., Simon, M., Slayman, C., Hunkapiller, M., Bolanos, R., Delcher, A., Dew, I., Fasulo, D., Flanigan, M., Florea, L., Halpern, A., Hannenhalli, S., Kravitz, S., Levy, S., Mobarry, C., Reinert, K., Remington, K., Abu-Threideh, J., Beasley, E., Biddick, K., Bonazzi, V., Brandon, R., Cargill, M., Chandramouliswaran, I., Charlab, R., Chaturvedi, K., Deng, Z., Di Francesco, V., Dunn, P., Eilbeck, K., Evangelista, C., Gabrielian, A. E., Gan, W., Ge, W., Gong, F., Gu, Z., Guan, P., Heiman, T. J., Higgins, M. E., Ji, R. R., Ke, Z., Ketchum, K. A., Lai, Z., Lei, Y., Li, Z., Li, J., Liang, Y., Lin, X., Lu, F., Merkulov, G. V., Milshina, N., Moore, H. M., Naik, A. K., Narayan, V. A., Neelam, B., Nusskern, D., Rusch, D. B., Salzberg, S., Shao, W., Shue, B.,

80

Aron B. Jaffe and Alan Hall

Sun, J., Wang, Z., Wang, A., Wang, X., Wang, J., Wei, M., Wides, R., Xiao, C., Yan, C., Yao, A., Ye, J., Zhan, M., Zhang, W., Zhang, H., Zhao, Q., Zheng, L., Zhong, F., Zhong, W., Zhu, S., Zhao, S., Gilbert, D., Baumhueter, S., Spier, G., Carter, C., Cravchik, A., Woodage, T., Ali, F., An, H., Awe, A., Baldwin, D., Baden, H., Barnstead, M., Barrow, I., Beeson, K., Busam, D., Carver, A., Center, A., Cheng, M. L., Curry, L., Danaher, S., Davenport, L., Desilets, R., Dietz, S., Dodson, K., Doup, L., Ferriera, S., Garg, N., Gluecksmann, A., Hart, B., Haynes, J., Haynes, C., Heiner, C., Hladun, S., Hostin, D., Houck, J., Howland, T., Ibegwam, C., Johnson, J., Kalush, F., Kline, L., Koduru, S., Love, A., Mann, F., May, D., McCawley, S., McIntosh, T., McMullen, I., Moy, M., Moy, L., Murphy, B., Nelson, K., Pfannkoch, C., Pratts, E., Puri, V., Qureshi, H., Reardon, M., Rodriguez, R., Rogers, Y. H., Romblad, D., Ruhfel, B., Scott, R., Sitter, C., Smallwood, M., Stewart, E., Strong, R., Suh, E., Thomas, R., Tint, N. N., Tse, S., Vech, C., Wang, G., Wetter, J., Williams, S., Williams, M., Windsor, S., Winn-Deen, E., Wolfe, K., Zaveri, J., Zaveri, K., Abril, J. F., Guigo, R., Campbell, M. J., Sjolander, K. V., Karlak, B., Kejariwal, A., Mi, H., Lazareva, B., Hatton, T., Narechania, A., Diemer, K., Muruganujan, A., Guo, N., Sato, S., Bafna, V., Istrail, S., Lippert, R., Schwartz, R., Walenz, B., Yooseph, S., Allen, D., Basu, A., Baxendale, J., Blick, L., Caminha, M., Carnes-Stine, J., Caulk, P., Chiang, Y. H., Coyne, M., Dahlke, C., Mays, A., Dombroski, M., Donnelly, M., Ely, D., Esparham, S., Fosler, C., Gire, H., Glanowski, S., Glasser, K., Glodek, A., Gorokhov, M., Graham, K., Gropman, B., Harris, M., Heil, J., Henderson, S., Hoover, J., Jennings, D., Jordan, C., Jordan, J., Kasha, J., Kagan, L., Kraft, C., Levitsky, A., Lewis, M., Liu, X., Lopez, J., Ma, D., Majoros, W., McDaniel, J., Murphy, S., Newman, M., Nguyen, T., Nguyen, N., Nodell, M., Pan, S., Peck, J., Peterson, M., Rowe, W., Sanders, R., Scott, J., Simpson, M., Smith, T., Sprague, A., Stockwell, T., Turner, R., Venter, E., Wang, M., Wen, M., Wu, D., Wu, M., Xia, A., Zandieh, A., and Zhu, X. (2001). Science 291, 1304–1351. Vignal, E., De Toledo, M., Comunale, F., Ladopoulou, A., Gauthier-Rouviere, C., Blangy, A., and Fort, P. (2000). J. Biol. Chem. 275, 36457–36464. Vincent, S., Jeanteur, P., and Fort, P. (1992). Mol. Cell Biol. 12, 3138–3148. Westwick, J. K., Lambert, Q. T., Clark, G. J., Symons, M., Van Aelst, L., Pestell, R. G., and Der, C. J. (1997). Mol. Cell Biol. 17, 1324–1335. Whitehead, I., Kirk, H., and Kay, R. (1995). Oncogene 10, 713–721. Whitehead, I. P., Abe, K., Gorski, J. L., and Der, C. J. (1998). Mol. Cell. Biol. 18, 4689–4697. Whitehead, I. P., Lambert, Q. T., Glaven, J. A., Abe, K., Rossman, K. L., Mahon, G. M., Trzaskos, J. M., Kay, R., Campbell, S. L., and Der, C. J. (1999). Mol. Cell Biol. 19, 7759– 7770. Wu, W. J., Erickson, J. W., Lin, R., and Cerione, R. A. (2000). Nature 405, 800–804. Wu, W. J., Lin, R., Cerione, R. A., and Manor, D. (1998). J. Biol. Chem. 273, 16655–16658. Yeramian, P., Chardin, P., Madaule, P., and Tavitian, A. (1987). Nucleic Acids Res. 15, 1869. Yoshioka, K., Matsumura, F., Akedo, H., and Itoh, K. (1998). J. Biol. Chem. 273, 5146–5154. Yuan, B. Z., Miller, M. J., Keck, C. L., Zimonjic, D. B., Thorgeirsson, S. S., and Popescu, N. C. (1998). Cancer Res. 58, 2196–2199. Zheng, Y., Fischer, D. J., Santos, M. F., Tigyi, G., Pasteris, N. G., Gorski, J. L., and Xu, Y. (1996). J. Biol. Chem. 271, 33169–33172. Zheng, Y., Olson, M. F., Hall, A., Cerione, R. A., and Toksoz, D. (1995). J. Biol. Chem. 270, 9031–9034. Zohn, I. E., Klinger, M., Karp, X., Kirk, H., Symons, M., Chrzanowska-Wodnicka, M., Der, C. J., and Kay, R. J. (2000). Oncogene 19, 3866–3877. Zohn, I. E., Symons, M., Chrzanowska-Wodnicka, M., Westwick, J. K., and Der, C. J. (1998). Mol. Cell Biol. 18, 1225–1235. Zong, H., Raman, N., Mickelson-Young, L. A., Atkinson, S. J., and Quilliam, L. A. (1999). J. Biol. Chem. 274, 4551–4560.

The myc Oncogene: MarvelouslY Complex Sara K. Oster,* Cynthia S. W. Ho,* Erinn L. Soucie,* and Linda Z. Penn Division of Cellular and Molecular Biology Ontario Cancer Institute, Princess Margaret Hospital Department of Medical Biophysics University of Toronto Toronto, Ontario M5G 2M9

I. Introduction II. Myc and Cancer A. Deﬁning Myc Activation B. Detecting Activated Myc Expression in Human Tumors C. Myc Activation through Posttranslational Modiﬁcation D. Myc as an Anticancer Target E. Summary and Perspectives III. Biological Activities of Myc A. Cell Cycle B. Cell Growth C. Block of Differentiation D. Apoptosis E. Cellular Transformation F. Genomic Instability G. Angiogenesis H. Summary and Perspectives IV. Functions of Myc as a Regulator of Transcription A. Identiﬁcation of Target Genes B. Deﬁning Bona Fide Myc Target Genes C. Essential Regions D. Mechanisms of Activation E. Mechanisms of Repression F. Summary and Perspectives V. Identiﬁcation of Myc-Interacting Proteins A. Impact on Myc Function and Activity B. Coactivators C. Transcriptional Regulators D. Tumor Suppressors E. Cytoplasmic Proteins F. The Myc Family and c-Myc Isoforms G. Summary and Perspectives VI. Overall Perspectives References ∗ Authors

contributed equally to this work.

Advances in CANCER RESEARCH 0065-230X/02 $35.00

81

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

82

Sara K. Oster et al.

The activated product of the myc oncogene deregulates both cell growth and death check points and, in a permissive environment, rapidly accelerates the affected clone through the carcinogenic process. Advances in understanding the molecular mechanism of Myc action are highlighted in this review. With the revolutionary developments in molecular diagnostic technology, we have witnessed an unprecedented advance in detecting activated myc in its deregulated, oncogenic form in primary human cancers. These improvements provide new opportunities to appreciate the tumor subtypes harboring deregulated Myc expression, to identify the essential cooperating lesions, and to realize the therapeutic potential of targeting Myc. Knowledge of both the breadth and depth of the numerous biological activities controlled by Myc has also been an area of progress. Myc is a multifunctional protein that can regulate cell cycle, cell growth, differentiation, apoptosis, transformation, genomic instability, and angiogenesis. New insights into Myc’s role in regulating these diverse activities are discussed. In addition, breakthroughs in understanding Myc as a regulator of gene transcription have revealed multiple mechanisms of Myc activation and repression of target genes. Moreover, the number of reported Myc regulated genes has expanded in the past few years, inspiring a need to focus on classifying and segregating bona ﬁde targets. Finally, the identity of Myc-binding proteins has been difﬁcult, yet has exploded in the past few years with a plethora of novel interactors. Their characterization and potential impact on Myc function are discussed. The rapidity and magnitude of recent progress in the Myc ﬁeld strongly suggests that this marvelously complex molecule will soon be unmasked. C 2002 Academic Press.

I. INTRODUCTION Deregulation of the cellular myc protooncogene is one of the strongest activators of carcinogenesis. Clearly, understanding its role in tumor etiology will mark a key advance in the cancer problem. The goals of this review are to focus on advances, key challenges, and outstanding issues in the Myc ﬁeld that have emerged at the start of this new millennium. The speciﬁc topics covered include the role of Myc in cancer, the biological activities of Myc, the functions of Myc as a regulator of gene transcription, and the identiﬁcation of Myc-interacting proteins. Recent articles will be highlighted to reﬂect the aims of the periodical.

II. MYC AND CANCER Activation of c-, N-, and L-myc (Fig. 1) signiﬁcantly affects the germination, expansion, and evolution of numerous diverse human cancers (Dang, 1999; Nesbit et al., 1999). The power and plasticity of this oncogene family to drive transformation reinforce the need to better understand the molecular mechanism(s) of action as well as to improve methods to detect and

The myc Oncogene

83

Fig. 1 c-Myc isoforms and Myc family members. (A) The three c-Myc isoforms result from alternative translational start sites. c-Myc-1 is initiated at an upstream cryptic CUG while c-Myc-S is initiated at downstream internal AUGs. c-Myc-2 is initiated at the consensus AUG and is the most abundantly expressed isoform found in proliferating tissues and tumors. The small circles denote the locations of known phosphorylation sites. (B) The Myc family includes c-, N-, L-, S-, and B-Myc. Myc family members are highly conserved through evolution and share the greatest degree of homology at the N-terminal domain (NTD) Myc Box regions (MB1 and MB2) that are unique to the Myc family, and at the C-terminal domain (CTD) basic helix– loop–helix–leucine zipper motif (bHLHLZ). By contrast, amino acids in the NTD excluding MB1 and MB2, and the central region amino acids are less conserved. The MB1 and MB2 sequences encode functions that are important for Myc’s ability to regulate gene transcription. The bHLHLZ is responsible for site-speciﬁc DNA-binding and heterodimerization with Myc’s primary partner Max. Importantly, c-, N-, and L-Myc family members have transforming activities while B- and S-Myc are inhibitory.

84

Sara K. Oster et al.

quantify oncogenic myc in tumor cells. Activation of myc is deﬁned as deregulation of the normal, highly controlled expression pattern of the myc protooncogene (Cole, 1986; DePinho et al., 1991; Kelly and Siebenlist, 1986; Rabbitts, 1985; Spencer and Groudine, 1991). This has been difﬁcult to measure, as deregulation can occur by a variety of mechanisms and the resulting expression patterns of activated Myc protein are amorphous. Distinguishing between cells that harbor normal or activated Myc protein has been a challenge that has plagued the ﬁeld since the myc oncogene was ﬁrst identiﬁed in the early 1980s, yet advances in molecular diagnostic technology have begun to address this long-standing problem. The recent fundamental shift in both deﬁning and measuring Myc activation and its impact will be discussed.

A. Deﬁning Myc Activation In nontransformed cells, expression of the c-myc protooncogene is exquisitely responsive to the extracellular milieu and generally reﬂects the growth state of the cell (Henriksson and Luscher, 1996). A detailed overview of the complexities of myc regulation is beyond the scope of this article and will only be discussed in the context of the carcinogenic process. In normal, nondividing cells, Myc levels are low, but as an immediate early growth response gene, expression is rapidly elevated following exposure to a variety of growth stimuli. In dividing cells, Myc expression is maintained at a relatively constant intermediate level throughout the cell cycle. Exposure of cells to agents that block cell proliferation stimulates the immediate and rapid down-regulation of Myc expression. Supraphysiologic levels of Myc expression are not tolerated and are thought to be lethal in nontransformed cells. In normal cells Myc expression is so highly controlled and immediately responsive to the proliferative cues of the cellular environment that it has been termed “the intracellular sentinel of the extracellular milieu.” In contrast to the highly regulated state of Myc expression in nontransformed cells, activation of the myc oncogene in tumor cells is synonymous with deregulated expression of the intact Myc protein (Henriksson and Luscher, 1996; Nesbit et al., 1999). In its oncogenic form, Myc may be constitutively expressed at levels ranging from moderate to extreme and is nonresponsive to external signals. Alternatively, the highly regulated pattern of Myc expression can remain intact, but the level of expression exceeds the usual ceiling for the given cell type. Mutations of the Myc protein product are not necessary for activation, although they have been documented within the activated c-myc allele of Burkitt’s lymphoma (Albert et al., 1994; Axelson et al., 1995; Bhatia et al., 1993; Yano et al., 1993). Interestingly, in neuroblastoma, where N-myc ampliﬁcation is a hallmark of aggressive

The myc Oncogene

85

disease, investigations have suggested that point mutations of the coding region are not apparent (Feng et al., 1997; Hogarty and Brodeur, 1999). Because mutations in the coding region are not diagnostic of an activated myc allele, molecular pathologists have relied on the presence of gross chromosomal abnormalities, such as translocation or ampliﬁcation, of the myc locus to strictly deﬁne activation of this oncogene in tumor cells (Hecht and Aster, 2000; Klein, 1999; Liao and Dickson, 2000; Thorner and Squire, 1998). Although this diagnostic criterion is deﬁnitive, it is also highly restrictive and has resulted in an underestimate of the numbers of tumors harboring deregulated Myc protein. Myc activation can occur by a wide range of direct and indirect mechanisms. Direct activation can result from disruption of any one of the multiple regulatory mechanisms that maintain control of Myc expression in normal cells. For example, Myc deregulation can occur through stabilization of myc mRNA transcripts or through enhanced initiation of translation due to mutation of the internal ribosomal entry site (Bernasconi et al., 2000; Chappell et al., 2000). In addition, Myc activation can occur as an indirect consequence of an activated upstream signaling cascade (Barker et al., 2000; Behrens, 2000; Chen et al., 2001a; Kolligs et al., 2000; Yang et al., 2001a; Barone and Courtneidge, 1995; Blake et al., 2000; Bowman et al., 2001; Chiariello et al., 2001). Importantly, new insights suggest that whether myc is activated as a direct or indirect consequence is largely extraneous. The assumption is that the elevated Myc levels in the tumor cell contribute to the carcinogenic evolution of the transformed clone (Herms et al., 2000). Clearly this hypothesis requires further investigation. The key is to evaluate whether deregulated Myc expression can distinguish tumor cells from normal cells and serve as a therapeutic target to speciﬁcally block tumor-cell growth or trigger tumor-cell death (Herms et al., 2000). This intriguing possibility can be directly addressed with the recent advancements in both Myc molecular diagnostics and therapeutics.

B. Detecting Activated Myc Expression in Human Tumors Recent appraisal strategies to detect myc activation in tumor cells rely on both elevated expression of myc mRNA and genetic alterations of the myc locus. This reﬁnement in deﬁnition of deregulated myc is now possible with the development of new technologies that increase sensitivity and speciﬁcity of detection and add a level of quantiﬁcation with greater dynamic range than was previously possible. In the past, an activated myc allele was primarily diagnosed by time-consuming, labor-intensive karyotype or Southern blot analysis.Technological advances to detect genetic alterations of the myc locus

86

Sara K. Oster et al.

include the application of comparative genomic hybridization (CGH), ﬂuorescence in situ hybridization (FISH), and spectral karyotype (SKY) analyses. In addition, to evaluate expression levels of myc mRNA, precise excision of tumor material using laser-capture microdissection or fresh needle aspirates is used in combination with real-time reverse transcription–polymerase chain reaction (RT-PCR) as well as expression proﬁling using microarray technology. Using these approaches, rapid, quantiﬁable, high-throughput screening of myc activation is now possible (Glockner et al., 2000; Jung et al., 2000; Mitas et al., 2001; Schmittgen et al., 2000; Taylor et al., 2000; Thiesse et al., 2000). Application of these new diagnostic approaches has already signiﬁcantly advanced the ﬁeld at many levels. First, diagnosis of deregulated myc can now be achieved with a clinically useful turnaround time. Results from these approaches can be used in a timely manner to better evaluate both disease prognosis and choice of therapeutic approach. With the customization of cancer diagnosis and treatment fast approaching, evaluating the activation status of the myc oncogenes will be pivotal and signiﬁcantly affect individualized care. Myc activation is often associated with late-stage and/or poor-prognostic disease, and evidence of this genetic lesion may suggest that a more aggressive treatment strategy is appropriate (El Gedaily et al., 2001; Kaltz-Wittmer et al., 2000; Kraehn et al., 2001; Minard et al., 2000; Obara et al., 2001; Pagnano et al., 2001; Sarbia et al., 2001). Second, analysis of myc deregulation has suggested that this feature may segregate with particular tumor subtypes and may be used as a diagnostic marker for speciﬁc diseases (Kraehn et al., 2001; Obara et al., 2001). Segregation of myc activation with speciﬁc disease states will likely become thematic and expand signiﬁcantly in coming years. This is particularly conceivable with the rapid application of expression proﬁling using a microarray approach to further distinguish tumor subtypes that show increased myc expression (Huang et al., 2000; Watatani et al., 2000). Third, insight into the mechanism of transformation will be forthcoming with the identiﬁcation of cooperating lesions that are consistently associated with myc deregulation in a tumor subtype (Cohn et al., 2000; Fiche et al., 2000; Fruhwald et al., 2000; George and Squire, 2000; Janocko et al., 2001; Janoueix-Lerosey et al., 2000; Matthay, 2000; Teitz et al., 2000). Moreover, solid tumor sections often show evidence of cellular heterogeneity, and it will be interesting to determine whether patches of tumor cells harboring an activated myc allele will evolve and subsequently expand over time to dominate the mass (Squire et al., 1996; Vandesompele et al., 1998). Fourth, novel activation mechanisms that can increase Myc dosage have been identiﬁed, including unbalanced translocations (PadillaNash et al., 2001). Fifth, to better understand the role of Myc in human cancer, the activation status of both a downstream Myc target gene and Myc itself has been evaluated in patient samples. The linkage in expression of

The myc Oncogene

87

speciﬁc Myc target genes with Myc activation in human cancers may be a valuable approach to evaluate the molecular role Myc holds in carcinogenesis and deserves further investigation (Bergmann et al., 2001; Drissi et al., 2001; Kumamoto et al., 2001; Latil et al., 2000; Poremba et al., 2000; Sagawa et al., 2001; Sasaki et al., 2001; Streutker et al., 2001). Finally, as expected, myc deregulation in human cancer has been shown to occur at a higher rate than previously determined (El Gedaily et al., 2001; Kaltz-Wittmer et al., 2000; Kraehn et al., 2001; Latil et al., 2000; Mathew et al., 2001). For example, the incidence of Myc ampliﬁcation in medulloblastoma was on average ∼8%; however, analysis has shown that Myc activation due to mRNA overexpression occurs in 42% of tumors and is associated with particular subtypes of this primitive neuroectodermal tumor (Brown et al., 2000; Gilhuis et al., 2000; Herms et al., 2000; Reardon et al., 2000; Scheurlen et al., 1998). With advances in molecular pathology, signiﬁcant insight into the molecular role as well as the diagnostic prevalence and relevance of Myc deregulation in speciﬁc human cancers promises to be immediately forthcoming and highly instructive. Despite these advances, many outstanding questions regarding myc activation remain unresolved. Ultimately measuring changes at the DNA and RNA level is insufﬁcient as it assumes that these alterations result in elevated Myc protein levels. Expression of Myc protein is largely assayed using an immunohistochemical approach (Pich et al., 2000; Ricaniadis et al., 2001). This approach is ideal when directly quantiﬁed or combined with RT-PCR (Kraehn et al., 2001). A ﬂow cytometric approach has also been described but has been rarely adopted (Zheng et al., 2000). Technological advances to rapidly measure both quantitative and qualitative parameters of Myc protein in tumor cells would ﬁll this gap and further advance the ﬁeld. Indeed, given that the protein product is identical in nontransformed and transformed cells, it has been assumed that the function of Myc protein remains uniform in the two cell settings. However, there is emerging evidence that the precise level of Myc protein expression can have a direct impact on function (Bazarov et al., 2001). These observations and those of others further suggest that precise measurement of Myc protein levels will be informative in both normal and tumor cells.

C. Myc Activation through Posttranslational Modiﬁcation Further evidence to show that Myc activation can occur at multiple levels comes from studies on Myc protein stabilization. The Myc protein is among a group of short-lived proteins whose destructions are regulated by the

88

Sara K. Oster et al.

ubiquitin/26S proteasome pathway (Bahram et al., 2000; Bonvini et al., 1998; Gregory and Hann, 2000; Gross-Mesilaty et al., 1998; Salghetti et al., 1999). Myc is a dynamically regulated phosphoprotein with least 10 major phosphorylation sites spanning the length of the protein (Henriksson et al., 1993; Lutterbach and Hann, 1994, 1997; Peukert et al., 1997; Seth et al., 1992, 1993). However, the role of these modiﬁcations remains ill deﬁned and has been largely unexplored in primary malignant cells harboring activated Myc. Recently, several groups have identiﬁed posttranslational modiﬁcations of Myc that may be responsible for signaling protein turnover. Myc proteins with mutations at or near T58, which abolish phosphorylation at this site are expressed at a high frequency in Burkitt’s lymphoma and have an increased transformation potential in cell culture systems (Albert et al., 1994; Axelson et al., 1995; Bhatia et al., 1993; Henriksson et al., 1993; Pulverer et al., 1994; Yano et al., 1993). These earlier observations were a prelude to the recently described role of T58 in the regulation of Myc ubiquitination and protein turnover. Several groups have shown that mutations at T58 result in inefﬁcient ubiquitination and a stabilization of Myc protein (Bahram et al., 2000; Gregory and Hann, 2000; Niklinski et al., 2000; Salghetti et al., 1999; Sears et al., 2000). The protein kinase GSK-3 is most likely responsible for the phosphorylation of Myc at T58 that requires prior phosphorylation of the nearby S62 residue (Henriksson et al., 1993; Lutterbach and Hann, 1994; Pulverer et al., 1994; Sears et al., 2000). The precise role and upstream kinases involved in S62 phosphorylation, however, remain controversial (Henriksson et al., 1993; Lutterbach and Hann, 1999; Pulverer et al., 1994; Sears et al., 2000). Higher order regulation of these two phosphorylation events has been reported to occur via a bifurcated ras signaling pathway to promote the accumulation of Myc protein (Sears et al., 1999, 2000). Alternative mechanisms for the regulation of Myc protein stability have also been suggested. The highly conserved MB1 and MB2 regions contain signals for Myc degradation via the proteasome pathway and can act as autonomous degradation signals in both yeast and mammalian cells (Flinn et al., 1998; Gregory and Hann, 2000; Salghetti et al., 1999). Multiple conserved lysine residues are encoded within MB2 and suggest potential sites for ubiquitination and a PEST sequence at amino acids 226–270 of Myc has been shown to be necessary for rapid protein turnover, but not ubiquitination (Gregory and Hann, 2000). Although the PEST sequence encodes a putative CKII phosphorylation consensus sequence, in vivo regulation through phosphorylation at this site has not been conﬁrmed. Interestingly, wild-type Myc protein is stabilized and ubiquitination of Myc is blocked during mitosis (Gregory and Hann, 2000). Coincidentally, Myc becomes hyperphosphorylated during this phase of the cell cycle (Luscher and Eisenman, 1992; Lutterbach and Hann, 1994). However, whether this hyperphosphorylated state of Myc is the cause or consequence of increased protein stability is

The myc Oncogene

89

unknown. Taken together, it is likely that multiple mechanisms exist to regulate Myc protein stability. These may be regulated in a cell type-dependent manner or be inﬂuenced by inherent genetic factors that can differ between cell systems, including normal versus tumor cells (Bahram et al., 2000; Gregory and Hann, 2000). Interestingly, Myc is alternatively modiﬁed at T58 by O-GlcNAc glycosylation, suggesting additional control beyond the binary model of phosphorylation at this site (Chou et al., 1995a, 1995b). The impact of this unique and dynamic modiﬁcation of proteins has not been clearly elucidated (Wells et al., 2001). It is known that O-GlcNAc modiﬁcation of Sp1 can prevent its degradation by the proteasome and correlates with its transcriptional potential and ability to interact with cofactors (Du et al., 2000; Han and Kudlow, 1997; Roos et al., 1997). Similarly, the DNA binding ability of p53 has been reported to be regulated through glycosylation (Shaw et al., 1996). Together, these studies suggest that T58 of the Myc protein may exist in three states: glycosylated, phosphorylated, or unmodiﬁed. The impact of each of these modiﬁcations on Myc activity or protein stability has yet to be fully addressed. Overall, recent advances have begun to unveil the functional signiﬁcance of Myc posttranslational modiﬁcation with a particular emphasis on the role of phosphorylation. In building new models to describe Myc protein regulation and its impact on the pathology of cancer, one must be mindful of the following observations. First, Myc phosphorylation can alter depending on the immortalization status of the cell system in which it is studied (Lutterbach and Hann, 1997). By extension, such ﬁndings also vary amid endogenous, regulated Myc and ectopically expressed versions of the protein in the same cell type (Lutterbach and Hann, 1999). Thus, the relevant consequences of Myc phosphorylation and the signaling pathways described to date may be confounded by protein-expression and cell-type differences. Nonetheless, this evidence shows T58 mutations can have a signiﬁcant impact on Myc protein stability, and mutations at this loci are of pathological signiﬁcance, although clearly there are other mechanisms at work. Together these different regulatory events, along with those described in the previous section, may have either additive or synergistic control over the amount of Myc protein present in a cell at any given point. Clearly more effort in these areas will provide further insight into the role and mechanisms of Myc activation in the evolution of malignant transformations.

D. Myc as an Anticancer Target We have entered an exciting and revolutionary phase in the design of novel anticancer therapeutics. With the power to identify the genetic abnormalities that are the root cause of an individual’s cancer, we will be in a strong position

90

Sara K. Oster et al.

to customize treatment options to speciﬁcally target the patient’s tumor cells and spare the neighboring normal cells. Given the high prevalence of Myc activation in a broad range of human cancers, agents targeting Myc promise to have enormous impact on the clinical management of disease. These drugs would be utilized in combination with a cocktail of mechanistically distinct anticancer agents to successfully combat and eradicate tumor cells. There are two distinct approaches that are presently under investigation: those that “inhibit” and those that “exploit” Myc deregulation to kill tumor cells in a sensitive and speciﬁc manner (Penn, 2001).

1. INHIBITING MYC ACTIVITY Myc is an attractive therapeutic target, as many tumor cells are dependent upon Myc expression for continued proliferation (Pelengaris et al., 1999). Moreover, blocking Myc expression leads to a rapid growth inhibitory response due to the short half-life of both myc mRNA and Myc protein. In addition, it has been shown that even small decreases in Myc expression can affect cell transformation (Bazarov et al., 2001). The majority of effort is presently focused on the use of antisense molecules (Akie et al., 2000; Chen et al., 2001b; Clark, 2000; Ebinuma et al., 2001; Hudziak et al., 2000; Junghans et al., 2000; McGufﬁe et al., 2000; Schmutzler and Koehrle, 2000). Peptide inhibitors directed to the Myc carboxyl end are also under investigation (Boffa et al., 2000; Pescarolo et al., 2001). Another approach is to introduce ectopic Mad family members to antagonize Myc action (Bejarano et al., 2000; Gagandeep et al., 2000; Taj et al., 2001). One group has identiﬁed a speciﬁc cocktail of cytokines that overcome Myc and drive tumor cell differentiation (Oberg et al., 2001). Others have shown that iron chelators will suppress N-myc expression in neuroblastoma cells (Fan et al., 2001). Clearly, novel approaches to directly target Myc expression, Myc activity, and Myc/protein or Myc/DNA interactions hold enormous promise for inhibiting tumor cell proliferation or triggering tumor cell apoptosis.

2. EXPLOITING MYC ACTIVITY Myc activation readily distinguishes tumor cells from their neighboring normal cells and can be used to target tumor-speciﬁc cell destruction (Penn, 2001; Garrett and Workman, 1999; Lowe and Lin, 2000). It is now well acknowledged that cytostatic agents will not cause collateral damage to nontransformed cells, yet can trigger apoptosis in tumor cells harboring an activated myc allele (Evan et al., 1992). Efforts to exploit this feature of Myc activation are underway (Rossler et al., 2001). This heightened sensitivity to apoptosis is also evident in response to vesicular stomatitis virus infection,

The myc Oncogene

91

and this microorganism is under development as a therapeutic approach (Balachandran et al., 2001). Another mechanism to exploit the presence of Myc activation is to develop Myc-speciﬁc prodrug activation systems. With this approach, tumor cells harboring an activated myc allele are infected with adenovirus carrying a Myc-responsive promoter driving an enzyme that catalyzes the conversion of a prodrug to its active form. Subsequent exposure to prodrug leads to tumor-speciﬁc drug conversion and sensitivity to the active therapeutic agent (Iyengar et al., 2001; Nishino et al., 2001; Patterson and Harris, 1999; Pawlik et al., 2000). Although many of the Myc-speciﬁc therapeutic approaches show promise, this ﬁeld is in its infancy. One of the major challenges with these approaches lies in achieving speciﬁc delivery of the Myc inhibitor to the nucleus of tumor cells in vivo. Methods of delivery presently under investigation include replication-incompetent adenovirus and liposome delivery (Chen et al., 2001b; Kondoh et al., 2000). Novel approaches in tumor-speciﬁc drug delivery should address this important issue and further increase the therapeutic index of this class of agent. Indeed, the drug delivery issue is generally applicable to most modulators that target speciﬁc proteins, such as anti-p53, and is not restricted to agents that target Myc (Penn, 2001; Garrett and Workman, 1999; Lowe and Lin, 2000). The development of agents that inhibit or exploit Myc activation will expand further as our fundamental knowledge of Myc biological functions increases.

E. Summary and Perspectives Myc can be deregulated by a plethora of activation mechanisms ranging from myc gene ampliﬁcation to alterations in Myc protein stability. It is well established that Myc protein functions in a dose-dependent manner and that even small increases in expression can have profound impact and confer a strong growth advantage on the affected clone. Advances in diagnostic technologies have signiﬁcantly increased the sensitivity and speciﬁcity of detection of myc deregulation within tumor cells in a rapid and reproducible manner. This progress will signiﬁcantly advance our understanding of the role of Myc in tumorigenesis and promises to directly affect the clinical management of patients. In addition, the knowledge gained will provide novel insights into the nature of the lesions that collaborate with Myc to drive the transformation process as well as lead to better understanding of Myc’s mechanism(s) of action. Analysis suggests that the presence of deregulated Myc is common, especially among particular tumor subtypes. This further supports the rationale of devising additional strategies to directly exploit this unique feature of tumor cells for the design of novel therapeutics to target activated Myc.

92

Sara K. Oster et al.

III. BIOLOGICAL ACTIVITIES OF MYC In response to signals from the cellular microenvironment, Myc can regulate a wide array of distinct biological activities (Fig. 2). Myc has been shown to induce cell cycle progression, enable cell growth, block differentiation, potentiate apoptosis, drive transformation, activate genomic instability, and stimulate angiogenesis. Clearly Myc is a multifunctional protein that holds a pivotal role in regulating numerous cellular processes essential for the control of cell proliferation. It is believed that Myc can direct such a disparate set of biological activities by functioning as a regulator of gene transcription. Thus the working hypothesis is that Myc regulates a speciﬁc subset of target genes to stimulate separate downstream signaling cascades that then execute each distinct biological activity (Fig. 2). Advances in understanding the role of Myc in regulating these activities will be discussed.

A. Cell Cycle Several lines of evidence strongly advocate an essential role for Myc in the transition from G0 to S phase of the cell cycle. Myc is an early growth response gene whose induction following exposure to mitogens such as platelet-derived growth factor (PDGF) is required for cells to traverse G1. This was ﬁrst shown when microinjection of antisense c-myc was able to block G1 transition (Heikkila et al., 1987). More recently, it has been shown that loss of c-myc using a conditional knockout approach also signiﬁcantly compromises G0 to S phase progression (de Alboran et al., 2001). Historically, the G1 phase of the cell cycle has been subdivided into two discrete and sequential phases known as competence and progression. To successfully pass through these check points, speciﬁc signals provided by growth and survival factors are essential. It has been shown that Myc expression, along with activated MEK leading to Erk activation, could effectively substitute for PDGF in the early phase of G1 and render cells competent for further passage through the later G1 progression phase (Jones and Kazlauskas, 2001). The interdependence of PDGF signaling with Myc activity has been further reinforced by ﬁndings suggesting that a feedback loop intimately links these two molecules. PDGF stimulation has been shown to activate src, which signals through Stat3 and/or Rho GTPases, leading to up-regulation of c-myc expression (Barone and Courtneidge, 1995; Blake et al., 2000; Bowman et al., 2001; Chiariello et al., 2001). Myc in turn represses PDGF-β receptor expression at the level of gene transcription in a negative feedback loop to regulate mitogen signaling (Izumi et al., 2001; Oster et al., 2000). Evidence showing that Myc collaborates with Ras signaling to promote transformation

Fig. 2 Myc biological activities. Myc responds to changes in the cellular microenvironment to regulate many known and unknown downstream target genes. Through its function as a transcription factor, Myc affects many disparate biological activities. Myc can drive cell proliferation by regulating cell cycle, cell growth, differentiation, genomic stability, and angiogenesis. Importantly, Myc can also drive the apoptotic process that can act as a fail-safe mechanism to hinder the expansion of a clone harboring activated Myc proteins. The working hypothesis is that Myc directs distinct downstream genetic programs to drive each of these processes (see text for further detail).

94

Sara K. Oster et al.

further supports this PDGF–src–myc cascade. An additional role for Myc in the late progression phase of G1 has also been proposed (Jones and Kazlauskas, 2001). Indeed, Myc expression in the absence of mitogen can fully drive G0 to S phase transition, an activity that is dependent upon the conserved MB2 region of Myc (Daksis et al., 1994; Eilers et al., 1989). Myc may also have a role in phases other than G1. Unlike other immediate early response genes, Myc expression is maintained throughout subsequent phases of the cell cycles (Hann et al., 1985). Indeed, a role for Myc in the G2 phase is supported from analysis of the c-myc null Rat1 ﬁbroblasts, which exhibit a prolonged G2 as well as G1 phase of the cell cycle (Mateyak et al., 1997). The question of Myc’s role, if any, in permitting cells to traverse these other phases of the cell cycle remains unanswered and requires investigation. At a minimum, Myc is essential for the G0 to S phase progression of the cell cycle. At a molecular level Myc appears to regulate cell cycle progression by promoting coordinate changes in expression of a large number of genes. This likely explains why expression of a single gene cannot substitute for Myc in driving cell proliferation (Berns et al., 2000a; Nikiforov et al., 2000). Deregulated myc expression leads to direct or indirect activation of cyclin D1, D2, E, and A, as well as CDK4 and Cdc25A. Myc also suppresses cell cycle check-point genes such as gadd45 and gadd153, as well as cyclin kinase inhibitors p15INK4b, p21Cip1, and p27Kip1 (Dang, 1999; Grandori et al., 2000). The cascade triggered by Myc is unique at a molecular level and yet has features that functionally overlap with other cell cycle regulators such as Ras and E2F. Indeed, one of the roles of Myc is to activate G1 cyclin/cdk complexes and subsequently inactivate the restriction point characterized by the product of the retinoblastoma gene, pRB (Beier et al., 2000; Bouchard et al., 1998; Elend and Eilers, 1999; Santoni-Rugiu et al., 2000). This is further supported by observations that Myc can stimulate the expression of the Id2 protein that inhibits pRB function (Lasorella et al., 2000). In addition, Myc induces E2F-1 and E2F-2, which in turn regulate an independent, parallel pathway that plays an important synergistic role with Myc in cell cycle control (Beier et al., 2000; Berns et al., 2000b; Santoni-Rugiu et al., 2000). The E2F family regulates many target genes, including those required for replication and mitotic activities (Humbert et al., 2000; Ishida et al., 2001; Muller et al., 2001). At a minimum, Myc coordinates the molecular events required for the cell to traverse G1 to S phase by regulating both D and E cdk/cyclin kinase complexes as well as the E2F transcription factors and their downstream cascades. Clearly, since Ras, E2F, and Myc are not functionally redundant (Beier et al., 2000; Berns et al., 2000b; Conner et al., 2000; Obaya et al., 1999; Santoni-Rugiu et al., 2000), additional Myc target genes are operable downstream of Myc to control cell cycle progression.

The myc Oncogene

95

B. Cell Growth Experimental evidence suggests that Myc can also increase the proliferative potential of a cell by regulating cell growth. Cell growth is deﬁned as the accumulation of cell mass or cell size, which is a prerequisite for cellular division. This pathway appears to be distinct, yet intimately linked to Myc’s ability to drive cell cycle progression (Beier et al., 2000; Schmidt, 1999). Studies of Drosophila dmyc, the ortholog of vertebrate myc, have shown that decreased expression of dMyc resulted in cells with diminished size, whereas overexpression led to cells that were larger than normal (Johnston et al., 1999). Importantly, overall cell number and cell cycle time were not affected by altered Myc levels, suggesting that Myc regulation of cell growth was uncoupled from its effects on the cell cycle. Evidence to support a role for Myc in the regulation of cell growth in mammalian cells has also been reported. Murine B cells expressing the Eμ-myc transgene have increased rates of protein synthesis and are enlarged throughout development as well as during all phases of the cell cycle (Iritani and Eisenman, 1999). Moreover, in an independent study, B cells that could not be triggered to transit the cell cycle because of an inhibitory lesion could still be induced to grow in size in response to Myc activation (Schuhmacher et al., 1999). Similar ﬁndings have been observed in response to ectopic Myc expression in mouse hepatocytes in vivo (Kim et al., 2000). Moreover, studies in p27 Kip1 nullizygous mouse embryo ﬁbroblasts (MEFs) have shown that Myc regulation of cell growth can be uncoupled from the cell cycle-regulatory effects of Myc (Beier et al., 2000). Further advances in this area have been forthcoming in Drosophila, where additional genes have been shown to regulate cell growth in a similar manner to dMyc, including dRas1, as well as dPI3K and other members of the insulin receptor pathway (Prober and Edgar, 2001; Stocker and Hafen, 2000; Weinkove and Leevers, 2000). It will be interesting to determine how these growth-regulatory pathways intersect and whether they fall into similar or different complementation groups. The precise mechanism employed by Myc to regulate cell growth has not yet been delineated; however, candidate target genes have been identiﬁed that regulate protein synthesis as well as cell metabolism. They include genes such as eIF4E and eIF2α, as well as RNA helicase MrDb, IRP-2, and H-ferritin (Grandori et al., 2000; Schmidt, 1999). With the power of P-element insertion, genetic screens in Drosophila for both suppressors and enhancers of the dMyc phenotype should unveil the pathway(s) regulated by Myc to control cell growth. Indeed, by this approach novel downstream targets of Myc may also be identiﬁed and, once veriﬁed as Myc-regulated genes in mammalian cells, may shed light on this newly discovered biological activity of Myc.

96

Sara K. Oster et al.

C. Block of Differentiation Members of the Myc family have signiﬁcant impact on cellular programs of differentiation. In a wide variety of differentiating systems, four recurring themes have emerged. First, each member of the Myc family has a distinct pattern of developmental gene expression, suggesting nonredundant contributions in vivo (Bull et al., 2001; DePinho et al., 1991; Douglas et al., 2001; Lemaitre et al., 1996; Morgenbesser and DePinho, 1994). Second, Myc-family genes are preferentially expressed in proliferating cellular compartments (Bull et al., 2001; Foley et al., 1998; Hurlin et al., 1995; McArthur et al., 1998; Queva et al., 1998). Third, down-regulation of c-myc expression is a consistent observation in cells triggered to undergo differentiation (Chang et al., 2000a; Gonda and Metcalf, 1984; Henriksson and Luscher, 1996; Larsson et al., 1994). Fourth, ectopic expression of myc effectively blocks differentiation in a wide variety of cell types both in vitro and in vivo (Brewer, 2000; Facchini and Penn, 1998; Iritani and Eisenman, 1999). The differentiation program is so exquisitely sensitive to the levels of myc expression that down-regulation of myc expression is itself sufﬁcient to trigger differentiation in some cell systems (Bacon and Wickstrom, 1991; Facchini and Penn, 1998). More often, suppression of Myc expression is an essential component to a full genetic program of differentiation (Chang et al., 2000a). This suppression of c-myc is mediated directly or indirectly by the “master regulator” of differentiation for that cell type. For example, C/EBPα is a transcription factor that is essential for myoblast commitment to the granulocytic lineage and also plays a role in the differentiation of hepatocytes and adipocytes. C/EBPα is thought to drive the differentiation program, in part, by forming a complex with E2F/pRB and directly repressing c-myc transcription through the E2F-binding site in the proximal promoter region of the c-myc gene (Johansen et al., 2001). Indeed, the absence of Myc is often one of the ﬁrst measurable events triggered by negative growth regulatory signals and is often essential for cells to exit the cell cycle and enable processes such as differentiation to proceed (Facchini and Penn, 1998; Henriksson and Luscher, 1996). Conversely, ectopic Myc expression blocks most programs of differentiation. It remains unclear whether Myc blocks differentiation by directly repressing transcription of the so-called master regulators or whether this effect is an indirect consequence of Myc-driven cell proliferation. Indeed, structure–function analysis maps both activities to similar regions of the Myc protein that include MB2 and the bHLHLZ motifs (Freytag et al., 1990). Evidence that Myc can directly repress master regulators such as C/EBPα has been suggested. Myc can repress C/EBPα transactivation function and can inhibit C/EBPα promoter activity as measured by reporter assays (Constance et al., 1996; Li et al., 1994). Whether it is a direct or indirect effect of Myc,

The myc Oncogene

97

inhibiting cell differentiation is a powerful biological activity that directly contributes to transformation. The precise mechanism of Myc suppression of differentiation has not yet been fully explored and is an area of research that requires more attention.

D. Apoptosis Another facet of Myc activity is the ability to potentiate apoptosis triggered by most death agonists. The molecular mechanism by which Myc exerts this effect has not been fully elucidated, although several models have been described (Breckenridge and Shore, 2000; Evan and Vousden, 2001; Nesbit et al., 1999; Prendergast, 1999; Thompson, 1998). Insight into this issue has been forthcoming in recent years and has been obtained by three main approaches: genetic dissection of the regions of Myc important for biological activities including apoptosis, identiﬁcation of bona ﬁde Myc target genes, and determination of rate-limiting steps in the apoptotic pathways activated downstream of Myc. Several groups have reported that different subregions of the Myc NTD control distinct biological functions, including apoptosis (Chang et al., 2000b; Conzen et al., 2000; Nesbit et al., 2000). Overall, these studies support the notion that Myc’s ability to drive apoptosis may be genetically distinct from its ability to drive proliferation and transformation, and that the regulation of certain downstream target genes, but not others, may be required to elicit Myc’s apoptotic response. Moreover, the MB2 region of Myc as well as key phosphorylation sites within the Myc NTD appear to be important regulatory sites involved in modulating the apoptotic potential of Myc (Chang et al., 2000b; Conzen et al., 2000; Nesbit et al., 2000; Noguchi et al., 1999, 2001). These regions of Myc have been implicated in the regulation of Myc protein stability and protein interactions (for more detail see Section V, Table III, and Sakamuro and Prendergast, 1999) and could in turn have an impact on the downstream genetic program induced by Myc under apoptotic conditions. Until Myc target genes essential for apoptosis are identiﬁed, however, this issue cannot be directly addressed. Indeed, the search for putative Myc target genes involved in apoptosis has been an area of intense research. The advent of microarray technology has led to a number of reports describing a plethora of potential target genes that are either activated or repressed in response to Myc (Boon et al., 2001; Coller et al., 2000; Guo et al., 2000; Kim et al., 2001; Neiman et al., 2001; Nesbit et al., 2000; O’Hagan et al., 2000; Schuhmacher et al., 2001). However, despite the many new candidate genes, none so far have had any obvious role in directly regulating the apoptotic machinery, although further characterization of these genes may prove more revealing. A second approach has

98

Sara K. Oster et al.

been to investigate the potential role of Myc in the regulation of genes known to be important during the apoptotic process. Preliminary evidence suggests that Myc may regulate key apoptotic molecules such as the proapoptotic BH3-containing molecule Bax, as well as the ligand to the CD95 death receptor (see Table I for further details) (Fulda et al., 1999; Kasibhatla et al., 2000; Mitchell et al., 2000). This regulation is not universal, as neither we nor others have observed changes in Bax protein levels when comparing control cells and cells expressing an activated myc allele, whether under growth or under apoptotic conditions (Duelli and Lazebnik, 2000; Soucie et al., 2001). However, Myc is unable to sensitize cells to undergo apoptosis in low-serum conditions in Bax null cells, implicating this molecule as a downstream effector during Myc-induced apoptosis (Mitchell et al., 2000). Myc has been reported to repress transcription of the caveolin-1 gene (Park et al., 2001; Timme et al., 2000). Ectopic expression of caveolin-1, a major structural component of caveolae, appears to abrogate the apoptotic response upon serum withdrawal in the context of deregulated Myc expression (Timme et al., 2000). Interestingly, Myc has also been reported to be involved in the down-regulation of another scaffolding protein, E-cadherin (Evan and Vousden, 2001). The effect of this repression would presumably be the detachment of adherent cells and the consequent induction of apoptosis. Therefore, both activated and repressed Myc target genes may be involved in eliciting the proapoptotic activity downstream of Myc’s function as a transcription factor, but their identities remain largely unexplored. Functional approaches have also yielded important insight into the role of Myc during apoptosis in terms of identifying the rate-limiting step required for Myc to drive the apoptotic process. Myc appears to play an important role at the mitochondria downstream of a variety of disparate apoptotic triggers (Duelli and Lazebnik, 2000; Juin et al., 1999; Soucie et al., 2001). The mitochondria are a nodal point for many apoptotic pathways and can serve as part of an ampliﬁcation loop to bolster the apoptotic signal within a cell (Desagher and Martinou, 2000; Ferri and Kroemer, 2001). First insight into Myc’s role at the mitochondria came from studies showing that Myc induced cytochrome c release from the mitochondria during apoptosis and that ectopic addition of cytochrome c sensitized cells to undergo apoptosis in a manner similar to that resulting from the addition of Myc (Juin et al., 1999). More recently, we and others have shown that the rate-limiting step during apoptosis in the presence of Myc lies upstream of cytochrome c release and involves functional activation of Bax. Using a cell-fusion-complementation assay with cells expressing either ectopic E1A or Myc, it has been shown that these oncogenes may facilitate cytochrome c release from the mitochondria by at least two pathways: one involving the regulation of Bax translocation to the mitochondria downstream of DNA damage, and the second directing the release of cytochrome c from the mitochondria (Duelli and Lazebnik,

The myc Oncogene

99

2000). Analysis of Rat1 c-myc null cells shows that different death stimuli induce apoptosis with characteristic features such as nuclear condensation, membrane blebbing, PARP cleavage, and DNA fragmentation; however, this Myc-independent apoptosis is not inhibited by Bcl-2. This work has shown that Bax translocates to the mitochondria independently of Myc expression. In the absence of Myc, Bax is not activated and cytochrome c is not released from the mitochondria. Reintroduction of Myc into the c-myc null cells restores Bax activation, cytochrome c release, and inhibition by Bcl-2 (Soucie et al., 2001). These results demonstrate a role for Myc in the regulation of Bax activation during apoptosis. Moreover, this Myc-independent apoptosis provides evidence that signaling pathways exist that can circumvent Bax activation and cytochrome c release to trigger caspase activation. Apoptosis initiated by the DNA damaging agent etoposide during the G2 phase of the cell cycle can be dependent upon Myc expression, and this apoptotic defect can be rescued by ectopic expression of cyclin A in c-myc null cells (Adachi et al., 2001). It has also been reported that Myc-mediated apoptosis is abrogated in both caspase-9 and caspase-3 null cells, likely because these effector caspases lie downstream of the mitochondria during death induction (Hakem et al., 1998; Kuida et al., 1998; Woo et al., 1998). Thus, mechanistically, Myc increases the cellular competence to die by enhancing disparate apoptotic signals at a common mitochondrial ampliﬁcation step involving Bax activation and cytochrome c release. The mechanism of Myc-mediated Bax activation remains unclear. It is possible that Myc translocates to the mitochondria in response to apoptotic stimuli and directly activates Bax. A similar mechanism has been described for the TR3/Nur77 transcription factor that can induce apoptosis by translocating to the mitochondria and inducing cytochrome c release (Li et al., 2000). Alternatively, Myc may indirectly control Bax activation and cytochrome c release, perhaps by regulating an upstream activator of Bax such as Bid or caspase-8. Indeed, caspase-8 is frequently inactivated in childhood neuroblastomas with an ampliﬁcation of N-myc, providing indirect evidence that loss of caspase-8 could cooperate with Myc to allow for tumor progression by blocking the apoptotic arm of Myc signaling and thus allowing Myc to drive cellular proliferation unchecked (Teitz et al., 2000). This would be consistent with a role for caspase-8 upstream of Bax activation and downstream of Myc during apoptosis. In light of these facts, it will be important to determine the Myc apoptotic phenotype in caspase-8-deﬁcient cells. In summary, recent data are consistent with a model in which Myc derepresses at least one level of apoptotic control, removing a rate-limiting event in the regulation of apoptosis and rendering cells more suceptible to death in response to a diverse set of stimuli. Interestingly, Myc expression has been shown to increase in response to some apoptotic stimuli (Guenette et al., 1994; Strange et al., 1992). Whether or not this induction of myc is essential

100

Sara K. Oster et al.

to the apoptotic response, it suggests that regulation of the myc gene may also play a role downstream of some apoptotic signaling cascades and should be further investigated. By contrast, in some cells the absence of Myc is essential for apoptosis to ensue (Donjerkovic et al., 2000; Liu et al., 2000). Most notably, receptor-ligand-induced death in B-cell lymphomas is accompanied by the necessary down-regulation of Myc (Carey et al., 2000; Donjerkovic et al., 2000). An essential role for Myc in promoting cell survival of WEHI 231 cells in response to a variety of apoptotic stimuli has also been reported (Arsura et al., 2000; Wu et al., 1996). In most cells, however, Myc functions as a proapoptotic molecule, and the ability of Myc to sensitize cells to a wide array of apoptotic stimuli has been well documented (Prendergast, 1999). Moreover, as Myc has been implicated in both p53-dependent and -independent apoptotic signaling, it will be important to elucidate the precise role of Myc in regulating both of these processes. This is further discussed in the following section.

E. Cellular Transformation The ability of Myc to induce cellular transformation has been established using both animal models and tissue-culture systems. In the latter, the transformation potential of an ectopic Myc allele is measured by both focus formation and anchorage-independent growth of primary embryonic cells when coexpressed with a cooperating oncogene such the activated form of ras (Land et al., 1983; Ruley, 1990). By contrast, Myc alone is sufﬁcient to induce these transformed phenotypes in immortalized Rat1a cells that have incurred unknown genetic changes that can cooperate with Myc for transformation (Stone et al., 1987). Structure–function analysis has revealed the necessity of an intact MB2 region of Myc to promote cellular transformation, whereas deletion of MB1 has less impact on Myc’s ability to transform cells in culture (Conzen et al., 2000). Interestingly, the point mutation at conserved Myc phosphorylation site T58A had no signiﬁcant effect on Myc’s transforming potential, whereas T58I mutants were defective in Myc/Ras cotransformation assays in rat embryo ﬁbroblasts (Chang et al., 2000b). The steric impact of these substitutions with respect to Myc structure or protein interactions has yet to be assessed but may explain these apparent differences in the ability of Myc to promote transformation in the context of such mutations. Together, these studies highlight the importance of the Myc transcription factor during cellular transformation. Although genes regulated downstream of Myc that are important for transformation have not been fully elucidated, ectopic expression of known target genes such as odc, HMG-I/Y, and the combination of LDH-A and rcl is able to transform Rat1a cells and as such has been linked to Myc’s ability to transform cells in culture (Auvinen et al., 1992; Lewis et al., 2000; Moshier et al., 1993; Wood et al., 2000a).

The myc Oncogene

101

Studies in mouse models show that the strong oncogenic potential of Myc is sufﬁcient to initiate the transformation process upon spontaneous deregulation in vivo. An advantage to studying mouse models of transformation is the potential to identify relevant aspects of the oncogene-induced phenotype that may not be seen in tissue-culture systems. Examples of this would include the ability of an oncogene to dictate the angiogenic or metastatic potential of a tumor or to determine important tissue-speciﬁc constraints on individual tumor types initiated at different physiologically relevant loci. Previous studies have relied on mouse models in which the Myc gene is constitutively deregulated in transgenic animals (Jamerson et al., 2000; Morgenbesser and DePinho, 1994). More recent mouse models have taken advantage of inducible Myc systems to investigate the immediate effects of Myc deregulation in otherwise normal, fully developed adult tissues, thereby mimicking relevant tumor modalities (Pelengaris et al., 2000). Interestingly, ectopic activation of Myc in these animals resulted in dramatically different physiological outcomes depending on the nature of the tissue in which Myc was activated. These differences appear to be dictated by the balance between Myc-mediated proliferation and apoptosis. For example, ectopic activation of Myc in the skin’s dermal layer or in hematopoetic precursors was shown to induce hyperproliferation of undifferentiated cells. In the skin, these precancerous lesions remained mostly benign because of the inherent shedding mechanisms that prevented accumulation of these abnormal cells in the epidermal layers (Pelengaris et al., 1999). By contrast, hyperproliferating T-cells expressing the activated myc allele eventually gave rise to lymphomas (Blyth et al., 2000; Felsher and Bishop, 1999a). In both cases, deinduction of the myc transgene led to regression of the cancerous tissue, showing the dependence on Myc for maintenance of these malignancies. Moreover, malignant cells from these animals remained sensitive to apoptosis under low-serum conditions when cultured in vitro (Blyth et al., 2000). Therefore, the apoptotic arm of the Myc pathway can be suppressed in vivo, likely because of the wealth of survival factors provided by the surrounding tissue microenvironments. These ﬁndings are consistent with a similar study using a tetracycline-responsive myc allele expressed in the mammary epithelium of transgenic mice (D’Cruz et al., 2001). In this case, expression of the ectopic myc allele resulted in the formation of invasive mammary adenocarcinomas, most of which fully regressed upon myc deinduction. Notably, tumors bearing secondary mutations in ras did not regress, suggesting that once these tumors progressed beyond a certain point, reversion to a nontransformed state was no longer possible. In separate experiments, activation of Myc targeted to pancreatic β cells resulted in extensive apoptosis, quickly leading to complete ablation of islet cells and the subsequent onset of diabetes (Pelengaris et al., 2000). However, in cooperation with an antiapoptotic signal, activation of Myc resulted in β-cell hyperproliferation. These results suggest that the proliferative arm

102

Sara K. Oster et al.

of the Myc pathway is not necessarily predominant and thus is not able to initiate tumor formation in all tissues. Together, these studies further support a model wherein Myc is able to drive both growth and apoptosis, the balance of which is dictated by the cellular milieu. Where Myc-induced proliferation is sustainable, a single lesion in the myc oncogene alone can be sufﬁcient for tumor initiation in vivo, highlighting Myc’s importance during the transformation process. Alternatively, where Myc-induced apoptosis prevails, cooperating lesions to disarm this fail-safe mechanism are required in order for the tumor to progress. In a model well supported by experimental evidence, Myc acts in concert with other oncogenes that suppress apoptosis, thereby allowing Myc to drive cell proliferation in an unrestricted manner (Evan and Vousden, 2001; Hueber and Evan, 1998; Prendergast, 1999). Historically, oncogenes that collaborate with Myc in tumorigenesis have included bcl-2, ras, raf, pim-1, tgfα, and v-abl (Adams and Cory, 1992). Myc has also been shown to trigger apoptosis in a p53-dependent and -independent manner (Amanullah et al., 2000; Prendergast, 1999). As such, it has been shown that part of Myc’s ability to drive the p53-dependent apoptotic process depends on the induction of p19ARF. In cultured primary MEFs Myc activates the p19ARF–Mdm-2–p53 tumor suppressor pathway, enhancing p53-dependent apoptosis but ultimately selecting for surviving immortalized cells that have sustained either p53 or biallelic p19ARF deletion (Zindy et al., 1998). p53 and p19ARF also potentiate Myc-mediated apoptosis in primary B-cell cultures, and spontaneous inactivation of the p19ARF–p53 pathway occurs frequently in tumors arising in Eμ-c-myc transgenic mice (Eischen et al., 1999; Henriksson et al., 2001). It has been shown that Bmi-1, one of the polycomb genes, is a negative upstream regulator of the p19ARF pathway, and thus Bmi-1 can cooperate with Myc by inhibiting the p53-dependent apoptotic pathway (Jacobs et al., 1999). The inability of Myc to access the apoptotic pathway via p19ARF, in this case, allows for Myc to drive cell proliferation in an unrestricted manner, leading to tumor formation. By a similar mechanism, the T-box member TBX2 has been reported to contribute to breast cancer development by attenuating the induction of p19ARF by oncogenes such as Myc (Jacobs et al., 2000). Interestingly, DAP kinase, a Ca2+/calmodulin-regulated Ser/Thr kinase that carries a death domain, has been implicated in the modulation of p19ARF downstream of Myc (Raveh et al., 2001). DAP kinase has also been shown to induce apoptosis in a p53-dependent manner and to participate in the apoptotic responses triggered by IFN-γ , TNF, and CD95 ligand (Cohen et al., 1999). DAP kinase is up-regulated in response to hyperproliferation signals, and overexpression of this kinase has been shown to limit the transforming potential of Myc and Ras in vitro. In the absence of DAP kinase, Myc potentiation of apoptosis is compromised but remains evident, suggesting that Myc regulates the apoptotic response through multiple effector

The myc Oncogene

103

pathways. Indeed, a second Ca2+/calmodulin dependent kinase, Pnck, is highly overexpressed in a subtype of human breast cancers and in cell lines derived from mouse mammary tumors initiated by Myc and Fgf-3, but not those initiated by Neu or H-Ras (Gardner et al., 2000a,b). As such, the deregulated expression of this gene may have important and as yet unknown ramiﬁcations for the progression of breast tumors also harboring an activated myc allele. One of the most potent Myc cooperating oncogenes is bcl-2, which functions as a global inhibitor of apoptosis. Myc-mediated apoptosis triggered by serum withdrawal is also blocked by the IGF-1 survival pathway and Akt activation, which leads to deregulated Bcl-2 activity (Harrington et al., 1994; Kauffmann-Zeh et al., 1997). More recently, Twist has been identiﬁed as an important downstream mediator of the IGF-1-induced antiapoptotic effect. Twist and a related factor Dermo1, members of the bHLH family of transcription factors, were originally identiﬁed as suppressors of apoptosis in a functional screen designed to clone novel inhibitors of Myc-mediated apoptosis (Maestro et al., 1999). Mechanistically, Twist has been reported to reduce the expression of p19ARF and thus may indirectly mediate the activity of p53 during apoptosis (Dupont et al., 2001). Moreover, Twist is known to inhibit myogenic differentiation and is deregulated in 50% of rhabdomyosarcomas (Maestro et al., 1999). Together, these results provide evidence supporting a role for Twist as an oncogene that can abrogate the apoptotic function of p19ARF and thereby conspire with Myc in tumorigenesis. By a similar functional screening approach, we have identiﬁed a novel cullin that confers protection against Myc-mediated apoptosis (Penn Lab, unpublished data). The cullins are a family of genes that are components of the E3 ubiquitin protein ligases that mediate the critical step of substrate recognition. Like Twist and Dermo1, the overall potency of this novel cullin in inhibiting apoptosis is not as robust as that of Bcl-2. This may be a consequence of multiple, nonredundant mechanisms of apoptosis that can be exploited when the cell encounters a block in any one death cascade, whereas Bcl-2 appears to act at a downstream point common to almost all of these pathways (Adams and Cory, 2001). However, the protection afforded by these molecules may still be sufﬁcient to stay the apoptotic process long enough to select for further mutations that would allow for the expansion of a malignant clone.

F. Genomic Instability Myc has been ascribed a role in mediating the integrity of the cellular genome. Founding observations in this ﬁeld of research have noted that transient or constitutive Myc overexpression results in chromosomal and extrachromosomal gene ampliﬁcation and rearrangement (Mai et al., 1999;

104

Sara K. Oster et al.

Mai and Jalava, 1994; Taylor and Mai, 1998). Advances have been made in understanding the mechanism by which Myc is able to promote this genetic instability that has since been observed in both cultured cells and in vivo animal models. The current paradigm is that Myc uncouples DNA synthesis from mitosis (Li and Dang, 1999). It has been shown that cells overexpressing Myc can bypass check points upon spindle disruption in the G2/M phase of the cell cycle, thereby inducing endoreduplication and apoptosis. As such, cells that incur secondary mutations that abrogate the apoptotic response can survive with an abnormal DNA content. Not surprisingly, either loss of p53 or overexpression of Bcl-XL can cooperate with Myc to propagate polyploidy in diverse cell lines (Li and Dang, 1999; Minn et al., 1996; Yin et al., 1999). Interestingly, Myc induction of polyploidy does not necessarily result in chromosomal damage per se; rather, the ability of Myc to hasten G1 transit under mutagenic conditions may compromise genomic integrity (Felsher and Bishop, 1999b). Downstream effector molecules, which could enable Myc to bypass these check points, have been proposed, namely cdk2 and cyclins A and B1 (Adachi et al., 2001; Li and Dang, 1999; Yin et al., 2001). However, further studies are needed before any ﬁrm conclusions regarding the molecular mechanism of this Myc activity can be made. Moreover, there is evidence to suggest that Myc may operate to promote tumorigenesis by invoking genomic instability in a “hit-and-run” mechanism, as seen in the persistent tumorigenic phenotypes of Rat1a cells after transient activation of Myc (Felsher and Bishop, 1999b). Such a mechanism supports the notion that while most cells that have acquired deleterious genetic abnormalities would not survive, cells in which these genetic changes confer a growth advantage are selected for, leading to clonal expansion. As such, the loss of tumor suppressor activity may also cooperate with Myc to promote genetic instability during malignant transformation by circumventing these normal suicidal tendencies and allowing for more malignant clones to expand.

G. Angiogenesis To arrive at its full malignant potential a growing tumor must develop angiogenic ability (Bouck et al., 1996; Hanahan and Folkman, 1996; Hanahan and Weinberg, 2000). It has been argued that distinct molecular mechanisms operate to induce the angiogenic switch in different tumor types. Oncogenes such as ras, src, jun, and p53 regulate the balance of angiogenic activators and inhibitors in some systems (Hanahan and Weinberg, 2000). More recently, Myc has been implicated in the regulation of the angiogenic phenotype during tumorigenesis. Several groups have reported the down-regulation of antiangiogenic molecules upon overexpression of Myc. At least three novel

The myc Oncogene

105

endothelial cell growth inhibitors were reported that were down-regulated in conditioned media derived from N-Myc overexpressing neuroblastoma cell lines (Fotsis et al., 1999). One of these factors has since been identiﬁed as activin A, a factor that can inhibit angiogenesis in vivo (Breit et al., 2000). Moreover, activation of Myc in rat ﬁbroblasts has been associated with the down-regulation of thrombospondin-1 (Tsp-1), another negative regulator of angiogenesis (Janz et al., 2000; Ngo et al., 2000; Tikhonenko et al., 1996). Down-regulation occurs at the level of mRNA or protein destabilization, although the precise role of Myc in this mechanism remains unclear. It has also been suggested that Myc is sufﬁcient for activating the angiogenic switch in vivo. Activation of the inducible MycER chimera in tumors derived from matrigel-implanted immortalized Rat1a cells induced tumor neovascularization and suppression of Tsp-1 (Ngo et al., 2000). However, the occurrence of secondary lesions that could have enabled this switch was not ruled out by these experiments. A separate line of evidence supports a contrasting model wherein Myc’s oncogenic potential is limited through its repression of vascular endothelial growth factor (VEGF) expression. A correlation between Myc up-regulation and VEGF suppression is recurrent across diverse cell systems (Barr et al., 2000). Overexpression of Myc has also been shown to decrease tumor establishment of small-cell lung cancer cells in mice (Barr et al., 2000). These results may indicate that the down-regulation of VEGF concurrent with Mycinduced transformation limits the growth potential of such tumors in vivo. Alternatively, as was seen upon activation of Myc in β-pancreatic cells, tumor development may be limited in this case by the dominant apoptotic function of Myc in the lung tissue. Interestingly, papillomas induced upon ectopic activation of Myc in keratinocytes were characterized by an increase in VEGF secretion when passaged in vitro, and these cancerous lesions were highly vascularized in vivo (Pelengaris et al., 2000). Together, these data suggest that the role of Myc in triggering angiogenesis may be dictated by the trophic microenvironment of the cell. Indeed, angiogenesis is activated in the early to mid-stage of most tumors and is not speciﬁc to those in which Myc has been deregulated, indicating the multiple strategies employed by malignant cells to turn on this switch during the progression of cancer.

H. Summary and Perspectives Accumulating evidence strongly supports the hypothesis that Myc regulates numerous and diverse biological activities by functioning as a regulator of speciﬁc bona ﬁde target genes that then stimulate independent signaling cascades to effect a particular biological activity (Fig. 2). Historically, one of the ﬁrst indications that this model was valid stemmed from analysis of

106

Sara K. Oster et al.

collaborating oncogenes, such as Bcl-2. These lesions are often potent inhibitors of Myc-regulated apoptosis, and yet these molecules do not hinder Myc’s ability to drive tumor growth. This concept is also exempliﬁed by reports detailing Myc regulation of the D-type cyclins (Bouchard et al., 1999; Perez-Roger et al., 1999). In particular, cyclin D2 has been shown to be a direct target for Myc activation downstream of growth factor stimulation. These ﬁndings are important, not only in the quest to understand Myc’s role in the cell cycle progression, but indirectly to appreciate the role of Myc in apoptosis. The loss of either cyclin D1 or D2 alleles in primary ﬁbroblasts blocks Myc’s ability to drive proliferation and subsequent immortalization of these cells. By contrast, deletion of these genes does not affect Myc’s ability to induce apoptosis in low serum conditions. Indeed, evidence to further suggest that Myc’s biological activities can be uncoupled through speciﬁc mutation of N-terminal phosphorylation sites has emerged (Chang et al., 2000b). Experiments to address whether these regions are important for known or novel protein interactions and the downstream consequences of these interactions on gene regulation will be of seminal importance and provide further insight into the mechanism(s) by which Myc is able to regulate these distinct biological activities. It remains unclear whether Myc is constitutively regulating all downstream effector pathways or whether each pathway only becomes fully activated upon signaling from a second stimulus. For example, Myc alone is not sufﬁcient to induce apoptosis, but rather a combination of both Myc expression and a growth inhibitory signal is required to trigger cell death. Initiation of apoptotic signaling cascades by diverse stimuli may have an impact on the Myc protein itself and/or alter Myc’s genetic program to elicit a full apoptotic response. This cooperativity in signaling is not limited to apoptosis. Myc and p53 collaborate to drive cyclin B1 transcription, which can play a role in regulating genomic instability (Yin et al., 2001). Moreover, the combination of Myc and Ras can potentiate Myc induction of E2F transcription (Sears et al., 1997) to drive cell-cycle progression. Indeed, Myc expression is tightly linked to the extracellular milieu, so it would not be too surprising to learn that Myc function may also be inﬂuenced by the cellular environment. The incidence of deregulated Myc expression is high in many tumor types, and Myc’s role as an oncogene has been well established. In a review article, Hanahan and Weinberg describe six hallmarks of cancer that are acquired by malignant cells during the process of tumorigenesis: self-sufﬁciency in growth signals, insensitivity to antigrowth signals, evasion of apoptosis, limitless replicative potential, sustained angiogenesis, and tissue invasion and metastasis (Hanahan and Weinberg, 2000). It has been surmised that mutations in certain key regulatory genes, such as myc, can promote several of these capabilities concurrently, fast-tracking the cell toward a cancerous state. As described, cells harboring a deregulated myc allele possess a strong

The myc Oncogene

107

proliferative potential by virtue of Myc’s ability to promote both cell cycle progression and cell growth and to override growth inhibitory controls. The ability of Myc to promote cell proliferation, however, is undermined by its inherent ability to promote the apoptotic process. Nevertheless, Myc can be seen to cooperate with other oncogenic mutations that inhibit the apoptotic arm of Myc function, thereby allowing tumor progression to continue unchecked. Further evidence that Myc deregulation plays a key role during tumorigenesis comes from recent reports that Myc may be involved in eliciting the angiogenic phenotype in tumors and can promote the destabilization of the cellular genome. Thus, the deregulated expression of Myc can signiﬁcantly contribute to the aggressive onset of further mutation and tumor progression. Together, these recent advances have led to new paradigms for Myc as an initiator, a cooperator, and an ampliﬁer of malignant transformation.

IV. FUNCTIONS OF MYC AS A REGULATOR OF TRANSCRIPTION As previously discussed, Myc is thought to mediate its biological activities by regulating the expression of speciﬁc target genes. A number of genes have been identiﬁed as potential Myc-regulated targets that are involved in a wide variety of cellular processes (For activated and repressed genes see Table I and Table II, respectively; see also Dang, 1999; Facchini and Penn, 1998; Grandori et al., 2000; Peters and Taparowsky, 1998). Strikingly, the change in gene expression as a result of regulation by Myc is often only three- to ﬁvefold, and no single target gene is able to fully reconstitute Myc’s ability to drive cellular proliferation (Berns et al., 2000a; Nikiforov et al., 2000). Indeed, it is likely that Myc alters the expression of multiple target genes to induce any given activity within the cell. This is demonstrated by Myc’s role in cell-cycle progression. Myc has been shown to be an integral component in the G0 to S phase transition (Amati et al., 1998; Bouchard et al., 1998; Obaya et al., 1999), and many target genes have been identiﬁed to date that can play a role at this critical phase of the cell cycle. This includes the up-regulation of cdc25A, cyclin D2, odc, cad, and Id2 and the repression of p27 Kip1, p21Cip1, p15Ink4b, gadd45, and gas1 (Bello-Fernandez et al., 1993; Bouchard et al., 1999; Galaktionov et al., 1996; Gartel et al., 2001; Lasorella et al., 2000; Lee et al., 1997; Marhin et al., 1997; Miltenberger et al., 1995; Perez-Roger et al., 1999; Staller et al., 2001; Yang et al., 2001b). It is likely that Myc regulates multiple subsets of genes under certain cellular conditions to elicit speciﬁc genetic programs. As Myc activities are global, regulation of true target genes is likely to be conserved in all cell types. However, changes in gene expression downstream of these primary targets

Table I Putative Myc Activated Genes with Supportive Experimental Evidence Strongly supportive evidence Activated gene target

Function

AdMLP

Adenoviral major late promoter

bax

Apoptosis

BN51

RNA pol III co-factor

cad

Regulated Not Myc in response regulated regulated to MycER in Myc in primary activation null cells cells

Myc present at promoter √

Suggestive evidence De novo Promoter– Myc protein Timely Cell line reporter Nuclear associated synthesis kinetics of survey constructs run-on expression unnecessary regulation completed assayed accomplished assessed √

in vitro G

√ in vitro G √ in vivo D √

√

√

√ √

n.i

Mitchell et al., 2000 √

Greasley et al., 2000

Q √

Pyrimidine biosynthesis

Q&A

√

√

√

in vivo C x in vitro G

Miltenberger et al., 1995 Boyd et al., 1998 Bush et al., 1998 Eberhardy et al., 2000

108 cdc2 cdc25a

√

Cell cycle Cell cycle phosphatase

cdk4

Cyclin dependent kinase

cyclin A

Cell cycle

√ Q&A

x Q&A

x Q √ A

√

√

in vitro G

in vitro G

√

√

√ √

√

References Li et al., 1994

Born et al., 1994

√

Galaktionov et al., 1996 Perez-Roger et al., 1997 Ben-Yosef et al., 1998 Bush et al., 1998

√

Hermeking et al., 2000 x

√

Jansen-Durr et al., 1993 Domashenko et al., 1997 Perez-Roger et al., 1997

cyclinD2

Cell cycle

√

√

√

√

√

√

Bouchard et al., 1999 Perez-Roger et al., 1999

Q

cyclinE

Cell cycle

√

√

√

Q&A

109

E2F-1

Cell cycle and apoptosis

E2F-2

Cell cycle and apoptosis

E2F-3a

Cell cycle and apoptosis

ECA39/

Branched-chain amino acid

BCAT1

aminotransferase

eIF-2α

Protein synthesis

√

√

x

Leone et al., 1997 Sears et al., 1997

√ √ √

x

in vitro G

in vitro G

√

√

√

√

Sears et al., 1997

Adams et al., 2000 √

Q&A

√

Protein synthesis

√ Q

Fas Ligand/ CD95 L

Apoptosis

√ A

x Q&A

Benvenisty et al., 1992 Ben-Yosef et al., 1996 Ben-Yosef et al., 1998 Bush et al., 1998

√

√

√

√

in vitro G

√

√

√

√

√

√

Rosenwald et al., 1993

Q eIF4E

Jansen-Durr et al., 1993 Domashenko et al., 1997 Perez-Roger et al., 1997

√

in vitro G

in vitro G

√

√

Rosenwald et al., 1993 Jones et al., 1996 Bush et al., 1998 √

Genestier et al., 1999 Brunner et al., 2000 Kasibhatla et al., 2000

√ , result is indicative of a Myc regulated gene; x, result does not indicate a Myc regulated gene; Q, expression was compared between quiescent and serum stimulated cells; A, expression was assessed in asynchronous cells; G, in vitro gel shift; D, in vivo DNA footprint analysis; C, in vivo ChIP analysis; n.i., not interpretable; see text for experimental details.

Table I

(continued ) Strongly supportive evidence

Activated gene target HMG-I/Y

Function Architectural transcription factor

hsp70

Heat shock protein

hTERT

Telomere maintenance

110

id2

IRP-2

LDH-A

MrDb

Proliferation and differentiation Iron regulatory protein Anaerobic glycolysis RNA helicase

Regulated Not Myc in response regulated regulated to MycER in Myc in primary activation null cells cells √ Q

√

√

odc

Nucleolar phospho-protein Polyamine synthesis

in vitro G

√

√

√

√

in vitro G

√ in vitro G √ in vivo C

A

√

in vivo C

√

√

√

References Wood et al., 2000a

√

√

√

√

√

√

√

√

Kingston et al., 1984 Kaddurah-Daouk et al., 1987 Taira et al., 1992 Ahn et al., 1998 Greenberg et al., 1999 Oh et al., 1999 Xu et al., 2001 Lasorella et al., 2000

Q √

√

Wu et al., 1999

Q √ A √ Q

nucleolin

De novo Promoter– Myc Myc protein Timely Cell line reporter Nuclear associated present at synthesis kinetics of survey constructs run-on expression promoter unnecessary regulation completed assayed accomplished assessed √

x Q

Suggestive evidence

x Q&A

√

in vitro G

x Q&A

√

√

√

√

√

√

√

Shim et al., 1997 Bush et al., 1998 √

√

n.i

√

Grandori et al., 1996 Bush et al., 1998 Greasley et al., 2000

Q √ Q

x Q&A

√ in vitro G √ in vivo C

√

√

√

√

√

Bello-Fernandez et al., 1993 Wagner et al., 1993 Domashenko et al., 1997 Ben-Yosef et al., 1998 Bush et al., 1998 Eberhardy et al., 2000

P48/ISGF3γ

Component of interferon regulated txn factor ISGF3

p53

Tumor suppressor

Prothymosin α

Cellular proliferation

√

√

√

in vitro G

x Q&A

√ Q&A

x Q&A

√

√

in vitro G

√ in vitro G √ in vivo D

√

√

√

√

√

√

√

√

√

√

111 rcc1

rcl1

Cellular proliferation Growth and transformation

Thymidine kinase

DNA metabolism enzyme

tmp

Tumor-associated membrane glycoprotein

Weihua et al., 1997

Q

√ A

x Q&A

√

in vitro G

√ A √ Q&A √

√

in vitro G

√

Eilers et al., 1991 Gaubatz et al., 1994 Domashenko et al., 1997 Ben-Yosef et al., 1998 Bush et al., 1998 Walhout et al., 1998 Tsuneoka et al., 1997 Bush et al., 1998 Lewis et al., 1997

x in vitro G √

√

Reisman et al., 1993 Roy et al., 1994 Ben-Yosef et al., 1998 Bush et al., 1998

Pusch et al., 1997 n.i.

√

√

√

Ben-Porath et al., 1999

Q

√ , result is indicative of a Myc regulated gene; x, result does not indicate a Myc regulated gene; Q, expression was compared between quiescent and serum stimulated cells; A, expression was assessed in asynchronous cells; G, in vitro gel shift; D, in vivo DNA footprint analysis; C, in vivo ChIP analysis; n.i., not interpretable; see text for experimental details.

Table II Putative Myc Repressed Genes with Supportive Experimental Evidence Strongly supportive evidence Repressed gene target

Function

112

α3β1 integrin

Adhesion

λ5

Expressed in pre-B cells

activin A

Angiogenesis inhibitor

AdMLP

Adenoviral major late promoter

Adrenomedullin

Vasodilator

Suggestive evidence

Regulated Not Myc De novo Promoter– Myc in response regulated regulated Myc protein Timely Cell line reporter Nuclear associated to MycER in Myc in primary present at synthesis kinetics of survey constructs run-on expression activation null cells cells promoter unnecessary regulation completed assayed accomplished assessed √

√

Judware and Culp, 1995 Judware and Culp, 1997 √

in vitro G

√

Mai and Martensson, 1995

√

Breit et al., 2000

√ √

References

√

√

√

√

Li et al., 1994 Wang et al., 1999

A c/ebp-α

Transcription factor

caveolin-1

Scaffolding protein, component of caveolae

√

√

Q √

x

√

√

√

√

Li et al., 1994 Antonson et al., 1995 Park et al., 2001 Timme et al., 2000

Q c-myc

Oncogene

√

√

Q

A

√

n.i.

√

√

√

√

Penn et al., 1990 Facchini et al., 1997 Oster et al., 2000

Collagen genes

Extracellular matrix proteins

cyclinD1

Cell cycle

gadd153

Growth arrest and DNA damage inducible gene

gadd45

Growth arrest and DNA damage inducible

gas1

H-ferritin

Membrane associated protein, blocks S-phase entry Iron chelator

113

Immunoglobulin κ

Immune response

LFA-1 complex (αL RNA)

Lymphocyte function associated antigen-1

MHC/HLA class 1

Cell surface antigens

MT-1

Metallothionein I

Ndr1/NDRG1

Cell growth and differentiation

neu

Growth factor receptor tyrosine kinase

√

√

√

x Q √

√

√

√

Philipp et al., 1994

√

√

√

√

√

Yang et al., 1991

Chen et al., 1996 Amundson et al., 1998 √

Marhin et al., 1997

Q √

√

√

√

Lee et al., 1997

Q

√

√

Q

A

√

√

√

Wu et al., 1999

√ √

√

√

Sigvardsson et al., 1994 √

Inghirami et al., 1990

√

Versteeg et al., 1988, Lenardo,1989

√ √ √

√

in vitro G

in vitro G

√

√

Kaddurah-Daouk et al., 1987 √

Shimono et al., 1999 Suen and Hung, 1991

√ , result is indicative of a Myc regulated gene; x, result does not indicate a Myc regulated gene; Q, expression was compared between quiescent and serum stimulated cells; A, expression was assessed in asynchronous cells; G, in vitro gel shift; D, in vivo DNA footprint analysis; C, in vivo ChIP analysis; n.i., not interpretable; see text for experimental detail.

Table II (continued ) Strongly supportive evidence Repressed gene target

Function

Neural cell adhesion molecule

Cell–cell adhesion

p15Ink4b

Cell cycle

p21Cip1

Cell cycle

Suggestive evidence

Regulated Not Myc De novo Promoter– Myc in response regulated regulated Myc protein Timely Cell line reporter Nuclear associated to MycER in Myc in primary present at synthesis kinetics of survey constructs run-on expression activation null cells cells promoter unnecessary regulation completed assayed accomplished assessed √

√

√

A

A

√

√ in vitro G √ in vivo C

√

√ √

√

√

Staller et al., 2001

√

Claassen and Hann, 2000 Gartel et al., 2001

A p27Kip1 pdgfβ receptor

√

Cell cycle Growth factor receptor

TdT

Terminal deoxynucleotidyl transferase

vegf

Vascular growth factor

√

√

Q&A

Q&A

√

√

√

√

in vitro G in vivo C

n.i.

√

√

√

√

√

Yang et al., 2001 Oster et al., 2000 Izumi et al., 2001 Penn lab, unpublished data

√

in vitro G

√

References Akeson and Bernards, 1990

Mai and Martensson, 1995 √

Barr et al., 2000

A

√ , result is indicative of a Myc regulated gene; x, result does not indicate a Myc regulated gene; Q, expression was compared between quiescent and serum stimulated cells; A, expression was assessed in asynchronous cells; G, in vitro gel shift; D, in vivo DNA footprint analysis; C, in vivo ChIP analysis; n.i., not interpretable; see text for experimental detail.

The myc Oncogene

115

may vary in different tissues because of the nature of the cell and/or the microenvironment. Thus, to truly understand the role of Myc in cellular physiology it is essential to identify universal bona ﬁde target genes. By investigating the functions of these true Myc targets it will be possible to determine how Myc can induce such diverse biological outcomes in a wide variety of cell types. These targets will also provide a basis to mechanistically dissect how Myc functions as a transcriptional regulator.

A. Identiﬁcation of Target Genes To identify Myc-regulated genes, several groups have taken advantage of array technologies (Boon et al., 2001; Coller et al., 2000; Guo et al., 2000; Kim et al., 2001; Neiman et al., 2001; Nesbit et al., 2000; O’Hagan et al., 2000; Schuhmacher et al., 2001). One of the primary goals in adopting this powerful approach is to identify groups of coregulated genes that function coordinately to regulate a common biological activity. This outcome would provide strong preliminary evidence that Myc is an important upstream regulator of the particular cell function revealed by groups of regulated targets. Interestingly, the data collected from array studies to date have resulted in very few overlapping Myc-regulated targets, with the exception of numerous ribosomal genes. The identiﬁcation of such diverse groups of genes without signiﬁcant overlap may be due to the different cell systems and arrays analyzed. However, it is also possible that genes regulated as a downstream consequence of Myc activity as well as the Myc-speciﬁc targets are being captured by the array approach, obscuring the identity of bona ﬁde, direct Myc targets. Common patterns of coordinately regulated genes between the different array experiments may only become apparent with the conﬁrmation of bona ﬁde Myc targets. To date, steps to determine whether the regulation of genes identiﬁed in the various array experiments are indeed Myc-speciﬁc have not yet been conducted. To this end, a discussion of the experimental procedures that help to deﬁne bona ﬁde Myc target genes is outlined in the following section.

B. Deﬁning Bona Fide Myc Target Genes One approach to delineate the unique contribution of Myc to cellular physiology is to identify the select subset of genes that are dependent upon Myc as a regulator of gene transcription. We deﬁne a bona ﬁde Myc-regulated gene as one that is regulated by Myc function alone and not as a downstream consequence of Myc activity, such as cell-cycle progression. Clearly, genes indirectly regulated downstream of Myc activity have critical roles in Myc function within the cellular context. However, to understand Myc’s global

116

Sara K. Oster et al.

role in cell physiology, we will focus on the experimental evidence that, in combination, distinguishes direct, bona ﬁde targets of Myc from indirect, downstream-regulated genes. This can be accomplished using a repertoire of experimental tools developed to assess gene regulation, and evidence can be grouped by those results that strongly support the identity of a bona ﬁde target and those that lend additional suggestive evidence that the gene is regulated by Myc. The experimental evidence that supports each putative Myc target gene is provided in Table I and Table II.

1. STRONGLY SUPPORTIVE EVIDENCE There are many experimental systems in use that provide evidence strongly supporting the notion that a gene is regulated by Myc. One key piece of evidence is regulation of the target gene downstream of an inducible Myc allele, the most common of which is the MycER system (Eilers et al., 1989; Littlewood et al., 1995). This system is particularly valuable in identifying genes regulated by Myc activation in the absence of a full mitogenic response in quiescent cells. In addition, target gene regulation upon induction of the MycER protein in asynchronous cells ensures that regulation is independent of cell cycle position. Gene regulation upon activation of MycER in the presence of translational inhibitors such as cycloheximide (CHX) provides further support for the notion of direct target regulation by Myc. We expect that a direct target would be regulated under these conditions, while indirectly regulated genes that require de novo protein synthesis would not. A problem with this approach is that CHX alone often affects gene expression, making these experiments uninterpretable. A third issue is whether a target gene is regulated in Myc null cells (Mateyak et al., 1997). The expression of a Myc target gene mRNA will remain unchanged over the duration of a complete cell cycle in synchronized Myc null cells if regulation is dependent on Myc function. Also, assessing the differences in basal levels of target gene expression between asynchronously growing Myc null cells and Myc reconstituted cells can show that the gene is misregulated in the absence of Myc. An important caveat when working with these cells is that genetic alterations may have occurred during the knockout procedure that allow the cells to continue cycling in the absence of Myc and may mask the requirement for Myc in the regulation of some targets but not all. Indeed, this may also be true of many immortalized cell lines; thus it is essential to determine that a target gene is regulated by Myc in primary cells. Lastly, identiﬁcation of the Myc protein at the target promoter can help to establish Myc regulation of a particular gene. Previously, such results were obtained using in vitro gel shifts, which can be difﬁcult to interpret and in which false positives are common. More recently, the powerful chromatin immunoprecipitation (ChIP) technique has been employed to show that Myc is present at the endogenous regulatory region of a putative target gene in vivo. Using

The myc Oncogene

117

these experimental systems, it has been is possible to identify targets that are dependent on Myc for their regulation.

2. SUGGESTIVE EVIDENCE Additional experimental evidence can also suggest the identity of Mycregulated genes and can be used in combination with the approaches listed previously. To assess the likelihood of Myc regulation of a putative target, the kinetics of regulation can be investigated. The up-regulation of a Myc activated gene would be expected to occur very shortly after the induction of Myc expression, whereas down-regulation of a repressed gene would vary depending on the half-life of the mRNA for that target. Showing that the endogenous target is regulated in Myc-ampliﬁed tumor cell lines or in immortalized cell lines overexpressing exogenous Myc can suggest that Myc plays a role in its regulation. In addition, promoter–reporter constructs can be used to investigate the level of regulation of a target promoter in response to Myc as well as the region(s) essential for response. However, the regulation of the target in the context of endogenous chromatin architecture cannot be adequately assessed by this method. Nuclear run-on experiments can determine if target gene regulation is occurring at the transcriptional level and thus show differences in mRNA stability compared to transcriptional regulation. Finally, the expression of the target gene can be compared with Myc levels in various environments where Myc expression is well deﬁned, such as in differentiating cell systems, embryonic development, tissues from transgenic mouse models, and cell systems in which antisense Myc is employed. An overview of genes reportedly regulated by Myc along with supportive and suggestive experimental evidence is provided (Table I and II). These tables are offered as a reference guide to the experimental procedures currently being implemented in the characterization of bona ﬁde Myc targets and the evidence used to support the identity of individual regulated genes. Given the magnitude of this exercise, we apologize in advance for omissions. As new and more sensitive technologies are developed, our ability to segregate bona ﬁde Myc targets will continue to improve. To identify novel Myc-regulated genes, we have undertaken a microarray analysis of gene expression in response to Myc and have identiﬁed 53 putative targets by their differential regulation in cells overexpressing Myc. Using the aforementioned techniques, we were able to discount 36 of 53 possible targets as being indirectly regulated by Myc function. This included the many ribosomal genes that were regulated as a result of cell-cycle progression, and not by Myc per se (Penn laboratory, in preparation). This trend will likely hold true for many of the ribosomal genes that were identiﬁed in other studies. Moreover, many of the bona ﬁde target genes identiﬁed in our microarray analysis were identiﬁed as differentially regulated in other screens as well, highlighting the idea that bona ﬁde Myc targets are universally regulated.

118

Sara K. Oster et al.

Indeed, each of the genes listed as differentially expressed in those cell systems must be analyzed further to determine whether it is truly Myc-regulated. Once this has been accomplished, we will be better able to assess the functional consequences of regulation of these genes with respect to Myc activities and we will undoubtedly learn more about Myc function and mechanism(s) of action.

C. Essential Regions Myc can both activate and repress transcription of multiple target genes. The Myc protein has two main functional domains that play a role in gene regulation: an NTD, including the highly conserved MB1 and MB2 regions, and a CTD that contains a bHLHLZ motif, which is required for all known functions of the Myc protein. It has been shown that Myc heterodimerizes with its partner protein Max at the bHLHLZ to allow for binding at speciﬁc DNA elements, E-boxes (Luscher and Larsson, 1999). This interaction is necessary for induction of apoptosis, cell cycle progression and transformation (Amati et al., 1993a,b). Myc/Max association is also required for activation of gene expression (Amati et al., 1992) as well as for Myc autosuppression (Facchini et al., 1997). Although it is clear that the CTD is required for Myc’s ability to regulate gene expression, the required regions in the NTD are less well deﬁned. Analysis of deletion and point mutations within the NTD has indicated that Myc repression and activation can be uncoupled. These results suggest that the region surrounding and including MB2 is essential for repression, while small deletions spanning the entire NTD have differential effects on transactivation potential, suggesting that no one region within the NTD regulates all activation (Brough et al., 1995; Chang et al., 2000b; Conzen et al., 2000; Kato et al., 1990; Lee et al., 1996). Indeed, our own results indicate that small deletions and point mutations within the NTD do not greatly affect activation of the target gene cad, but others suggest that mutations near MB1 abrogate activation of odc (Conzen et al., 2000; Penn laboratory, in preparation). By contrast, small deletions surrounding MB2 have a signiﬁcant effect on Myc’s ability to differentially repress target genes (Penn laboratory, in preparation). The regions involved in Myc-mediated activation or repression are likely involved in protein–protein interactions that are necessary for these functions.

D. Mechanisms of Activation The expression of most genes in eukaryotic cells is mediated by RNA polymerase II (polII), but a number of regulated steps are required for transcription to occur. DNA in the cell is usually tightly wound around histones,

The myc Oncogene

119

and chromatin remodeling is required to allow transcription factors to access the promoter regions of genes. A preinitiation complex then forms at the promoter, providing a single-stranded template for the polymerase by melting the DNA, which then allows polII to clear the promoter and begin transcribing the gene. Each of these events is highly regulated, and the protein complex at the promoter is constantly modiﬁed by the addition or removal of particular subunits to allow each speciﬁc process to occur. Recent advances support the notion that Myc is likely involved in many distinct points of gene regulation to activate its targets. Speciﬁcally, evidence suggests that activation of target genes by Myc involves at least two regulatory steps, namely chromatin remodeling and promoter clearance (Fig. 3). Indeed, Myc has been implicated as playing a role in two different forms of chromatin remodeling: histone acetylation and ATP-dependent remodeling.

1. HISTONE ACETYLATION It has been suggested that Myc can contribute to chromatin remodeling at activated targets to allow for an increase in transcription. Increased histone acetylation is often found at genes that are actively transcribed and is believed to allow the preinitiation complex to access these active promoters (Gregory et al., 2001). Myc was ﬁrst implicated in this process when it was shown to interact with TRRAP in an MB2-dependent manner (McMahon et al., 1998). The yeast homolog of TRRAP, Tra1p, is a component of a histone acetyltransferase (HAT) complex, SAGA (Grant et al., 1998). This complex contains a number of proteins, including the HAT GCN5. The human homolog, hGCN5, has been characterized and shown to interact in a complex with Myc through TRRAP (McMahon et al., 2000). Fusion of hGCN5 HAT to a transformation-defective MB2 deletion mutant of Myc can partially rescue the ability of the Myc mutant to induce transformation, strongly suggesting that recruitment of HAT activity is an important component of Myc function. In addition, Myc has been shown to interact with other molecules whose functions have been implicated in chromatin remodeling, namely TIP48 and TIP49 (Wood et al., 2000b). Interaction of these molecules with Myc has been shown to require the Myc NTD and occurs independently of TRRAP binding. TIP48 and 49 have ATP-hydrolyzing activity as well as suspected helicase activity and have been shown to be required for focus formation by Myc and Ras in a primary cotransformation assay (Wood et al., 2000b). Intriguingly, a HAT complex termed NuA4 has been identiﬁed (Allard et al., 1999) that contains the yeast homologs to TIP48, TIP49, TRRAP, and the HAT TIP60 (Ikura et al., 2000). It will be interesting to determine whether Myc is involved in recruiting various HAT complexes to activate transcription at different target genes.

120

Sara K. Oster et al.

Fig. 3 Mechanisms of transcriptional activation by Myc. Myc may activate expression of target genes by at least three mechanisms: (A) recruiting histone acetyltransferase (HAT) complexes to the promoter, (B) recruiting ATP-dependent chromatin remodeling complexes to the promoter, and (C) inducing promoter clearance of the RNA polymerase (polII) at the target promoter (see text for further detail).

The histone acetylation hypothesis is also supported by an analysis of a series of promoters by ChIP in which Myc binding to regulatory regions has been associated with an increase of histone H4 acetylation, but not histone H3 acetylation (Frank et al., 2001). Interestingly, binding of Myc and acetylation of these promoters was not always sufﬁcient to induce transcription, indicating that other points of regulation may also be integral to the regulation of a subset of Myc target genes. Additionally, Myc binding to the promoter of cyclin D2 resulted in an increase of both H3 and H4 acetylation. This increase in histone acetylation was dependent on an interaction

The myc Oncogene

121

between Myc and TRRAP, indicating that the acetylation complex includes the TRRAP protein (Bouchard et al., 2001). Indeed, these studies suggest that Myc mediates the activation of some targets by recruiting HAT activity to the regulatory regions of these genes. The exact complexes and their protein components that are recruited to each promoter remain to be determined. It has been proposed that Myc activation of gene expression can be regulated by competition for Max with another family of interacting proteins, the Mad family (Grandori et al., 2000). These proteins are mainly repressors of transcription and they do so by interacting with Sin3a and Sin3b, resulting in recruitment of histone deacetylases (HDACs) to the DNA, which leads to an inhibition of transcription. Thus, the hypothesis is that Myc activates gene expression by competitively binding Max to recruit HATs while the Mad family members bind to Max and recruit HDACs to repress gene expression. This is supported by an analysis of histone acetylation and deacetylation upon differentiation of HL60 cells at two Myc-activated promoters, hTERT and cyclin D2 (Xu et al., 2001; Bouchard et al., 2001). ChIP analysis has shown that acetylation of histones was increased at these promoters upon Myc binding in nondifferentiated cells. Differentiation of the HL60 cells led to decreased Myc binding and reduced acetylation of these promoters, which corresponded with an increase in Mad protein binding. However, evidence suggests that regulation of some targets may not be as clear-cut as this model would imply (Baudino and Cleveland, 2001). This may be due to the differences in the activation mechanism of speciﬁc targets or factors speciﬁc to each cellular environment.

2. ATP-DEPENDENT CHROMATIN REMODELING A second mechanism of chromatin remodeling that Myc may be involved in is ATP-dependent remodeling. ATP-dependent reorganization of the DNA exposes naked DNA by inducing nucleosomes to slide along the DNA (Whitehouse et al., 1999). INI1/hSNF5, which has been shown to interact at the Myc CTD (Cheng et al., 1999), is a key component of the SWI/SNF complex that remodels chromatin in an ATP-dependent manner. It was shown that functional components of the SWI/SNF complex are necessary for Myc-mediated activation of a luciferase reporter under the control of a promoter containing multiple E-boxes. It has also been suggested that competitive interference of INI1/hSNF5 binding to Myc through interaction with another CTD-binding protein, ORC1, leads to reduced activation of transcription (Takayama et al., 2000b), further implicating this as a possible mode of regulation of gene activation. More recently, Myc has been seen to interact with BAF53, which is an integral component of the SWI/SNF complex (Park et al., 2001b). Although the precise functional consequence of this interaction has not yet been

122

Sara K. Oster et al.

determined, it is possible that this interaction may mediate ATP-dependent chromatin remodeling.

3. PROMOTER CLEARANCE A study of an activated target, the cad gene, has indicated that histone acetylation was not signiﬁcantly increased at the promoter upon Myc activation (Eberhardy et al., 2000). Further analysis by ChIP indicated that the polII complex was present at the promoter before activation by Myc (Eberhardy and Farnham, 2001). In fact, activation of the cad promoter by Myc seemed to correlate with phosphorylation of the CTD of polII, indicating that Myc may be involved in promoter clearance and elongation at this target gene. This was complemented by in vitro evidence that Myc can interact with cdk9 and cyclin T1, components of the polII CTD kinase P-TEFb, and activator bypass experiments that suggest that direct recruitment of cyclin T1 to the promoter can substitute for Myc function in activating the cad gene. These results suggest a third possible mechanism in which Myc is able to activate the expression of target genes. Together, these studies suggest that Myc’s role in target gene activation involves multiple points in the regulation of transcription, which may depend on the cellular environment and/or the proteins recruited to speciﬁc target genes. It is of interest to identify regions of the Myc NTD that are involved in activation of speciﬁc target genes and determine the protein interactions that occur in these regions to aid in the understanding of the multiple mechanisms of Myc activation. Assessing the mechanisms of regulation at speciﬁc target genes, especially those involved in the transforming phenotype of the Myc protein, is of critical importance. In this way it may be possible to target these speciﬁc mechanisms to inhibit tumor progression by Myc.

E. Mechanisms of Repression The mechanisms of Myc repression of gene expression have been less well deﬁned than the mechanisms of activation. However, evidence supporting the notion that repression of gene expression is related to the transforming ability of Myc has spurred a ﬂurry of research in this area (Claassen and Hann, 1999; Facchini and Penn, 1998). It has long been debated whether Myc repression is an indirect consequence of Myc activation or a separate function of the Myc protein. Analysis of Myc mutants has provided indirect evidence that repression and activation can be uncoupled, suggesting that they are indeed separate functions. More recently, Myc was determined to be present at the promoters of repressed targets (Staller et al., 2001; Penn laboratory, in preparation), proving that Myc does indeed function directly in the down-regulation of these genes. In fact, certain deletion mutants of Myc

The myc Oncogene

123

are able to repress speciﬁc target genes, but not others, while still maintaining activation capability, indicating that there may be more than one mechanism of repression (Oster et al., 2000; Penn laboratory, in preparation). It is likely that these different mechanisms of repression depend on the architecture of the target promoter and the transcription factors and enhancers that inﬂuence expression of these genes. Myc’s partner protein Max may also play a role in Myc repression of gene expression. Evidence to date suggests that the mechanism of Myc down-regulation of many targets that contain initiator elements (Inr) is through this regulatory site. However, Myc also represses genes that lack an Inr; thus, there are likely multiple mechanisms by which Myc mediates repression (Fig. 4). The role of Max and the mechanism(s) of Myc repression will be discussed in the following sections.

Fig. 4 Mechanisms of transcriptional repression by Myc. Myc may repress the expression of target genes by two related mechanisms that depend on the architecture of the target promoter. (A) Inr-dependent repression involves Myc interacting with Inr-binding transcription factors to inhibit target gene activation. (B) Inr-independent repression involves Myc interacting with transcription factors at enhancer elements to inhibit the target gene activation. The role of Max and the characteristics of the complex of proteins that interact with the transcription factors are as yet not determined (see text for further detail).

124

Sara K. Oster et al.

1. THE ROLE OF MAX The Myc CTD is essential for repression of target genes but the importance of Max binding has been a relatively unstudied event. Indeed, Max seems to be important for Myc’s biological activities, and it is hypothesized that repression as well as activation of gene expression mediates these activities; thus it is not surprising that interaction with Max could be essential for repression. We have previously published that Myc/Max binding is required for Myc autosuppression by analysis of Myc and Max mutants that bind speciﬁcally with one another but are unable to interact with their endogenous counterparts (Facchini et al., 1997). Whether Myc/Max heterodimerization is required for repression of other targets has not been speciﬁcally addressed. However, more recent experiments have suggested that Max may be necessary, as its presence has been detected at repressed promoters along with Myc by ChIP analysis (Staller et al., 2001; Penn laboratory, unpublished data). Additionally, in transient transfection experiments Max appeared to enhance Myc repression of p27 (Yang et al., 2001b). As more evidence accumulates, it will likely become apparent that Max is integral to the repression function of Myc. Myc/Max heterodimers are known to bind E-box elements, and notably, Myc-repressed genes identiﬁed to date have lacked E-boxes in their regulatory regions. Thus, Myc and Max must be acting by some alternate mechanism to down-regulate these targets. Therefore, the role of Max in gene repression remains to be determined. However, we can speculate that heterodimerization with Max may stabilize the Myc protein, or perhaps that Max acts to recruit Myc to speciﬁc target genes.

2. INR DEPENDENT REPRESSION Many of the genes that can be repressed by Myc contain Inr elements within their promoters. It has been shown that numerous transcription factors can bind Inr sequences, which are found at or near the start site of transcription. It was originally shown that Myc could repress the adenoviral major late promoter (AdMLP) by a mechanism that was dependent on the Inr in the promoter (Li et al., 1994). It was subsequently shown that Myc could interact with transcriptional activators that bind to Inrs, such as YY1 and TFII-I, leading to an inhibition of reporter plasmid transcription (Roy et al., 1993; Shrivastava et al., 1993). A similar mechanism has been established for the p15INK4b promoter, another Myc-repressed target. p15INK4b gene expression is activated by the binding of the Miz-1 transcription factor to the Inr where Miz-1 recruits p300 to the promoter, leading to an increase in p15INK4b transcription. Myc has been shown to interact with Miz-1 to inhibit the formation of the Miz-1/p300 complex at the p15INK4b promoter, thereby inhibiting expression (Staller et al., 2001). Interestingly, it has also

The myc Oncogene

125

been shown by ChIP that Max can bind the p15INK4b promoter, and it has been suggested that a ternary complex between Myc, Max, and Miz-1 is formed. However, further analysis is required to distinguish whether these molecules bind independently to different regions of the promoter or bind separately at different stages of the cell cycle, or if they do indeed bind to form a ternary complex in vivo. It is likely that most of the target genes that contain Inrs are repressed by Myc through a similar mechanism, namely, through the interaction of Myc with Miz-1 or other Inr-binding transcription factors to inhibit binding events that lead to activation, and thereby suppress expression of the gene. Indeed, Myc has been implicated in the repression of most of its Inr-containing targets by a mechanism that is dependent on the Inr (Li et al., 1994; Mai and Martensson, 1995; Park et al., 2001; Staller et al., 2001; Wang et al., 1999; Yang et al., 2001b).

3. INR INDEPENDENT REPRESSION It is interesting to note that Myc also represses genes that do not contain Inrs, including gadd45 and pdgfβr, and thus these targets must be repressed by an Inr-independent mechanism. Analysis of serial deletions of these promoter regions suggests that Myc requires the minimal promoter of these genes for its repressive effect (Marhin et al., 1997; Izumi et al., 2001). The best studied of these genes is pdgfβr, whose activation is dependent on binding of the trimeric NF-Y activator to CCAAT boxes in the promoter. It has been shown that Myc can bind NF-YB and C, but not A, to interfere with activation of pdgfβr by NF-Y (Izumi et al., 2001). There is some controversy as to whether or not Myc interaction inhibits the NF-Y complex binding to DNA (Izumi et al., 2001; Taira et al., 1999). However, Myc interaction with NF-Y results in the decreased expression of pdgfβr. Indeed, we have observed Myc binding to the pdgfβr promoter upon serum stimulation of quiescent cells by ChIP analysis. Moreover, we have also observed the presence of Max at this promoter, implying that it is likely involved in Myc-mediated repression of pdgfβr (Penn laboratory, unpublished data). Interestingly, the promoter region of gadd45 bears many similarities to the pdgfβr promoter and is therefore likely to be regulated in a similar manner. Another Myc target, p21Cip1, has also been shown to be repressed in an Inr-independent manner. Although it has not been formally addressed, the proposed mechanism is similar to that suggested for gadd45 and pdgfβr as it involves Myc binding to other transcription factors, in this case Sp1/Sp3, to inhibit gene activation (Gartel et al., 2001). The mechanism of Myc repression is unquestionably distinct from that of activation. It is likely that there are multiple mechanisms of Myc-mediated repression, which depend on the characteristics of the target promoter and

126

Sara K. Oster et al.

the speciﬁc transcription factors that are important for activation of these genes. The degree of speciﬁcity with which Myc is able to repress certain targets is a particularly interesting question, as there does not seem to be any consensus Myc DNA-binding element within the promoter regions of all Myc-repressed targets, nor consistent interaction with any particular transcription factor. Indeed, it is possible that the elusive role of Max in Myc repression may be to recruit Myc to these promoters. As Myc repression has been implicated as being involved in the transformation process (Xiao et al., 1998; Brough et al., 1995; Lee et al., 1996), identifying the mechanism repression of speciﬁc targets may provide a means to inhibit this activity of the Myc protein.

F. Summary and Perspectives As Myc mediates its many biological activities by regulating gene transcription, identifying novel bona ﬁde target genes and elucidating their function will undoubtedly aid in our understanding the role of Myc in controlling cellular activities. In addition to providing a better understanding of the nature of the biological cascades triggered by Myc to effect the different activities, the bona ﬁde target genes will serve as critical experimental tools to evaluate the association of Myc function as a regulator of gene transcription with biological outcome. One hypothesis is that a similar molecular mechanism of gene regulation is activated by Myc to coordinately regulate the cohort of effector molecules required to drive a particular biological activity. This model allows for speciﬁcity of outcome through targeted interaction with certain proteins and, in turn, speciﬁc gene promoters. Only with the bona ﬁde target genes will this simple linear model be able to be evaluated.

V. IDENTIFICATION OF MYC-INTERACTING PROTEINS Myc participates in protein interactions that are key to its molecular function. Indeed, the identiﬁcation of Max as a required partner of Myc proteins was pivotal and at the time led to a revolution in our understanding of Myc function. Until recently, no other protein interactors had been identiﬁed that are so intimately linked to Myc activity. Recent advances have shown that Myc collaborates with a variety of other cellular factors at both its NTD and CTD to mediate its many biological activities. However, the role of protein interactions in Myc function has evaded full evaluation to date because of challenges in working with the Myc protein that are imposed by its inherent

The myc Oncogene

127

biochemical properties. Nevertheless, perseverance in the ﬁeld has brought forth a number of new insights leading to a better understanding of how protein interactions mediate Myc function, particularly in the area of target gene activation and the identiﬁcation of cellular factors that are important for Myc’s transforming potential. Indeed, there is a particular urgency to identify cellular factors that mediate Myc’s tremendous potency as a transforming agent so that speciﬁc protein targets can be identiﬁed and precise inhibitors can be designed to disrupt these tumorigenic interactions. The Myc-interacting proteins identiﬁed to date are summarized in Table III along with the experimental procedures used in their characterization. The net effect of interaction, in terms of its speciﬁc impact on Myc function and activity, is also noted as a quick reference to details that are elaborated within the text. Recent advances have led to the rapid expansion in the number of known Myc interactions, and in light of these developments new models of the molecular function of Myc activity can now be formulated.

A. Impact on Myc Function and Activity The cellular functions of the Myc-interacting proteins represent a wide range of biochemical activities that reﬂects the many biological pathways in which Myc function is involved (Table III). As such, it is difﬁcult to absolutely classify this panel of proteins based on any one set of criteria. However, several themes can be drawn from the repertoire of Myc interactions documented to date. These can include the following broadly deﬁned categories: interactions with coactivator proteins, interactions with other transcriptional regulators, interactions with putative tumor suppressor proteins, and interactions that occur outside the nucleus (Fig. 5). In addition, some protein interactors bind differentially to speciﬁc members of the Myc family. As more Myc interactors are identiﬁed, new categories will no doubt become apparent. The ultimate goal of this work is to determine how Myc mediates its various biological activities through interaction with other cellular factors, which will expand the current models of Myc function to include these important regulatory elements. A review of Myc interactions in the context of these categories and their relevance to Myc function and activity will be discussed in the following section.

B. Coactivators One mechanism by which transcription factors activate gene expression is through recruitment of coactivator proteins to promoter regions (Fry and Peterson, 2001; Roberts, 2000). It is believed that coactivator proteins

Table III Summary of Experimental Procedures used in Characterizing Myc-Interacting Proteins and Their Impact on Myc Function and Activity Impact on Myc Interacting protein α-Tubulin

AMY-1

Cellular function Structural protein, microtubules Unknown

c-Mye and/or N-Myc

Y2H

c

In vivo

In vitro

Exo

Endo

√

√

√

√

√

√

Colocalization exo

endo

Function An

Rn

Activity T

√

c

√

A

References Alexandrova et al., 1995 Niklinski et al., 2000

cyto √

P

+

Taira et al., 1998

−

Gaubatz et al., 1995 Sakamuro et al., 1996

nuc AP2

Transcription factor

c

BIN1

Tumor suppressor

c

BRCA1

Tumor suppressor

c

√ √

√

√

√

√

√

√

√

−

−

See text for further details and references

−

Wang et al., 1998

BRCA1

128

CDC6

cdr2

Prereplication complex subunit

c

Onconeuronal antigen

c

√

√

√

√

√

√

cdr2 INI1/hSNF5

SWI/SNF subunit

√

−

Takayama et al., 2000a

−

Okano et al., 1999

+

Cheng et al., 1999

nuc

c

√

√

√

√

√

√

√

cyto

cyto

√

INI1 Miz-1

Transcriptional activator

c, N

√

√

+

Peukert et al., 1997

nuc MM-1

Unknown

c

MSSP1

DNA replication

c

NF-Y

Trimeric transcription factor

c

IFN-inducible protein, cytokine signaling

c, N

Nmi

ORC1

Origin recognition complex (ORC) subunit

√

√

−

√

√

√

√

√

√

√

√

√

−

Izumi et al., 2001

√ √

√

Niki et al., 2000 Taira et al., 1999

Bao and Zervos, 1996 Bannasch et al., 2001

N-Myc c

Mori et al., 1998 +

√

−

Takayama et al., 2000b Beijersbergen et al., 1994

p107

pRB-family, cell cycle regulator

c

p202a

IFN-inducible 200-family, inhibits proliferation

c

p21

CKI, mediates p53-induced growth arrest

c

P-TEFb

RNA polymerase II kinase, elongation initiator

c

Pag

peroxiredoxin family, oxidative stress response, putative tumor suppressor

c

129

Pam

Similar to RCC1 involved in the function of chromatin

c only

Sp1/Sp3

Transcription factor

c

TBP

TATA-box binding protein, basal transcription machinery

c

TFII-I

Transcription initiation factor, basal transcription machinery

c

TIP48/49

ATPase/helicase motifs

φ

√

ATM/PI3K-related protein, recruits HATs

c, N

Yaf2

YY-1-associated factor (Yaf), zinc-ﬁnger protein

c, N

YY-1

Multifunctional transcription factor

√

√

√

√

√

√

√

−

Gu et al., 1994 Hoang et al., 1995 Haas et al., 1997

− −

− +

√

√

+ −

√

√

√

√

√

−

√

BP

√

Gartel et al., 2001 Hateboer et al., 1993 √

Maheswaran et al., 1994 McEwan et al., 1996

Roy et al., 1993

√

√

√

√

√

+

Wood et al., 2000b

+

McMahon et al., 1998 McMahon et al., 2000 Bannasch et al., 2001 Madge and Schwab, 2001

+

N-Myc √ c

YY-1

Mu et al., 2001

Guo et al., 1998

√

BP

Wang et al., 2000

Eberhardy et al., 2001

√

φ

−

Kitaura et al., 2000

√

c, N

TRRAP

√

√

√

−

−

Shrivastava et al., 1996 Shrivastava et al., 1993

Abbreviations: Y2H, yeast two-hybrid screen using c-Myc protein as bait unless otherwise noted; φ, λgt11 phage library screen; BP, biochemical puriﬁcation; In vitro, in vitro binding expts; In vivo, in vivo coimmunoprecipitation expts; Colocalization, in vivo colocalization expts; exo refers to instances where one or both protein partners were introduced exogenously; endo refers to use of endogenous protein; nuc, nuclear colocalization; cyto, cytoplasmic colocalization; +, increase in Myc function/activity; −, decrease in Myc function/activity; An, activation function; Rn, repression function; T, transformation activity; P, proliferation activity; A, apoptotic activity.

130

Sara K. Oster et al.

Fig. 5 Myc-interacting proteins. Myc interactions generally map to two main functional regions, the NTD and CTD, and can be divided into four broadly deﬁned functional categories: coactivator proteins, transcriptional regulators, tumor suppressors, and cytoplasmic proteins. Interactors that could not be classiﬁed by these categories are listed as unknown. Myc-interacting proteins that can bind to differential amino acids depending on the Myc family member are denoted by an asterisk (*). A solid line denotes the minimal amino acids required for each interaction; a broken line indicates amino acids that are important but not absolutely required for binding. Triangles indicate amino acids that when mutated lead to either loss of the interaction or loss of Myc regulation by the interaction (see text for further detail).

The myc Oncogene

131

function to induce an open chromatin conformation that allows the basal transcription machinery to access relevant DNA sites. It has been hypothesized that Myc recruits coactivator proteins to target promoters. Indeed, the identiﬁcation of Myc-binding proteins that have coactivator activity, both in histone acetylation and ATP-dependent chromatin remodeling, supports this hypothesis. An early study indicated that the NTD of Myc bound a cellular factor that was required for transformation (Brough et al., 1995). This factor was later identiﬁed as the TRRAP protein by biochemical puriﬁcation (McMahon et al., 1998). The importance of TRRAP is underscored by the essential nature of its yeast homolog, Tra1, which is a component of the SAGA and NuA4 chromatin remodeling complexes in yeast (Allard et al., 1999). TRRAP appears to be required for Myc-Ras cotransformation, and this effect was thought to be mediated in part by recruitment of HAT activity (McMahon et al., 1998). Indeed, TRRAP can recruit hGCN5 HAT activity to Myc, and it was later shown that partial restoration of cotransformation activity to a nontransforming Myc mutant could be achieved by direct fusion to the HAT domain of hGCN5 (McMahon et al., 2000). In addition, recruitment of TRRAP and hGCN5 has been shown to be essential for E2F-dependent transcription (Lang et al., 2001). This strongly suggests that TRRAP functions as an essential binding partner to transcriptional activators, such as E2F and Myc, and is required for activation of gene expression (see Section IV.D.1 for further detail). However, Myc has been shown to interact with components of the P-TEFb kinase in vitro, which can induce transcript elongation rather than transcriptional activation per se (Eberhardy and Farnham, 2001). This suggests that mechanisms other than histone acetylation is relevant in certain promoter contexts (see Section IV.D.3 for further detail). It will also be interesting to determine if Myc can interact with PAF400, which is a protein that is almost identical to TRRAP. PAF400 was isolated independently during characterization of proteins found in complex with PCAF in vivo and appears to be involved in recruitment of p300/CBP HAT activity (Vassilev et al., 1998). Thus, if Myc and PAF400 can interact, it remains possible that alternate HAT activities could be important to Myc activation function and transformation activity. The TIP48 and TIP49 proteins have also been shown to interact speciﬁcally with Myc using a biochemical puriﬁcation strategy similar to that which identiﬁed TRRAP (Wood et al., 2000b). Myc interacted with both TIP48 and TIP49 in vivo and in a manner that was independent of TRRAP binding. Interestingly, a single amino acid mutation of Myc at W136 was able to disrupt Myc and TIP49 interaction in vivo. Both TIP49 and TIP48 have homology to proteins that display ATP-dependent DNA helicase activities and have essential nonredundant functions in yeast cell growth (Wood et al., 2000b). The exact nature of TIP49 and TIP48 function in relation to Myc activity remains to be clariﬁed; however, TIP49 was shown to be

132

Sara K. Oster et al.

important for Myc-Ras cotransformation (Wood et al., 2000b). Thus it is likely that Myc function in target gene activation and Myc’s ability to promote transformation may require the ATPase and helicase activities of TIP proteins. A novel interactor, protein associated with Myc (Pam), may also have a role in transcriptional activation by Myc (Guo et al., 1998). Interaction between Myc and Pam was conﬁrmed in vivo, and interestingly, Pam appears to interact speciﬁcally with c-Myc and not with N-Myc proteins. Although the biological signiﬁcance of this interaction is unknown, Pam has homology to regulator of chromosome condensation (RCC1), a protein factor that may play a role in the function of chromatin (Guo et al., 1998). A CTD fragment of Myc was identiﬁed as an INI1 interactor in a yeast two-hybrid (Y2H) screen (Cheng et al., 1999). INI1/hSNF5 is a component of the SWI/SNF ATP-dependent chromatin remodeling complex that is required for the expression of many genes and the full activity of a number of transcription factors (Yudkovsky et al., 1999). Myc and INI1 interaction can be detected in vivo and use of a dominant negative INI1 peptide inhibited activation by Myc. This indicates that Myc activation may require SWI/SNF recruitment (see Section IV.D.2 for further detail). Further support for the importance of INI1/hSNF binding in Myc function was provided by an independent study identifying an in vitro interaction between Myc and the human homolog of the yeast origin recognition complex-1 (ORC1), which may be involved in gene silencing (Takayama et al., 2000b). ORC1 repressed Myc activation by competing with INI1/hSNF5 for interaction with Myc. It remains to be determined whether the competitive binding seen in vitro between ORC1 and INI1/hSNF5 for Myc will translate to an in vivo setting. Myc interaction with CDC6, a component of the prereplication complex that includes ORC1, was also demonstrated (Takayama et al., 2000a). However, unlike ORC1, CDC6 appears to inhibit Myc activation by interfering with Myc/Max heterodimerization and not by blocking INI1/hSNF and Myc association. It will be important to determine whether other members of the SWI/SNF complex have the potential to interact with Myc and whether SWI/SNF recruitment, similar to recruitment of HAT activity, plays a role in Myc-mediated transformation. Several other proteins also appear to inﬂuence Myc activation by a mechanism(s) that has yet to be determined. A Y2H screen using the CTD of N-Myc identiﬁed Yaf2 as an interactor (Bannasch et al., 2001). Interaction between N-Myc and Yaf2 was mapped to a short stretch of amino acids in the central region of N-Myc. In addition, colocalization experiments showed that Yaf2 was found along with N-Myc in the nucleus of neuroblastoma cells and led to enhanced N-Myc activation (Bannasch et al., 2001). Yaf2 also appears to interact with c-Myc, and interestingly, amino acids important for this interaction map to the NTD of c-Myc. More interestingly, in contrast to

The myc Oncogene

133

its impact on N-Myc function, Yaf2 inhibits c-Myc activation and transformation (Madge and Schwab, 2001). It is attractive to speculate that differential binding and regulation of speciﬁc Myc family members by proteins such as Yaf2 could account for some of the functional differences seen between family members in disease states such as neuroblastoma. The Nmi protein was likewise identiﬁed as an interactor by Y2H screening using N-Myc (Bao and Zervos, 1996). Nmi can also interact with c-Myc, and a later report showed that, similar to Yaf2, c-Myc interaction with Nmi appeared to require different amino acids than N-Myc and Nmi interaction (Bannasch et al., 1999; Bao and Zervos, 1996). Nmi is an interferoninducible protein that can associate with members of the signal transducer and activator of transcription (STAT) family of proteins (Chen et al., 2000; Zhu et al., 1999). Nmi increases STAT-mediated transcription in response to cytokine signaling by enhancing association of p300/CBP coactivator proteins (Zhu et al., 1999). The functional consequences of interaction between Myc and Nmi have yet to be determined, and it will be interesting to see if Nmi can affect Myc family protein function in a similar manner to STATs by recruitment of coactivators.

C. Transcriptional Regulators Myc interacts with components of the basal transcription machinery such as TBP (Hateboer et al., 1993; Maheswaran et al., 1994; McEwan et al., 1996) and TFII-I (Roy et al., 1993). However, the mechanisms by which Myc activates or represses gene expression cannot be fully explained by interactions with basal transcription factors alone. Myc appears to interact with a number of other transcriptional regulators that can result in the suppression of Myc activation through direct and in some cases indirect mechanisms. Conversely, there is also evidence to suggest that Myc can act as an antagonist of activation by other transcription factors through direct protein interaction. The interaction between Myc and YY-1 was ﬁrst documented a number of years ago (Shrivastava et al., 1993). More recently, this interaction has been demonstrated in vivo and appears to be modulated by changes in endogenous Myc levels, where YY-1 interaction was detected preferentially in cells with high Myc expression (Shrivastava et al., 1996). YY-1 is a multifunctional transcriptional regulator that, depending on context, can act as an activator, initiator, or repressor of gene expression. Myc can antagonize both the activation and repression functions of YY-1. However, whether direct interaction is required for this effect remains to be established (Shrivastava et al., 1993). Interestingly, YY-1 can inhibit activation and transformation by Myc in a manner that is independent of interaction (Austen et al., 1998). Thus,

134

Sara K. Oster et al.

the role of direct interaction with YY-1 in regulating Myc activity is unclear, though it remains possible that some effects of Myc function are mediated through modulation of YY-1 activity. Similarly, direct and indirect mechanisms of Myc regulation have also been demonstrated for AP-2, another transcription factor that can negatively regulate Myc activation (Gaubatz et al., 1995). AP-2 indirectly antagonized Myc activation by competing with Myc/Max heterodimers for DNA binding at target promoters that contain overlapping sites (Gaubatz et al., 1995). In the absence of overlapping DNA binding sites, AP-2 could directly inhibit Myc activation through speciﬁc interaction at the Myc CTD, which does not preclude Myc/Max heterodimerization but does prevent Myc/Max complexes from binding DNA (Gaubatz et al., 1995). Interestingly, AP-2 expression was shown to strongly suppress Myc-induced apoptosis in vitro, and AP-2 null mice displayed signiﬁcantly enhanced apoptosis in the kidneys (Moser et al., 1997). Together, these data point to a role for AP-2 in regulating Myc function and activity, but whether AP-2 inﬂuences Myc’s biological activities directly or indirectly in vivo requires further clariﬁcation. The novel Myc-interacting Zn ﬁnger protein-1 (Miz-1) was identiﬁed as an interactor of both c-Myc and N-Myc proteins (Peukert et al., 1997; Schneider et al., 1997). Miz-1 can activate transcription and has potent growth-arrest activity on cells in culture. These functions can be inhibited by interaction with Myc, possibly by a mechanism that involves Myc-induced sequestration of Miz-1 to discrete subnuclear foci (Peukert et al., 1997). Miz-1 has also been implicated in stabilizing Myc protein, as deletion of the Miz-1 binding region in Myc resulted in decreased protein half-life while overexpression of Miz-1 led to increased Myc stability, suggesting that Miz-1 is involved in regulating Myc protein turnover (Salghetti et al., 1999). More recently, it was proposed that Myc/Max/Miz-1 could form a ternary complex in vivo where interaction with Myc may inhibit Miz-1 activation by preventing it from contacting coactivator proteins such as p300/CBP (Staller et al., 2001). The exact conﬁguration of this ternary complex and whether Myc can interfere directly with the recruitment of coactivators by Miz-1 needs further investigation. Two independent reports have demonstrated an interaction between Myc and the trimeric NF-Y transcription complex (Izumi et al., 2001; Taira et al., 1999). However, whether Myc and NF-Y interaction occurs at the CTD or NTD of Myc remains to be clariﬁed. Both studies agree that Myc antagonizes NF-Y activation, but whether Myc disrupts DNA binding by the NF-Y complex (Taira et al., 1999) or whether it directly inhibits the NF-Y subunit responsible for activation (Izumi et al., 2001) to achieve this suppression is also unclear. In addition, NF-Y can interact with a number of coactivator proteins, including hGCN5, p300/CBP, and PCAF (Mantovani, 1999); thus it is possible that Myc antagonizes NF-Y function by interfering with its

The myc Oncogene

135

ability to recruit HATs. However, further experiments are required before a deﬁnitive mechanism can be assigned to Myc and NF-Y interaction in vivo. It has been shown that Myc can also interact with the Sp1 transcription factor (Gartel et al., 2001). Study of the p21Cip1 promoter, which is a repressed Myc target, revealed the presence of several Sp1 binding sites and a potential Inr element. The idea that Myc interacts with Sp1 came from the observation that the antagonistic effect of these proteins on p21Cip1 expression did not appear to occur through the Inr. Instead, Myc was shown to bind with Sp1 in vivo, where it is thought to suppress Sp1 activation by inhibiting its ability to bind DNA (Gartel et al., 2001). The suppression of Sp1 by Myc is particularly intriguing because it suggests that Myc may repress certain target genes by direct inhibition of the transcription factors responsible for their expression even when an Inr sequence is present (see Section IV.E for further detail on Inr-dependent and -independent mechanisms). Interestingly, a report has identiﬁed the p21Cip1 protein as a Myc interactor, suggesting a mechanism of reciprocal inhibition between these proteins (Kitaura et al., 2000). The details of the p21Cip1 and Myc interaction will be discussed in the following sections.

D. Tumor Suppressors The balance between oncogene and tumor suppressor function maintains homeostatic cell growth and death in vivo, whereas gain-of-function mutations to oncogenes or loss-of-function mutations to tumor suppressors contribute to the stepwise progression of cancer. Physical interaction between tumor suppressors and oncoproteins is one possible mechanism by which tumor suppressors may mediate their negative effects on growth and transformation. The evidence for Myc interaction with proteins that have tumor suppressor activity is discussed below. An early report of interaction between the pRB and Myc was unable to be conﬁrmed in vivo (Rustgi et al., 1991). However, an in vivo interaction between Myc and p107, a member of the pRB family, has been demonstrated (Beijersbergen et al., 1994; Gu et al., 1994; Hoang et al., 1995). Although pRB is the only family member with proven tumor suppressor activity, the other family members, including p107, share many biochemical similarities and have signiﬁcant functional overlap with pRB (Classon and Dyson, 2001). Myc expression was able to relieve p107-induced growth arrest but had no effect on pRB-induced growth arrest, providing further support for a functional interaction between Myc and p107 in vivo. In addition, interaction with p107 resulted in inhibition of Myc activation (Beijersbergen et al., 1994). In one survey, Burkitt’s lymphoma-derived mutant Myc alleles, which were deﬁcient for phosphorylation at T58 and enhanced in their ability to transform, were found to be defective for suppression by p107 but,

136

Sara K. Oster et al.

interestingly, were still able to bind p107 in a manner similar to wild-type Myc (Hoang et al., 1995). The ﬁnding that these mutants were able to escape inhibition by p107 is consistent with their increased transforming phenotype and also supports the idea that speciﬁc phosphorylation events play a role in modulating p107 inhibition of Myc function. It is attractive to speculate that p107 modulates Myc’s transforming potential by inhibiting Myc-mediated regulation of certain gene targets involved in tumorigenesis. Myc interaction with another cell-cycle regulator, the p21Cip1 cyclin kinase inhibitor, has been reported from an in vitro screen of a select panel of known cell-cycle regulators (Kitaura et al., 2000). Interaction between Myc and p21Cip1 was then demonstrated in vivo and mapped to the Myc NTD. Binding of p21Cip1 appears to inhibit Myc activation through disruption of Myc/Max heterodimerization; however, the mechanism by which p21Cip1 bound at the NTD can inhibit Max interaction at the CTD remains to be deﬁned. It is also possible that Myc regulates p21Cip1 function by competing it from interaction with proliferating cell nuclear antigen (PCNA), since PCNA binds to overlapping amino acids in p21Cip1 (Kitaura et al., 2000). PCNA-dependent DNA synthesis is inhibited by p21Cip1 interaction, and it is attractive to speculate that Myc-induced cell cycle progression is mediated in part through alleviation of p21Cip1 inhibition, either through direct interaction or by antagonizing Sp1-mediated activation of expression as previously described. The murine p202a protein has been identiﬁed by Y2H and conﬁrmed in vivo as an interacting factor at the Myc CTD (Wang et al., 2000). p202a is part of the 200-family of interferon-inducible proteins that may contribute to the growth-suppressive activity of interferons, and the human homolog, Absent In Melanoma (AIM-2), has been identiﬁed as a possible tumor suppressor (Johnstone and Trapani, 1999). Myc activation was inhibited by p202a, and similarly, expression of several endogenous Myc-activated gene targets was decreased in cells overexpressing p202a (Wang et al., 2000). Introduction of p202a also inhibited apoptosis and transformation by Myc and is consistent with the observation that p202a can inhibit the apoptosis and anchorage-independent growth of human prostate tumor cells (Yan et al., 1999). p202a also inhibits the activity of several other transcription factors by interfering with their ability to bind DNA, and likewise, expression of p202a appears to block Myc/Max heterodimerization (Wang et al., 2000). Interaction between Myc and p202a is interesting in that it suggests a possible link between interferon signaling and suppression of Myc function as a mechanism by which these cellular factors may negatively regulate growth. Another tumor suppressor protein that inhibits Myc function, perhaps through interfering with Myc/Max association, is the breast cancer susceptibility gene 1 (BRCA1), which interacts with the CTD of Myc. Myc and BRCA1 interaction was ﬁrst detected by Y2H and then demonstrated in vivo

The myc Oncogene

137

(Wang et al., 1998). Loss or inactivation of BRCA1, along with BRCA2, results in increased risk of breast and ovarian cancer and has been implicated in the familial inheritance of these cancers (Arver et al., 2000; Welcsh and King, 2001). BRCA1 interaction inhibits Myc activation and suppresses Myc-Ras cotransformation (Wang et al., 1998). Importantly, Myc plays a signiﬁcant role in the progression of breast and ovarian tumors (Liao and Dickson, 2000); thus the mechanism by which BRCA1 can inhibit Myc transformation warrants further investigation. Whether loss of BRCA1 enhances tumorigenesis by Myc in vivo remains to be established; however, cancer therapies that could mimic BRCA1 inhibition would be valuable tools in the treatment of these familial cancers. The Bridging Integrator Protein-1 (BIN1), formerly known as BoxInteracting Protein-1, is a novel Myc-interacting protein that is thought to have putative tumor suppressor activity. BIN1 inhibited Myc-Ras cotransformation, and similar to p107, mutation at T58 resulted in abrogation of BIN1 inhibition while having no effect on binding (Sakamuro et al., 1996). However, BIN1 is found in at least seven alternate splice variants that are differentially regulated in various tissues and also differ in their functional proﬁles (Ge et al., 1999; Mao et al., 1999; Wechsler-Reya et al., 1997, 1998). The ability of the various individual splice variants to interact with Myc has not been established. BIN1 is frequently deleted in cultured cell lines and in a wide variety of tumors, which supports its classiﬁcation as a tumor suppressor (Sakamuro et al., 1996; Ge et al., 1999, 2000a,b; Livezey et al., 2000). BIN1 can inhibit growth, mediate differentiation, and induce apoptosis in a number of cellular backgrounds (Galderisi et al., 1999; Ge et al., 2000a; Mao et al., 1999; Wechsler-Reya et al., 1998; DuHadaway et al., 2001; Elliott et al., 2000; Hogarty et al., 2000). As the functional intricacies of these highly spliced proteins continue to emerge, it will be of great interest to determine whether speciﬁc Myc functions could be regulated differentially by various BIN1 isoforms and whether the variant forms of BIN1 have alternative and possibly antagonistic functions. Another putative tumor suppressor, Pag, a member of the peroxiredoxin family that is involved in oxidative stress response, has been identiﬁed as a protein that speciﬁcally interacted with Myc in vivo (E. Prochownik, personal communication). Pag is located in a chromosomal region that is often deleted in cancers, including leukemia, breast cancer, and neuroblastoma. Interestingly, Pag appears to promote increase in cell mass and confers a proapoptotic phenotype that is reminiscent of ectopic Myc overexpression; however, whether these effects involve binding to Myc is not clear. Importantly, coexpression of Pag can inhibit Myc-Ras cotransformation. Also interesting is the differential effect that Pag appears to have on Myc activation, where Pag expression can antagonize the expression of some but not all Myc target genes tested (E. Prochownik, personal communication). Whether Pag

138

Sara K. Oster et al.

modulates Myc activity through direct or indirect mechanisms remains to be clariﬁed. However, it is attractive to speculate that proteins such as Pag have the ability to selectively target certain Myc functions as a way to regulate speciﬁc aspects of Myc activity. This raises the exciting possibility that speciﬁc Myc activities, such as transformation, could be functionally separated by interaction, which would provide precise therapeutic targets for suppression of Myc activities.

E. Cytoplasmic Proteins Myc protein is normally localized to the nucleus, where it is thought to function as a transcriptional regulator. However, Myc can be detected in the cytoplasm in certain cell lines and tumor samples, as well as during development and cellular processes such as differentiation and serum withdrawal-induced growth arrest (Lemaitre et al., 1995; Liao and Dickson, 2000; Niklinski et al., 2000; Vriz et al., 1992; Wakamatsu et al., 1993). Thus, interaction with cytoplasmic proteins may serve to relocate Myc to the cytoplasm as a way to down-regulate its nuclear function. Myc interaction with α-tubulin was demonstrated in vitro and in vivo using endogenous proteins (Alexandrova et al., 1995). Colocalization experiments have shown that Myc could be visualized in association with microtubules in vivo. A later report demonstrated that interaction with α-tubulin appeared to be dependent on the phosphorylation status of Myc, where hyperphosphorylation of Myc during mitosis or constitutive phosphorylation of Myc mutants found in Burkitt’s lymphoma led to the disruption of α-tubulin binding (Niklinski et al., 2000). It was thought that interaction of wild-type Myc with α-tubulin allowed for its sequestration to the cytoplasm and correlated with rapid Myc protein turnover. In contrast, mutant Myc proteins deﬁcient for α-tubulin binding were increased in stability (Niklinski et al., 2000). It will be important to determine whether interaction with α-tubulin can regulate Myc transformation and, more importantly, whether loss of α-tubulin interaction leads to a more aggressive transformed phenotype in vivo. Another cytoplasmic protein named Associate of MYC-1 (AMY-1) is translocated to the nucleus during early S-phase when Myc expression is elevated and returns to the cytoplasm following S-phase as Myc protein levels decay (Taira et al., 1998). Noninteracting AMY-1 mutants do not translocate to the nucleus, suggesting that change in subcellular localization of AMY-1 is dependent upon interaction with Myc. Interaction between Myc and AMY-1 leads to enhanced activation by Myc/Max complexes in vitro (Taira et al., 1998). Since Myc appears to induce nuclear translocation of AMY-1 speciﬁcally at S-phase, it is possible that Myc recruits AMY-1 function to mediate speciﬁc changes in gene expression during this point of the cell cycle.

The myc Oncogene

139

The Purkinje neuron antigen cdr2, which is also localized to the cytoplasmic compartment of cells, was determined to speciﬁcally interact with the CTD of Myc (Okano et al., 1999). Cdr2 is expressed almost exclusively in Purkinje neurons under normal conditions, but can be found in breast or ovarian tumors where expression leads to a disorder called paraneoplastic cerebellar degeneration (PCD) (Okano et al., 1999). PCD is characterized by production of antibodies to cdr2, which accumulate in the sera and cerebrospinal ﬂuid, leading to an immune response and subsequent cerebellar degeneration. Interaction with cdr2 induces Myc translocation to the cytoplasm, which is supported by the colocalization of these proteins within the cytoplasm of Purkinje neurons (Okano et al., 1999). Importantly, cdr2 interaction leads to suppression of Myc activation, and taken together, these results suggest that cdr2 can down-regulate the nuclear functions of Myc by relocating it to the cytoplasm. It is possible that cdr2 antibodies found in PCD patients antagonize the cdr2-dependent cytoplasmic translocation of Myc, possibly leading to an overall increase in nuclear Myc activity and progression of PCD (Okano et al., 1999). It is interesting that cdr2 becomes expressed in breast and ovarian tumors, and the signiﬁcance of this observation in terms of counteracting Myc function in the progression of these cancers warrants further investigation.

F. The Myc Family and c-Myc Isoforms Certain protein interactors can bind differentially to speciﬁc members of the Myc family (Table III). In particular, the Myc family members (Fig. 1) have overlapping but not completely redundant functions, as suggested by temporal differences in expression during development and by spatially different expression patterns in various tissues and organ systems (Bull et al., 2001; DePinho et al., 1991; Douglas et al., 2001; Morgenbesser and DePinho, 1994). A second level of complexity exists in that different Myc isoforms occur as a result of alternative translational start sites (Fig. 1). The c-Myc isoforms, like members of the Myc family, also appear to be differentially expressed and have biologically distinguishable activities (Hann et al., 1994; Spotts et al., 1997; Xiao et al., 1998). It is possible that select interactions speciﬁcally targeting certain Myc family members and/or isoforms could account for some of their nonredundant activity. Conversely, interactions common to Myc proteins would account for their similar functions and mechanisms of action at the molecular level. Furthermore, the lack of binding to certain protein partners would account for the loss of activity or dominantnegative effect of some Myc family members and isoforms. Indeed, it is likely that individual Myc molecules fulﬁll certain unique functional niches, which is supported by their observed differences during normal development and

140

Sara K. Oster et al.

in the progression of certain cancers. A comprehensive protein interaction proﬁle of the individual Myc family members and isoforms would be highly informative. Such information would be useful in the design of therapies that could target individual Myc proteins in disease states where their particular function is most relevant.

G. Summary and Perspectives The interactions described in Table III represent the many biochemical pathways in which Myc function appears to play an important role. It will be of critical interest to clarify how the speciﬁcity of Myc interactions is maintained in vivo and whether distinct interactors are responsible for recruiting Myc function to a particular biological process. Also intriguing is the recurrent role that coactivator proteins have in transformation by Myc, and it will be important to determine which of these interactors is key to Mycmediated tumorigenesis in vivo. Myc also appears to directly antagonize the function of several other transcription factors. It would be of great interest to clarify the mechanism that dictates the speciﬁcity of these interactions in vivo such that inhibition by Myc is directed only to the intended target promoters. Conversely, the binding of certain proteins with tumor suppressor function inhibits Myc activation; these interactions also appear to suppress transformation by Myc. It will be important to determine whether the loss or inactivation of interacting tumor suppressors enhances Myc-mediated tumor progression in vivo. It is possible that when required, Myc induces the translocation of speciﬁc cytoplasmic factors for particular nuclear processes; thus the impaired transit to and from the nucleus of Myc or its protein partners could lead to deregulation of Myc activity and contribute to certain pathological states. Recent advances have led to a sudden expansion in the number of known Myc interactors, and exciting new insights into Myc biology are on the horizon as the work to elucidate the molecular mechanisms behind the current interactions continues to progress. However, full understanding of Myc function at the molecular level and the way in which interactions contribute to Myc-mediated pathologies will only be achieved by the continued identiﬁcation of Myc interactors in vivo. A better understanding of the molecular interplay between Myc and its interacting partners promises to reveal new avenues by which to target Myc’s activity in cancer.

VI. OVERALL PERSPECTIVES Myc has remarkable inﬂuence over numerous, diverse biological activities as well as a deﬁning and singular impact on tumorigenesis. The goal of this

The myc Oncogene

141

review was not only to highlight recent advances in the ﬁeld, but also to foreshadow future directions in light of this perspective. Clearly, the identiﬁcation of bona ﬁde Myc target genes and novel Myc-interacting proteins will continue to update the current models of Myc function at the molecular level. Moreover, understanding the downstream consequences of such regulation at the genetic level will begin to bridge the gap between Myc function as a transcription factor and Myc’s biological activities. The advent of new technologies will also aid in our ability to better deﬁne the signiﬁcance of these events at the cellular level for the purposes of cancer diagnosis and treatment. Clearly, great strides in delineating the mechanisms of Myc function have been achieved in recent years; however, the complexity of Myc biology remains to be fully unraveled.

ACKNOWLEDGMENTS We thank our colleagues who generously contributed data prior to publication. We also thank Jeremy Squire, David Sealey, Wilson Marhin, and the Penn laboratory for helpful comments. We apologize to those whose contributions have not been cited because of space constraints. Support from the National Cancer Institute of Canada with funds from the Canadian Cancer Society and The Terry Fox Foundation are gratefully acknowledged.

REFERENCES Adachi, S., Obaya, A. J., Han, Z., Ramos-Desimone, N., Wyche, J. H., and Sedivy, J. M. (2001). Mol. Cell. Biol. 21, 4929–4937. Adams, J. M., and Cory, S. (1992). Cancer Surv. 15, 119–141. Adams, J. M., and Cory, S. (2001). Trends Biochem. Sci. 26, 61–66. Adams, M. R., Sears, R., Nuckolls, F., Leone, G., and Nevins, J. R. (2000). Mol. Cell. Biol. 20, 3633–3639. Ahn, S. G., Jeong, S. Y., Rhim, H., and Kim, I. K. (1998). Biochim. Biophys. Acta 1448, 115– 125. Akeson, R., and Bernards, R. (1990). Mol. Cell. Biol. 10, 2012–2016. Akie, K., Dosaka-Akita, H., Murakami, A., and Kawakami, Y. (2000). Antisense Nucleic Acid Drug Dev. 10, 243–249. Albert, T., Urlbauer, B., Kohlhuber, F., Hammersen, B., and Eick, D. (1994). Oncogene 9, 759–763. Alexandrova, N., Niklinski, J., Bliskovsky, V., Otterson, G. A., Blake, M., Kaye, F. J., and Zajac-Kaye, M. (1995). Mol. Cell. Biol. 15, 5188–5195. Allard, S., Utley, R. T., Savard, J., Clarke, A., Grant, P., Brandl, C. J., Pillus, L., Workman, J. L., and Cote, J. (1999). EMBO J. 18, 5108–5119. Amanullah, A., Liebermann, D. A., and Hoffman, B. (2000). Oncogene 19, 2967–2977. Amati, B., Alevizopoulos, K., and Vlach, J. (1998). Front. Biosci. 3, D250–D268. Amati, B., Brooks, M. W., Levy, N., Littlewood, T. D., Evan, G. I., and Land, H. (1993a). Cell 72, 233–245.

142

Sara K. Oster et al.

Amati, B., Dalton, S., Brooks, M. W., Littlewood, T. D., Evan, G. I., and Land, H. (1992). Nature 359, 423–426. Amati, B., Littlewood, T. D., Evan, G. I., and Land, H. (1993b). EMBO J. 12, 5083–5087. Amundson, S. A., Zhan, Q., Penn, L. Z., and Fornace, A. J., Jr. (1998). Oncogene 17, 2149– 2154. Antonson, P., Pray, M. G., Jacobsson, A., and Xanthopoulos, K. G. (1995). Eur. J. Biochem. 232, 397–403. Arsura, M., Hofmann, C. S., Golay, J., Introna, M., and Sonenshein, G. E. (2000). Blood 96, 1013–1020. Arver, B., Du, Q., Chen, J., Luo, L., and Lindblom, A. (2000). Semin. Cancer Biol. 10, 271– 288. Austen, M., Cerni, C., Luscher-Firzlaff, J. M., and Luscher, B. (1998). Oncogene 17, 511–520. Auvinen, M., Paasinen, A., Andersson, L. C., and Holtta, E. (1992). Nature 360, 355–358. Axelson, H., Henriksson, M., Wang, Y., Magnusson, K. P., and Klein, G. (1995). Eur. J. Cancer 31A, 2099–2104. Bacon, T. A., and Wickstrom, E. (1991). Oncogene Res. 6, 21–32. Bahram, F., von der Lehr, N., Cetinkaya, C., and Larsson, L. G. (2000). Blood 95, 2104–2110. Balachandran, S., Porosnicu, M., and Barber, G. N. (2001). J. Virol. 75, 3474–3479. Bannasch, D., Weis, I., and Schwab, M. (1999). Oncogene 18, 6810–6817. Bannasch, D., Madge, B., and Schwab, M. (2001). Oncogene 41, 5913–5919. Bao, J., and Zervos, A. S. (1996). Oncogene 12, 2171–2176. Barker, N., Morin, P. J., and Clevers, H. (2000). Adv. Cancer Res. 77, 1–24. Barone, M. V., and Courtneidge, S. A. (1995). Nature 378, 509–512. Barr, L. F., Campbell, S. E., Diette, G. B., Gabrielson, E. W., Kim, S., Shim, H., and Dang, C. V. (2000). Cancer Res. 60, 143–149. Baudino, T. A., and Cleveland, J. L. (2001). Mol. Cell. Biol. 21, 691–702. Bazarov, A. V., Adachi, S., Li, S. F., Mateyak, M. K., Wei, S., and Sedivy, J. M. (2001). Cancer Res. 61, 1178–1186. Behrens, J. (2000). Ann. NY Acad. Sci. 910, 21-33; discussion 33–35. Beier, R., Burgin, A., Kiermaier, A., Fero, M., Karsunky, H., Saffrich, R., Moroy, T., Ansorge, W., Roberts, J., and Eilers, M. (2000). EMBO J. 19, 5813–5823. Beijersbergen, R., Hijmans, E. M., Zhu, L., and Bernards, R. (1994). Eur. Mol. Biol. Org. J. 13, 4080–4086. Bejarano, M. T., Albihn, A., Cornvik, T., Brijker, S. O., Asker, C., Osorio, L. M., and Henriksson, M. (2000). Exp. Cell Res. 260, 61–72. Bello-Fernandez, C., Packham, G., and Cleveland, J. L. (1993). Proc. Natl. Acad. Sci. USA 90, 7804–7808. Ben-Porath, I., Yanuka, O., and Benvenisty, N. (1999). Mol. Cell. Biol. 19, 3529–3539. Ben-Yosef, T., Yanuka, O., and Benvenisty, N. (1996). Oncogene 13, 1859–1866. Ben-Yosef, T., Yanuka, O., Halle, D., and Benvenisty, N. (1998). Oncogene 17, 165–171. Benvenisty, N., Leder, A., Kuo, A., and Leder, P. (1992). Genes Dev. 6, 2513–2523. Bergmann, E., Wanzel, M., Weber, A., Shin, I., Christiansen, H., and Eilers, M. (2001). Int. J. Cancer 95, 176–783. Bernasconi, N. L., Wormhoudt, T. A., and Laird-Offringa, I. A. (2000). Am. J. Respir. Cell. Mol. Biol. 23, 560–565. Berns, K., Hijmans, E. M., Koh, E., Daley, G. Q., and Bernards, R. (2000a). Oncogene 19, 3330–3334. Berns, K., Martins, C., Dannenberg, J. H., Berns, A., te Riele, H., and Bernards, R. (2000b). Oncogene 19, 4822–4827. Bhatia, K., Huppi, K., Spangler, G., Siwarski, D., Iyer, R., and Magrath, I. (1993). Nat. Genet. 5, 56–61.

The myc Oncogene

143

Blake, R. A., Broome, M. A., Liu, X., Wu, J., Gishizky, M., Sun, L., and Courtneidge, S. A. (2000). Mol. Cell. Biol. 20, 9018–9027. Blyth, K., Stewart, M., Bell, M., James, C., Evan, G., Neil, J. C., and Cameron, E. R. (2000). Oncogene 19, 773–782. Boffa, L. C., Scarﬁ, S., Mariani, M. R., Damonte, G., Allfrey, V. G., Benatti, U., and Morris, P. L. (2000). Cancer Res. 60, 2258–2262. Bonvini, P., Nguyen, P., Trepel, J., and Neckers, L. M. (1998). Oncogene 16, 1131–1139. Boon, K., Caron, H. N., van Asperen, R., Valentijn, L., Hermus, M. C., van Sluis, P., Roobeek, I., Weis, I., Voute, P. A., Schwab, M., and Versteeg, R. (2001). EMBO J. 20, 1383–1393. Born, T. L., Frost, J. A., Schonthal, A., Prendergast, G. C., and Feramisco, J. R. (1994). Mol. Cell. Biol. 14, 5710–5718. Bouchard, C., Staller, P., and Eilers, M. (1998). Trends Cell Biol. 8, 202–206. Bouchard, C., Thieke, K., Maier, A., Saffrich, R., Hanley-Hyde, J., Ansorge, W., Reed, S., Sicinski, P., Bartek, J., and Eilers, M. (1999). EMBO J. 18, 5321–5333. Bouchard, C., Dittrich, O., Kiermaier, A., Dohmann, K., Menkel, A., Eilers, M., and Luscher, B. (2001). Submitted for publication. Bouck, N., Stellmach, V., and Hsu, S. C. (1996). Adv. Cancer Res. 69, 135–174. Bowman, T., Broome, M. A., Sinibaldi, D., Wharton, W., Pledger, W. J., Sedivy, J. M., Irby, R., Yeatman, T., Courtneidge, S. A., and Jove, R. (2001). Proc. Natl. Acad. Sci. USA 98, 7319–7324. Boyd, K. E., Wells, J., Gutman, J., Bartley, S. M., and Farnham, P. J. (1998). Proc. Natl. Acad. Sci. USA 95, 13887–13892. Breckenridge, D. G., and Shore, G. C. (2000). Crit. Rev. Eukaryot. Gene Expr. 10, 273–280. Breit, S., Ashman, K., Wilting, J., Rossler, J., Hatzi, E., Fotsis, T., and Schweigerer, L. (2000). Cancer Res. 60, 4596–4601. Brewer, G. (2000). J. Biol. Chem. 275, 33336–33345. Brough, D. E., Hoffman, T. J., Ellmood, K. B., Townley, R. A., and Cole, M. (1995). Mol. Cell. Biol. 15, 1536–1544. Brown, H. G., Kepner, J. L., Perlman, E. J., Friedman, H. S., Strother, D. R., Duffner, P. K., Kun, L. E., Goldthwaite, P. T., and Burger, P. C. (2000). J. Neuropathol. Exp. Neurol. 59, 857–865. Brunner, T., Kasibhatla, S., Pinkoski, M. J., Frutschi, C., Yoo, N. J., Echeverri, F., Mahboubi, A., and Green, D. R. (2000). J. Biol. Chem. 275, 9767–9772. Bull, J. J., Muller-Rover, S., Patel, S. V., Chronnell, C. M., McKay, I. A., and Philpott, M. P. (2001). J. Invest. Dermatol. 116, 617–622. Bush, A., Mateyak, M., Dugan, K., Obaya, A., Adachi, S., Sedivy, J., and Cole, M. (1998). Genes Dev. 12, 3797–3802. Carey, G. B., Donjerkovic, D., Mueller, C. M., Liu, S., Hinshaw, J. A., Tonnetti, L., Davidson, W., and Scott, D. W. (2000). Immunol. Rev. 176, 105–115. Chang, D. H., Angelin-Duclos, C., and Calame, K. (2000a). Nat. Immunol. 1, 169–176. Chang, D. W., Claassen, G. F., Hann, S. R., and Cole, M. D. (2000b). Mol. Cell Biol. 20, 4309–4319. Chappell, S. A., LeQuesne, J. P., Paulin, F. E., deSchoolmeester, M. L., Stoneley, M., Soutar, R. L., Ralston, S. H., Helfrich, M. H., and Willis, A. E. (2000). Oncogene 19, 4437–4440. Chen, C., Nussenzweig, A., Guo, M., Kim, D., Li, G. C., and Ling, C. C. (1996). Oncogene 13, 1659–1665. Chen, C. R., Kang, Y., and Massague, J. (2001a). Proc. Natl. Acad. Sci. USA 98, 992–999. Chen, J., Shpall, R. L., Meyerdierks, A., Hagemeier, M., Bottger, E. C., and Naumovski, L. (2000). J. Biol. Chem. 275, 36278–36284. Chen, J. P., Lin, C., Xu, C. P., Zhang, X. Y., Fu, M., Deng, Y. P., Wei, Y., and Wu, M. (2001b). J. Gastroenterol. Hepatol. 16, 22–28.

144

Sara K. Oster et al.

Cheng, S. W., Davies, K. P., Yung, E., Beltran, R. J., Yu, J., and Kalpana, G. V. (1999). Nat. Genet. 22, 102–105. Chiariello, M., Marinissen, M. J., and Gutkind, J. S. (2001). Nat. Cell Biol. 3, 580–586. Chou, T. Y., Dang, C. V., and Hart, G. W. (1995a). Proc. Natl. Acad. Sci. USA 92, 4417–4421. Chou, T. Y., Hart, G. W., and Dang, C. V. (1995b). J. Biol. Chem. 270, 18961–18965. Claassen, G. F., and Hann, S. R. (1999). Oncogene 18, 2925–2933. Claassen, G. F., and Hann, S. R. (2000). Proc. Natl. Acad. Sci. USA 97, 9498–9503. Clark, R. E. (2000). Leukemia 14, 347–355. Classon, M., and Dyson, N. (2001). Exp. Cell Res. 264, 135–147. Cohen, O., Inbal, B., Kissil, J. L., Raveh, T., Berissi, H., Spivak-Kroizaman, T., Feinstein, E., and Kimchi, A. (1999). J. Cell. Biol. 146, 141–148. Cohn, S. L., London, W. B., Huang, D., Katzenstein, H. M., Salwen, H. R., Reinhart, T., Madaﬁglio, J., Marshall, G. M., Norris, M. D., and Haber, M. (2000). J. Clin. Oncol. 18, 3604–3613. Cole, M. D. (1986). Annu. Rev. Genet. 20, 361–384. Coller, H. A., Grandori, C., Tamayo, P., Colbert, T., Lander, E. S., Eisenman, R. N., and Golub, T. R. (2000). Proc. Natl. Acad. Sci. USA 97, 3260–3265. Conner, E. A., Lemmer, E. R., Omori, M., Wirth, P. J., Factor, V. M., and Thorgeirsson, S. S. (2000). Oncogene 19, 5054–5062. Constance, C. M., Morgan, J. I. T., and Umek, R. M. (1996). Mol. Cell. Biol. 16, 3878–3883. Conzen, S. D., Gottlob, K., Kandel, E. S., Khanduri, P., Wagner, A. J., O’Leary, M., and Hay, N. (2000). Mol. Cell. Biol. 20, 6008–6018. Daksis, J. I., Lu, R. Y., Facchini, L. M., Marhin, W. W., and Penn, L. J. (1994). Oncogene 9, 3635–3645. Dang, C. V. (1999). Mol. Cell Biol. 19, 1–11. D’Cruz, C. M., Gunther, E. J., Boxer, R. B., Hartman, J. L., Sintasath, L., Moody, S. E., Cox, J. D., Ha, S. I., Belka, G. K., Golant, A., Cardiff, R. D., and Chodosh, L. A. (2001). Nat. Med. 7, 235–239. de Alboran, I. M., O’Hagan, R. C., Gartner, F., Malynn, B., Davidson, L., Rickert, R., Rajewsky, K., DePinho, R. A., and Alt, F. W. (2001). Immunity 14, 45–55. DePinho, R. A., Schreiber-Agus, N., and Alt, F. W. (1991). Adv. Cancer Res. 57, 1–46. Desagher, S., and Martinou, J. C. (2000). Trends Cell Biol. 10, 369–377. Domashenko, A. D., Latham, K. E., and Hatton, K. S. (1997). Mol. Reprod. Dev. 47, 57–65. Donjerkovic, D., Carey, G. B., Mueller, C. M., Liu, S., and Scott, D. W. (2000). Curr. Top. Microbiol. Immunol. 252, 151–159. Douglas, N. C., Jacobs, H., Bothwell, A. L., and Hayday, A. C. (2001). Nat. Immunol. 2, 307–315. Drissi, R., Zindy, F., Roussel, M. F., and Cleveland, J. L. (2001). J. Biol. Chem. 11, 11. Du, X. L., Edelstein, D., Rossetti, L., Fantus, I. G., Goldberg, H., Ziyadeh, F., Wu, J., and Brownlee, M. (2000). Proc. Natl. Acad. Sci. USA 97, 12222–12226. Duelli, D. M., and Lazebnik, Y. A. (2000). Nat. Cell Biol. 2, 859–862. DuHadaway, J. B., Sakamuro, D., Ewert, D. L., and Prendergast, G. C. (2001). Cancer Res. 61, 3151–3156. Dupont, J., Fernandez, A. M., Helman, L., and LeRoith, D. (2001). J. Biol. Chem. 25, 25. Eberhardy, S. R., and Farnham, P. J. (2001). J. Biol. Chem. in press. Eberhardy, S. R., D’Cunha, C. A., and Farnham, P. J. (2000). J. Biol. Chem. 275, 33798– 33805. Ebinuma, H., Saito, H., Kosuga, M., Wakabayashi, K., Saito, Y., Takagi, T., Nakamoto, N., Okuyama, T., and Ishii, H. (2001). J. Cell Physiol. 188, 56–66. Eilers, M., Picard, D., Yamamoto, K. R., and Bishop, J. M. (1989). Nature 340, 66–68. Eilers, M., Schirm, S., and Bishop, J. M. (1991). EMBO J. 10, 133–141.

The myc Oncogene

145

Eischen, C. M., Weber, J. D., Roussel, M. F., Sherr, C. J., and Cleveland, J. L. (1999). Genes Dev. 13, 2658–2669. El Gedaily, A., Bubendorf, L., Willi, N., Fu, W., Richter, J., Moch, H., Mihatsch, M. J., Sauter, G., and Gasser, T. C. (2001). Prostate 46, 184–190. Elend, M., and Eilers, M. (1999). Curr. Biol. 9, R936–R938. Elliott, K., Ge, K., Du, W., and Prendergast, G. C. (2000). Oncogene 19, 4669–4684. Evan, G. I., and Vousden, K. H. (2001). Nature 411, 342–348. Evan, G. I., Wyllie, A. H., Gilbert, C. S., Littlewood, T. D., Land, H., Brooks, M., Waters, C. M., Penn, L. Z., and Hancock, D. C. (1992). Cell 69, 119–128. Facchini, L. M., and Penn, L. Z. (1998). FASEB J. 12, 633–651. Facchini, L. M., Chen, S., Marhin, W. W., Lear, J. N., and Penn, L. Z. (1997). Mol. Cell. Biol. 17, 100–114. Fan, L., Iyer, J., Zhu, S., Frick, K. K., Wada, R. K., Eskenazi, A. E., Berg, P. E., Ikegaki, N., Kennett, R. H., and Frantz, C. N. (2001). Cancer Res. 61, 1073–1079. Felsher, D. W., and Bishop, J. M. (1999a). Mol. Cell 4, 199–207. Felsher, D. W., and Bishop, J. M. (1999b). Proc. Natl. Acad. Sci. USA 96, 3940–3944. Feng, L., Miyake, M., Seeger, R. C., and Bogenmann, E. (1997). Int. J. Cancer 71, 1116–1117. Ferri, K. F., and Kroemer, G. (2001). Bioessays 23, 111–115. Fiche, M., Avet-Loiseau, H., Maugard, C. M., Sagan, C., Heymann, M. F., Leblanc, M., Classe, J. M., Fumoleau, P., Dravet, F., Mahe, M., and Dutrillaux, B. (2000). Int. J. Cancer 89, 403–410. Flinn, E. M., Busch, C. M., and Wright, A. P. (1998). Mol. Cell. Biol. 18, 5961–5969. Foley, K. P., McArthur, G. A., Queva, C., Hurlin, P. J., Soriano, P., and Eisenman, R. N. (1998). EMBO J. 17, 774–785. Fotsis, T., Breit, S., Lutz, W., Rossler, J., Hatzi, E., Schwab, M., and Schweigerer, L. (1999). Eur. J. Biochem. 263, 757–764. Frank, S. R., Schroeder, M., Fernandez, P., Taubert, S., and Amati, B. (2001). Submitted for publication. Freytag, S. O., Dang, C. V., and Lee, W. M. (1990). Cell Growth Differ. 1, 339–343. Fruhwald, M. C., O’Dorisio, M. S., Rush, L. J., Reiter, J. L., Smiraglia, D. J., Wenger, G., Costello, J. F., White, P. S., Krahe, R., Brodeur, G. M., and Plass, C. (2000). J. Med. Genet. 37, 501–509. Fry, C. J., and Peterson, C. L. (2001). Curr. Biol. 11, R185–R197. Fulda, S., Lutz, W., Schwab, M., and Debatin, K. M. (1999). Oncogene 18, 1479–1486. Gagandeep, S., Ott, M., Nisen, P. D., DePinho, R. A., and Gupta, S. (2000). J. Gene Med. 2, 117–127. Galaktionov, K., Chen, X., and Beach, D. (1996). Nature 382, 511–517. Galderisi, U., Di Bernardo, G., Cipollaro, M., Jori, F. P., Piegari, E., Cascino, A., Peluso, G., and Melone, M. A. (1999). J. Cell Biochem. 74, 313–322. Gardner, H. P., Ha, S. I., Reynolds, C., and Chodosh, L. A. (2000a). Cancer Res. 60, 5571– 5577. Gardner, H. P., Rajan, J. V., Ha, S. I., Copeland, N. G., Gilbert, D. J., Jenkins, N. A., Marquis, S. T., and Chodosh, L. A. (2000b). Genomics 63, 279–288. Garrett, M. D., and Workman, P. (1999). Eur. J. Cancer 35, 2010–2030. Gartel, A. L., Ye, X., Goufman, E., Shianov, P., Hay, N., Najmabadi, F., and Tyner, A. L. (2001). Proc. Natl. Acad. Sci. USA 98, 4510–4515. Gaubatz, S., Imhof, A., Dosch, R., Werner, O., Mitchell, P., Buettner, R., and Eilers, M. (1995). EMBO J. 14, 1508–1519. Gaubatz, S., Meichle, A., and Eilers, M. (1994). Mol. Cell. Biol. 14, 3853–3862. Ge, K., DuHadaway, J., Du, W., Herlyn, M., Rodeck, U., and Prendergast, G. C. (1999). Proc. Natl. Acad. Sci. USA 96, 9689–9694.

146

Sara K. Oster et al.

Ge, K., Duhadaway, J., Sakamuro, D., Wechsler-Reya, R., Reynolds, C., and Prendergast, G. C. (2000a). Int. J. Cancer 85, 376–383. Ge, K., Minhas, F., Duhadaway, J., Mao, N. C., Wilson, D., Buccafusca, R., Sakamuro, D., Nelson, P., Malkowicz, S. B., Tomaszewski, J., and Prendergast, G. C. (2000b). Int. J. Cancer 86, 155–161. Genestier, L., Kasibhatla, S., Brunner, T., and Green, D. R. (1999). J. Exp. Med. 189, 231–239. George, R. E., and Squire, J. A. (2000). In “Neuroblastoma” (G. M. Brodeur, R. Sawada, Y. Tsuchida, and P. A. Voute, eds.), pp. 85–100. Elsevier, New York. Gilhuis, H. J., Anderl, K. L., Boerman, R. H., Jeuken, J. M., James, C. D., Raffel, C., Scheithauer, B. W., and Jenkins, R. B. (2000). Clin. Neurol. Neurosurg. 102, 203–209. Glockner, S., Lehmann, U., Wilke, N., Kleeberger, W., Langer, F., and Kreipe, H. (2000). Pathobiology 68, 173–179. Gonda, T. J., and Metcalf, D. (1984). Nature 310, 249–251. Grandori, C., Cowley, S. M., James, L. P., and Eisenman, R. N. (2000). Annu. Rev. Cell. Dev. Biol. 16, 653–699. Grandori, C., Mac, J., Siebelt, F., Ayer, D. E., and Eisenman, R. N. (1996). EMBO J. 15, 4344–4357. Grant, P. A., Schieltz, D., Pray-Grant, M. G., Yates, J. R. 3rd, and Workman, J. L. (1998). Mol. Cell 2, 863–867. Greasley, P. J., Bonnard, C., and Amati, B. (2000). Nuc. Acids Res. 28, 446–453. Greenberg, R. A., O’Hagan, R. C., Deng, H., Xiao, Q., Hann, S. R., Adams, R. R., Lichtsteiner, S., Chin, L., Morin, G. B., and DePinho, R. A. (1999). Oncogene 18, 1219–1226. Gregory, M. A., and Hann, S. R. (2000). Mol. Cell. Biol. 20, 2423–2435. Gregory, P. D., Wagner, K., and Horz, W. (2001). Exp. Cell Res. 265, 195–202. Gross-Mesilaty, S., Reinstein, E., Bercovich, B., Tobias, K. E., Schwartz, A. L., Kahana, C., and Ciechanover, A. (1998). Proc. Natl. Acad. Sci. USA 95, 8058–8063. Gu, W., Bhatia, K., Magrath, I. T., Dang, C. V., and Dalla-Favera, R. (1994). Science 264, 251–254. Guenette, R. S., Daehlin, L., Mooibroek, M., Wong, K., and Tenniswood, M. (1994). J. Androl. 15, 200–211. Guo, Q., Xie, J., Dang, C. V., Liu, E. T., and Bishop, J. M. (1998). Proc. Natl. Acad. Sci. USA 95, 9172–9177. Guo, Q. M., Malek, R. L., Kim, S., Chiao, C., He, M., Ruffy, M., Sanka, K., Lee, N. H., Dang, C. V., and Liu, E. T. (2000). Cancer Res. 60, 5922–5928. Haas, K., Staller, P., Geisen, C., Bartek, J., Eilers, M., and Moroy, T. (1997). Oncogene 15, 179–192. Hakem, R., Hakem, A., Duncan, G. S., Henderson, J. T., Woo, M., Soengas, M. S., Elia, A., de la Pompa, J. L., Kagi, D., Khoo, W., Potter, J., Yoshida, R., Kaufman, S. A., Lowe, S. W., Penninger, J. M., and Mak, T. W. (1998). Cell 94, 339–352. Han, I., and Kudlow, J. E. (1997). Mol. Cell. Biol. 17, 2550–2558. Hanahan, D., and Folkman, J. (1996). Cell 86, 353–364. Hanahan, D., and Weinberg, R. A. (2000). Cell 100, 57–70. Hann, S. R., Dixit, M., Sears, R. C., and Sealy, L. (1994). Genes Dev. 8, 2441–2452. Hann, S. R., Thompson, C. B., and Eisenman, R. N. (1985). Nature 314, 366–369. Harrington, E. A., Bennett, M. R., Fanidi, A., and Evan, G. I. (1994). EMBO J. 13, 3286–3295. Hateboer, G., Timmers, H. T., Rustgi, A. K., Billaud, M., van’t Veer, L. J., and Bernards, R. (1993). Proc. Natl. Acad. Sci. USA 90, 8489–8493. Hecht, J. L., and Aster, J. C. (2000). J. Clin. Oncol. 18, 3707–3721. Heikkila, R., Schwab, G., Wickstrom, E., Loke, S. L., Pluznik, D. H., Watt, R., and Neckers, L. M. (1987). Nature 328, 445–449. Henriksson, M., and Luscher, B. (1996). Adv. Cancer Res. 68, 109–182.

The myc Oncogene

147

Henriksson, M., Bakardjiev, A., Klein, G., and Luscher, B. (1993). Oncogene 8, 3199–3209. Henriksson, M., Selivanova, G., Lindstrom, M., and Wiman, K. G. (2001). Apoptosis 6, 133– 137. Hermeking, H., Rago, C., Schuhmacher, M., Li, Q., Barrett, J. F., Obaya, A. J., O’Connell, B. C., Mateyak, M. K., Tam, W., Kohlhuber, F., Dang, C. V., Sedivy, J. M., Eick, D., Vogelstein, B., and Kinzler, K. W. (2000). Proc. Natl. Acad. Sci. USA 97, 2229–2234. Herms, J., Neidt, I., Luscher, B., Sommer, A., Schurmann, P., Schroder, T., Bergmann, M., Wilken, B., Probst-Cousin, S., Hernaiz-Driever, P., Behnke, J., Hanefeld, F., Pietsch, T., and Kretzschmar, H. A. (2000). Int. J. Cancer 89, 395–402. Hoang, A. T., Lutterbach, B., Lewis, B. C., Yano, T., Chou, T. Y., Barrett, J. F., Raffeld, M., Hann, S. R., and Dang, C. V. (1995). Mol. Cell. Biol. 15, 4031–4042. Hogarty, M. D., and Brodeur, G. M. (1999). Int. J. Cancer 80, 630–631. Hogarty, M. D., Liu, X., Thompson, P. M., White, P. S., Sulman, E. P., Maris, J. M., and Brodeur, G. M. (2000). Med. Pediatr. Oncol. 35, 559–562. Huang, H., Colella, S., Kurrer, M., Yonekawa, Y., Kleihues, P., and Ohgaki, H. (2000). Cancer Res. 60, 6868–6874. Hudziak, R. M., Summerton, J., Weller, D. D., and Iversen, P. L. (2000). Antisense Nucleic Acid Drug Dev. 10, 163–176. Hueber, A. O., and Evan, G. I. (1998). Trends Genet. 14, 364–367. Humbert, P. O., Verona, R., Trimarchi, J. M., Rogers, C., Dandapani, S., and Lees, J. A. (2000). Genes Dev. 14, 690–703. Hurlin, P. J., Foley, K. P., Ayer, D. E., Eisenman, R. N., Hanahan, D., and Arbeit, J. M. (1995). Oncogene 11, 2487–2501. Ikura, T., Ogryzko, V. V., Grigoriev, M., Groisman, R., Wang, J., Horikoshi, M., Scully, R., Qin, J., and Nakatani, Y. (2000). Cell 102, 463–473. Inghirami, G., Grignani, F., Sternas, L., Lombardi, L., Knowles, D. M., and Dalla-Favera, R. (1990). Science 250, 682–686. Iritani, B. M., and Eisenman, R. N. (1999). Proc. Natl. Acad. Sci. USA 96, 13180–13185. Ishida, S., Huang, E., Zuzan, H., Spang, R., Leone, G., West, M., and Nevins, J. R. (2001). Mol. Cell. Biol. 21, 4684–4699. Iyengar, R. V., Pawlik, C. A., Krull, E. J., Phelps, D. A., Burger, R. A., Harris, L. C., Potter, P. M., and Danks, M. K. (2001). Cancer Res. 61, 3045–3052. Izumi, H., Molander, C., Penn, L. Z., Ishisaki, A., Kohno, K., and Funa, K. (2001). J. Cell Sci. 114, 1533–1544. Jacobs, J. J., Keblusek, P., Robanus-Maandag, E., Kristel, P., Lingbeek, M., Nederlof, P. M., van Welsem, T., van de Vijver, M. J., Koh, E. Y., Daley, G. Q., and van Lohuizen, M. (2000). Nat. Genet. 26, 291–299. Jacobs, J. J., Scheijen, B., Voncken, J. W., Kieboom, K., Berns, A., and van Lohuizen, M. (1999). Genes Dev. 13, 2678–2690. Jamerson, M. H., Johnson, M. D., and Dickson, R. B. (2000). Oncogene 19, 1065–1071. Janocko, L. E., Brown, K. A., Smith, C. A., Gu, L. P., Pollice, A. A., Singh, S. G., Julian, T., Wolmark, N., Sweeney, L., Silverman, J. F., and Shackney, S. E. (2001). Cytometry 46, 136–149. Janoueix-Lerosey, I., Penther, D., Thioux, M., de Cremoux, P., Derre, J., Ambros, P., Vielh, P., Benard, J., Aurias, A., and Delattre, O. (2000). Genes Chromosomes Cancer 28, 276–284. Jansen-Durr, P., Meichle, A., Steiner, P., Pagano, M., Finke, K., Botz, J., Wessbecher, J., Draetta, G., and Eilers, M. (1993). Proc. Natl. Acad. Sci. USA 90, 3685–3689. Janz, A., Sevignani, C., Kenyon, K., Ngo, C. V., and Thomas-Tikhonenko, A. (2000). Nucleic Acids Res. 28, 2268–2275. Johansen, L. M., Iwama, A., Lodie, T. A., Sasaki, K., Felsher, D. W., Golub, T. R., and Tenen, D. G. (2001). Mol. Cell. Biol. 21, 3789–3806.

148

Sara K. Oster et al.

Johnston, L. A., Prober, D. A., Edgar, B. A., Eisenman, R. N., and Gallant, P. (1999). Cell 98, 779–790. Johnstone, R. W., and Trapani, J. A. (1999). Mol. Cell. Biol. 19, 5833–5838. Jones, S. M., and Kazlauskas, A. (2001). Nat. Cell Biol. 3, 165–172. Jones, R. M., Branda, J., Johnston, K. A., Polymenis, M., Gadd, M., Rustgi, A., Callanan, L., and Schmidt, E. V. (1996). Mol. Cell. Biol. 16, 4754–4764. Judware, R., and Culp, L. A. (1995). Oncogene 11, 2599–2607. Judware, R., and Culp, L. A. (1997). Oncogene 14, 1341–1350. Juin, P., Hueber, A. O., Littlewood, T., and Evan, G. (1999). Genes Dev. 13, 1367–1381. Jung, R., Soondrum, K., and Neumaier, M. (2000). Clin. Chem. Lab. Med. 38, 833–836. Junghans, M., Kreuter, J., and Zimmer, A. (2000). Nucleic Acids Res. 28, E45. Kaddurah-Daouk, R., Greene, J. M., Baldwin, A. S. Jr., and Kingston, R. E. (1987). Genes Dev. 1, 347–357. Kaltz-Wittmer, C., Klenk, U., Glaessgen, A., Aust, D. E., Diebold, J., Lohrs, U., and Baretton, G. B. (2000). Lab. Invest. 80, 1455–1464. Kasibhatla, S., Beere, H. M., Brunner, T., Echeverri, F., and Green, D. R. (2000). Curr. Biol. 10, 1205–1208. Kato, G. J., Barrett, J., Villa-Garcia, M., and Dang, C. V. (1990). Mol. Cell. Biol. 10, 5914– 5920. Kauffmann-Zeh, A., Rodriguez-Viciana, P., Ulrich, E., Gilbert, C., Coffer, P., Downward, J., and Evan, G. (1997). Nature 385, 544–548. Kelly, K., and Siebenlist, U. (1986). Annu. Rev. Immunol. 4, 317–338. Kim, S., Li, Q., Dang, C. V., and Lee, L. A. (2000). Proc. Natl. Acad. Sci. USA 97, 11198–11202. Kim, S., Zeller, K., Dang, C. V., Sandgren, E. P., and Lee, L. A. (2001). Anal. Biochem. 288, 141–148. Kingston, R. E., Baldwin, A. S. Jr., and Sharp, P. A. (1984). Nature 312, 280–282. Kitaura, H., Shinshi, M., Uchikoshi, Y., Ono, T., Iguchi-Ariga, S. M., and Ariga, H. (2000). J. Biol. Chem. 275, 10477–10483. Klein, G. (1999). Curr. Top. Microbiol. Immunol. 246, 161–167. Kolligs, F. T., Kolligs, B., Hajra, K. M., Hu, G., Tani, M., Cho, K. R., and Fearon, E. R. (2000). Genes Dev. 14, 1319–1331. Kondoh, M., Matsuyama, T., Suzuki, R., Mizuguchi, H., Nakanishi, T., Nakagawa, S., Tsutsumi, Y., Nakanishi, M., Sato, M., and Mayumi, T. (2000). Biol. Pharm. Bull. 23, 1011– 1013. Kraehn, G. M., Utikal, J., Udart, M., Greulich, K. M., Bezold, G., Kaskel, P., Leiter, U., and Peter, R. U. (2001). Br. J. Cancer 84, 72–79. Kuida, K., Haydar, T. F., Kuan, C. Y., Gu, Y., Taya, C., Karasuyama, H., Su, M. S., Rakic, P., and Flavell, R. A. (1998). Cell 94, 325–337. Kumamoto, H., Kinouchi, Y., and Ooya, K. (2001). J. Oral Pathol. Med. 30, 231–236. Land, H., Parada, L. F., and Weinberg, R. A. (1983). Nature 304, 596–602. Lang, S. E., McMahon, S. B., Cole, M. D., and Hearing, P. (2001). J. Biol. Chem. 276, 32627– 32634. Larsson, L. G., Pettersson, M., Oberg, F., Nilsson, K., and Luscher, B. (1994). Oncogene 9, 1247–1252. Lasorella, A., Noseda, M., Beyna, M., and Iavarone, A. (2000). Nature 407, 592–598. Latil, A., Vidaud, D., Valeri, A., Fournier, G., Vidaud, M., Lidereau, R., Cussenot, O., and Biache, I. (2000). Int. J. Cancer 89, 172–176. Lee, L. A., Dolde, C., Barrett, J., Wu, C. S., and Dang, C. V. (1996). J. Clin. Invest. 97, 1687– 1695. Lee, T. C., Li, L., Philipson, L., and Ziff, E. B. (1997). Proc. Natl. Acad. Sci. USA 94, 12886– 12891.

The myc Oncogene

149

Lemaitre, J. M., Bocquet, S., Buckle, R., and Mechali, M. (1995). Mol. Cell. Biol. 15, 5054– 5062. Lemaitre, J. M., Buckle, R. S., and Mechali, M. (1996). Adv. Cancer Res. 70, 95–144. Lenardo, M., Rustgi, A. K., Schievella, A. R., and Bernards, R. (1989). EMBO J. 8, 3351–3355. Leone, G., DeGregori, J., Sears, R., Jakoi, L., and Nevins, J. R. (1997). Nature 387, 422–426. Lewis, B. C., Prescott, J. E., Campbell, S. E., Shim, H., Orlowski, R. Z., and Dang, C. V. (2000). Cancer Res. 60, 6178–6183. Lewis, B. C., Shim, H., Li, Q., Wu, C. S., Lee, L. A., Maity, A., and Dang, C. V. (1997). Mol. Cell. Biol. 17, 4967–4978. Li, Q., and Dang, C. V. (1999). Mol. Cell. Biol. 19, 5339–5351. Li, H., Kolluri, S. K., Gu, J., Dawson, M. I., Cao, X., Hobbs, P. D., Lin, B., Chen, G., Lu, J., Lin, F., Xie, Z., Fontana, J. A., Reed, J. C., and Zhang, X. (2000). Science 289, 1159– 1164. Li, L. H., Nerlov, C., Prendergast, G., MacGregor, D., and Ziff, E. B. (1994). EMBO J. 13, 4070–4079. Liao, D. J., and Dickson, R. B. (2000). Endocr. Relat. Cancer 7, 143–164. Littlewood, T. D., Hancock, D. C., Danielian, P. S., Parker, M. G., and Evan, G. I. (1995). Nucleic Acids Res. 23, 1686–1690. Liu, H., Lo, C. R., Jones, B. E., Pradhan, Z., Srinivasan, A., Valentino, K. L., Stockert, R. J., and Czaja, M. J. (2000). J. Biol. Chem. 275, 40155–40162. Livezey, K. W., Negorev, D., and Simon, D. (2000). Cancer Genet. Cytogenet. 116, 35–39. Lowe, S.W., and Lin, A.W. (2000). Carcinogenesis 21, 485–495. Luscher, B., and Eisenman, R. N. (1992). J. Cell. Biol. 118, 775–784. Luscher, B., and Larsson, L. G. (1999). Oncogene 18, 2955–2966. Lutterbach, B., and Hann, S. R. (1994). Mol. Cell. Biol. 14, 5510–5522. Lutterbach, B., and Hann, S. R. (1997). Oncogene 14, 967–975. Lutterbach, B., and Hann, S. R. (1999). J. Cell. Biochem. 72, 483–491. Madge, B., and Schwab, M. (2001). Submitted for publication. Maestro, R., Dei Tos, A. P., Hamamori, Y., Krasnokutsky, S., Sartorelli, V., Kedes, L., Doglioni, C., Beach, D. H., and Hannon, G. J. (1999). Genes Dev. 13, 2207–2217. Maheswaran, S., Lee, H., and Sonenshein, G. E. (1994). Mol. Cell. Biol. 14, 1147–1152. Mai, S., and Jalava, A. (1994). Nucleic Acids Res. 22, 2264–2273. Mai, S., and Martensson, I. L. (1995). Nucleic Acids Res. 23, 1–9. Mai, S., Hanley-Hyde, J., Rainey, G. J., Kuschak, T. I., Paul, J. T., Littlewood, T. D., Mischak, H., Stevens, L. M., Henderson, D. W., and Mushinski, J. F. (1999). Neoplasia 1, 241–252. Mantovani, R. (1999). Gene 239, 15–27. Mao, N. C., Steingrimsson, E., DuHadaway, J., Wasserman, W., Ruiz, J. C., Copeland, N. G., Jenkins, N. A., and Prendergast, G. C. (1999). Genomics 56, 51–58. Marhin, W. W., Chen, S., Facchini, L. M., Fornace, A. J., Jr., and Penn, L. Z. (1997). Oncogene 14, 2825–2834. Mateyak, M. K., Obaya, A. J., Adachi, S., and Sedivy, J. M. (1997). Cell Growth Differentiation 8, 1039–1048. Mathew, P., Valentine, M. B., Bowman, L. C., Rowe, S. T., Nash, M. B., Valentine, V. A., Cohn, S. L., Castleberry, R. P., Brodeur, G. M., and Look, A. T. (2001). Neoplasia 3, 105–109. Matthay, K. K. (2000). J. Clin. Oncol. 18, 3591–3594. McArthur, G. A., Laherty, C. D., Queva, C., Hurlin, P. J., Loo, L., James, L., Grandori, C., Gallant, P., Shiio, Y., Hokanson, W. C., Bush, A. C., Cheng, P. F., Lawrence, Q. A., Pulverer, B., Koskinen, P. J., Foley, K. P., Ayer, D. E., and Eisenman, R. N. (1998). Cold Spring Harb. Symp. Quant. Biol. 63, 423–433. McEwan, I. J., Dahlman-Wright, K., Ford, J., and Wright, A. P. (1996). Biochemistry 35, 9584– 9593.

150

Sara K. Oster et al.

McGufﬁe, E. M., Pacheco, D., Carbone, G. M., and Catapano, C. V. (2000). Cancer Res. 60, 3790–3799. McMahon, S. B., Van Buskirk, H. A., Dugan, K. A., Copeland, T. D., and Cole, M. D. (1998). Cell 94, 363–374. McMahon, S. B., Wood, M. A., and Cole, M. D. (2000). Mol. Cell. Biol. 20, 556–562. Miltenberger, R. J., Sukow, K. A., and Farnham, P. J. (1995). Mol. Cell. Biol. 15, 2527–2535. Minard, V., Hartmann, O., Peyroulet, M. C., Michon, J., Coze, C., Defachelle, A. S., Lejars, O., Perel, Y., Bergeron, C., Boutard, P., Leverger, G., Stephan, J. L., Thyss, A., Chastagner, P., Couillault, G., Devalck, C., Lutz, P., Mechinaud, F., Millot, F., Plantaz, D., Rialland, X., and Rubie, H. (2000). Br. J. Cancer 83, 973–979. Minn, A. J., Boise, L. H., and Thompson, C. B. (1996). Genes Dev. 10, 2621–2631. Mitas, M., Mikhitarian, K., Walters, C., Baron, P. L., Elliott, B. M., Brothers, T. E., Robison, J. G., Metcalf, J. S., Palesch, Y. Y., Zhang, Z., Gillanders, W. E., and Cole, D. J. (2001). Int. J. Cancer 93, 162–171. Mitchell, K. O., Ricci, M. S., Miyashita, T., Dicker, D. T., Jin, Z., Reed, J. C., and El-Deiry, W. S. (2000). Cancer Res. 60, 6318–6325. Morgenbesser, S. D., and DePinho, R. A. (1994). Semin. Cancer Biol. 5, 21–36. Mori, K., Maeda, Y., Kitaura, H., Taira, T., Iguchi-Ariga, S. M. M., and Ariga, H. (1998). J. Biol. Chem. 273, 29794–29800. Moser, M., Pscherer, A., Roth, C., Becker, J., Mucher, G., Zerres, K., Dixkens, C., Weis, J., Guay-Woodford, L., Buettner, R., and Fassler, R. (1997). Genes Dev. 11, 1938–1948. Moshier, J. A., Dosescu, J., Skunca, M., and Luk, G. D. (1993). Cancer Res. 53, 2618–2622. Muller, H., Bracken, A. P., Vernell, R., Moroni, M. C., Christians, F., Grassilli, E., Prosperini, E., Vigo, E., Oliner, J. D., and Helin, K. (2001). Genes Dev. 15, 267–285. Neiman, P. E., Ruddell, A., Jasoni, C., Loring, G., Thomas, S. J., Brandvold, K. A., Lee, R., Burnside, J., and Delrow, J. (2001). Proc. Natl. Acad. Sci. USA 98, 6378–6383. Nesbit, C. E., Tersak, J. M., Grove, L. E., Drzal, A., Choi, H., and Prochownik, E. V. (2000). Oncogene 19, 3200–3212. Nesbit, C. E., Tersak, J. M., and Prochownik, E. V. (1999). Oncogene 18, 3004–3016. Ngo, C. V., Gee, M., Akhtar, N., Yu, D., Volpert, O., Auerbach, R., and Thomas-Tikhonenko, A. (2000). Cell Growth Differ. 11, 201–210. Niki, T., Izumi, S., Saegusa, Y., Taira, T., Takai, T., Iguchi-Ariga, S. M. M., and Ariga, H. (2000). Genes Cells. 5, 127–141. Nikiforov, M. A., Kotenko, I., Petrenko, O., Beavis, A., Valenick, L., Lemischka, I., and Cole, M. D. (2000). Oncogene 19, 4828–4831. Niklinski, J., Claassen, G., Meyers, C., Gregory, M. A., Allegra, C. J., Kaye, F. J., Hann, S. R., and Zajac-Kaye, M. (2000). Mol. Cell. Biol. 20, 5276–5284. Nishino, K., Osaki, T., Kumagai, T., Kijima, T., Tachibana, I., Goto, H., Arai, T., Kimura, H., Funakoshi, T., Takeda, Y., Tanio, Y., and Hayashi, S. (2001). Int. J. Cancer 91, 851–856. Noguchi, K., Kitanaka, C., Yamana, H., Kokubu, A., Mochizuki, T., and Kuchino, Y. (1999). J. Biol. Chem. 274, 32580–32587. Noguchi, K., Kokubu, A., Kitanaka, C., Ichijo, H., and Kuchino, Y. (2001). Biochem. Biophys. Res. Commun. 281, 1313–1320. Obara, K., Yokoyama, M., Asano, G., and Tanaka, S. (2001). Int. J. Oncol. 18, 233–239. Obaya, A. J., Mateyak, M. K., and Sedivy, J. M. (1999). Oncogene 18, 2934–2941. Oberg, F., Wu, S., Bahram, F., Nilsson, K., and Larsson, L. G. (2001). Leukemia 15, 217–227. Oh, S., Song, Y. H., Kim, U. J., Yim, J., and Kim, T. K. (1999). Biochem. Biophys. Res. Commun. 263, 361–365. O’Hagan, R. C., Schreiber-Agus, N., Chen, K., David, G., Engelman, J. A., Schwab, R., Alland, L., Thomson, C., Ronning, D. R., Sacchettini, J. C., Meltzer, P., and DePinho, R. A. (2000). Nat. Genet. 24, 113–119.

The myc Oncogene

151

Okano, H. J., Park, W. Y., Corradi, J. P., and Darnell, R. B. (1999). Genes Dev. 13, 2087–2097. Oster, S. K., Marhin, W. W., Asker, C., Facchini, L. M., Dion, P. A., Funa, K., Post, M., Sedivy, J. M., and Penn, L. Z. (2000). Mol. Cell Biol. 20, 6768–6778. Padilla-Nash, H. M., Heselmeyer-Haddad, K., Wangsa, D., Zhang, H., Ghadimi, B. M., Macville, M., Augustus, M., Schrock, E., Hilgenfeld, E., and Ried, T. (2001). Genes Chromosomes Cancer 30, 349–363. Pagnano, K. B., Vassallo, J., Lorand-Metze, I., Costa, F. F., and Saad, S. T. (2001). Am. J. Hematol. 67, 84–92. Park, D. S., Razani, B., Lasorella, A., Schreiber-Agus, N., Pestell, R. G., Iavarone, A., and Lisanti, M. P. (2001). Biochemistry 40, 3354–3362. Park, J., Wood, M. A., and Cole, M. D. (2001b). Submitted. Patterson, A., and Harris, A. L. (1999). Drugs Aging 14, 75–90. Pawlik, C. A., Iyengar, R. V., Krull, E. J., Mason, S. E., Khanna, R., Harris, L. C., Potter, P. M., Danks, M. K., and Guichard, S. M. (2000). Mol. Ther. 1, 457–463. Pelengaris, S., Littlewood, T., Khan, M., Elia, G., and Evan, G. (1999). Mol. Cell 3, 565–577. Pelengaris, S., Rudolph, B., and Littlewood, T. (2000). Curr. Opin. Genet. Dev. 10, 100– 105. Penn, L. Z. (2001). Curr. Opin. Invest. Drugs 2, 648–692. Penn, L. J., Brooks, M. W., Laufer, E. M., and Land, H. (1990). EMBO J. 9, 1113–1121. Perez-Roger, I., Kim, S. H., Grifﬁths, B., Sewing, A., and Land, H. (1999). EMBO J. 18, 5310– 5320. Perez-Roger, I., Solomon, D. L., Sewing, A., and Land, H. (1997). Oncogene 14, 2373–2381. Pescarolo, M. P., Bagnasco, L., Malacarne, D., Melchiori, A., Valente, P., Millo, E., Bruno, S., Basso, S., and Parodi, S. (2001). FASEB J. 15, 31–33. Peters, M. A., and Taparowsky, E. J. (1998). Crit. Rev. Eukaryot. Gene Expr. 8, 277–296. Peukert, K., Staller, P., Schneider, A., Carmichael, G., Hanel, F., and Eilers, M. (1997). EMBO J. 16, 5672–5686. Philipp, A., Schneider, A., Vasrik, I., Finke, K., Xiong, Y., Beach, D., Alitalo, K., and Eilers, M. (1994). Mol. Cell. Biol. 14, 4032–4043. Pich, A., Margaria, E., and Chiusa, L. (2000). J. Clin. Oncol. 18, 2948–2956. Poremba, C., Scheel, C., Hero, B., Christiansen, H., Schaefer, K. L., Nakayama, J., Berthold, F., Juergens, H., Boecker, W., and Dockhorn-Dworniczak, B. (2000). J. Clin. Oncol. 18, 2582–2592. Prendergast, G. C. (1999). Oncogene 18, 2967–2987. Prober, D. A., and Edgar, B. A. (2001). Curr. Opin. Genet. Dev. 11, 19–26. Pulverer, B. J., Fisher, C., Vousden, K., Littlewood, T., Evan, G., and Woodgett, J. R. (1994). Oncogene 9, 59–70. Pusch, O., Soucek, T., Hengstschlager-Ottnad, E., Bernaschek, G., and Hengstschlager, M. (1997). DNA Cell Biol. 16, 737–747. Queva, C., Hurlin, P. J., Foley, K. P., and Eisenman, R. N. (1998). Oncogene 16, 967–977. Rabbitts, T. H. (1985). Trends Genet. 1, 327–331. Raveh, T., Droguett, G., Horwitz, M. S., DePinho, R. A., and Kimchi, A. (2001). Nat. Cell Biol. 3, 1–7. Reardon, D. A., Jenkins, J. J., Sublett, J. E., Burger, P. C., and Kun, L. K. (2000). Pediatr. Neurosurg. 32, 187–191. Reisman, D., Elkind, N. B., Roy, B., Beamon, J., and Rotter, V. (1993). Cell Growth Differ. 4, 57–65. Ricaniadis, N., Kataki, A., Agnantis, N., Androulakis, G., and Karakousis, C. P. (2001). Eur. J. Surg. Oncol. 27, 88–93. Roberts, S. G. (2000). Cell. Mol. Life Sci. 57, 1149–1160. Roos, M. D., Su, K., Baker, J. R., and Kudlow, J. E. (1997). Mol. Cell. Biol. 17, 6472–6480.

152

Sara K. Oster et al.

Rosenwald, I. B., Rhoads, D. B., Callanan, L. D., Isselbacher, K. J., and Schmidt, E. V. (1993). Proc. Natl. Acad. Sci. USA 90, 6175–6178. Rossler, J., Schwab, M., Havers, W., and Schweigerer, L. (2001). Biochem. Biophys. Res. Commun. 281, 272–276. Roy, B., Beamon, J., Balint, E., and Reisman, D. (1994). Mol. Cell. Biol. 14, 7805–7815. Roy, A. L., Carruthers, C., Gutjahr, T., and Roeder, R. G. (1993). Nature 365, 359–361. Ruley, H. E. (1990). Cancer Cells 2, 258–268. Rustgi, A. K., Dyson, N., and Bernards, R. (1991). Nature 352, 541–544. Sagawa, Y., Nishi, H., Isaka, K., Fujito, A., and Takayama, M. (2001). Cancer Lett. 168, 45–50. Sakamuro, D., and Prendergast, G. C. (1999). Oncogene 18, 2942–2954. Sakamuro, D., Elliott, K. J., Wechsler-Reya, R., and Prendergast, G. C. (1996). Nat. Genet. 14, 69–77. Salghetti, S. E., Kim, S. Y., and Tansey, W. P. (1999). EMBO J. 18, 717–726. Santoni-Rugiu, E., Falck, J., Mailand, N., Bartek, J., and Lukas, J. (2000). Mol. Cell Biol. 20, 3497–3509. Sarbia, M., Arjumand, J., Wolter, M., Reifenberger, G., Heep, H., and Gabbert, H. E. (2001). Am. J. Clin. Pathol. 115, 835–840. Sasaki, H., Sato, Y., Kondo, S., Fukai, I., Kiriyama, M., Yamakawa, Y., and Fujii, Y. (2001). Cancer Lett. 168, 191–195. Scheurlen, W. G., Schwabe, G. C., Joos, S., Mollenhauer, J., Sorensen, N., and Kuhl, J. (1998). J. Clin. Oncol. 16, 2478–2485. Schmidt, E. V. (1999). Oncogene 18, 2988–2996. Schmittgen, T. D., Zakrajsek, B. A., Mills, A. G., Gorn, V., Singer, M. J., and Reed, M. W. (2000). Anal. Biochem. 285, 194–204. Schmutzler, C., and Koehrle, J. (2000). Eur. J. Endocrinol. 143, 15–24. Schneider, A., Peukert, K., Eilers, M., and Hanel, F. (1997). Curr. Top. Microbiol. Immunol. 224, 137–146. Schuhmacher, M., Kohlhuber, F., Holzel, M., Kaiser, C., Burtscher, H., Jarsch, M., Bornkamm, G. W., Laux, G., Polack, A., Weidle, U. H., and Eick, D. (2001). Nucleic Acids Res. 29, 397–406. Schuhmacher, M., Staege, M. S., Pajic, A., Polack, A., Weidle, U. H., Bornkamm, G. W., Eick, D., and Kohlhuber, F. (1999). Curr. Biol. 9, 1255–1258. Sears, R., Leone, G., DeGregori, J., and Nevins, J. R. (1999). Mol. Cell 3, 169–179. Sears, R., Nuckolls, F., Haura, E., Taya, Y., Tamai, K., and Nevins, J. R. (2000). Genes Dev. 14, 2501–2514. Sears, R., Ohtani, K., and Nevins, J. R. (1997). Mol. Cell. Biol. 17, 5227–5235. Seth, A., Gonzalez, F. A., Gupta, S., Raden, D. L., and Davis, R. J. (1992). J. Biol. Chem. 267, 24796–24804. Seth, A., Gupta, S., and Davis, R. J. (1993). Mol. Cell. Biol. 13, 4125–4136. Shaw, P., Freeman, J., Bovey, R., and Iggo, R. (1996). Oncogene 12, 921–930. Shim, H., Dolde, C., Lewis, B. C., Wu, C. S., Dang, G., Jungmann, R. A., Dalla-Favera, R., and Dang, C. V. (1997). Proc. Natl. Acad. Sci. USA 94, 6658–6663. Shimono, A., Okuda, T., and Kondoh, H. (1999). Mech. Dev. 83, 39–52. Shrivastava, A., Saleque, S., Kalpana, G. V., Artandi, S., Goff, S. P., and Calame, K. (1993). Science 262, 1889–1892. Shrivastava, A., Yu, J., Artandi, S., and Calame, K. (1996). Proc. Natl. Acad. Sci. USA 93, 10638–10641. Sigvardsson, M., Johansson, K., Larsson, L. G., Nilsson, K., and Leanderson, T. (1994). Leukemia 8, 1157–1163. Soucie, E. L., Annis, M. G., Sedivy, J., Filmus, J., Leber, B., Andrews, D. W., and Penn, L. Z. (2001). Mol. Cell Biol. 21, 4725–4736.

The myc Oncogene

153

Spencer, C. A., and Groudine, M. (1991). Adv. Cancer Res. 56, 1–48. Spotts, G. D., Patel, S. V., Xiao, Q., and Hann, S. R. (1997). Mol. Cell. Biol. 17, 1459–1468. Squire, J. A., Thorner, P., Marrano, P., Parkinson, D., Ng, Y. K., Gerrie, B., Chilton-Macneill, S., and Zielenska, M. (1996). Mol. Diagn. 1, 281–289. Staller, P., Peukert, K., Kiermaier, A., Seoane, J., Lukas, J., Karsunky, H., Moroy, T., Bartek, J., Massague, J., Hanel, F., and Eilers, M. (2001). Nat. Cell Biol. 3, 392–399. Stocker, H., and Hafen, E. (2000). Curr. Opin. Genet. Dev. 10, 529–535. Stone, J., de Lange, T., Ramsay, G., Jakobovits, E., Bishop, J. M., Varmus, H., and Lee, W. (1987). Mol. Cell. Biol. 7, 1697–1709. Strange, R., Li, F., Saurer, S., Burkhardt, A., and Friis, R. R. (1992). Development 115, 49–58. Streutker, C. J., Thorner, P., Fabricius, N., Weitzman, S., and Zielenska, M. (2001). Pediatr. Dev. Pathol. 4, 62–67. Suen, T. C., and Hung, M. C. (1991). Mol. Cell. Biol. 11, 354–362. Taira, T., Maeda, J., Onishi, T., Kitaura, H., Yoshida, S., Kato, H., Ikeda, M., Tamai, K., Iguchi-Ariga, S. M., and Ariga, H. (1998). Genes Cells 3, 549–565. Taira, T., Negishi, Y., Kihara, F., Iguchi-Ariga, S. M., and Ariga, H. (1992). Biochim. Biophys. Acta 1130, 166–174. Taira, T., Sawai, M., Ikeda, M., Tamai, K., Iguchi-Ariga, S. M., and Ariga, H. (1999). J. Biol. Chem. 274, 24270–24279. Taj, M. M., Tawil, R. J., Engstrom, L. D., Zeng, Z., Hwang, C., Sanda, M. G., and Wechsler, D. S. (2001). Prostate 47, 194–204. Takayama, M., Taira, T., Iguchi-Ariga, S. M., and Ariga, H. (2000a). FEBS Lett. 477, 43–48. Takayama, M. A., Taira, T., Tamai, K., Iguchi-Ariga, S. M., and Ariga, H. (2000b). Genes Cells 5, 481–490. Taylor, C., and Mai, S. (1998). Cancer Detect. Prev. 22, 350–356. Taylor, C. P., Bown, N. P., McGuckin, A. G., Lunec, J., Malcolm, A. J., Pearson, A. D., and Sheer, D. (2000). Br. J. Cancer 83, 40–49. Teitz, T., Wei, T., Valentine, M. B., Vanin, E. F., Grenet, J., Valentine, V. A., Behm, F. G., Look, A. T., Lahti, J. M., and Kidd, V. J. (2000). Nat. Med. 6, 529–535. Thiesse, P., Hany, M. A., Combaret, V., Ranchere-Vince, D., Bouffet, E., and Bergeron, C. (2000). Eur. J. Cancer 36, 1544–1551. Thompson, E. B. (1998). Annu. Rev. Physiol. 60, 575–600. Thorner, P. S., and Squire, J. A. (1998). Pediatr. Dev. Pathol. 1, 337–365. Tikhonenko, A. T., Black, D. J., and Linial, M. L. (1996). J. Biol. Chem. 271, 30741–30747. Timme, T. L., Goltsov, A., Tahir, S., Li, L., Wang, J., Ren, C., Johnston, R. N., and Thompson, T. C. (2000). Oncogene 19, 3256–3265. Tsuneoka, M., Nakano, F., Ohgusu, H., and Mekada, E. (1997). Oncogene 14, 2301–2311. Vandesompele, J., Van Roy, N., Van Gele, M., Laureys, G., Ambros, P., Heimann, P., Devalck, C., Schuuring, E., Brock, P., Otten, J., Gyselinck, J., De Paepe, A., and Speleman, F. (1998). Genes Chromosomes Cancer 23, 141–152. Vassilev, A., Yamauchi, J., Kotani, T., Prives, C., Avantaggiati, M. L., Qin, J., and Nakatani, Y. (1998). Mol. Cell 2, 869–875. Versteeg, R., Noordermeer, I. A., Kruse-Wolters, M., Ruiter, D. J., and Schrier, P. I. (1988). EMBO J. 7, 1023–1029. Vriz, S., Lemaitre, J. M., Leibovici, M., Thierry, N., and Mechali, M. (1992). Mol. Cell. Biol. 12, 3548–3555. Wagner, A. J., Meyers, C., Laimins, L. A., and Hay, N. (1993). Cell Growth Differ. 4, 879– 883. Wakamatsu, Y., Watanabe, Y., Shimono, A., and Kondoh, H. (1993). Neuron 10, 1–9. Walhout, A. J., van der Vliet, P. C., and Timmers, H. T. (1998). Biochim. Biophys. Acta 1397, 189–201.

154

Sara K. Oster et al.

Wang, H., Liu, C., Lu, Y., Chatterjee, G., Ma, X. Y., Eisenman, R. N., and Lengyel, P. (2000). J. Biol. Chem. 275, 27377–27385. Wang, Q., Zhang, H., Kajino, K., and Greene, M. I. (1998). Oncogene 17, 1939–1948. Wang, X., Peters, M. A., Utama, F. E., Wang, Y., and Taparowsky, E. J. (1999). Mol. Endocrinol. 13, 254–267. Watatani, M., Inui, H., Nagayama, K., Imanishi, Y., Nishimura, K., Hashimoto, Y., Yamauchi, E., Hojo, T., Kotsuma, Y., Yamato, M., Matsunami, N., and Yasutomi, M. (2000). Surg. Today 30, 516–522. Wechsler-Reya, R., Sakamuro, D., Zhang, J., Duhadaway, J., and Prendergast, G. C. (1997). J. Biol. Chem. 272, 31453–31458. Wechsler-Reya, R. J., Elliott, K. J., and Prendergast, G. C. (1998). Mol. Cell. Biol. 18, 566–575. Weihua, X., Lindner, D. J., and Kalvakolanu, D. V. (1997). Proc. Natl. Acad. Sci. USA 94, 7227–7232. Weinkove, D., and Leevers, S. J. (2000). Curr. Opin. Genet. Dev. 10, 75–80. Welcsh, P. L., and King, M. C. (2001). Hum. Mol. Genet. 10, 705–713. Wells, L., Vosseller, K., and Hart, G. W. (2001). Science 291, 2376–2378. Whitehouse, I., Flaus, A., Cairns, B. R., White, M. F., Workman, J. L., and Owen-Hughes, T. (1999). Nature 400, 784–787. Woo, M., Hakem, R., Soengas, M. S., Duncan, G. S., Shahinian, A., Kagi, D., Hakem, A., McCurrach, M., Khoo, W., Kaufman, S. A., Senaldi, G., Howard, T., Lowe, S. W., and Mak, T. W. (1998). Genes Dev. 12, 806–819. Wood, L. J., Mukherjee, M., Dolde, C. E., Xu, Y., Maher, J. F., Bunton, T. E., Williams, J. B., and Resar, L. M. (2000a). Mol. Cell. Biol. 20, 5490–5502. Wood, M. A., McMahon, S. B., and Cole, M. D. (2000b). Mol. Cell 5, 321–330. Wu, K. J., Polack, A., and Dalla-Favera, R. (1999). Science 283, 676–679. Wu, M., Arsura, M., Bellas, R. E., FitzGerald, M. J., Lee, H., Schauer, S. L., Sherr, D. H., and Sonenshein, G. E. (1996). Mol. Cell. Biol. 16, 5015–5025. Xiao, Q., Claassen, G., Shi, J., Adachi, S., Sedivy, J., and Hann, S. R. (1998). Genes Dev. 12, 3803–3808. Xu, D., Popov, N., Hou, M., Wang, Q., Bjorkholm, M., Gruber, A., Menkel, A. R., and Henriksson, M. (2001). Proc. Natl. Acad. Sci. USA 98, 3826–3831. Yan, D. H., Wen, Y., Spohn, B., Choubey, D., Gutterman, J. U., and Hung, M. C. (1999). Oncogene 18, 807–811. Yang, B. S., Geddes, T. J., Pogulis, R. J., de Crombrugghe, B., and Freytag, S. O. (1991). Mol. Cell. Biol. 11, 2291–2295. Yang, S. Z., Kohno, N., Yokoyama, A., Kondo, K., Hamada, H., and Hiwada, K. (2001a). Int. J. Oncol. 18, 541–548. Yang, W., Shen, J., Wu, M., Arsura, M., FitzGerald, M., Suldan, Z., Kim, D. W., Hofmann, C. S., Pianetti, S., Romieu-Mourez, R., Freedman, L. P., and Sonenshein, G. E. (2001b). Oncogene 20, 1688–1702. Yano, T., Sander, C. A., Clark, H. M., Dolezal, M. V., Jaffe, E. S., and Raffeld, M. (1993). Oncogene 8, 2741–2748. Yin, X. Y., Grove, L., Datta, N. S., Long, M. W., and Prochownik, E. V. (1999). Oncogene 18, 1177–1184. Yudkovsky, N., Logie, C., Hahn, S., and Peterson, C. L. (1999). Genes Dev. 13, 2369–2374. Zheng, Z. H., Ye, R. G., Bergstrom, J., and Lindholm, B. (2000). Adv. Perit. Dial. 16, 31–35. Zhu, M., John, S., Berg, M., and Leonard, W. J. (1999). Cell 96, 121–130. Zindy, F., Eischen, C. M., Randle, D. H., Kamijo, T., Cleveland, J. L., Sherr, C. J., and Roussel, M. F. (1998). Genes Dev. 12, 2424–2433.

Genetic Requirements for the Episomal Maintenance of Oncogenic Herpesvirus Genomes Christopher M. Collins and Peter G. Medveczky Department of Medical Microbiology and Immunology and the H. Lee Mofﬁtt Cancer Center University of South Florida Tampa FL 33612-4799

I. Introduction II. Latent Episomal Genomes of Oncogenic Herpesviruses III. The Role of EBNA-1 and OriP in EBV Episomal Replication A. EBNA-1 General Characteristics B. OriP General Characteristics C. EBNA-1 and OriP in Episomal Replication and Maintenance D. Cellular Proteins Mediating Binding of the EBNA-1 Viral DNA Complex to Chromosomes IV. The Role of LANA1 and Terminal Repeats in Episomal Replication of KSHV and HVS A. The LANA1 Gene and Its Protein Product B. LANA1 and the KSHV Genome Colocalize in the Nucleus C. KSHV and HVS LANA1 and Terminal Repeats Are Sufficient for Episomal Replication D. HVS Terminal Repeats Are Essential for Episomal Replication of the Viral Genome E. LANA1 Is a DNA Binding Protein That Directly Interacts with the Terminal Repeats F. Is the Stable Association of the LANA1–Viral DNA Complex with Chromosomes Mediated by Cellular Proteins or by Direct Binding to Heterochromatin? V. Conclusions References

Herpesviruses are large double-stranded DNA viruses that are characterized by lifelong latency. Epstein–Barr virus (EBV), the recently discovered Kaposi’s sarcoma associated herpesvirus (KSHV), also referred to as human herpesvirus-8 (HHV-8), and the simian Herpesvirus saimiri (HVS) are associated with malignant lymphoproliferative diseases. These viruses establish latent infection in lymphoid cells. During latency only a few viral genes are expressed and the viral genome persists as a multicopy circular episome. The episome contains repetitive sequences that serve as multiple cooperative binding sites for the viral DNA binding proteins Epstein–Barr virus nuclear antigen 1 (EBNA-1) of

Advances in CANCER RESEARCH 0065-230X/02 $35.00

155

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

156

Christopher M. Collins and Peter G. Medveczky

EBV and latency-associated nuclear antigen (LANA1) of KSHV and HVS, which are expressed during latency. The oligomerized proteins associate with the viral genome and tether it to host chromosomes, assuring continual lifelong persistence of the virus. C 2002 Academic Press.

I. INTRODUCTION The ﬁrst goal of this review is to summarize the most recent and important ﬁndings on the latent episomal replication of EBV published during the past year. A comprehensive and outstanding review by Leight and Sugden on this topic has been published (Leight and Sugden, 2000). The second aim of this work is to focus on and compare episomal latent replication of KSHV and HVS with that of EBV. The eight human herpesviruses have been classiﬁed into the three subgroups α, β, and γ (Roizman et al., 1981). Gamma, or lymphotropic, herpesviruses are implicated as causative agents of malignant diseases. They have been further classiﬁed to γ -1 and γ -2 subgroups. EBV represents the γ -1 subgroup, which is also referred to as the genus Lymphocryptovirus. EBV infects epithelial and lymphoid cells (for review, see Kieff, 1996). EBV infection is common and typically asymptomatic in young children. Another, more severe manifestation of the primary infection is a self-limiting lymphoproliferative disease called infectious mononucleosis that is often seen in adolescents and young adults. EBV is also associated with several malignant diseases that develop years or decades after primary infection. Examples of EBV-associated cancers include African Burkitt’s lymphoma, Hodgkin’s disease, T cell lymphoma, nasopharyngeal and gastric carcinoma, and lymphomas in the immunosuppressed including AIDS. KSHV and HVS belong to the gamma-2 subgroup, which is also referred to as the genus Rhadinovirus. KSHV is implicated as a causative agent of Kaposi’s sarcoma, AIDS-associated primary effusion lymphoma (PEL), and a subset of cases of the lymphoproliferative disorder multicentric Castleman’s disease (for review, see Boshoff and Chang, 2001). HVS is a ubiquitous agent of squirrel monkeys (Saimiri sciureus). Virus can be reproducibly isolated from the peripheral blood of apparently healthy animals by cocultivation of T cells with owl monkey kidney cells (Melendez et al., 1968). Among the human herpesviruses, KSHV is the closest relative of HVS. Many of the HVS (open reading frames) ORFs are collinear with those of KSHV as revealed by DNA sequencing (Albrecht et al., 1992; Russo et al., 1996). Both viruses encode several genes with signiﬁcant homology to host sequences. The architecture of both viral genomes is similar as they encode a large number of terminal tandem repeats. HVS induces malignant T cell lymphomas in various species of marmosets, owl monkeys, and New Zealand White rabbits (Melendez et al., 1969).

Episomal Maintenance of Oncogenic Herpesviruses

157

II. LATENT EPISOMAL GENOMES OF ONCOGENIC HERPESVIRUSES Most studies on latency of gamma herpesviruses utilized lymphoblastoid cell lines that were derived from lymphomas or were established by in vitro infection/immortalization of peripheral blood lymphocytes. Several B lymphoblastoid cell lines have been derived from EBV-associated African Burkitt’s lymphomas. EBV, which is typically obtained from supernatants of the producer B95-8 cell line, can also immortalize peripheral blood B lymphocytes in vitro. In vivo or in vitro immortalized cell lines can be cultured indeﬁnitely. Similarly, PEL cell lines that carry latent KSHV have been successfully established (Arvanitakis et al., 1996; Nador et al., 1996; Said et al., 1996), although no in vitro immortalization of B lymphocytes has been reported yet. HVS is more suitable to experimental manipulation than EBV or KSHV because it can be cultured in monolayers of owl monkey kidney cells, yielding as many as 106 infectious virus per ml. Unlike the other two viruses, HVS immortalizes T lymphocytes rather than B lymphocytes and can immortalize them either in vivo or in vitro. Gamma herpesviruses contain terminal repeats, which permit joining of the ends of the linear virion DNA by homologous recombination after infection of lymphoid cells. After infection and circularization of the viral genome, the viral DNA of EBV, KSHV, and HVS persists in lymphoid cells in a latent episomal circular form without production of signiﬁcant amounts of infectious virus. The copy number of episomes varies from 10 to about 100. EBV and KSHV episomes are replicated by host DNA polymerases, although these agents encode their own polymerases. This is demonstrated by studies using antiviral drugs (Medveczky et al., 1997). The copy number of the episomes did not decrease after prolonged treatment with high concentrations of anti-herpesvirus drugs; however, TPA-induced linear DNA synthesis (representing unit length virion DNA) was strongly suppressed by some of these drugs (Medveczky et al., 1997; Sullivan et al., 1984).

III. THE ROLE OF EBNA-1 AND ORIP IN EBV EPISOMAL REPLICATION A. EBNA-1 General Characteristics EBNA-1 from the prototypical B95-8 strain is a 641 amino acid protein that is capable of dimerization and has DNA-sequence-speciﬁc binding activity (Ambinder et al., 1991; Baer et al., 1984). The carboxy terminus of

158

Christopher M. Collins and Peter G. Medveczky

Fig. 1 Schematic representation of EBNA-1 showing domains involved in dimerization, DNA binding and linking, chromosome binding, and nuclear localization.

EBNA-1 contains the domains responsible for DNA binding and dimerization, and DNA binding activity is dependent on dimerization of EBNA-1 (Fig. 1) (Ambinder et al., 1991; Chen et al., 1993). The crystal structure of the DNA-binding domain of EBNA-1 revealed that it consists of a core domain and a ﬂanking domain (Bochkarev et al., 1995). The core domain spans amino acids 504–604 and mediates dimerization, while the ﬂanking domain spans amino acids 461–503 and mediates DNA binding (Bochkarev et al., 1995, 1996, 1998). The core domain is very similar in structure to the DNA binding and dimerization domain of the E2 protein of bovine papillomavirus, although they share no sequence similarity (Bochkarev et al., 1995). The DNA-binding and dimerization domains alone are not sufﬁcient to support replication and actually act as dominant negative inhibitors of these functions (Kirchmaier and Sugden, 1997). A large repeat of glycine-glycine-alanine, most of which can be deleted without compromising replication function, is found in the amino terminus of EBNA-1 (Fig. 1) (Yates and Camiolo, 1988; Yates et al., 1985). This

Episomal Maintenance of Oncogenic Herpesviruses

159

domain inhibits ubiquitin/proteasome-dependent degradation, thereby inhibiting presentation of EBNA-1 derived peptides on MHC class I (Levitskaya et al., 1997). EBNA-1 contains a nuclear localization signal located between amino acids 379 and 386 (Fig. 1) (Ambinder et al., 1991) that binds the nuclear transporter karyopherin alpha 2 (Fischer et al., 1997). EBNA-1 deletion mutants unable to bind karyopherin alpha 2 are also deﬁcient in replication function (Kim et al., 1997). When a KRP to GAG mutation at amino acids 379–381 was introduced in the NLS, karyopherin alpha 2 binding was partially abrogated and EBNA-1 was partially redistributed to the cytoplasm (Kim et al., 1997). Replication function was not affected, however, indicating that sufﬁcient levels of EBNA-1 were still transported to the nucleus. That EBNA-1 is essential for latent infection has been demonstrated by an EBNA-1 knockout virus created by introducing a frameshift mutation in the EBNA 1 gene (Lee et al., 1999). EBNA-1-deﬁcient EBV could readily be isolated after superinfection of EBV positive cell lines. However, mutant virus was unable to immortalize B cells or establish latent infection.

B. OriP General Characteristics Origin for plasmid replication (oriP) was ﬁrst identiﬁed by its ability to support replication of plasmids in cells transformed by EBV (Lupton and Levine, 1985; Yates et al., 1984). OriP consists of two elements, a dyad symmetry element (DS) and a family of repeats (FR), that are separated by approximately 1 kb of DNA (Reisman et al., 1985). DS contains four lowafﬁnity EBNA-1 binding sites, two of which are in tandem and two that are in dyad symmetry (Ambinder et al., 1990; Baer et al., 1984; Rawlins et al., 1985). FR consists of 20 tandem copies of a 30 bp repeat that EBNA-1 binds to with high afﬁnity (Ambinder et al., 1990; Baer et al., 1984). The EBNA-1 binding sites in oriP are occupied during most of the cell cycle as revealed by DNA footprinting (Hsieh et al., 1993; Niller et al., 1995). In plasmids containing oriP, replication initiates at or near DS (Gahn and Schildkraut, 1989). Replication is bidirectional, but when the leftward moving replication fork reaches FR, it stalls and terminates (Gahn and Schildkraut, 1989). The fork moving in the other direction moves around the plasmid until it reaches the stalled fork, and replication is terminated. Plasmids that contain FR but not DS are not replicated (Krysan et al., 1989; Middleton and Sugden, 1994). However, Rep∗ , a 298 bp sequence between nucleotides 9370 and 9668 of the EBV genome, can support replication of plasmids containing FR but not DS at 96 hours posttransfection, but with much lower efﬁciency (Kirchmaier and Sugden, 1998). Some cellular DNAs can also support replication of plasmids lacking DS in the presence EBNA-1

160

Christopher M. Collins and Peter G. Medveczky

(Krysan and Calos, 1991; Krysan et al., 1989). Replication initiates at multiple sites within these cellular sequences. The EBV genome is replicated one time per cell cycle (Adams et al., 1989; Yates and Guan, 1991). Replication occurs in the nucleus during S phase of the cell cycle and the episomes are faithfully partitioned to daughter cells, similarly to replication of host cell chromosomes (Hampar et al., 1974; Kirchmaier and Sugden, 1995). Cellular chromosomes undergo replication licensing during G1 phase of the cell cycle, during which time a prereplicative complex is formed at replication origins, making the chromosome competent for the next round of replication (for review, see Newlon, 1997). Plasmids containing oriP have been shown to require a cell cycle window that includes early G1 phase for replication (Shirakata et al., 1999). By using sucrose density gradient centrifugation, an oriP containing plasmid was found to exist in a different state at G1 than in G2/M (Shirakata et al., 1999). These ﬁndings suggest that replication licensing may be involved in replication of the EBV genome.

C. EBNA-1 and OriP in Episomal Replication and Maintenance 1. REPLICATION OF THE EBV GENOME EBNA-1 lacks ATPase and helicase activity (Frappier and O’Donnell, 1991b; Middleton and Sugden, 1994) and has not been shown to interact with any components of the host cell replication machinery (Aiyar et al., 1998). Therefore, the contribution of EBNA-1 to replication must be indirect. EBNA-1 binds the replicative single-strand DNA-binding protein (RPA) (Zhang et al., 1998), but it is not known how this contributes to replication. EBNA-1 dimers are able to link the regions of DNA to which it speciﬁcally binds through homotypic interaction, with the intervening sequence forming a loop (Frappier et al., 1994; Frappier and O’Donnell, 1991a, 1992; Middleton and Sugden, 1992; Su et al., 1991). When EBNA-1 dimers bind to both elements of oriP, the interactions between these dimers stabilize EBNA-1 on DS (Frappier et al., 1994; Su et al., 1991). Three independent domains within EBNA-1 are able to link DNA (Fig. 1) (Mackey et al., 1995), and the interaction between EBNA-1 dimers is dependent on DNA binding (AvolioHunter and Frappier, 1998). The ability of EBNA-1 to link DNA directly correlates with its ability to support replication, but how this is accomplished is not known (Mackey and Sugden, 1999). There are two conﬂicting papers about the requirement of EBNA-1 in the replication of oriP plasmids. Aiyar et al. found that oriP plasmids were replicated in the absence of EBNA-1 for at least two rounds of replication (Aiyar et al., 1998). Furthermore, the presence of EBNA-1 increased the

Episomal Maintenance of Oncogenic Herpesviruses

161

amount of replicated oriP plasmid DNA by only 1.6-fold. These newly synthesized plasmids were rapidly lost when EBNA-1 was not present. From these ﬁndings, it was suggested that EBNA-1 did not play a role in replication, but was only involved in maintaining replicated DNA. However, Yates et al. found that EBNA-1 was required for replication, and the minimal replicator was two properly spaced EBNA-1 binding sites (Yates et al., 2000). Although it is not clear what role DS plays in replication of oriP plasmids, two experiments reveal that DS is not essential for replication of the intact EBV genome. By using 2D gel electrophoresis, Little and Schildkraut showed that replication initiates at multiple sites in addition to DS in the intact EBV genome (Little and Schildkraut, 1995). In the Raji EBV strain, replication initiated primarily in a large region extending leftward of oriP (Little and Schildkraut, 1995). Norio et al. showed that EBV mutants that lack DS are able to establish latent infection in BL30 cells (Norio et al., 2000). Initiation of replication within oriP and the stalling of replication forks that normally occurs in the vicinity of DS could not be detected. These results indicate that in the EBV genome, there are multiple sites where replication initiates, and although DS is a replication origin, it is not essential.

2. MAINTENANCE OF THE EBV GENOME Although the role that EBNA-1 plays in replication of the EBV genome is not understood, it is clear that EBNA-1 plays a role in maintenance through its interaction with FR. Plasmids that contain FR are retained at a signiﬁcantly higher level in cells that express EBNA-1 than in cells that do not (Middleton and Sugden, 1994). EBNA-1 binds mitotic chromatin (Luka et al., 1977; Ohno et al., 1977; Petti et al., 1990; Reedman and Klein, 1973), and EBV episomes are closely associated with metaphase chromosomes (Harris et al., 1985). By binding both FR and chromosomal DNA, EBNA-1 is able to tether the viral genome to the host cell chromosome, ensuring proper partitioning to daughter cells. The chromosome binding domains of EBNA-1 have been mapped to three independent locations (Fig. 1) (Marechal et al., 1999). Two experiments demonstrate that the ability of EBNA-1 to tether episomes to cellular chromosomes is essential for maintenance. Kanda et al. showed that EBNA-1 is required to tether plasmids containing FR to chromosomes, but DS was dispensable for this function (Kanda et al., 2001). OriP binding function remained intact in EBNA-1 mutants lacking amino acids 16–372, but the ability to bind chromosomes was abolished. Importantly, the ability of EBNA-1 mutants to tether episomes to chromosomes directly correlated with the ability to support replication. Hung et al. (2001) made EBNA-1 chimeras by replacing the ﬁrst 378 amino acids of EBNA-1 with either histone H1-2 or high-mobility group I amino acids 1–90, two cellular proteins that have DNA-binding characteristics

162

Christopher M. Collins and Peter G. Medveczky

similar to those of EBNA-1. Both of these tethered oriP plasmids to mitotic chromosomes and functioned similar to wild-type EBNA-1 in long-term maintenance.

D. Cellular Proteins Mediating Binding of the EBNA-1 Viral DNA Complex to Chromosomes EBNA-1 has been shown to interact with two cellular proteins that may mediate binding of the EBNA-1 viral DNA complex to host chromosome, p32/TAP (Wang et al., 1997) and EBNA-1 binding protein 2 (EBP2) (Shire et al., 1999). p32/TAP associates with both linking regions of EBNA-1 while EBP2 associates with linking region 2. EBNA-1 mutants that were unable to bind either of these proteins were also unable to maintain oriP plasmids in cell culture (Shire et al., 1999; Van Scoy et al., 2000; Wang et al., 1997). It is not known what the signiﬁcance of the interaction between EBNA-1 and p32/TAP is, but EBP2 appears to play a central role in maintenance of the EBV genome. EBNA1 mutants lacking amino acids 325–376, which have been shown to be involved in binding to mitotic chromosomes (Marechal et al., 1999), were also unable to bind EBP2. These mutants were, however, able to support replication of oriP plasmids, suggesting that it was the segregation function of EBNA-1 that was disrupted (Shire et al., 1999). Wu et al. showed that EBP2 binds to mitotic chromosomes, and the staining pattern of EBP2 was indistinguishable from that of EBNA-1 (Wu et al., 2000). EBNA-1 mutants lacking amino acids 325–376 were not able to localize to mitotic chromosomes (Wu et al., 2000), supporting the hypothesis that EBP2 mediates the binding of EBNA-1 to chromosomes. Furthermore, expression of EBNA-1 in Saccharomyces cerevisiae is not sufﬁcient to support segregation of plasmids containing FR, but when human EBP2 was introduced into the system, segregation was rescued (Kapoor et al., 2001), indicating that the interaction of EBNA-1 with EBP2 is critical for the maintenance function of EBNA-1.

IV. THE ROLE OF LANA1 AND TERMINAL REPEATS IN EPISOMAL REPLICATION OF KSHV AND HVS A. The LANA1 Gene and Its Protein Product KSHV encodes a fascinating multifunctional nuclear protein designated as the latency associated nuclear antigen, abbreviated in the literature as LANA, LNA1, or LANA1; we will use the term LANA1 in this review as a second latent protein has been identiﬁed (Rivas et al., 2001). LANA1

Episomal Maintenance of Oncogenic Herpesviruses

163

is expressed in all primary effusion lymphoma (PEL) cells and serves as a useful diagnostic tool to identify KSHV seropositive patients (Gao et al., 1996; Kedes et al., 1996; Lennette et al., 1996; Simpson et al., 1996). Sera from Kaposi’s sarcoma patients contains antibodies to this protein as shown by immunoﬂuorescence and Western blot methods (Gao et al., 1996; Kedes et al., 1996; Lennette et al., 1996; Simpson et al., 1996). Immunoﬂuorescence assays show accumulation of the LANA1 protein in a characteristic nuclear speckled pattern (Gao et al., 1996; Moore et al., 1996). LANA1 is encoded by open reading frame ORF73, which is located near the right end of the unique sequence. Three mRNA species initiating upstream of the LANA1 ORF around nucleotides 127880 and 127886 from a TATA-less promoter have been described (Kedes et al., 1997). A large unspliced and a smaller spliced transcript encode the LANA open reading frame. The third bicistronic mRNA initiating at the same locus encodes ORF72 (v-cyclin) and ORF71 (v-FLIP) (Kedes et al., 1997). The LANA1 locus is expressed under both stringently latent and lytic conditions in PEL and Kaposi’s sarcoma spindle cells (Kedes et al., 1997; Rainbow et al., 1997). As will be discussed in detail later, LANA1 directly binds the TR in vitro and in vivo and mediates stable episomal replication. Consistent with these ﬁndings, computer-assisted analysis of the LANA1 protein shows three domains encoding charged basic amino acids that could serve as DNA-binding domains (Fig. 2). The fact that there are three possible domains suggests that these domains might serve different functions, or even bind different sequences (e.g., in the TR and unique region of the KSHV genome, or cellular DNA). There are also histidine and cysteine residues clustered at the C and N termini, suggesting the presence of zinc ﬁnger DNA-binding motifs. Zinc ﬁnger motifs are also known to be present in DNA-binding proteins (Alberts et al., 1995). LANA1 also contains a long leucine zipper. Leucine

Fig. 2 Putative DNA binding domains and experimentally identiﬁed nuclear localization, dimerization and speckled nuclear concentration domains of LANA. B.D., binding domain.

164

Christopher M. Collins and Peter G. Medveczky

zipper motifs allow oligomerization of DNA-binding proteins, so LANA1 may also form oligomers with itself or heterodimers with host proteins while speciﬁcally interacting with DNA (Alberts et al., 1995). Data of Schwam et al. show that LANA1 forms dimers, and the C-terminal amino acids are required for this function in vitro (Schwam et al., 2000). The C terminus is also required for accumulation of LANA complexes in foci, causing a speckled appearance when nuclei are stained with anti-LANA1 ﬂuorescent antibodies. Both the N- and the C-terminal regions contain nuclear localization signals (Piolot et al., 2001; Rainbow et al., 1997; Russo et al., 1996). HVS encodes a positional homolog of KSHV LANA1 (Albrecht et al., 1992). Alignment of KSHV and HVS LANA1 shows a high level of conservation of the central glutamate-rich domain and several potential phosphorylation sites. HVS RNA corresponding to the LANA1 open reading frame is expressed in persistently infected cells and stringently latent T cell lines (unpublished results).

B. LANA1 and the KSHV Genome Colocalize in the Nucleus Ballestas et al. and Cotter and Robertson reported that the LANA1 protein colocalizes with viral episomes (Ballestas et al., 1999; Cotter and Robertson, 1999), providing the ﬁrst line of evidence that LANA1 may function in episomal maintenance. Confocal microscopy using immunoﬂuorescence and in situ hybridization was employed and showed that LANA1 and the viral genome colocalize in PEL cell nuclei. Approximately 40 LANA-speciﬁc dots were observed, corresponding to the copy number of KSHV genomes in PEL cells. These experiments also suggested that many LANA molecules associate with each single viral episome because of the intensity and speckled appearance of ﬂuorescent staining. These confocal microscopic studies showed that LANA1 complexes associate with both interphase and mitotic chromosomes, which is consistent with a role for LANA1 in episomal maintenance.

C. KSHV and HVS LANA1 and Terminal Repeats Are Sufficient for Episomal Replication As shown in Fig. 3A, the KSHV genome consists of a unique central region (approximately 140 kb), which encodes more than 80 open reading frames (Russo et al., 1996). The unique sequences are ﬂanked by approximately 40 units of 801 bp tandem terminal repeats of noncoding DNA that has an unusually high G+C content (Russo et al., 1996). Until recently, no function has been experimentally ascribed to these repeats.

Episomal Maintenance of Oncogenic Herpesviruses

165

Fig. 3 Summary of studies toward identiﬁcation of the minimal essential region required for the maintenance of plasmids by LANA1 of KSHV. (A) Schematic summary of data from three studies (Ballestas et al., 1999; Cotter and Robertson, 1999; Medveczky et al., 2001). (B) Schematic depiction of the LANA1 expression vector used to identify ARS of both KSHV and HVS (Collins et al., 2001; Medveczky et al., 2001).

The overall organization of the HVS genome is similar to that of KSHV. It consists of a shorter unique sequence that is about 113 kb long (referred to as L-DNA, about 35% G+C content) and is ﬂanked by tandem repeats (often referred to as H-DNA, about 71% G+C) (Albrecht et al., 1992). The H-DNA contains consensus packaging and cleavage sites allowing for the generation and packaging of full-length genomes (Stamminger et al., 1987). At least 75 open reading frames, which are likely to be expressed as proteins, are encoded by the L-DNA of HVS (Albrecht et al., 1992).

166

Christopher M. Collins and Peter G. Medveczky

The genomes of KSHV and HVS encode a set of genes commonly described in all herpesviruses, including immediate-early transactivator genes, several enzymes involved in viral DNA metabolism, structural proteins such as capsid components, and various glycoproteins (Albrecht et al., 1992; Russo et al., 1996). These common herpesvirus genes function during the lytic replication cycle. To map the minimum essential region required for episomal replication two methods have been developed. Ballestas and Kaye transfected the B cell line BJAB with a LANA1 expression cassette and a stable LANA1 protein-producing clone was established (Ballestas and Kaye, 2001). Two cosmids, Z6 (containing several terminal repeats and the left unique 33,391 nucleotides) and Z8 (coordinates from 72,904 to 107,974) were transfected into LANA1 expressing BJAB cells. After several weeks in culture under antibiotic selection, episomal persistence was detected by the method of Gardella (Gardella et al., 1984). Figure 3A schematically illustrates that transfected cells stably maintained Z6 cosmids, but did not maintain Z8 cosmids, indicating that LANA1 mediates episomal replication through DNA sequences within Z6. Subsequent studies localized the sequence essential for episomal replication to the terminal repeats (Ballestas and Kaye, 2001). Even a single repeat unit was sufﬁcient for stable episomal replication in LANA1 expressing BJAB cells. Ballestas and Kaye also showed that uninfected B cells lacking LANA1 are not able to support episomal replication of KSHV DNA (Ballestas and Kaye, 2001). These investigations, however, did not extend the search for autonomously replicating DNA sequences to the entire genome, and it was unclear if the unique sequences that were not tested also contain autonomously replicating sequence (ARS) activity. By using transfection of overlapping cosmids that represent the entire KSHV genome into the KSHV positive BCBL-1 PEL cell line, the most efﬁcient cis-acting element(s) of KSHV were again localized to sequences containing terminal repeats, and no unique sequences were able to persist as episomes (Medveczky et al., 2001). Further sublocalization showed that plasmid constructs containing only terminal repeats were able to stably replicate in virus-transformed cells as tested by the Gardella method (Gardella et al., 1984) and by digesting episomal DNA extracted by the Hirt method (Hirt, 1967) with the restriction enzymes Dpn I and MboI (Medveczky et al., 2001). Further proof of the role of LANA1 and the terminal repeats in episomal replication was provided by construction and transfection of a LANA1 expression vector containing terminal repeats into monolayer 293 cells. Figure 3B shows the open reading frame coding for LANA1 cloned in pBK CMV (Stratagene) under the control of the CMV promoter (Medveczky et al., 2001). To generate the pLANA1 episomal expression plasmids, one, two, three, and four units of terminal repeats were inserted into the Mlu I

Episomal Maintenance of Oncogenic Herpesviruses

167

unique site of the pBKCMV-LANA1 expression vector (Fig. 3B). Subfragments of the TR were also cloned into the Mlu I site of pBKCMVLANA1, as indicated in Fig. 3A. G418 resistant 293 clones were tested for episomal replication of the transfected plasmids. Episomal DNA was isolated from the cultures by the method of Hirt (Hirt, 1967) and digested with either Dpn I or Mbo I. Plasmid DNA that was resistant to Dpn I but sensitive to Mbo I digestion was detected by Southern blotting, indicating that these plamsids were replicated by the mammalian replication machinery (Medveczky et al., 2001). Figure 3A summarizes the ﬁndings from these two groups (Ballestas and Kaye, 2001; Medveczky et al., 2001). pBKCMVLANA1 constructs containing various terminal repeat fragments replicated as episomes in 293 cells. The minimum cis-acting region sufﬁcient for stable episomal replication was narrowed down to a 424 bp fragment (nucleotides 548–171 in the terminal repeats). It was also noted that the copy number of episomal DNA was much lower with a single repeat unit than observed with plasmids encoding multiple repeats (Medveczky et al., 2001). LANA1 and the terminal repeats of HVS are also sufﬁcient for episomal replication. The open reading frame encoding HVS LANA1 was cloned into the pBKCMV expression vector under the control of the CMV promoter along with terminal repeats as indicated in Fig. 3B (Collins et al., 2001). The LANA1-terminal repeat constructs were stably replicated in 293 cells as the KSHV-LANA-TR vectors were, indicating that HVS LANA1 and the terminal repeats are sufﬁcient for episomal DNA replication. Interestingly, constructs containing only the LANA expression cassette and no terminal repeats also replicated as episomes for a limited time and at a very low copy number (this phenomenon was not observed with KSHV). A cis-acting ARS element located within the unique region of the genome was also identiﬁed in the Herpesvirus saimiri genome (Kung and Medveczky, 1996). A 2 kb fragment near the left end of the genome contains a dyad symmetry element. This 2 kb fragment was shown to replicate autonomously in virally transformed T cells. However, it was not found to be essential for episomal replication as deletion of this fragment did not abolish episomal replication of mutant virus in T cells (Kung and Medveczky, 1996). The left end of the KSHV genome shares no homology with the Herpesvirus saimiri genome and encodes no similar dyad symmetry element (Albrecht et al., 1992; Russo et al., 1996).

D. HVS Terminal Repeats Are Essential for Episomal Replication of the Viral Genome As mentioned earlier, the viral genome of HVS consists of a long unique stretch of DNA referred to as L-DNA that is characterized by low G+C

168

Christopher M. Collins and Peter G. Medveczky

content. The L-DNA is ﬂanked by noncoding terminal repeats of unusually high G+C content that are referred to as H-DNA. The 113 kb L-DNA was cloned into an F replicon-based Escherichia coli vector as a single Not-I restriction fragment (Collins et al., 2001). While the L-DNA has no Not-I recognition sites, each terminal repeat unit contains a single Not-I site. So all the terminal repeats were absent in the Bac clone as shown in Fig. 4, with the exception of 1443 nucleotides remaining attached to the termini of the 113 kb Not I fragment. Cloned DNA was found to be infectious, and the ends of the progeny viral genome consisted of ampliﬁed tandem alternating repeats of vector and a single rearranged unit of H-DNA. Detailed analysis showed that the genomic ends of the novel progeny virus contained only 4–5 repeat units of H-DNA and vector. In comparison, there are 30–40 terminal in the wild-type virus genome (Albrecht et al., 1992). Figure 4 illustrates that when T cells were infected with these “H-less” viruses, they were unable to form episomal circular viral DNA, which is the latent form of the viral genome. To reconstruct the wild type, recombinant viruses containing the normal high number of H-DNA repeats were generated. T cells infected with these rescued viruses contained high copies of episomal DNA. These data show that intact and multiple terminal repeats are essential components for episomal replication in latently infected T cells.

Fig. 4 Schematic illustration of construction of an infectious clone of HVS and subsequent analysis of the role of TR in episomal replication in T cells. HVS was cloned in pBeloBac as a single Not I restriction fragment (Collins et al., 2001). Viral DNA was infectious but progeny virus was unable to establish circular episomes in T cells. Virus with rescued TR established latent episomes as wild-type indicating that multiple TR are essential for the maintenance of the latent genome.

Episomal Maintenance of Oncogenic Herpesviruses

169

E. LANA1 Is a DNA Binding Protein That Directly Interacts with the Terminal Repeats Data from three laboratories indicate that LANA binds KSHV DNA. Cotter et al. used large radiolabeled DNA fragments representing the entire viral genome and compared their binding to LANA1 by performing immunoprecipitation with myc-tagged LANA1 (Cotter and Robertson, 1999). All KSHV DNA fragments showed some afﬁnity for the LANA1 construct, but LANA1 bound the terminal repeats with slightly more afﬁnity. Similar results were obtained with GST-tagged LANA1 (Medveczky et al., 2001). Ballestas and Kaye developed several assays to investigate the possible interaction of LANA1 with the terminal repeats (Ballestas and Kaye, 2001). Again, all of the various labeled DNA fragments showed some afﬁnity for LANA1 in immunoprecipitation experiments, but the most signiﬁcant binding activity was narrowed down to a 132 bp Sau 3A-Ava II fragment. Overlapping oligonucleotides were used to further analyze the 132 bp region. Among the 13 oligonucleotides, a 20-nucleotide fragment displayed the highest and most speciﬁc afﬁnity to the LANA1 fusion protein as tested by bandshift and supershift assays (Ballestas and Kaye, 2001). To determine whether or not LANA1 associates with KSHV DNA in vivo, chromatin immunoprecipitation was performed (Medveczky et al., 2001). KSHV infected PEL cells were treated with formaldehyde to cross-link DNA, and sonicated cell extracts were immunoprecipitated with either a speciﬁc anti-LANA1 or a control antibody. The immunoprecipitated DNA was puriﬁed, labeled, and used as probes on Southern blots that contained DNA fragments representing the entire viral genome. The LANA1 antibody selectively immunoprecipitated the 801 base-pair terminal repeat, although longer exposures revealed the possible interaction of LANA with several other regions of the KSHV genome. The control antibody did not immunoprecipitate any detectable KSHV DNA fragments. In this study, direct association of LANA1 with the terminal repeats was further demonstrated by chromosome immunoprecipitation using UV cross-linking (Medveczky et al., 2001).

F. Is the Stable Association of the LANA1–Viral DNA Complex with Chromosomes Mediated by Cellular Proteins or by Direct Binding to Heterochromatin? Although it is certain that LANA1 is involved in tethering the viral genome to chromosomes, it is unclear how and where this complex attaches to the host nuclear architecture. LANA1 may mediate binding of the LANA1

170

Christopher M. Collins and Peter G. Medveczky

complex to a chromosome-associated protein. Alternatively, LANA1 may simultaneously bind both viral and cellular DNA. The KSHV LANA1 protein appears to be a multifunctional protein since it has been shown to bind several host proteins. LANA1 associates with the tumor suppressor protein p53 and inhibits p53 transcriptional activity as well as p53 mediated apoptosis (Friborg et al., 1999). LANA1 has also been shown to associate with the retinoblastoma protein (Rb) (Radkov et al., 2000), the members of the mSin 3 corepressor complex (Krithivas et al., 2000), and RING3 (Platt et al., 1999). The loss of p53 and retinoblastoma protein function is implicated in various human tumors and viral oncogenesis. It is unclear whether these binding activities of LANA1 are related to the episomal replication function at all, or more likely, play a role in transcriptional activation and/or immortalization as proposed. LANA1 aslo binds histone H1, which may tether the KSHV genomeLANA1 complex to chromosomes. Association with histones is an attractive hypothesis; however, it is inconsistent with ﬁndings of Szekely et al. (1999). Szekely et al. demonstrated by high-resolution confocal microscopy that LANA1 associates preferentially with the border of heterochromatin in interphase nuclei. So the problem with the histone model is that these proteins are not speciﬁc for heterochromatin regions and associate with all nuclear DNA. Therefore, an alternative model may be the valid one: LANA1 simultaneously binds both viral and cellular DNA. Similar to the terminal repeats, heterochromatin contains repetitive sequences with relatively high G+C. Therefore, it is tempting to speculate that LANA1 associates with chromosomes through direct DNA binding without the need for any cellular accessory protein.

V. CONCLUSIONS Despite the signiﬁcant sequence divergence between γ -1 and γ -2 herpesviruses, many of the basic mechanisms involved in the maintenance of latent genomes are fundamentally similar. A single viral protein, either EBNA-1 or LANA1, mediates persistence of episomes of both classes of gamma herpesvirus. Both proteins form homodimers and the dimerization domains are encoded by a C-terminal region. Both EBNA-1 and LANA1 are sequencespeciﬁc DNA-binding proteins that recognize viral DNA. The speciﬁc viral protein binding sequences, FR in EBV and TR in γ -2 herpesviruses, are highly repetitive. Multiple protein complexes accumulate on a single viral episome, and multiple viral proteins probably simultaneously interact with chromosomal proteins (or possibly DNA), leading to stable anchoring to chromosomes, faithful segregation during cell division, and latent, lifelong

Episomal Maintenance of Oncogenic Herpesviruses

171

maintenance of the viral genome. A role for the DS elements in episome replication of the viral genome has been proposed but these elements are not essential. Both EBV and the HVS DS can be deleted from the viral genome without the loss of latent replication of knockout virus. It is therefore likely, as has been shown with EBV, that multiple alternative origins of replication exist in the viral genome. Although the dyad symmetry elements represent recruitment sites for the host DNA replication machinery, they are different from those found in small DNA viruses. Unlike small circular papova viral genomes, which contain precise and essential origins of DNA replication, herpesviruses seem to replicate by the rules governing cellular DNA replication as no speciﬁc origins have been found in the mammalian cellular genome and DNA replication initiates at broad regions (Hamlin et al., 1994). Several important issues regarding herpesvirus latency remain unresolved and are likely to be the focus of future studies. The role of DS elements in the virus life cycle needs to be redeﬁned. The LANA1 DNA-binding motifs are unknown. Is there more than one DNA-binding motif or is there only a single one? Is LANA1 involved in replication? Although it is likely that LANA1 is essential for maintenance of the latent viral genome, no genetic evidence is available yet that proves this hypothesis. Finally, for both classes of gamma herpesviruses, more studies should be conducted to identify the mechanism of interaction between the viral DNA–protein complex and host chromosomes because comprehensive understanding of the precise mechanism of latent replication should bring about substantial practical beneﬁts. Such knowledge could lead to rational design of antiviral compounds targeting different essential viral elements. Further development of these drugs could possibly lead to the elimination of latent herpesvirus genomes from patients suffering from the severe illnesses associated with EBV and KSHV infection.

ACKNOWLEDGMENTS This work was supported by grants from the National Institutes of Health National Cancer Institute (RO1 CA75895 and RO1 CA76586).

REFERENCES Adams, A., Pozos, T. C., and Purvey, H. V. (1989). Int. J. Cancer 44, 560–564. Aiyar, A., Tyree, C., and Sugden, B. (1998). EMBO J. 17, 6394–6403. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. (1995). “Molecular Biology of the Cell.” Garland Publishing, New York.

172

Christopher M. Collins and Peter G. Medveczky

Albrecht, J. C., Nicholas, J., Biller, D., Cameron, K. R., Biesinger, B., Newman, C., Wittmann, S., Craxton, M. A., Coleman, H., and Fleckenstein, B. (1992). J. Virol. 66, 5047–5058. Ambinder, R. F., Mullen, M. A., Chang, Y. N., Hayward, G. S., and Hayward, S. D. (1991). J. Virol. 65, 1466–1478. Ambinder, R. F., Shah, W. A., Rawlins, D. R., Hayward, G. S., and Hayward, S. D. (1990). J. Virol. 64, 2369–2379. Arvanitakis, L., Mesri, E. A., Nador, R. G., Said, J. W., Asch, A. S., Knowles, D. M., and Cesarman, E. (1996). Blood 88, 2648–2654. Avolio-Hunter, T. M., and Frappier, L. (1998). Nucleic Acids Res. 26, 4462–4470. Baer, R., Bankier, A. T., Biggin, M. D., Deininger, P. L., Farrell, P. J., Gibson, T. J., Hatfull, G., Hudson, G. S., Satchwell, S. C., and Seguin, C. (1984). Nature 310, 207–211. Ballestas, M. E., and Kaye, K. M. (2001). J. Virol. 75, 3250–3258. Ballestas, M. E., Chatis, P. A., and Kaye, K. M. (1999). Science 284, 641–644. Bochkarev, A., Barwell, J. A., Pfuetzner, R. A., Bochkareva, E., Frappier, L., and Edwards, A. M. (1996). Cell 84, 791–800. Bochkarev, A., Barwell, J. A., Pfuetzner, R. A., Furey, W., Jr., Edwards, A. M., and Frappier, L. (1995). Cell 83, 39–46. Bochkarev, A., Bochkareva, E., Frappier, L., and Edwards, A. M. (1998). J. Mol. Biol. 284, 1273–1278. Boshoff, C., and Chang, Y. (2001). Ann. Rev. Med. 52, 453–470. Chen, M. R., Middeldorp, J. M., and Hayward, S. D. (1993). J. Virol. 67, 4875–4885. Collins, C. M., Medveczky, M. M., Lund, T., and Medveczky, P. G. (2001). Submitted. Cotter, M. A., 2nd, and Robertson, E. S. (1999). Virology 264, 254–264. Fischer, N., Kremmer, E., Lautscham, G., Mueller-Lantzsch, N., and Grasser, F. A. (1997). J. Biol. Chem. 272, 3999–4005. Frappier, L., and O’Donnell, M. (1991a). Proc. Natl. Acad. Sci. USA 88, 10875–10879. Frappier, L., and O’Donnell, M. (1991b). J. Biol. Chem. 266, 7819–7826. Frappier, L., and O’Donnell, M. (1992). J. Virol. 66, 1786–1790. Frappier, L., Goldsmith, K., and Bendell, L. (1994). J. Biol. Chem. 269, 1057–1062. Friborg, J., Jr., Kong, W., Hottiger, M. O., and Nabel, G. J. (1999). Nature 402, 889–894. Gahn, T. A., and Schildkraut, C. L. (1989). Cell 58, 527–535. 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). Nat. Med. 2, 925–928. Gardella, T., Medveczky, P., Sairenji, T., and Mulder, C. (1984). J. Virol. 50, 248–254. Hamlin, J. L., Mosca, P. J., and Levenson, V. V. (1994). Biochim. Biophys. Acta 1198, 85–111. Hampar, B., Tanaka, A., Nonoyama, M., and Derge, J. G. (1974). Proc. Nat. Acad. Sci. USA 71, 631–633. Harris, A., Young, B. D., and Grifﬁn, B. E. (1985). J. Virol. 56, 328–332. Hirt, B. (1967). J. Mol. Biol. 26, 365–369. Hsieh, D. J., Camiolo, S. M., and Yates, J. L. (1993). EMBO J. 12, 4933–4944. Hung, S. C., Kang, M. S., and Kieff, E. (2001). Proc. Natl. Acad. Sci. USA 98, 1865–1870. Kanda, T., Otter, M., and Wahl, G. M. (2001). Mol. Cell. Biol. 21, 3576–3588. Kapoor, P., Shire, K., and Frappier, L. (2001). EMBO J. 20, 222–230. Kedes, D. H., Lagunoff, M., Renne, R., and Ganem, D. (1997). J. Clin. Invest. 100, 2606–2610. Kedes, D. H., Operskalski, E., Busch, M., Kohn, R., Flood, J., and Ganem, D. (1996). Nat. Med. 2, 918–924. Kieff, E. (1996). In “Virology” (B. N. Feilds, D. M. Knipe, and P. M. Howley, eds.), pp. 2343– 2397. Lippincott-Raven, Philadelphia. Kim, A. L., Maher, M., Hayman, J. B., Ozer, J., Zerby, D., Yates, J. L., and Lieberman, P. M. (1997). Virology 239, 340–351.

Episomal Maintenance of Oncogenic Herpesviruses

173

Kirchmaier, A. L., and Sugden, B. (1995). J. Virol. 69, 1280–1283. Kirchmaier, A. L., and Sugden, B. (1997). J. Virol. 71, 1766–1775. Kirchmaier, A. L., and Sugden, B. (1998). J. Virol. 72, 4657–4666. Krithivas, A., Young, D. B., Liao, G., Greene, D., and Hayward, S. D. (2000). J. Virol. 74, 9637–9645. Krysan, P. J., and Calos, M. P. (1991). Mol. Cell. Biol. 11, 1464–1472. Krysan, P. J., Haase, S. B., and Calos, M. P. (1989). Mol. Cell. Biol. 9, 1026–1033. Kung, S. H., and Medveczky, P. G. (1996). J. Virol. 70, 1738–1744. Lee, M. A., Diamond, M. E., and Yates, J. L. (1999). J. Virol. 73, 2974–2982. Leight, E. R., and Sugden, B. (2000). Rev. Med. Virol. 10, 83–100. Lennette, E. T., Blackbourn, D. J., and Levy, J. A. (1996). Lancet 348, 858–861. Levitskaya, J., Sharipo, A., Leonchiks, A., Ciechanover, A., and Masucci, M. G. (1997). Proc. Natl. Acad. Sci. USA 94, 12616–12621. Little, R. D., and Schildkraut, C. L. (1995). Mol. Cell. Biol. 15, 2893–2903. Luka, J., Siegert, W., and Klein, G. (1977). J. Virol. 22, 1–8. Lupton, S., and Levine, A. J. (1985). Mol. Cell. Biol. 5, 2533–2542. Mackey, D., and Sugden, B. (1999). Mol. Cell. Biol. 19, 3349–3359. Mackey, D., Middleton, T., and Sugden, B. (1995). J. Virol. 69, 6199–6208. Marechal, V., Dehee, A., Chikhi-Brachet, R., Piolot, T., Coppey-Moisan, M., and Nicolas, J. C. (1999). J. Virol. 73, 4385–4392. Medveczky, M. M., Fejer, G., Spain-Santana, T. A., Horvath, E., Lane, B., Chang, Y., Moore, P. S., Chandran, B., and Medveczky, P. G. (2001). Submitted. Medveczky, M. M., Horvath, E., Lund, T., and Medveczky, P. G. (1997). AIDS 11, 1327–1332. Melendez, L. V., Daniel, M. D., Garcia, F. G., Fraser, C. E., Hunt, R. D., and King, N. W. (1969). Lab. Anim. Care. 19, 372–377. Melendez, L. V., Daniel, M. D., Hunt, R. D., and Garcia, F. G. (1968). Lab. Anim. Care 18, 374–381. Middleton, T., and Sugden, B. (1992). J. Virol. 66, 489–495. Middleton, T., and Sugden, B. (1994). J. Virol. 68, 4067–4071. Moore, P. S., Gao, S. J., Dominguez, G., Cesarman, E., Lungu, O., Knowles, D. M., Garber, R., Pellett, P. E., McGeoch, D. J., and Chang, Y. (1996). J. Virol. 70, 549–558. Nador, R. G., Cesarman, E., Chadburn, A., Dawson, D. B., Ansari, M. Q., Sald, J., and Knowles, D. M. (1996). Blood 88, 645–656. Newlon, C. S. (1997). Cell 91, 717–720. Niller, H. H., Glaser, G., Knuchel, R., and Wolf, H. (1995). J. Biol. Chem. 270, 12864–12868. Norio, P., Schildkraut, C. L., and Yates, J. L. (2000). J. Virol. 74, 8563–8574. Ohno, S., Luka, J., Lindahl, T., and Klein, G. (1977). Proc. Nat. Acad. Sci. USA 74, 1605–1609. Petti, L., Sample, C., and Kieff, E. (1990). Virology 176, 563–574. Piolot, T., Tramier, M., Coppey, M., Nicolas, J. C., and Marechal, V. (2001). J. Virol. 75, 3948–3959. Platt, G. M., Simpson, G. R., Mittnacht, S., and Schulz, T. F. (1999). J. Virol. 73, 9789–9795. Radkov, S. A., Kellam, P., and Boshoff, C. (2000). Nat. Med. 6, 1121–1127. Rainbow, L., Platt, G. M., Simpson, G. R., Sarid, R., Gao, S. J., Stoiber, H., Herrington, C. S., Moore, P. S., and Schulz, T. F. (1997). J. Virol. 71, 5915–5921. Rawlins, D. R., Milman, G., Hayward, S. D., and Hayward, G. S. (1985). Cell 42, 859–868. Reedman, B. M., and Klein, G. (1973). Int. J. Cancer 11, 499–520. Reisman, D., Yates, J., and Sugden, B. (1985). Mol. Cell. Biol. 5, 1822–1832. Rivas, C., Thlick, A. E., Parravicini, C., Moore, P. S., and Chang, Y. (2001). J. Virol. 75, 429–438. Roizman, B., Carmichael, L. E., Deinhardt, F., de-The, G., Nahmias, A. J., Plowright, W., Rapp, F., Sheldrick, P., Takahashi, M., and Wolf, K. (1981). Intervirology 16, 201–217.

174

Christopher M. Collins and Peter G. Medveczky

Russo, J. J., Bohenzky, R. A., Chien, M. C., Chen, J., Yan, M., Maddalena, D., Parry, J. P., Peruzzi, D., Edelman, I. S., Chang, Y., and Moore, P. S. (1996). Proc. Natl. Acad. Sci. USA 93, 14862–14867. Said, W., Chien, K., Takeuchi, S., Tasaka, T., Asou, H., Cho, S. K., de Vos, S., Cesarman, E., Knowles, D. M., and Koefﬂer, H. P. (1996). Blood 87, 4937–4943. Schwam, D. R., Luciano, R. L., Mahajan, S. S., Wong, L., and Wilson, A. C. (2000). J. Virol. 74, 8532–8540. Shirakata, M., Imadome, K. I., and Hirai, K. (1999). Virology 263, 42–54. Shire, K., Ceccarelli, D. F., Avolio-Hunter, T. M., and Frappier, L. (1999). J. Virol. 73, 2587– 2595. Simpson, G. R., Schulz, T. F., Whitby, D., Cook, P. M., Boshoff, C., Rainbow, L., Howard M. R., Gao, S. J., Bohenzky, R. A., Simmonds, P., Lee, C., de Ruiter, A., Hatzakis, A., Tedder, R. S., Weller, I. V., Weiss, R. A., and Moore, P. S. (1996). Lancet 348, 1133–1138. Stamminger, T., Honess, R. W., Young, D. F., Bodemer, W., Blair, E. D., and Fleckenstein, B. (1987). J. Gen. Virol. 68, 1049–1066. Su, W., Middleton, T., Sugden, B., and Echols, H. (1991). Proc. Natl. Acad. Sci. USA 88, 10870–10874. Sullivan, J. L., Medveczky, P., Forman, S. J., Baker, S. M., Monroe, J. E., and Mulder, C. (1984). New Engl. J. Med. 311, 1163–1167. Szekely, L., Kiss, C., Mattsson, K., Kashuba, E., Pokrovskaja, K., Juhasz, A., Holmvall, P., and Klein, G. (1999). J. Gen. Virol. 80, 2889–2900. Van Scoy, S., Watakabe, I., Krainer, A. R., and Hearing, J. (2000). Virology 275, 145–157. Wang, Y., Finan, J. E., Middeldorp, J. M., and Hayward, S. D. (1997). Virology 236, 18–29. Wu, H., Ceccarelli, D. F., and Frappier, L. (2000). EMBO Rep. 1, 140–144. Yates, J. L., and Camiolo, S. M. (1988). Cancer Cells 6, 197–205. Yates, J. L., and Guan, N. (1991). J. Virol. 65, 483–488. Yates, J. L., Camiolo, S. M., and Bashaw, J. M. (2000). J. Virol. 74, 4512–4522. Yates, J., Warren, N., Reisman, D., and Sugden, B. (1984). Proc. Nat. Acad. Sci. USA 81, 3806–3810. Yates, J. L., Warren, N., and Sugden, B. (1985). Nature 313, 812–815. Zhang, D., Frappier, L., Gibbs, E., Hurwitz, J., and O’Donnell, M. (1998). Nucleic Acids Res. 26, 631–637.

Treatment of Epstein–Barr Virus-Associated Malignancies with Specific T Cells 1,2

1,2,3

Stephen Gottschalk, Helen E. Heslop, and Cliona M. Rooney1,2,4 1 Center for Cell and Gene Therapy Departments of 2Pediatrics,3Medicine 4 and Molecular Virology and Microbiology Baylor College of Medicine Houston, Texas

I. Introduction A. EBV-Associated Malignancies B. Rationale for Adoptive Immunotherapy with EBV-Specific Cytotoxic T Lymphocytes II. EBV-Associated Lymphoproliferative Disease A. Treatment with Donor T Cells B. Treatment with EBV-Specific Cytotoxic T Lymphocytes III. Hodgkin’s disease IV. Nasopharyngeal Carcinoma V. Burkitt’s Lymphoma VI. EBV-Associated Non-Hodgkin’s Lymphoma in HIV Patients VII. Conclusions References

Latent Epstein–Barr virus (EBV) infection is associated with a heterogeneous group of malignancies, including Burkitt’s lymphoma, Hodgkin’s disease, nasopharyngeal carcinoma, and lymphoproliferative disease (LPD). The development of adoptive immunotherapies for these malignancies is being fueled by the successful generation of allogeneic donor derived EBV-speciﬁc cytotoxic T cells (CTL) for the prevention and treatment of EBV-LPD after hematopoietic stem cell transplantation. This approach is being extended to EBV-LPD after solid organ transplantation by use of autologous and haploidentical EBV-speciﬁc CTL. For other EBV-associated malignancies, there is only limited clinical experience with EBV-speciﬁc CTL. With few exceptions, only patients with recurrent Hodgkin’s disease have been treated with autologous EBV-speciﬁc CTL, and although the results have been promising, they do not include cures. Lack of CTL efﬁcacy may reﬂect either down-regulation of immunodominant EBV proteins, which are major CTL targets, or the presence of inhibitory cytokines. Further improvement of EBV-speciﬁc CTL therapy for Hodgkin’s disease will require improved methods to activate and expand CTL speciﬁc for the latent EBV genes expressed in Hodgkin’s disease and to genetically

Advances in CANCER RESEARCH 0065-230X/02 $35.00

175

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

176

Stephen Gottschalk et al.

modify the expanded CTL to render them resistant to inhibitory cytokines. If effective, such strategies could be applied not only to other EBV-associated malignancies, but also to a broad range of human tumors with deﬁned tumor antigens and similar immune evasion strategies. C 2002 Academic Press.

I. INTRODUCTION Epstein–Barr virus (EBV) was the ﬁrst human virus implicated in oncogenesis. It was originally isolated in 1964 from Burkitt’s lymphoma cells and since then has been linked to a heterogeneous group of malignant and nonmalignant disorders (Hsu and Glaser, 2000; Epstein et al., 1964). EBV is a latent herpesvirus that infects over 90% of all human populations (Cohen, 2000). Primary infection with EBV usually occurs through the oropharynx, where mucosal epithelial cells and/or B cells become productively infected (Faulkner et al., 2000). The virus produced in these cells may then infect neighboring epithelial cells and B cells circulating through the mucosa-associated lymphoid tissues. Primary infection usually results in a mild, self-limiting illness that is followed by lifelong virus latency in B cells. In normal seropositive individuals, EBV latency in B cells is tightly controlled by EBV-speciﬁc T cells (Rickinson and Moss, 1997), and studies using tetramer technology have shown that high frequencies of EBV-speciﬁc CD8-positive T cells persist long-term (Tan et al., 1999). EBV-speciﬁc CD4-positive T cells may play also an important role in the cellular immune response to EBV (Bickham et al., 2001; Munz et al., 2000; Nikiforow et al., 2001; van Baarle et al., 2001).

A. EBV-Associated Malignancies All EBV-associated malignancies involve the virus’s latent cycle, and four patterns of latent gene expression have been described. In healthy individuals the virus persists episomally in resting memory B cells, and of the almost 100 viral proteins, only LMP2 is expressed (Babcock et al., 2000; ThorleyLawson and Babcock, 1999; Tierney et al., 1994). In addition small nonpolyadenylated viral RNAs termed EBERs 1 and 2 are present, and this type of latency has been designated type “0.” Three other distinct types of EBV latency have been characterized in a heterogeneous group of malignancies (Table I) (Hsu and Glaser, 2000; Kanno et al., 1999; Kerr et al., 1992; Young et al., 2000). All are EBER positive, but the EBV latent protein expression varies. Latency type III is expressed in lymphoblastoid cell lines (LCL), which can be readily produced by infecting B cells in vitro with EBV. These cells express the entire array of nine EBV latency proteins:

177

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

Table I Common EBV-Associated Malignancies

Malignancy Burkitt’s lymphoma Gastric adenocarcinoma Hodgkin’s disease Nasopharyngeal carcinoma Peripheral T/NK-cell lymphoma Lymphoproliferative disease after solid organ or stem cell transplantation HIV-associated immunoblastic lymphomaa a

Latent gene expression

Pattern of latency

EBNA-1, BARFO

I

EBNA-1, LMP1, LMP2, BARFO

II

EBNAs 1, 2, 3A, 3B, 3C, LP LMP1, LMP2, BARFO

III

Not all lymphomas are latency type III (see Section VI for details).

EBNAs 1, 2, 3A, 3B, 3C, LP, BARF0, and the two viral membrane proteins LMP1 and LMP2. This pattern of EBV gene expression characterizes the EBV-associated lymphoproliferative diseases (EBV-LPD) occur in individuals severely immunocompromised by solid-organ or stem-cell transplantation, congenital immunodeﬁciency, or human immunodeﬁciency virus (HIV) infection. Latency type II is the hallmark of EBV-positive Hodgkin’s disease, nasopharyngeal carcinomas, and peripheral T/NK-cell lymphomas. The EBV proteins expressed in these malignancies are EBNA-1, BARF0, LMP1, and LMP2. In latency type I, found in EBV-positive Burkitt’s lymphoma and gastric adenocarcinomas, only EBNA-1 and BARF0 are expressed. In all types of latency spontaneous viral replication occurs at very low frequency. Since antiviral agents like acyclovir only prevent productive viral replication and do not affect latency (Yao et al., 1989a,b), these agents are of limited therapeutic value. However, long-term prophylactic administration of high-dose acyclovir may decrease the incidence of EBV-associated malignancies in HIV patients and EBV-LPD after solid organ transplantation by limiting intercellular virus transmission (Darenkov et al., 1997; Fong et al., 2000).

B. Rationale for Adoptive Immunotherapy with EBV-Specific Cytotoxic T Lymphocytes Until recently, EBV vaccine development was impeded by the lack of a suitable animal model, so that EBV vaccines are not available (Khanna et al., 1999b; Moss et al., 1996). Most vaccination studies have been performed in cottontop tamarins, a species of New World monkeys that develop B-cell lymphomas after intraperitoneal injection of EBV (Epstein et al., 1985; Wilson et al., 1999). This model lacks certain key features of human EBV

178

Stephen Gottschalk et al.

disease considered to be important for vaccine development: (1) the routes of virus entry differ; (2) the symptoms and signs of acute human EBV infection are absent; and (3) virus does persist not long-term in B cells. To overcome these limitations, Moghaddan et al. (1997) developed a rhesus monkey model for acute and persistent EBV infection that takes advantage of the rhesus lymphocryptovirus, which belongs to the same herpesvirus subgroup as EBV. This animal model reproduces the salient features of human EBV infection, such as oral transmission, lymphadenopathy, atypical lymphocytosis, serological responses to EBV proteins, and viral latency in B cells, and therefore this model has the potential to become a useful platform on which to test different vaccination strategies. Although potentially ideal for preventing EBV-associated malignancies, vaccines providing lifelong immunity against primary EBV infection may not be feasible because the type of immunity required to avoid repeated infection through mucosal surfaces is not clearly deﬁned. Moreover, such vaccines would likely not be cost effective for the general population, since EBV-associated malignancies are rare outside speciﬁc regions such as southern China or equatorial Africa (Moss et al., 1996). The most promising vaccine strategies for the immunotherapy of EBV-associated malignancies seek to elicit or boost the EBV-speciﬁc cellular immune response against EBV latency. Persons likely to beneﬁt from this approach are EBV-seronegative patients who are scheduled to undergo solid-organ transplantation or seropositive patients who have an EBV-associated malignancy with a low tumor burden or are in remission. Two clinical trials to evaluate this strategy are under way. In one study, EBV-positive nasopharyngeal carcinoma patients are being vaccinated with MHC class I-restricted EBV peptide epitopes; in the other, HLA B8-positive, EBV-seronegative volunteers are being vaccinated with an HLA B8-restricted peptide epitope from EBNA-3A. Efﬁcacy data are not yet available for either study (A. B. Rickinson, personal communication; Khanna et al., 1999b). Active vaccination with EBV peptides or peptide-loaded dendritic cells is unlikely to be the optimal method to enhance EBV-speciﬁc cytotoxic T-lymphocyte (CTL) responses for patients who develop EBV-associated malignancies, such as EBV-LPD or Hodgkin’s disease, since they are immunocompromised. Thus, immunotherapy with infusion of ex vivo-activated tumor-speciﬁc CTL seems to be more promising. Recipients of allogeneic hematopoietic stem cell transplants receive highdose chemotherapy and/or radiation to completely ablate their immune systems prior to transplantation. The immune ablation increases the risk for serious infectious complications, such as EBV- and cytomegalovirus (CMV)associated diseases, during reconstitution of the immune system by donor B and T cells. Thus immunotherapeutic strategies to correct B- and T-cell deﬁciencies after transplantation might be of clinical beneﬁt. Passive immunization with immunoglobulin infusions is not helpful in preventing of EBV

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

179

or CMV disease after hematopoietic stem cell transplantation, and active immunization is not feasible because of the patients’ severe immunosuppression. Infusion of in vitro-expanded antigen-speciﬁc CTL, as pioneered by Riddell et al. in Seattle, has secured impressive protection against CMV disease (Riddell et al., 1992; Riddell and Greenberg, 2000). The CMV-speciﬁc CD8-positive CTL clones reconstituted CMV-speciﬁc immune responses without adverse effects, and none of the patients developed CMV disease. However, the CTL did not persist long-term except in patients who either endogenously recovered CMV-speciﬁc CD4-positive T helper cells or were coinfused with CMV-speciﬁc CD4-positive T-cell clones, underscoring the need for such cells in the maintenance of CD8-positive CTL populations (Walter et al., 1995). EBV-LPD after hematopoietic stem-cell transplantation is another attractive model for cellular immunotherapy because the malignant cells are highly immunogenic and are sensitive to immune-mediated killing. LCL, which can be generated in vitro from the donor, are phenotypically and antigenically identical to the tumor cells and provide both an effective antigen-presenting cell and a source of viral antigen for the generation of EBV-speciﬁc T cells in seropositive donors.

II. EBV-ASSOCIATED LYMPHOPROLIFERATIVE DISEASE EBV-LPD develops in patients with congenital or acquired immunodeﬁciencies, including severe combined immunodeﬁciency, X-linked lymphoproliferative disease, HIV infection, and immunosuppression due to transplantation of solid organs or hematopoietic stem cells. Most cases of EBV-LPD are lymphomas of B-cell origin, histologic high-grade non-Hodgkin’s lymphomas (NHL) of the immunoblastic or undifferentiated large-cell type that respond poorly to cytotoxic therapy (Orazi et al., 1997). In the setting of solid-organ transplantation, the reported incidence of EBV-LPD ranges from 1 to 25%, with the highest risk in seronegative recipients and patients receiving intensive immunosuppressive therapy.After hematopoietic stem-cell transplantation, the incidence of EBV-LPD varies with the transplant regimen and may be as high as 25% (Table II). Risk factors for the development of EBV-LPD include the use of stem cells from an HLA-mismatched family member or closely HLA-matched unrelated donor, T-cell depletion of the donor cells, intensive immunosuppression, and an underlying diagnosis of primary immunodeﬁciency (Gerritsen et al., 1996; Curtis et al., 1999; Heslop and Rooney, 1997; O’Reilly et al., 1997). The incidence is much lower when methods that also deplete B cells are employed. In a large review

180

Stephen Gottschalk et al.

Table II Incidence of EBV-LPD After Hematopoietic Stem Cell Transplantation Study

T-cell depletion

B-cell depletion

EBV-LPD (%)

— — —

8 14 11.5

—

26

O’Reilly et al. (1997) Skinner et al. (1998) Cavazzana-Calvo et al. (1998)

CD3 antibody E-rosetting CD6/CD8 or T10/B9 antibodies E-rosetting and CD34 selection E-rosetting CT-2 antibody CD2/CD7 antibodies

11 11 0

Hale et al. (1998) Liu et al. (1999)

CAMPATH-1 CD6/CD8 antibodies

— — CD19 and CD20 antibodies CAMPATH-1 CD20 antibody

Casper et al. (1995) Gerritsen et al. (1996) Heslop and Rooney (1997) Lucas et al. (1998)

<2 0

of patients who received hematopoietic stem-cell transplants treated with the Campath-1 antibody, which removes mature T and B cells (Hale et al., 1998), the incidence of EBV-LPD was less than 2%; a low incidence was also seen after elutriation, which removes over 90% of B cells from the donor graft (Gross et al., 1998). Addition of monoclonal antibodies for B-cell depletion to the T-cell depletion regimen for the prevention of EBV-LPD post stem-cell transplantation has also proven to be effective. Cavazzana-Calvo et al. (1998) reported the use of anti-CD19 or a combination of anti-CD19 and anti-CD20 monoclonal antibodies for the depletion of B cells in stem-cell products, and we have used anti-CD20 monoclonal antibodies for the same purpose (Liu et al., 1999). Both strategies prevented the development of EBV-LPD and acute GVHD without a reduction in the engraftment rate, as compared with historical controls. The onset of EBV-LPD seems to be preceded by a large increase in virus load as well as the proliferation of EBV infected B-cells. Studies have shown that frequent monitoring of the EBV-DNA load in peripheral blood is a valuable diagnostic test for early detection of EBV-LPD after both stem cell and solid-organ transplantation (Baldanti et al., 2000; Green et al., 2000; Rowe et al., 1997; Stevens et al., 2001). We found that a 2-to 3-log increase of EBV DNA levels is also a sensitive predictor of EBV-LPD development in recipients of hematopoietic stem-cell transplants (Rooney et al., 1995). The threshold levels of EBV-DNA suggestive of impending EBV-LPD vary according to the PCR method of quantifying viral DNA. It should be emphasized that not all patients with high EBV-DNA levels, especially those with a solid-organ transplant, develop EBV-LPD. However, a high virus load has a high positive prognostic value. More recent results indicate several distinct patterns of EBV latent gene expression in the memory B cells of “high-load EBV carriers” (Rose et al., 2001), with type III latency conferring the highest risk for EBV-LPD development. Thus, an elevated EBV-DNA load can lead

181

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

to early diagnosis of EBV-LPD with consequent reductions in mortality and treatment-related morbidity, although additional factors such as clinical signs and symptoms, as well as radiographic ﬁndings, must be taken into account before therapy is initiated.

A. Treatment with Donor T Cells Papadopoulos et al. (1994) at Memorial Sloan-Kettering Cancer Center originally reported on ﬁve patients who received donor leukocyte infusions for EBV-LPD. All had therapeutic responses, but three developed graftversus-host disease (GVHD) and two died from respiratory insufﬁciency. In an update of this experience, 17 of 19 patients responded positively to donor T cells (O’Reilly et al., 1997). Others have also documented success with this approach (Heslop et al., 1994; Lucas et al., 1998; Sasahara et al., 1998) but emphasize the risk of GVHD and, in some instances, a lower response rate to donor leukocyte infusions than reported by the SloanKettering group (Table III). In a series of 13 patients, who received T-celldepleted allogeneic stem-cell transplants at the University of Indiana, only four (31%) responded to donor leukocyte infusions (Lucas et al., 1998), one of whom died from acute GVHD and another from aspergillosis. Of the two patients who survived, one had been treated with EBV-speciﬁc CTL. Five of the nine patients with disease progression died within 10 days of receiving donor leukocytes, most likely because of advanced EBV-LPD at the time of Table III Treatment and Prophylaxis of EBV-LPD with Donor T Cells or EBV-Speciﬁc CTL after Hematopoietic Stem Cell Transplantation

Study

Cell product

Therapy or prophylaxis

Response

Gross et al. (1999) Heslop et al. (1994) Lucas et al. (1998) Nagafuji et al. (1998) O’Reilly et al. (1997)

Donor T cells Donor T cells Donor T cells Donor T cells Donor T cells

Therapy Therapy Therapy Therapy Therapy

0/3 1/1 4/13a 0/1 17/19

Sasahara et al. (1998)

Therapy

0/1

Gustafsson et al. (2000)

Donor T cells and EBV-speciﬁc CTL EBV-speciﬁc CTL

Prophylaxis

Rooney et al. (1998)

EBV-speciﬁc CTL

Prophylaxis

Rooney et al. (1998)

EBV-speciﬁc CTL

Therapy

1/6 developed LPD 0/39 developed LPD 2/3

a

One responding patient received EBV-speciﬁc CTL. NR, not reported.

GVHD NR 1/1 4/13 0/1 3 acute, 8 chronic 0/1 1/6 1/39 0/3

182

Stephen Gottschalk et al.

treatment, prompting the authors to advocate earlier diagnosis and initiation of therapy. The overall incidence of GVHD was 31%. Two case reports from Japan describe patients who failed to respond to donor T cells (Imashuku et al., 1998; Nagafuji et al., 1998), and in a review of experience at the University of Minnesota, none of three patients responded (Gross et al., 1999). Reasons for the discrepancies in response rates to donor leukocyte infusions (Table III) are unclear but may reﬂect different types of disease or better outcome with early diagnosis and treatment. Two strategies have been employed to reduce the risk of GVHD after immunotherapy with donor T cells. The ﬁrst strategy is to transduce T cells with a “suicide gene” so that cell death can be induced if adverse effects are noted (Tiberghien, 2001). Bonini et al. (1997) evaluated this approach by transducing donor T cells prior to infusion with a retroviral construct containing the herpes simplex virus thymidine kinase (HSV-tk) gene, which renders host cells sensitive to the cytotoxic effects of ganciclovir. Three of eight patients who developed GVHD following T-cell infusion were treated with ganciclovir, resulting in resolution in two patients and signiﬁcant reduction of clinical symptoms in one. Using the same approach, Tiberghien et al. (2001) have reported similar results for two out of three patients with GVHD who responded to ganciclovir. One of the limitations of the HSV-tk/ganciclovir strategy is its immunogenicity. Eight of 24 patients infused with HSV-tk-modiﬁed T cells developed a CD8-positive immune response to the kinase, resulting in selective elimination of HSV-tk-expressing cells (Thomis et al., 2001). To circumvent this problem, one group of investigators substituted a human Fas-based suicide switch for HSV-tk, and the initial results of in vitro testing have been human promising (Thomis et al., 2001). The second strategy to prevent GVHD after T-cell infusion is to administer EBV-speciﬁc CTL rather than unmanipulated donor T cells (Heslop and Rooney, 1997).

B. Treatment with EBV-Specific Cytotoxic T Lymphocytes In the majority of EBV-LPD cases that occur in hematopoietic stem-cell recipients, the transformed cells are of donor origin and express all latentcycle virus-associated antigens, providing excellent targets for virus-speciﬁc T cells. EBV-transformed B lymphoblastoid cell lines (LCL) likewise express all latent-cycle virus-associated antigens and several costimulatory molecules that facilitate CTL generation (Fig. 1). They can be readily prepared from any donor and provide a source of antigen-presenting cells that endogenously express the appropriate antigens for presentation of HLA class I–restricted epitopes. LCL also reactivate HLA class-II-restricted T cells

183

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

PBMC

Step 1: LCL generation Infection with EBV strain B95-8

EBV

LCL

LMP 1 EBNAs 1 2, 3A-C, LP LMP 2 BARFO

Step 2: CTL expansion Stimulation with irradiated LCL and expansion with IL-2 IL-2

PBMC

CTL

Step 3: Gene marking Transduction with neomycin retrovirus

Step 4: QA/QC

neo

Sterility, PCR, HLA type, Phenotype, Cytotoxicity neo

Fig. 1 Generation of clinical-grade, gene-marked EBV-speciﬁc CTL. Peripheral mononuclear cells (PBMC) are infected with the EBV laboratory strain B95-8 to generate lymphoblastoid cell lines (LCL). PBMC are then repeatedly stimulated with irradiated LCL, and interleukin 2 (IL-2) is added starting on day 14. Once sufﬁcient numbers of EBV-speciﬁc CTL are obtained, the cells are gene-marked with a retrovirus encoding the neomycin-resistance gene (neo), and quality-assurance (QA) and -control (QC) testing is performed.

either by endogenous lysosomal processing as by phagocytosis of dead cells. Because they bear the relevant tumor antigens for CTL testing, LCL also serve as excellent autologous target cells (Heslop and Rooney, 1997; O’Reilly et al., 1997). Since most donors are immune to EBV, they carry a high frequency of EBV-speciﬁc CTL precursors. The generation of EBV-speciﬁc CTL

184

Stephen Gottschalk et al.

from seropositive, healthy donors takes 8–12 weeks, of which 4–6 weeks are needed to generate sufﬁcient numbers of LCL for CTL stimulation. The resultant EBV-speciﬁc CTL are polyclonal and contain both CD4- and CD8positive EBV-speciﬁc T cells, which is considered advantageous since the presence of antigen-speciﬁc CD4-helper T cells is important for in vivo survival of cytotoxic CD8-positive T-cell populations (Riddell and Greenberg, 2000). In our study evaluating the efﬁcacy of prophylactic administration of EBV-speciﬁc CTL to hematopoietic stem-cell transplant recipients at high risk for EBV-LPD, 60 patients were treated and none developed lymphoproliferative disease, compared with an incidence of 11.5% in a comparable untreated historical control group (Rooney et al., 1998). The EBV-speciﬁc CTL were infused at a median of 88 days posttransplantation. The trial was initially designed as a dose escalation study, in which the ﬁrst 12 patients received either 4 × 107 cells/m2 (n = 6) or 1.2 × 108 cells/m2 (n = 6) over 4 weeks. At all dose levels there was evidence of CTL persistence and efﬁcacy; deescalation of the cell dose to a single injection of 2 × 107 cells/m2 resulted in the same outcome. Nine patients with high EBV-DNA levels in peripheral blood, predictive of the imminent onset of EBV-LPD, had a rapid decrease in EBV-DNA levels that correlated with an increase in EBV-speciﬁc CTL precursor frequency. To track the fate of EBV-speciﬁc CTL in recipients and to evaluate adverse effects, Heslop et al. (1996) marked cells with a retroviral vector encoding the neomycin-resistance gene before infusion. Gene-marked EBV-speciﬁc CTL persisted in the patients for longer than 7 years, and there was no toxicity attributable to expression of the marker gene. In one patient, reemergence of high EBV-DNA blood levels was accompanied by increased numbers of genemarked EBV-speciﬁc CTL followed by a decline of both measurements to baseline. These observations illustrate the importance of CTL in controlling EBV infection and the effectiveness of this strategy in preventing EBV-LPD. A study from Sweden has conﬁrmed the efﬁcacy of EBV-speciﬁc CTL in reducing the viral load in patients with high EBV-DNA levels after hematopoietic stem cell transplantation (Gustafsson et al., 2000). The patients received 4 weekly injections of 1 × 107 cells/m2; however, one out of six patients who received EBV-speciﬁc CTL subsequently developed overt EBV-LPD (Table III) and died of progressive disease. In vitro testing of the donor CTL line of this patient showed that it lacked a strong EBV-speciﬁc component, which may explain this failure of CTL therapy. lmmunotherapy with EBV-speciﬁc CTL was also used to treat three hematopoietic stem-cell recipients who developed overt lymphoma. Two of these patients responded well with complete regression of bulky tumors; one of the two had a biopsy-proven accumulation of gene-marked CTL in the disease site (Rooney et al., 1998). One of the responders required temporary mechanical ventilation because of airway compromise from the inﬂammatory

185

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

response at the disease site. The third patient did not respond and died of progressive disease 24 days after CTL therapy. This failure was attributed to a deletion in EBNA-3B in the tumor virus, which removed immunodominant epitopes and thereby caused resistance to killing by the CTL. Analysis of EBV polymorphisms demonstrated that before CTL infusion more than one virus was present, including a virus with wild-type EBNA-3B. After CTL infusion, only the virus with the EBNA-3B deletion could be detected, suggesting that the infused CTL had selected a resistant EBV strain in vivo (Gottschalk et al., 2001). Thus, escape mutants may be a problem even when polyclonal CTL lines are used, particularly when the tumor burden is large. In principle, the use of spontaneous donor-derived LCL rather than donor B cells transformed with the B95-8 laboratory strain of EBV would avoid this problem, although in our experience it has not been possible to obtain such LCL in a timely fashion from most donors. Finally, as in the case of our patient, a CTL escape mutation may be present only in a minority of donor B cells. In solid-organ transplant recipients who develop EBV-LPD, donor-derived T cells may be of limited value, since the tumor almost always arises in recipient B cells (Table IV). Unmanipulated allogeneic T cells from an HLAmatched sibling have been successfully used to treat CNS lymphoma in a lung transplant recipient (Emanuel et al., 1997). The limited availability of HLA-matched EBV-speciﬁc donor T cells, together with the high likelihood that most patients will have a haploidentical relative, has led to clinical investigation of haploidentical EBV-speciﬁc CTL (Orentas et al., 1998). In one case report, EBV-speciﬁc CTL derived from a haploidentical parent were administered to a renal transplant recipient with EBV-LPD, resulting in partial tumor regression. The observed increments in EBV-speciﬁc CTL were only transient, suggesting that the cells were eliminated because of recognition of MHC antigens on the nonshared haplotype. The patient then received autologous EBV-speciﬁc CTL, achieving a long-term remission (O’Reilly et al., 1998). EBV-speciﬁc autologous CTL have also been infused as prophylaxis in Table IV Treatment and Prophylaxis of EBV-LPD with T Cells or EBV-Speciﬁc CTL After Solid Organ Transplantation

Study

Cell product

Therapy or prophylaxis

Response

Comoli et al. (1997) Emanuel et al. (1997) Haque et al. (1998) Khanna et al. (1999) O’Reilly et al. (1998)

Autologous EBV-speciﬁc CTL HLA-identical sibling T cells Autologous EBV-speciﬁc CTL Autologous EBV-speciﬁc CTL Haploidentical parental EBV-speciﬁc CTL

Prophylaxis Therapy Prophylaxis Therapy Therapy

3/3a 1/1 3/3a 1/1 1/1b

a

Documented decrease of EBV DNA level in peripheral blood post EBV-speciﬁc CTL infusion. Partial reponse to haploidentical EBV-speciﬁc CTL, subsequent complete response to autologous EBVspeciﬁc CTL. b

186

Stephen Gottschalk et al.

solid-organ transplant recipients. Comoli et al. (1997) reported the infusion of autologous EBV-speciﬁc CTL in three heart transplant recipients. In all patients there was a 2- to 3-log decline of EBV-DNA levels in peripheral blood, and serial myocardial biopsies showed no evidence of organ rejection. Haque et al. (1998) described three patients who had received solidorgan transplants (two liver and one kidney). In each case EBV-DNA levels decreased and the numbers of EBV-CTL precursors increased following the infusions, and the effects persisted for up to 3 months. More recently, Khanna et al. (1999a) used autologous EBV-speciﬁc CTL to treat a renal transplant recipient with EBV-LPD and observed signiﬁcant regression following CTL infusion. The emergence of new lymphoma lesions requiring retreatment suggested that transferred CTL may not survive in solid-organ recipients who receive continuous immunosuppressive therapy. An additional cautionary note is that the patient died, with evidence of necrosis and hemorrhage found in a pulmonary vein at autopsy. This case history reinforces experience with inﬂammatory reactions in stem-cell transplant recipients and argues for caution when using adoptively transferred T cells to treat patients with bulky disease. Autologous lymphokine-activated killer (LAK) cells, generated from peripheral blood mononuclear cells (PBMC) by short-term culture in the presence of IL-2, have been used in adoptive immunotherapy for cancer patients with disappointing results overall (Bordignon et al., 1999). Four patients with overt EBV-LPD, who had had an incomplete or no response to discontinuation of immunosuppression and antiviral therapy within a 3-week period; received LAK cell infusions after solid-organ transplantation. (Nalesnik et al., 1998). All four patients had a complete response, but two developed controllable transplant rejections. One patient died of pneumonia 41 days after LAK-cell infusion; there was no evidence of EBV-LPD at autopsy. These results have not been duplicated. In addition to cellular immunotherapy, monoclonal antibodies have been used for the treatment of EBV-LPD. Of 58 patients treated with anti-CD21 and anti-CD24 murine monoclonal antibodies in a European multicenter study, 35 (61%) entered complete remission (Fischer et al., 1991), but these antibodies are no longer available. In addition murine monoclonal antibodies may cause anaphylaxis and the production of neutralizing human anti-mouse antibodies (HAMA). To reduce the incidence of HAMA, several manufacturers have developed chimeric murine/human monoclonal antibodies, one of the most successful examples being Rituximab (Rituxan; Genentech and IDEC Pharmaceuticals Corporation), an anti-CD20 monoclonal antibody (Maloney, 2000; Maloney et al., 1997; Levy, 2000). It has been used successfully in the treatment of CD20-positive B-cell NHL and is an effective agent for the treatment of EBV-LPD. In a multicenter retrospective analysis of 32 patients with EBV-LPD developing after solid-organ or hematopoietic stem-cell transplantation, Rituximab was well tolerated and the overall response rate was 69%, with 20 complete responses and two partial

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

187

responses (Milpied et al., 2000). Our own experience with Rituximab administered after hematopoietic stem-cell or solid-organ transplantation has been very favorable. All nine patients given the antibody after transplantation of hematopoietic stem cells (n = 5; Kuehnle et al., 2000) or a solid organ (three liver and one kidney) had a complete response. One patient, who had received high-dose steroids and anti-thymocyte immunoglobulin (ATG) for GVHD prior to Rituximab therapy, died of aspergillosis. Three of the ﬁve patients who underwent stem cell transplantation have now been followed for more than 1.5 years after Rituximab infusion, with no evidence of EBVLPD recurrence. Besides B-cell antibodies, anti-interleukin 6 (IL-6) monoclonal antibodies have been used to treat EBV-LPD (Haddad et al., 2001). In multicenter phase I and II clinical trials, 12 patients were treated with anti-IL-6 antibodies after solid organ transplantation, with ﬁve achieving complete remission and three partial remission. Further studies are needed to deﬁne the role of anti-IL-6 antibodies in therapy for EBV-LPD. Despite these promising results, the overall efﬁcacy of monoclonal antibodies in the treatment of EBV-LPD after solid-organ transplantation remains difﬁcult to assess, since the majority of studies lacked controls and were confounded by the use of other therapeutic agents. As with T-cell therapies for EBV-LPD, the success of monoclonal antibodies will most likely depend on early diagnosis and intervention. Such therapy seems to be a reasonable alternative in clinical situations where EBV-speciﬁc CTL are not available. However, the profound B-cell depletion after monoclonal antibody therapy with anti-CD20, -CD21, and -CD24 may exacerbate the immunodeﬁciency in transplant recipients, and a lack of EBV-infected B cells could potentially delay recovery of EBV-speciﬁc immunity, increasing the risk of EBVLPD late in the posttransplant period. Neither of these complications was observed in our transplant recipients who were treated with Rituximab. It is also possible that monoclonal antibody therapy results in selection of B cells negative for the targeted antigen, as reported in some lymphoma patients after anti-CD20 therapy (Davis et al., 1999).

III. HODGKIN’S DISEASE Hodgkin’s disease is a malignant neoplasm of lymphoreticular cell origin, and 40–50% of cases in immunocompetent patients are associated with expression of EBV-derived antigens in malignant Reed–Sternberg (H-RS) cells and their variants Herbst et al., 1990; Pallesen et al., 1991). Early-stage Hodgkin’s disease has an excellent prognosis with a disease-free survival rate of 80–95% (Diehl et al., 1999). Yet, among all childhood cancer survivors, the absolute risk for a second neoplasm was highest among survivors of Hodgkin’s disease (Neglia et al., 2001). Advanced-stage Hodgkin’s disease

188

Stephen Gottschalk et al.

has a 5-year progression-free survival rate of 50–65% when treated with standard chemotherapy regimens. Higher survival rates have been reported with the use of newer regimens, but deﬁnitive evaluation of these treatments requires additional follow-up (Conners et al., 1998; Diehl et al., 1999). Autologous hematopoietic stem-cell transplantation has been used as salvage therapy for relapsed patients with 5-year progression-free survival rates of 40–71%. Complications include restrictive lung disease, infections, and second malignancies in 5–24% of patients. It is therefore desirable to develop novel therapies that could improve disease-free survival in relapsed/refractory patients and that might ultimately reduce the incidence of long-term treatment-related complications in all patients. In contrast to EBV-LPD, only a limited number of EBV-derived antigens (EBNA-1, LMP1, LMP2, and BARF0) are present in cases of EBV-positive Hodgkin’s disease (Herbst et al., 1991; Oudejans et al., 1995). Nevertheless, these viral antigens provide legitimate targets for CTL immunotherapy. The use of autologous polyclonal EBV-speciﬁc CTL is being evaluated in two federally approved phase I clinical trials for patients with EBV-positive Hodgkin’s disease with multiple relapses (group A) or with minimal residual disease following autologous stem-cell transplant (group B) (Heslop et al., 2000; Roskrow et al., 1998). The expansion rate of EBV-speciﬁc CTLs of Hodgkin’s patients was found to be ∼10-fold lower than in healthy donors. The generation of sufﬁcient numbers of EBV-speciﬁc CTL for infusion in these patients required weekly stimulation with a mitogenic cocktail consisting of IL-2, allogeneic irradiated PBMC, LCL, and anti-CD3 beginning on culture day 21. In group A, eight patients have received 2 × 107 cells/m2 (n = 1), 4 × 107 cells/m2 (n = 6), or 1.2 × 108 cells/m2 (n = 1). In seven of these patients, CTL were marked with the neomycin resistance gene. Three patients in group B have received two unmarked CTL injections of 2 × 107 cells/m2. No immediate toxicities were seen in either group. In group A, two patients developed transient “ﬂu-like symptoms,” and another developed spinal cord compression due to progressive disease 1 week after CTL infusion and was withdrawn from the protocol. One patient had bulky mediastinal disease and died 2 months after CTL infusion; tumor erosion through the left upper lobe bronchus was found at autopsy. Five patients with aggressive disease at the time of CTL infusion survived 8–13 months, and two were alive at 18 and 24 months post infusion. The three patients in group B were alive and well 9–12 months after CTL infusion. The availability of tetrameric complexes of MHC class I molecules (tetramer) loaded with a speciﬁc peptide (Altman et al., 1996) allowed us to directly quantify EBV-speciﬁc CD8-positive T cells in four patients from group A, who were HLA-A2 positive. A tetramer capable of recognizing an HLA-A2 restricted epitope of LMP2, CLGGLLTMV (CLG), was used to determine the frequency of LMP2-speciﬁc CTL in the CTL products and in patient’s peripheral blood samples both pre and post infusion. An average of 0.67%

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

189

(0.02–1.61%) CD8-positive T cells were positive for this epitope in CTL lines, compared with 2% (0.3–5.57%) in healthy controls. The frequency of CLG tetramer-positive CD8-positive T cells increased at least 2- to 20-fold in peripheral blood samples obtained up to 9 months after CTL infusions, by comparison to preinfusion samples. Moreover, CLG tetramer-positive cells and gene-marked CTL localized to a malignant pleural effusion in one patient 3 weeks after CTL infusion. Additional evidence that EBV-speciﬁc CTL are able to accumulate at tumor sites came from autopsy ﬁndings in the patient who died of progressive disease with tumor erosion through the left upper lobe bronchus. Since CTL therapy for EBV-LPD caused morbidity from tissue inﬂammation (Section II.B), tumor sections were analyzed by in situ PCR for the presence of gene-marked CTL, which were identiﬁed with in the tumor (Fig. 2; see color insert), but not at the erosion site. These studies demonstrate that autologous EBV-speciﬁc CTLs are well tolerated in patients with Hodgkin’s disease and that they can localize to the tumor and persist for prolonged periods of time. However, none of the CTL-treated patients had a complete response. This failure could in part be due to inhibitory factors such as TGF-β, TARC, IL-10, IL-13, or Fas ligand produced by H-RS cells (Newcom and Gu, 1995; Poppema et al., 1998; Skinnider et al., 2001; van den Berg et al., 1999). Decreased MHC class I presentation of EBV-speciﬁc epitopes on H-RS cells is unlikely, since EBVpositive H-RS cells have an intact MHC class I presenting pathway (Murray et al., 1998; Lee et al., 1998). Finally, the lack of CTL efﬁcacy may simply be quantitative, in that the current method of EBV-speciﬁc CTL generation produces CTL lines that are dominated by clones reactive to viral proteins (Fig. 3), such as the immunodominant EBNAs 3A, 3B, and 3C, which are not expressed in Hodgkin’s disease (Rickinson and Moss, 1997). Of the four EBV-associated antigens expressed in H-RS cells, only LMP1 and LMP2 are potential antigens for the generation of antigen-speciﬁc EBV latent proteins EBNAs 3A, 3B, 3C LMP2 EBNA-2, LMP1, LP EBNA-1, BARFO

EBV-specific CD8-positive CTL

Fig. 3 Hierarchy of MHC class I-restricted EBV epitopes. The EBV-speciﬁc CD8-positive CTL response in normal healthy individuals is skewed toward EBNAs 3A, 3B, and 3C. EBV antigenspeciﬁc CTL detected with activity against EBV-infected target cells ( ). EBV antigen). speciﬁc CTL detected with no activity against EBV-infected target cells (

190

Stephen Gottschalk et al.

CD8-positive CTL. EBNA-1 is not processed for HLA class I presentation (Levitskaya et al., 1995), and attempts to identify BARF0-speciﬁc CTL have produced lymphocytes that recognized target cells expressing BARF0 at high levels but failed to recognize the antigen expressed by EBV in lymphoblastoid cell lines (Kienzle et al., 1998). LMP1-speciﬁc CTL clones are rare in EBV-positive donors, and LMP1 by itself displays considerable heterogeneity in its C terminus (Khanim et al., 1996; Khanna et al., 1998; Knecht et al., 1993). Thus, we and others have focused on the generation of LMP2speciﬁc CTL (Gahn et al., 2001; Ranieri et al., 1999; Su et al., 2001). In preclinical studies it was possible to generate LMP2-speciﬁc CTL from PBMC when dendritic cells were used as antigen-presenting cells expressing LMP2. In contrast to the standard LCL protocol, this DC-based CTL stimulation strategy led to preferential expansion of LMP2-speciﬁc CTLs. The generated CTL killed not only autologous ﬁbroblasts infected with a vaccinia construct expressing LMP2, but also autologous LCL, in which LMP2 is naturally expressed by EBV. On the basis of these encouraging in vitro results, we are currently modifying our clinical protocols to administer LMP2-speciﬁc CTL to patients with multiple relapsed EBV-positive Hodgkin’s disease.

IV. NASOPHARYNGEAL CARCINOMA Nasopharyngeal carcinoma (NPC) arises from the epithelial cells of the nasopharynx. Almost all nonkeratinizing and undifferentiated NPC are associated with EBV. However, other environmental or genetic factors must play an important role in oncogenesis, since the incidence of NPC varies 50- to 100-fold from southern China to western countries (Hsu and Glaser, 2000). Clinical experience with cellular immunotherapy for NPC is quite limited. As in Hodgkin’s disease only a limited number of EBV latent genes are expressed, namely EBNA-1, LMP1, LMP2, and BARF0. Lee et al. (2000) have investigated the CTL response to EBV in 21 healthy controls and 10 NPC patients in southern China by reactivating EBV-speciﬁc CTL in vitro from donor PBMC. The activated EBV-speciﬁc CTL were cloned and analyzed for function and epitope speciﬁcity. LMP2-speciﬁc CTLs were identiﬁed albeit at a low frequency in NPC patients as well as healthy controls. It was also shown that two NPC-derived cell lines were able to process and present endogenously synthesized proteins in an HLA class I–restricted fashion. These studies support the concept of immunotherapeutic strategies to boost or elicit LMP2-speciﬁc CTL responses in NPC patients. A clinical trial is being conducted to boost the EBV-speciﬁc CTL response against latent EBV proteins with peptide vaccination, but results are not yet available (A. B. Rickinson, personal communication).

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

191

V. BURKITT’S LYMPHOMA Burkitt’s lymphoma is a high-grade malignant small noncleaved cell lymphoma. Whereas nearly all endemic Burkitt’s lymphomas in equatorial Africa are associated with EBV, the virus has been implicated less often in sporadic cases, and in the United States only 20% of Burkitt’s lymphoma are EBV positive (Hsu and Glaser, 2000). Burkitt’s lymphoma cells evade the immune system by down-regulating the expression of EBV latency antigens, cell adhesion molecules, and MHC class I molecules. EBNA-1, one of only two expressed EBV antigens, is not processed for HLA class I presentation because of an internal glycine-alanine repeat region and thus cannot be recognized by HLA class I–restricted CD8-positive CTL (Levitskaya et al., 1995). The other EBV antigen, BARF0, is not expressed at high enough levels to be recognized by BARF0-speciﬁc CTL. Two groups have shown that the down-regulation of MHC class I molecules in Burkitt’s lymphoma cells can be reversed by CD40 cross-linking in vitro (Frisan et al., 1999; Khanna et al., 1997). Despite signiﬁcant up-regulation of MHC class I molecules and TAP-1 and TAP-2 (peptide transporters associated with antigen presentation), the expression of EBV-associated proteins was not increased, so that EBV-speciﬁc CTL therapy for Burkitt’s lymphoma may be problematic. However, the demonstration of EBNA-1-speciﬁc CD4-positive CTL provides impetus to explore the role of CD4 T cells in the control and therapy of Burkitt’s lymphoma (Bickham et al., 2001; Munz et al., 2000). Finally, the possibility of up-regulating the MHC class I presentation pathway in Burkitt’s lymphoma cells may open the opportunity to search for other lymphoma-speciﬁc antigens as targets for T-cell therapy.

VI. EBV-ASSOCIATED NON-HODGKIN’S LYMPHOMA IN HIV PATIENTS Patients infected with HIV are at high risk to develop NHL. Depending on certain histologic features, the EBV association ranges from 30% in systemic HIV-related Burkitt’s lymphoma (HIV-BL) to 70–80% in HIV-related immunoblastic lymphoma (HIV-IBL), and virtually all cases of primary central nervous system lymphoma (PCNSL) are EBV positive (Hamilton-Dutoit et al., 1993; 1991). In biopsies of EBV-associated HIV-NHL, there is considerable variation in the number of EBV-positive cells, and the pattern of EBV latent gene expression varies among tumor types as in immunocompetent individuals. In one study, 9 of 22 EBV-positive HIV-IBL showed a lymphoblastoid cell-like pattern of gene expression, as judged by positivity for

192

Stephen Gottschalk et al.

LMP1 and EBNA-2; 7 of the remaining cases were LMP1 positive/EBNA-2 negative; and 6 cases were negative for both antigens (Hamilton-Dutoit et al., 1993). Among 11 cases of EBV-positive HIV-BL, none was positive for EBNA-2 and 3 cases were positive for LMP1. Two other reports documented that the majority of EBV-associated PCNSL cases are positive for LMP1 and EBNA-2 (Auperin et al., 1994; Camilleri-Broet et al., 1995). Clearly, a number of HIV-NHL express EBV antigens that could be used as targets for EBV-speciﬁc CTL therapy. However, except for one case report there is no published clinical experience with T-cell therapy for EBVassociated HIV-NHL. The patient described by Wheatley et al. (1997) had an EBV-positive B-cell lymphoma that was refractory to chemotherapy, but regressed after treatment with EBV-speciﬁc CTL generated from archived autologous PBMC. In HIV-infected patients, the development of EBV-associated HIV-NHL is preceded by a loss of functional EBV-speciﬁc CTL (van Baarle et al., 2001), suggesting that strategies to boost the endogenous EBV-speciﬁc T-cell response might prevent lymphomas. In contrast to other HIV-related malignancies, the incidence of HIV-NHL has not declined after the introduction of highly active antiretroviral therapy (HAART). One explanation is that once EBV-speciﬁc CD4-positive T cells are depleted from the patient’s immunological repertoire, a full and sustained cellular immune response to EBV cannot be restored (Porcu and Caligiuri, 2000). Thus, more active strategies to boost the EBV-speciﬁc T-helper cell response to EBV are needed. Obtaining PBMC from HIV-infected patients early in the clinical course (when their EBV-speciﬁc cellular immunity is still intact) for later generation of EBV-speciﬁc T cells would be a reasonable option. Such a strategy would require that patients at high risk for the development of EBV-associated HIV-NHL be identiﬁed by monitoring EBV-DNA load and serially determining EBV-speciﬁc T-cell responses (Subklewe et al., 1999; Tan et al., 1999; Yang et al., 2000; van Baarle et al., 2001). If however, the HIV infection of the patient is not controlled with HAART therapy, the protection offered by infused EBV-speciﬁc T cells is likely to be short-lived.

VII. CONCLUSIONS Adoptive immunotherapy with EBV-speciﬁc CTL has proved to be a feasible strategy for reconstituting immunity and preventing or treating EBV-LPD in patients at high risk for this complication after hematopoietic stem-cell transplantation. Although the ex vivo expansion of clinical-grade antigenspeciﬁc T cells is time consuming and requires a laboratory facility operating under current Good Manufacturing Practices, its success has highlighted several advantages of this approach over conventional vaccination strategies. First, it allows for T-cell stimulation and expansion in the absence of

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

193

immunosuppressive medications or tumor-derived immunosuppressive factors. Second, the phenotype, activity, and speciﬁcity of expanded T cells can be analyzed prior to injection. Third, gene marking of ex vivo expanded T cells allows for tracking of injected T cells to assess persistence, efﬁcacy, and toxicity. In addition, infused T cells can be altered by changing their effector function or modifying their receptor speciﬁcity. For example, T cells can be genetically modiﬁed to render them resistant to inhibitory cytokines or to enhance their function by increasing the levels of cytotoxic cytokines they produce at local tumor sites. The demonstrated efﬁcacy of CTL therapy against EBV-LPD developing after hematopoietic stem-cell transplantation warrants further exploration of this technology for the prophylaxis and treatment of EBV-LPD after solidorgan transplantation. Novel methods for generating EBV-speciﬁc CTL from EBV-seronegative donors, who are at particular risk for the development of EBV-LPD after solid-organ transplantation, are needed. Procedures for more rapid generation of EBV-speciﬁc CTL would be especially advantageous in patients with EBV-LPD for whom CTL were not prepared in advance. Further improvement of EBV-speciﬁc CTL therapy for Hodgkin’s disease will require methods to activate and expand CTL that are speciﬁc for latent EBV genes expressed in Hodgkin’s disease and to genetically modify the expanded CTL to render them resistant to inhibitory cytokines. A similar approach should be feasible for nasopharyngeal carcinoma since the pattern of EBV latency gene expression is similar to that in Hodgkin’s disease. For Burkitt’s lymphoma the prospect of EBV-speciﬁc CTL therapy is more problematic, since EBNA-1 is not presented on MHC class I molecules. Even so, EBNA-1-speciﬁc CD4-positive T cells recognizing MHC class II-restricted epitopes may have therapeutic potential. Whether adoptive immunotherapy with EBV-speciﬁc CTL will prove effective for the majority of EBV-associated malignancies and improve longterm outcome is unclear. The results obtained so far are certainly sufﬁciently encouraging to justify continued active exploration of this approach. If effective, such strategies would have broad implications for the adoptive immunotherapy of a range of human tumors with deﬁned tumor antigens and similar immune evasion strategies.

ACKNOWLEDGMENTS We thank John W. Sixbey (Louisiana State Health Center, Shreveport, LA) for performing the in situ PCR and John Gilbert for expert scientiﬁc editing. This work was supported by National Institutes of Health grants RO1 CA61384 and RO1 CA74126. S.G. is the recipient of a Doris Duke clinical scientist development grant. H. E. H. is the recipient of a Doris Duke distinguished clinical scientist award.

194

Stephen Gottschalk et al.

REFERENCES Altman, J. D., Moss, P. A. H., Goulder, P. J. R., Barouch, D. H., McHeyzer-Williams, M. G., Bell, J. I., McMichael, A. J., and Davis, M. M. (1996). Phenotypic analysis of antigen-speciﬁc T lymphocytes. Science 274, 94–96. Auperin, I., Mikolt, J., Oksenhendler, E., Thiebaut, J. B., Brunet, M., Dupont, B., and Morinet, F. (1994). Primary central nervous system malignant non-Hodgkin’s lymphomas from HIV-infected and non-infected patients: expression of cellular surface proteins and Epstein–Barr viral markers. Neuropathol. Appl. Neurobiol. 20, 243–252. Babcock, J. G., Hochberg, D., and Thorley-Lawson, A. D. (2000). The expression pattern of Epstein–Barr virus latent genes in vivo is dependent upon the differentiation stage of the infected B cell. Immunity 13(4), 497–506. Baldanti, F., Grossi, P., Furione, M., Simoncini, L., Sarasini, A., Comoli, P., Maccario, R., Fiocchi, R., and Gerna, G. (2000). High levels of Epstein–Barr virus DNA in blood of solidorgan transplant recipients and their value in predicting posttransplant lymphoproliferative disorders. J. Clin. Microbiol. 38, 613–619. Bickham, K., Munz, C., Tsang, M. L., Larsson, M., Fonteneau, J. F., Bhardwaj, N., and Steinman, R. (2001). EBNA1-speciﬁc CD4+ T cells in healthy carriers of Epstein–Barr virus are primarily Th1 in function. J. Clin. Invest. 107, 121–130. Bonini, C., Ferrari, G., Verzeletti, S., Servida, P., Zappone, E., Ruggieri, L., Ponzoni, M., Rossini, S., Malvilio, F., Traversari, C., and Bordignon, C. (1997). HSV-TK gene transfer into donor lymphocytes for control of allogeneic graft versus leukemia. Science 276, 1719–1724. Bordignon, C., Carlo-Stella, C., Colombo, M. P., De Vincentiis, A., Lanata, L., Lemoli, R. M., Locatelli, F., Olivieri, A., Rondelli, D., Zanon, P., and Tura, S. (1999). Cell therapy: achievements and perspectives. Haematologica 84, 1110–1149. Camilleri-Broet, S., Davi, F., Feuillard, J., Bourgeois, C., Seilhean, D., Hauw, J. J., and Raphael, M. (1995). High expression of latent membrane protein 1 of Epstein–Barr virus and BCL-2 oncoprotein in acquired immunodeﬁciency syndrome-related primary brain lymphomas. Blood 86, 432–435. Casper, J., Camitta, B., Truitt, R., Baxter-Lowe, L. A., Bunin, N., Lawton, C., Murray, K., Hunter, J., Pietryga, D., Garbrecht, F., Keever Taylor, C., Drobyski, W., Horowitz, M., Flomenberg, N., and Ash, R. (1995). Unrelated bone marrow donor transplants for children with leukemia or myelodysplasia. Blood 85, 2354–2363. Cavazzana-Calvo, M., Bensoussan, D., Jabado, N., Haddad, E., Yvon, E., Moskwa, M., Tachet, des Combes, Buisson, M., Morand, P., Virion, J. M., Le Deist, F., and Fischer, A. (1998). Prevention of EBV-induced B-lymphoproliferative disorder by ex vivo marrow B-cell depletion in HLA-phenoidentical or non-identical T-depleted bone marrow transplantation. Br. J. Haematol. 103, 543–551. Cohen, J. I. (2000). Epstein–Barr virus infection. N. Engl. J. Med. 343, 481–492. Comoli, P., Locatelli, F., Gerna, G., Grossi, P., Vigano, M., and Maccario, R. (1997). Autologous EBV-speciﬁc cytotoxic T cells to treat EBV-associated post-transplant lymphoproliferative disease (PTLD). Blood 90, 249a. Conners, J. M., Reece, D. E., Diehl, V., and Engbert, A. (1998). In “Hematology.” (J. R. McArthur, G. P. Schechter, and S. L. Schrier, eds.), pp. 274–295. American Society of Hematology, Washington, D.C. Curtis, R. E., Travis, L. B., Rowlings, P. A., Socie, G., Kingma, D. W., Banks, P. M., Jaffe, E. S., Sale, G. E., Horowitz, M. M., Witherspoon, R. P., Shriner, D. A., Weisdorf, D. J., Kolb, H. J., Sullivan, K. M., Sobocinski, K. A., Gale, R. P., Hoover, R. N., Fraumeni, Jr., J. F., and Deeg, H. J. (1999). Risk of lymphoproliferative disorders after bone marrow transplantation: a multi-institutional study. Blood 94, 2208–2216.

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

195

Darenkov, I. A., Marcarelli, M. A., Basadonna, G. P., Friedman, A. L., Lorber, K. M., Howe, J. G., Crouch, J., Bia, M. J., Kliger, A. S., and Lorber, M. I. (1997). Reduced incidence of Epstein–Barr virus-associated posttransplant lymphoproliferative disorder using preemptive antiviral therapy. Transplantation 64, 848–852. Davis, T. A., Czerwinski, D. K., and Levy, R. (1999). Therapy of B-cell lymphoma with antiCD20 antibodies can result in the loss of CD20 antigen expression. Clin. Cancer Res. 5, 611–615. Diehl, V., Mauch, P., and Conners, J. M. (1999). In “Hematology” (G. P. Schechter, R. Hoffman, and S. L. Schrier, eds.), pp. 270–289. American Society of Hematology, Washington, D.C. Emanuel, D. J., Lucas, K. G., Mallory, G. B., Edwards-Brown, M. K., Pollok, K. E., Conrad, P. D., Robertson, K. A., and Smith, F. O. (1997). Treatment of posttransplant lymphoproliferative disease in the central nervous system of a lung transplant recipient using allogeneic leucocytes. Transplantation 63, 1691–1694. Epstein, M. A., Achong, B. G., and Barr, Y. M. (1964). Virus particles in cultured lymphoblasts from Burkitt’s lymphoma. Lancet 1, 702–703. Epstein, M. A., Morgan, A. J., Finerty, S., Randle, B. J., and Kirkwood, J. K. (1985). Protection of cotton top against Epstein–Barr virus-induced malignant lymphoma by a prototype subunit vaccine. Nature 318, 287–289. Faulkner, G. C., Krajewski, A. S., and Crawford, D. H. (2000). The ins and outs of EBV infection. Trends Microbiol. 8, 185–189. Fischer, A., Blanche, S., LeBidois, J., Bordigoni, P., Garnier, J. L., Neaudet, P., Morinet, F., Le Deist, F., Fischer, A-M., Griscelli, C., and Hirn, M. (1991). Anti-B-cell monoclonal antibodies in the treatment of severe B-cell lymphoproliferative syndrome following bone marrow and organ transplantation. N. Engl. J. Med. 324, 1451–1456. Fong, I. W., Ho, J., Toy, C., Lo, B., and Fong, M. W. (2000). Value of long-term administration of acyclovir and similar agents for protecting against AIDS-related lymphoma: case-control and historical cohort studies. Clin. Infect. Dis. 30, 757–761. Frisan, T., Donati, D., Cervenak, L., Wilson, J., Masucci, M. G., and Bejarano, M. T. (1999). CD40 cross-linking enhances the immunogenicity of Burkitt’s-lymphoma cell lines. Int. J. Cancer 83, 772–779. Gahn, B., Siller-Lopez, F., Pirooz, A. D., Yvon, E., Gottschalk, S., Longnecker, R., Brenner, M. K., Heslop, H. E., Aguilar-Cordova, E., and Rooney, C. M. (2001). Adenoviral gene transfer into dendritic cells efﬁciently ampliﬁes the immune response to the LMP2A-antigen: a potential treatment strategy for Epstein–Barr virus-positive Hodgkin’s lymphoma. Int. J. Cancer 93, 706–713. Gerritsen, E. J., Stam, E. D., Hermans, J., van den Berg, H., Haraldsson, A., van Tol, M. J., van den Bergh, R. L., Waaijer, J. L., Kroes, A. C., Kluin, P. M., and Vosssen, J. M. (1996). Risk factors for developing EBV-related B cell lymphoproliferative disorders (BLPD) after non-HLA-identical BMT in children. Bone Marrow Transplant. 18, 377– 382. Gottschalk, S., Ng, C. Y. C., Smith, C. A., Perez, M., Sample, C., Brenner, M. K., Heslop, H. E., and Rooney, C. M. (2001). An Epstein–Barr virus deletion mutant that causes fatal lymphoproliferative disease unresponsive to virus-speciﬁc T cell therapy. Blood 97, 835–843. Green, M., Bueno, J., Rowe, D., Mazariegos, G., Qu, L., Abu-Almagd, K., and Reyes, J. (2000). Predictive negative value of persistent low Epstein–Barr virus viral load after intestinal transplantation in children. Transplantation 70, 593–596. Gross, T. G., Hinrichs, S. H., Davis, J. R., Mitchell, D., Bishop, M. R., and Wagner, J. E. (1998). Depletion of EBV-infected cells in donor marrow by counterﬂow elutriation. Exp. Hematol. 26, 395–399.

196

Stephen Gottschalk et al.

Gross, T. G., Steinbuch, M., DeFor, T., Shapiro, R. S., McGlave, P., Ramsay, N. K. C., Wagner, J. E., and Filipovich, A. H. (1999). B cell lymphoproliferative disorders following hematopoietic stem cell transplantation: Risk factors, treatment and outcome. Bone Marrow Transplant. 23, 251–258. Gustafsson, A., Levitsky, V., Zou, J. Z., Frisan, T., Dalianis, T., Ljungman, P., Ringden, O., Winiarski, J., Ernberg, I., and Masucci, M. G. (2000). Epstein–Barr virus (EBV) load in bone marrow transplant recipients at risk to develop posttransplant lymphoproliferative disease: prophylactic infusion of EBV-speciﬁc cytotoxic T cells. Blood 95, 807–814. Haddad, E., Paczesny, S., Leblond, V., Seigneurin, J. M., Stern, M., Achkar, A., Bauwens, M., Delwail, V., Debray, D., Duvoux, C., Hubert, P., de Ligny, B. H., Wijdenes, J., Durandy, A., and Fischer, A. (2001). Treatment of B-lymphoproliferative disorder with a monoclonal anti-interleukin-6 antibody in 12 patients: a multicenter phase 1–2 clinical trial. Blood 97, 1590–1597. Hale, G., and Waldmann, H., and for CAMPATH users. (1998). Risks of developing Epstein– Barr virus-related lymphoproliferative disorders after T-cell-depleted marrow transplants. Blood 91, 3079–3083. Hamilton-Dutoit, S., Rea, D., Raphael, M., Sandvej, K., Delecluse, H., Gisselbrecht, C., Marelle, L., van Krieken, J, and Pallesen, G. (1993). Epstein–Barr virus-latent gene expression and tumor cell phenotype in aquired immunodeﬁciency syndrome-related non-Hodgkin’s lymphoma. Am. J. Pathol. 143, 1072–1085. Hamilton-Dutoit, S. J., Pallesen, G., Franzmann, M. B., Karkov, J., Black, F., Skinhoj, P., and Pedersen, C. (1991). AIDS-related lymphoma. Histopathology, immunophenotype, and association with Epstein–Barr virus as demonstrated by in situ nucleic acid hybridization. Am. J. Pathol. 138, 149–163. Haque, T., Amlot, P. L., Helling, N., Thomas, J. A., Sweny, P., Rolles, K., Burroughs, A. K., Prentice, H. G., and Crawford, D. H. (1998). Reconstitution of EBV-speciﬁc T cell immunity in solid organ transplant recipients. J. Immunol. 160, 6204–6209. Herbst, H., Dallenback, F., Hummel, M., Niedobitek, G., Pileri, S., Muller-Lantzsch, N., and Stein, H. (1991). Epstein–Barr virus latent membrane protein expression in Hodgkin and Reed-Sternberg cells. Proc. Natl. Acad. Sci. USA 88, 4766–4770. Herbst, H., Niedobitek, G., Kneba, M., Hummel, M., Finn, T., Anagnostopoulos, I., Bergholz, M., Krieger, G., and Stein, H. (1990). High incidence of Epstein–Barr virus genomes in Hodgkin’s disease. Am. J. Pathol. 137, 13–18. Heslop, H. E., and Rooney, C. M. (1997). Adoptive immunotherapy of EBV lymphoproliferative diseases. Immunol. Rev. 157, 217–222. Heslop, H., Rooney, C., Brenner, M., Krance, R., Carrum, G., Gahn, B., Bollard, C., Khan, S., Gee, A., Popat, U., Gresik, M., Przepiorka, D., Kuehnle, I., and Grilley, B. (2000). Administration of neomycin resistance gene-marked EBV-speciﬁc cytotoxic T-lymphocytes as therapy for patients receiving a bone marrow transplant for relapsed EBV-positive Hodgkin disease. Hum. Gene Ther. 11, 1465–1475. Heslop, H. E., Brenner, M. K., and Rooney, C. M. (1994). Donor T cells to treat EBV-associated lymphoma. N. Engl. J. Med. 331, 679–680. Heslop, H. E., Ng, C. Y. C., Li, C., Smith, C. A., Loftin, S. K., Krance, R. A., Brenner, M. K., and Rooney, C. M. (1996). Long-term restoration of immunity against Epstein–Barr virus infection by adoptive transfer of gene-modiﬁed virus-speciﬁc T lymphocytes. Nat. Med. 2, 551–555. Hsu, J. L., and Glaser, S. L. (2000). Epstein-barr virus-associated malignancies: epidemiologic patterns and etiologic implications. Crit. Rev. Oncol. Hematol. 34, 27–53. Imashuku, S., Goto, T., Matsumura, T., Naya, M., Yamori, M., Hojo, M., Hibi, S., and Todo, S. (1998). Unsuccessful CTL transfusion in a case of post-BMT Epstein–Barr virus-associated lymphoproliferative disorder (EBV-LPD). Bone Marrow Transplant. 20, 337–340.

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

197

Kanno, H., Ohsawa, M., Hashimoto, M., Iuchi, K., Nakajima, Y., and Aozasa, K. (1999). HLA-A alleles of patients with pyothorax-associated lymphoma: anti-Epstein–Barr virus (EBV) host immune responses during the development of EBV latent antigen-positive lymphomas. Int. J. Cancer 82, 630–634. Kerr, B. M., Lear, A. L., Rowe, M., Croom-Carter, D., Young, L. S., Rookes, S. M., Gallimore, P. H., and Rickinson, A. B. (1992). Three transcriptionally distinct forms of Epstein–Barr virus latency in somatic cell hybrids: Cell phenotype dependence of virus promoter usage. Virology 187, 189–201. Khanim, F., Yao, Q. Y., Niedobitek, G., Sihota, S., Rickinson, A. B., and Young, L. S. (1996). Analysis of Epstein–Barr virus gene polymorphisms in normal donors and in virus-associated tumors from different geographic locations. Blood 88, 3491–3501. Khanna, R., Bell, S., Sherritt, M., Galbraith, A., Burrows, S. R., Rafter, L., Clarke, B., Slaughter, R., Falk, M. C., Douglass, J., Williams, T., Elliott, S. L., and Moss, D. J. (1999a). Activation and adoptive transfer of Epstein–Barr virus-speciﬁc cytotoxic T cells in solid organ transplant patients with posttransplant lymphoproliferative disease. Proc. Natl. Acad. Sci. USA 96, 10391–10396. Khanna, R., Burrows, S. R., Nicholls, J., and Poulsen, L. M. (1998). Identiﬁcation of cytotoxic T cell epitopes within Epstein–Barr virus (EBV) oncogene latent membrane protein 1 (LMP1): evidence for HLA A2 supertype-restricted immune recognition of EBV-infected cells by LMP1-speciﬁc cytotoxic T lymphocytes. Eur. J. Immunol. 28, 451–458. Khanna, R., Cooper, L., Kienzle, N., Moss, D. J., Burrows, S. R., and Khanna, K. K. (1997). Engagement of CD40 antigen with soluble CD40 ligand up-regulates peptide transporter expression and restores endogenous processing function in Burkitt’s lymphoma cells. J. Immunol. 159, 5782–5785. Khanna, R., Moss, D. J., and Burrows, S. R. (1999b). Vaccine strategies against Epstein–Barr virus-associated diseases: lessons from studies on cytotoxic T-cell-mediated immune regulation. Immunol. Rev. 170, 49–64. Kienzle, N, Sculley, T. B., Poulson, L., Buck, M., Cross, S., Raab-Traub, N., and Khanna, R. (1998). Identiﬁcation of a cytotoxic T-lymphocyte response to the novel BARF0 protein of Epstein–Barr virus: a critical role for antigen expression. J. Virol 72, 6614–6620. Knecht, H., Bachmann, E., Brousset, P., Sandvej, K., Nadal, D., Bachmann, F., Odermatt, B. F., Delsol, G., and Pallesen, G. (1993). Deletions within the LMP1 oncogene of EBV are clustered in Hodgkin’s disease and identical to those observed in nasophayngeal carcinoma. Blood 82, 2937–2942. Kuehnle, I., Huls, M. H., Liu, Z., Semmelmann, M., Krance, R. A., Brenner, M. K., Rooney, C. M., and Heslop, H. E. (2000). CD20 monoclonal antibody (rituximab) for therapy of Epstein–Barr virus lymphoma after hemopoietic stem-cell transplantation. Blood 95, 1502– 1505. Lee, S. P., Chan, A. T., Cheung, S. T., Thomas, W. A., CroomCarter, D., Dawson, C. W., Tsai, C. H., Leung, S. F., Johnson, P. J., and Huang, D. P. (2000). CTL control of EBV in nasopharyngeal carcinoma , (NPC): EBV-speciﬁc CTL responses in the blood and tumors of NPC patients and the antigen-processing function of the tumor cells. J. Immunol. 165, 573–582. Lee, S. P., Constandinou, C. M., Thomas, W. A., Croom-Carter, D., Blake, N. W., Murray, P. G., Crocker, J., and Rickinson, A. B. (1998). Antigen presenting phenotype of Hodgkin Reed–Sternberg cells: analysis of the HLA class I processing pathway and the effects of interleukin-10 on Epstein-barr virus-speciﬁc cytotoxic T-cell recognition. Blood 92, 1020– 1030. Levitskaya, J., Coram, M., Levitsky, V., Imreh, S., Steigerwald-Mullen, P. M., Klein, G., Kurilla, M. G., and Masucci, M. G. (1995). Inhibition of antigen processing by the internal repeat region of the Epstein–Barr virus nuclear antigen-1. Nature 375, 685–688.

198

Stephen Gottschalk et al.

Levy, R. (2000). A perspective on monoclonal antibody therapy: where we have been and where we are going. Semin. Hematol 37, 43–46. Liu, Z., Wilson, J. M., Jones, M. C, Khan, S. P., Krance, R. A., Brenner, M. K., Rooney, C. M., Heslop, H. E., and Gee, A. P. (1999). Addition of B cell depletion of donor marrow with anti-CD20 antibody to a T cell depletion regimen for prevention of EBV Lymphoma after bone marrow transplant. Blood 94(10) Suppl 1, 638a. Lucas, K. G., Burton, R. L., Zimmerman, S. E., Wang, J., Cornetta, K. G., Robertson, K. A., Lee, C. H., and Emanuel, D. J. (1998). Semiquantitative Epstein–Barr virus , (EBV) polymerase chain reaction for the determination of patients at risk for EBV-induced lymphoproliferative disease after stem cell transplantation. Blood 91, 3654–3661. Maloney, G. D. (2000). Monoclonal antibodies in lymphoid neoplasia: principles for optimal combined therapy. Semin. Hematol. 37, 17–26. Maloney, D. G., Grillo-Lopez, A. J., White, C. A., Bodkin, D., Schilder, R. J., Neidhart, J. A., Janakiraman, N., Foon, K. A., Liles, T. M., Dallaire, B. K., Royston, I., Davis, T., and Levy, R. (1997). IDEC-C2B8 , (Rituximab) anti-CD20 monoclonal antibody therapy in patients with relapsed low-grade non-Hodgkin’s lymphoma. Blood 90, 2188–2195. Milpied, N., Vasseur, B., Parquet, N., Garnier, J. L., Antoine, C., Quartier, P., Carret, A. S., Bouscary, D., Faye, A., Bourbigot, B., Reguerre, Y., Stoppa, A. M., Bourquard, P., de Ligny, B. H., Dubief, F., Mathieu-Boue, A., and Leblond, V. (2000). Humanized anti-CD20 monoclonal antibody , (Rituximab) in post transplant B-lymphoproliferative disorder: a retrospective analysis on 32 patients. Ann. Oncol. 11 Suppl 1, 113–116. Moghaddam, A., Rosenzweig, M., Lee-Parritz, D., Annis, B., Johnson, R. P., and Wang, F. (1997). An animal model for acute and persistent Epstein–Barr virus infection. Science 276, 2030–2033. Moss, D. J., Schmidt, C., Elliott, S., Suhrbier, A., Burrows, S., and Khanna, R. (1996). Strategies involved in developing an effective vaccine for EBV-associated diseases. Adv. Cancer Res. 69, 213–245. Munz, C., Bickham, K. L., Subklewe, M., Tsang, M. L., Chahroudi, A., Kurilla, M. G., Zhang, D., O’Donnell, M., and Steinman, R. M. (2000). Human CD4(+) T lymphocytes consistently respond to the latent Epstein–Barr virus nuclear antigen EBNA1. J. Exp. Med 191, 1649– 1660. Murray, P. G., Constandinou, C. M., Crocker, J., Young, L. S., and Ambinder, R. F. (1998). Analysis of major histocompatibility complex class I, TAP expression, and LMP2 epitope sequence in Epstein–Barr virus-positive Hodgkin’s disease. Blood 92, 2477–2483. Nagafuji, K., Eto, T., Hayashi, S., Oshima, K., Maeda, Y., Gondo, H., Inamura, T., and Niho, Y. (1998). Donor lymphocyte transfusion for the treatment of Epstein–Barr virusassociated lymphoproliferative disorder of the brain. Bone Marrow Transplant. 21, 1155– 1158. Nalesnik, M. A., Rao, A. S., Furukawa, H., Pham, S., Zeevi, A., Fung, J. J., Klein, G., Grittsch, H. A., Elder, E., Whiteside, T. L., and Starzl, T. E. (1998). Autologous lymphokine-activated killer cell therapy of Epstein–Barr virus-positive and -negative lymphoproliferative disorders arising in organ transplant recipients. Transplantation 63, 1200–1205. Neglia, J. P., Friedman, D. L., Yasui, Y., Mertens, A. C., Hammond, S., Stovall, M., Donaldson, S. S., Meadows, A. T., and Robison, L. L. (2001). Second malignant neoplasms in ﬁve-year survivors of childhood cancer: childhood cancer survivor study. J. Natl. Cancer Inst. 93, 618– 629. Newcom, S. R., and Gu, L. (1995). Transforming growth factor beta 1 messenger RNA in Reed–Sternberg cells in nodular sclerosing Hodgkin’s disease. J. Clin. Pathol. 48, 160– 163. Nikiforow, S., Bottomly, K., and Miller, G. (2001). CD4+ T-cell effectors inhibit Epstein–Barr virus-induced B-cell proliferation. J. Virol. 75, 3740–3752.

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

199

O’Reilly, R. J., Small, T. N., Papadopoulos, E., Lucas, K., Lacerda, J., and Koulova, L. (1997). Biology and adoptive cell therapy of Epstein–Barr virus-associated lymphoproliferative disorders in recipients of marrow allografts. Immunol. Rev. 157, 195–216. O’Reilly, R. J., Small, T. N., Papadopoulos, E., Lucas, K., Lacerda, J., and Koulova, L. (1998). Adoptive immunotherapy for Epstein–Barr virus-associated lymphoproliferative disorders complicating marrow allografts. Springer Semin. Immunopathol. 20, 455–491. Orazi, A., Hromas, R. A., Neiman, R. S., Greiner, T. C., Lee, C. H., Rubin, L., Haskins, S., Heerema, N. A., Ghapure, V., Abonour, R., Srour, E. F., and Cornetta, K. (1997). Posttransplantation lymphoproliferative disorders in bone marrow transplant recipients are aggressive diseases with a high incidence of adverse histologic and immunobiologic features. Am. J. Clin. Pathol. 107, 419–429. Orentas, R. J., Lemas, M. V., Mullin, M. J., Colombani, P. M., Schwartz, K., and Ambinder, R. (1998). Feasibility of cellular adoptive immunotherapy for Epstein–Barr virus-associated lymphomas using haploidentical donors. J. Hematother. 7, 257–261. Oudejans, J. J., van den Brule, A. J., Jiwa, N. M., de Bruin, P. C., Ossenkoppele, G. J., van, der, V, Walboomers, J. M., and Meijer, C. J. (1995). BHRF1, the Epstein–Barr virus (EBV) homologue of the BCL-2 protooncogene, is transcribed in EBV-associated B-cell lymphomas and in reactive lymphocytes. Blood 86, 1893–1902. Pallesen, G., Hamilton-Dutoit, S. J., Rowe, M., and Young, L. S. (1991). Expression of Epstein–Barr virus latent gene products in tumour cells of Hodgkin’s disease. Lancet 337, 320–322. Papadopoulos, E. B., Ladanyi, M., Emanuel, D., MacKinnon, S., Boulad, F., Carabasi, M. H., Castro-Malaspina, H., Childs, B. H., Gillio, A. P., Small, T. N., Young, J. W., Kernan, N. A., and O’Reilly, R. J. (1994). Infusions of donor leukocytes to treat Epstein–Barr virus-associated lymphoproliferative disorders after allogeneic bone marrow transplantation. N. Engl. J. Med. 330, 1185–1191. Poppema, S., Potters, M., Visser, L., and van den Berg, A. M. (1998). Immune escape mechanisms in Hodgkin’s disease. Ann. Oncol. 9 Suppl 5, S21–S24. Porcu, P., and Caligiuri, M. A. (2000). Acquired immunodeﬁciency syndrome–related lymphomas: future directions. Semin. Oncol. 27, 454–462. Ranieri, E., Herr, W., Gambotto, A., Olson, W., Rowe, D., Robbins, P. D., Kierstead, L. S., Watkins, S. C., Gesualdo, L., and Storkus, W. J. (1999). Dendritic cells transduced with an adenovirus vector encoding Epstein–Barr virus latent membrane protein 2B: a new modality for vaccination. J. Virol. 73, 10416–10425. Rickinson, A. B., and Moss, D. J. (1997). Human cytotoxic T lymphocyte responses to Epstein– Barr virus infection. Annu. Rev. Immunol. 15, 405–431. Riddell, S. R., and Greenberg, P. D. (2000). T-cell therapy of cytomegalovirus and human immunodeﬁciency virus infection. J. Antimicrob. Chemother. 45 Suppl T3, 35–43. Riddell, S. R., Watanabe, K. S., Goodrich, J. M., Li, C. R, Agha, M. E., and Greenberg, P. D. (1992). Restoration of viral immunity in immunodeﬁcient humans by the adoptive transfer of T cell clones. Science 257, 238–241. Rooney, C. M., Loftin, S. K., Holladay, M. S., Brenner, M. K., Krance, R. A., and Heslop, H. E. (1995). Early identiﬁcation of Epstein–Barr virus-associated post-transplant lymphoproliferative disease. Br. J. Haematol. 89, 98–103. Rooney, C. M., Smith, C. A, Ng, C. Y. C., Loftin, S. K., Sixbey, J. W., Gan, Y-J., Srivastava, D-K., Bowman, L. C., Krance, R. A., Brenner, M. K., and Heslop, H. E. (1998). Infusion of cytotoxic T cells for the prevention and treatment of Epstein–Barr virus-induced lymphoma in allogeneic transplant recipients. Blood 92, 1549–1555. Rose, C., Green, M., Webber, S., Ellis, D., Reyes, J., and Rowe, D. (2001). Pediatric solid-organ transplant recipients carry chronic loads of Epstein–Barr virus exclusively in the immunoglobulin D-negative B-cell compartment. J. Clin. Microbiol. 39, 1407–1415.

200

Stephen Gottschalk et al.

Roskrow, M. A., Suzuki, N., Gan, Y-J., Sixbey, J. W., Ng, C. Y. C., Kimbrough, S., Hudson, M. M., Brenner, M. K., Heslop, H. E., and Rooney, C. M. (1998). EBV-speciﬁc cytotoxic T lymphocytes for the treatment of patients with EBV positive relapsed Hodgkin’s disease. Blood 91, 2925–2934. Rowe, D. T., Qu, L., Reyes, J., Jabbour, N., Yunis, E., Putnam, P., Todo, S., and Green, M. (1997). Use of quantitative competitive PCR to measure Epstein–Barr virus genome load in the peripheral blood of pediatric transplant patients with lymphoproliferative disorders. J. Clin. Microbiol. 35, 1612–1615. Sasahara, Y., Kawai, S., Itano, M., Kumaki, S., Fujie, H., Minegishi, M., Tsuchiya, S., and Konno, T. (1998). Epstein–Barr virus-associated lymphoproliferative disorder after unrelated bone marrow transplantation in a young child with Wiskott–Aldrich syndrome. Pediatr. Hematol. Oncol. 15, 347–352. Skinner, J. C., Gilbert, E. F., Hong, R., Bozdech, M. J., Flynn, B., Borcherding, W., Hafez, G. R., Arya, S., Trigg, M. E., Finlay, J. L., and Sondel, P. M. (1988). B cell lymphoproliferative disorders following T cell depleted allogeneic bone marrow transplantation. Am. J. Pediatr. Hematol./Oncol. 10, 112–119. Skinnider, B. F., Elia, A. J., Gascoyne, R. D., Trumper, L. H., von Bonin, F., Kapp, U., Patterson, B., Snow, B. E., and Mak, T. W. (2001). Interleukin 13 and interleukin 13 receptor are frequently expressed by Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma. Blood 97, 250–255. Stevens, S. J., Verschuuren, E. A., Pronk, I., van Der, Bij W., Harmsen, M. C., The, T. H., Meijer, C. J., van den Brule, A. J., and Middeldorp, J. M. (2001). Frequent monitoring of Epstein–Barr virus DNA load in unfractionated whole blood is essential for early detection of posttransplant lymphoproliferative disease in high-risk patients. Blood 97, 1165– 1171. Su, Z., Peluso, M. V., Raffegerst, S. H., Schendel, D. J., and Roskrow, M. A. (2001). The generation of LMP2a-speciﬁc cytotoxic T lymphocytes for the treatment of patients with Epstein–Barr virus-positive Hodgkin disease. Eur. J. Immunol. 31, 947–958. Subklewe, M., Chahroudi, A., Bickham, K., Larsson, M., Kurilla, M. G., Bhardwaj, N., and Steinman, R. M. (1999). Presentation of epstein-barr virus latency antigens to CD8(+), interferon-gamma-secreting, T lymphocytes. Eur. J. Immunol. 29, 3995–4001. Tan, L. C., Gudgeon, N., Annels, N. E., Hansasuta, P., O’Callaghan, C. A., Rowland-Jones, S., McMichael, A. J., Rickinson, A. B., and Callan, M. F. (1999). A re-evaluation of the frequency of CD8+ T cells speciﬁc for EBV in healthy virus carriers. J. Immunol. 162, 1827–1835. Thomis, D. C., Marktel, S., Bonini, C., Traversari, C., Gilman, M., Bordignon, C., and Clackson, T. (2001). A Fas-based suicide switch in human T cells for the treatment of graft-versus-host disease. Blood 97, 1249–1257. Thorley-Lawson, D. A., and Babcock, G. J. (1999). A model for persistent infection with Epstein–Barr virus: the stealth virus of human B cells. Life Sci 65, 1433–1453. Tiberghien, P. (2001). Use of suicide gene-expressing donor T-cells to control alloreactivity after haematopoietic stem cell transplantation. J. Intern. Med. 249, 369–377. Tiberghien, P., Ferrand, C., Lioure, B., Milpied, N., Angonin, R., Deconinck, E., Certoux, J. M., Robinet, E., Saas, P., Petracca, B., Juttner, C., Reynolds, C. W., Longo, D. L., Herve, P., and Cahn, J. Y. (2001). Administration of herpes simplex-thymidine kinase-expressing donor T cells with a T-cell-depleted allogeneic marrow graft. Blood 97, 63–72. Tierney, R. J., Steven, N., Young, L. S., and Rickinson, A. B. (1994). Epstein–Barr virus latency in blood mononuclear cells: Analysis of viral gene transcription during primary infection and in the carrier state. J. Virol. 68, 7374–7385. van Baarle, D., Hovenkamp, E., Callan, M. F. C., Wolthers, K. C., Kostense, S., Tan, L. C., Niesters, H. G. M., Osterhaus, A. D. M. E., McMichael, A. J., van Oers, M. H. J., and Miedema, F. (2001). Dysfunctional Epstein–Barr virus (EBV)–speciﬁc CD8+

Treatment of EBV-Associated Malignancies with Speciﬁc T Cells

201

T lymphocytes and increased EBV load in HIV-1 infected individuals progressing to AIDSrelated non-Hodgkin lymphoma. Blood 98, 146–155. van den Berg, A., Visser, L., and Poppema, S. (1999). High expression of the CC chemokine TARC in Reed-Sternberg cells. A possible explanation for the characteristic T-cell inﬁltratein Hodgkin’s lymphoma. Am. J. Pathol. 154, 1685–1691. Walter, E. A., Greenberg, P. D., Gilbert, M. J., Finch, R. J., Watanabe, K. S., Thomas, E. D., and Riddell, S. R. (1995). Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N. Engl. J. Med. 333, 1038–1044. Wheatley, H. G., McKinnon, K., Lilly, S., Smith, C., and Lyerly, H. K. (1997). Adoptive immunotherapy using autologous EBV-speciﬁc CTL in a HIV infected patient with refractory Epstein–Barr virus expressing B cell lymphoma. J. Acquir. Immune Deﬁc. Syndr. Hum. Retrovirol. 14 (4), A53. Wilson, A. D., Lovgren-Bengtsson, K., Villacres-Ericsson, M., Morein, B., and Morgan, A. J. (1999). The major Epstein–Barr virus (EBV) envelope glycoprotein gp340 when incorporated into Iscoms primes cytotoxic T-cell responses directed against EBV lymphoblastoid cell lines. Vaccine 17, 1282–1290. Yang, J., Lemas, V. M., Flinn, I. W., Krone, C., and Ambinder, R. F. (2000). Application of the ELISPOT assay to the characterization of CD8(+) responses to Epstein–Barr virus antigens. Blood 95, 241–248. Yao, Q. Y., Ogan, P., Rowe, M., Wood, M., and Rickinson, A. B. (1989a). Epstein–Barr virusinfected B cells persist in the circulation of acyclovir-treated virus carriers. Int. J. Cancer 43, 67–71. Yao, Q. Y., Ogan, P., Rowe, M., Wood, M., and Rickinson, A. B. (1989b). The Esptein-Barr virus : host balance in acute infectious mononucleosis patients receiving acyclovir anti-viral therapy. Int. J. Cancer 43, 61–66. Young, L. S., Dawson, C. W., and Eliopoulos, A. G. (2000). The expression and function of Epstein–Barr virus encoded latent genes. Mol. Pathol. 53, 238–247.

This Page Intentionally Left Blank

Role of Glycogen Synthase Kinase-3 in Cancer: Regulation by Wnts and Other Signaling Pathways Armen S. Manoukian and James R. Woodgett∗ Divisions of Experimental Therapeutics and Cellular and Molecular Biology Ontario Cancer Institute Toronto, Ontario, Canada, M5G 2M9

I. Origins of GSK-3 II. GSK-3 Functions: Clues From Other Species A. Lord of the Flies B. The Components III. Mammalian Wnt Signaling and Cancer A. Role of GBPs B. Role of APC IV. Regulation of GSK-3 by Other Signaling Pathways V. Other Targets of GSK-3 Involved in Cancer A. Regulation of Nuclear Factor-κB VI. GSK-3 and Cancer References

Although glycogen synthase kinase-3 (GSK-3) is but one of more than a thousand distinct serine/threonine kinases present in the mammalian genome, this enzyme has attracted attention for its role in a diverse range of cellular processes and its positioning at a nexus of several signaling pathways that are important in cancer and other human diseases. The association of GSK-3 with widely different functions, from glycogen metabolism to fruit ﬂy segmentation and slime mold differentiation, was initially perplexing. However, as the context of the biological processes involving this enzyme has been clariﬁed, unifying themes have emerged that begin to explain its pleiotropic nature. Unlike most protein kinases involved in signaling, GSK-3 is active in unstimulated, resting cells. Its activity is inactivated during cellular responses and its substrates therefore tend to be dephosphorylated. As more of these targets have been identiﬁed and the effects of their modiﬁcation by GSK-3 determined, most have been found to be functionally inhibited by GSK-3. Hence, this kinase appears to act as a general repressor, keeping its targets switched off or inaccessible under resting conditions. The rarity of this form of regulation is perhaps related to the diversity of its targets. Over the past decade, the importance of GSK-3 has been established by three signiﬁcant properties: its remarkable evolutionary conservation, allowing analysis in genetically tractable organisms; its ∗ Corresponding

author.

204

Armen S. Manoukian and James R. Woodgett

involvement in the Wnt/wingless signaling pathway; and its inhibition by agonists of the prosurvival phosphatidylinositol 3 kinase (PI3 K) pathway. This review covers recent advances in understanding the physiological roles of this enzyme, particularly in the context of cancer. C 2002 Academic Press.

I. ORIGINS OF GSK-3 GSK-3 has exhibited a split personality from the very beginning of its known existence. The name derives from studies by the Cohen group that classiﬁed various entities capable of phosphorylating glycogen synthase (GS), the rate-limiting enzyme of glycogen synthesis. At the time (in the late 1970s), cyclic AMP–dependent protein kinase (PKA) and phosphorylase kinase were known to phosphorylate and inactivate GS but did not account for all GS kinase activity present in skeletal muscle extracts. Part of the residual activity was due to an entity subsequently puriﬁed as GSK-3 (Embi et al., 1980). At the same time, an activity termed Factor A was identiﬁed as deﬁned by its effect on the activation of a latent protein phosphatase (Vandenheede et al., 1980). Puriﬁcation of GSK-3 to homogeneity proved that it was responsible for Factor-A-like activity (Hemmings et al., 1981). This preparation consisted primarily of a 51 kDa protein. Isolation of Factor A activity from cow brain, however, yielded two proteins of ∼52 and 46 kDa, the latter thought to be a regulatory subunit (Tung and Reed, 1989). Isolation of cDNA clones of GSK-3 revealed two highly related gene products that were termed GSK-3α (51 kDa) and GSK-3β (47 kDa) (Woodgett, 1990). Within the canonical protein kinase catalytic domain, these two isoforms are >95% identical, with similarity dropping off toward the peripheral region of the protein. The mass difference is due to an extra glycine-rich domain present near the amino terminus of GSK-3α. The two genes map to human chromosomes 19q13.2 (GSK-3α) and 3q13.3 (GSK-3β) (Hansen et al., 1997). In vitro, the two isoforms cannot be distinguished by biochemical parameters (Hughes et al., 1992), although the physiological functions of these enzymes do not entirely overlap (see later discussion). The molecular cloning of GSK-3 revealed a major surprise. Two independent groups working to isolate a gene involved in fate speciﬁcation in Drosophila isolated cDNAs encoding a protein kinase that was highly related to GSK-3 (in fact, the three cloning papers were published within 35 days of one another). One group termed the ﬂy gene zeste-white3 based on genetic mapping of its chromosomal location (Siegfried et al., 1990); the other termed it shaggy, in reference to the “disheveled” nature of one of the phenotypes (Bourouis et al., 1990). Proof that GSK-3 was the mammalian homolog of Zeste-white3/Shaggy (herein termed Zw3Sgg) derived from experiments in

Regulation of GSK-3

205

which expression of the mammalian protein in ﬂies mutant for zw3sgg rescued the phenotype (Siegfried et al., 1992; Ruel et al., 1993a). These data also suggested that GSK-3β was more efﬁcient than GSK-3α in complementing the mutation. Zw3Sgg is differentially spliced to yield several polypeptides, and several of these show higher similarity to GSK-3β than to GSK-3α. However, the expression level of GSK-3α was lower than that achieved for β in the ﬂies, possibly accounting for the reduced efﬁciency (Ruel et al., 1993a). There may be a trivial explanation for this since the 5 region of GSK-3α (encoding its amino terminus) is very GC rich, resulting in lower translation from some promoters. Unfortunately, the apparent failure of GSK-3α to complement the zw3sgg mutation has lead to neglect of this isoform with the majority of subsequent research being conducted on GSK-3β. As a consequence, there is a strong “β” bias in the literature. As will be discussed, it is now clear that mammalian GSK-3α can substitute for much (but not all) of the functionality of GSK-3β, especially in terms of Wnt signaling.

II. GSK-3 FUNCTIONS: CLUES FROM OTHER SPECIES GSK-3 is a highly evolutionarily conserved enzyme. Homologs have been identiﬁed in every eukaryote investigated to date. In many cases, multiple isoforms or related genes exist, but all share at least 70% identity in the catalytic domain. In several cases, expression of mammalian GSK-3 in simpler organisms has conﬁrmed functional similarity. In addition to the Drosophila rescue mentioned earlier, for example, injection of a kinase-inactive form of GSK-3α or β into Xenopus embryos mimics the effect of injection of the cognate Xenopus mutant (He et al., 1995). This functional conservation is important, for it has facilitated analysis of the physiological functions of the kinase in genetically tractable organisms and allowed a degree of conﬁdence that these functions are relevant to mammals. The role of GSK-3/Zw3Sgg in the Wingless (Wg) pathway will ﬁrst be described in Drosophila, since study of this species has revealed the lion’s share of the components. With this background, the mammalian homologs and their relationship to human cancer will then be elaborated.

A. Lord of the Flies Arguably the most insightful studies of the role of GSK-3 have issued from analysis of Zw3Sgg in fruit ﬂies. The groups investigating this gene were doing so owing to its effects on cell fate determination in the developing

206

Armen S. Manoukian and James R. Woodgett

embryo. Mutations in this gene generated phenotypes that were opposite to the effects of loss of several genes that had been ordered into a pathway by genetic analysis—the wingless (wg) pathway. Epistasis experiments with Zw3Sgg and other elements of the pathway not only established that the gene was a critical player, but also placed the gene product with respect to the components of the pathway. At the time, the biochemistry of the pathway lagged well behind the genetics, but this information provided important clues to the possible interactions between the products of these genes. The wingless (wg) gene in Drosophila is a member of the Wnt gene family. The Wnt gene family members encode secreted glycoproteins that have been shown to act as extracellular signals during the development of diverse organisms (reviewed in Cadigan and Nusse, 1997). Related family members have now been identiﬁed in a variety of organisms as well as within the same species (Nusse et al., 1991; Wodarz and Nusse, 1998). Because of the conservation of Wnt genes and their importance, the elucidation of Wnt gene function has been of paramount interest for diverse disciplines. These proteins are approximately 400 amino acid residues long, have a hydrophobic amino terminus, and are highly conserved at the protein sequence level within a pattern of 24 cysteine residues (Nusse and Varmus, 1992). In short, the sequence of the Wnts predicts that they act as secreted factors. Wnt-1, the ﬁrst Wnt gene identiﬁed, is a common site of insertion for the mouse mammary tumor virus (MMTV) (Nusse and Varmus, 1982). MMTV viral integration results in the ectopic expression of Wnt-1 in the mammary gland and subsequent transformation. Transgenic studies in mice have also demonstrated the ability of Wnt-1 to transform tissues (Tsukamoto et al., 1988; Kwan et al., 1992; Shackleford et al., 1993). Wnt-1 can transform cells via a paracrine mechanism. Nonresponsive cells transfected with Wnt-1 can cause morphological and proliferative changes in cocultured mammary epithelial cells that do not express Wnt-1 (Jue et al., 1992). The conservation of wg as a member of the Wnt gene family is exempliﬁed by its ability to substitute for Wnt-1 in the tissue culture transformation assay (Ramakrishna and Brown, 1993). As the Wnt genes encode ligand signals, the identiﬁcation of signal transduction components required for their function has been the focus of much research. In the developing Drosophila embryo, wg is transcribed in single-cell-wide stripes in a segmental pattern (Fig. 1; see color insert). However, the requirement for wg genetic activity extends well beyond the expression domain of its transcripts. As a result of the absence of wg function during embryogenesis, the polarity of segments in the larva is lost and a “lawn” of denticles is all that remains in the ventral region. Also, during embryogenesis, wg has been shown to be required for the maintenance of transcription of several other

Regulation of GSK-3

207

segment polarity genes, such as engrailed (en), in adjacent cells (DiNardo et al., 1988). Taken together, these data suggest that wg acts as a signal to instruct adjacent cell fates. In addition, cell biological analyses have shown that secreted wg protein (Wg) is detected several cell diameters away from the cells that transcribe wg (Gonzalez et al., 1991; van den Heuvel et al., 1993). Wg has also been detected inside en-expressing cells (Gonzalez et al., 1991). The relationship between en and wg has been explored further in tissue-culture experiments where isolated en-expressing cells maintain en-expression only when cocultured with Wg-expressing cells (Cumberledge and Krasnow, 1993). Imaginal discs of Drosophila larvae are epithelial sacs of cells that are destined to give rise to the adult ﬂy body. During early larval development, wg is required for the development of imaginal discs where it is widely expressed (Couso et al., 1993). Later on, wg is required for the speciﬁcation and the maintenance of the dorsal–ventral axis of discs. In wing imaginal discs, cell fate determination of the wing margin is under the control of wg signaling as well as other signaling pathways (Neumann and Cohen, 1996). Loss of wg at the end of the larval stage results in the loss of margin cell fates as judged by loss of the margin bristles (Couso et al., 1994). During this time, wg is expressed in a narrow row of cells that map to the presumptive wing margin and separate dorsal and ventral cell populations. This last function of wg has been quite useful in mosaic experiments. Using the presence or absence of margin bristles as an assay, genetic mosaic experiments have shown wg to act non-cell-autonomously, again indicating that wg encodes a secreted extracellular gene product (Morata and Lawrence, 1977). Studies in imaginal discs have also shown that Wg functions as a morphogen gradient during development (Zecca et al., 1996). What, then, are the mechanisms that underlie the response to Wg? Here, genetics again played an important role in setting the ground rules. As the wg phenotype in the embryo is stereotypical, it has been possible to identify wg signaling pathway genes through the analysis of mutants with phenotypes resembling the “lawn” pattern of denticles seen in wg mutant embryos (Fig. 2). Historically, this approach was successful at identifying several new components of wg signaling in Drosophila. Three genes were shown to have mutant phenotypes identical to that of wg: dishevelled (dsh), porcupine (porc), and armadillo (arm) (Perrimon et al., 1989; Wieschaus and Riggleman, 1987). Although wg is a zygotic gene and has a restricted pattern of expression, dsh, porc, and arm are expressed both maternally and zygotically and have a ubiquitous pattern of expression. Therefore, in order to observe embryonic phenotypes, germline clones (GLC) have to be induced in females to remove all maternal and zygotic expression of these genes (Perrimon et al., 1989). Just as in the case of wg itself, these genes were

Fig. 2 Phenotypes of wg and zw3sgg in Drosophila. The wild-type embryo hatches into a fully segmented larva (shown with a preparation to illuminate the cuticle). The cuticle of the wild-type larva is segmented, with distinct denticle (illuminated hairs in the photograph) and “naked” (devoid of hairs) regions within each segment. In a wg mutant embryo, all “naked” regions of the cuticle are lost and the remaining cuticular regions are fused into a general “lawn” of denticles. Ectopic expression of wg gives the opposite phenotype of general “naked” cuticle, and a zw3sgg mutant embryo has an identical phenotype.

Regulation of GSK-3

209

shown to be required for the maintenance of en expression in the embryo, as well as for other wg-dependent processes during development (reviewed in Klingensmith and Nusse, 1994). The difference in these mutants was revealed in the study of their behavior in genetic mosaics. The porc mutation (like wg) behaves in a non-cell-autonomous fashion, whereas dsh and arm behave cell-autonomously (Kadowaki et al., 1996; Klingensmith et al., 1994; Wieschaus and Riggleman, 1987). Simple interpretation of these tests suggested that as in the case of wg, porc was required outside the cell receiving the Wg signal whereas dsh and arm were required inside the cell. Since wg is expressed in a restricted single-cell-wide striped pattern, it was possible to ubiquitously express wg to generate a phenotype that can be interpreted to be the opposite of the wg mutant phenotype (Fig. 2). This was initially done by inducing the expression of wg from a heat-inducible transgene (HsWg). In embryos expressing HsWg, en expression expands and the resulting larva are devoid of denticles in the ventral region of the cuticle (Noordermeer et al., 1992). Instead, these larva have “naked” cuticle, which is part of the reiterating pattern of segmentation in wild-type larvae. Simplifying these phenotypes, we can say that in wg mutants the “naked” regions are deleted, leaving only denticles, and in HsWg embryos the denticles are deleted, leaving only the naked regions. As HsWg creates the opposite phenotype to that of wg, it was possible to perform genetic epistasis tests to determine relationship between the components of the wg genetic pathway. In the absence of porc, HsWg can still specify “naked” cuticle and maintain en expression (Noordermeer et al., 1994). This suggests that porc functions upstream of wg as we predicted earlier. HsWg is, however, completely ineffective in the absence of either dsh or arm, suggesting that these genes function downstream of wg (Noordermeer et al., 1994). As Zw3Sgg has a phenotype similar to that of HsWg (Fig. 2), it was postulated that in zw3 mutants, the wg signaling pathway is constitutively active (Siegfried et al., 1994). It was therefore possible to make double mutants with zw3 and each of the wg pathway genes. Double mutant epistatic analysis placed zw3 function downstream of wg, porc, and dsh but upstream of arm (Siegfried et al., 1994). Therefore, the genetics have played an important role in setting the ground rules. The epistasis data indicated Zw3Sgg functions in an opposite manner to gene products that are both upstream and downstream of it, suggesting that Zw3Sgg normally acts to repress its target(s) and that it is, itself, regulated by signal-induced inactivation. Without the genetic clues, it is doubtful that such a complicated scenario would have been contemplated. However, the essence of this “double-negative” model for Zw3Sgg regulation in the Wg pathway has been borne out by subsequent studies— although a number of interesting twists and players have been added over time.

210

Armen S. Manoukian and James R. Woodgett

B. The Components Wg is a secreted protein with ligand-like properties such as disulﬁde bonds. Several gene products are required for its function, including some that operate within the cell secreting the protein. Media conditioned by cells engineered to overexpress Wg can be added to resting cells and elicit biochemical effects in those cells, an effect dependent on the presence of Wg in the conditioned medium, since the effects are inhibited by anti-Wg antibodies. Wg interacts with more than one moiety on the surface of its target cells. Diffusion of Wg in embryo cell layers is extremely limited and its half-life in an unbound state is very short. These effects are caused by binding of Wg to heparin sulfate proteoglycans (HSPs) on the cell surface. Inactivation of genes encoding HSP synthetic enzymes results in a phenotype similar to loss of Wg. Similarly, injection of embryos with heparinase results in rapid loss of Wg protein and ectopic denticle formation. Thus, free Wg appears to be rapidly degraded but transport of HSP-bound Wg is slow. These factors conspire to tightly localize the Wg sphere of inﬂuence and constrain the responder cells to those in the local area of the secreting cells. Binding of Wg to HSPs does not trigger a signal. For this to occur, Wg must bind to a serpentine receptor-like molecule termed Frizzled 2 (Frz2) (Bhanot et al., 1996). Wg associates with an extracellular, N-terminal, cysteine-rich domain on Frz2 and, by so doing, initiates an intracellular signal. In several species, decoy proteins have been identiﬁed that interfere with Wg binding to Frz2. This ligand–receptor interaction is therefore tightly regulated. Downstream of the Frz2 transmembrane receptor is a phosphoprotein termed Dishevelled (Dsh). The phosphorylation state of Dsh is sensitive to Wg. However, the meaning of this change is unclear as Dsh has no obvious catalytic function and changes in Dsh phosphorylation do not directly correlate with Wg signaling. Investigations of protein kinases acting on Dsh have revealed an associated kinase, termed DAK, casein kinase II and casein kinase Iε (Sakanaka et al., 1999, 2000; Willert et al., 1997; Peters et al., 1999). Whether or how the activity of these enzymes is regulated by Wg is not known, nor is their effect on Dsh function. However, overexpression of CKIε can mimic some of the downstream consequences of Wg signaling (Sakanaka et al., 1999; Peters et al., 1999; McKay et al., 2001). Dsh contains three potential interaction motifs, a DIX domain, a DEP domain, and a PDZ domain. The DIX motif is shared with another component of the pathway, Axin, and the two proteins can associate through this domain. Both the DIX and PDZ domains are required for Dsh function in the Wg response (Yanagawa et al., 1995). The DEP domain is shared with a Caenorhabditis elegans protein, Egl-10, and pleckstrin. A role for this part of the protein in Wg functions is unclear, but Dsh moonlights in regulating another signaling system, the Drosophila Jun kinase pathway that regulates planar polarity (Boutros et al., 1998). An intact DEP domain is necessary for this second pathway.

Regulation of GSK-3

211

Downstream of Dsh lies Zw3sgg. The fact that this is a well-characterized serine kinase strongly suggested that it inactivated its target by phosphorylating it. The problem was that the only candidate target was Armadillo (Arm), and despite much effort, Zw3sgg could not be enticed to phosphorylate Arm in vitro. Yet, Arm phosphorylation could be reduced by Wg expression in embryos and expression of Zw3Sgg increased Arm phosphorylation. Despite this conundrum, the phosphorylation sites on Arm were identiﬁed as a pair of serine residues near the N terminus of the protein. When these residues were phosphorylated, the stability of Arm was decreased. This effect was shown to be due to phosphorylation-directed ubiquitination of Arm. In ﬂies, Arm phosphorylation results in its recognition by the E3 component of the ubiquitin-processing team, which allows the E2-ubiquitin molecule to be coupled to lysine residues in Arm. Once tagged, the modiﬁed Arm is rapidly digested by the 26S proteosome. The E1 and E2 components of ubiquitin ligation are not substrate-speciﬁc and are common to all ubiquitinylations. There are many E3 proteins, however, and their binding speciﬁcity dictates the molecules to be modiﬁed. The E3 protein that binds Arm is termed Slimb (Jiang and Struhl, 1998). The phenotype of slimb mutant embryos is consistent with activated Wg signaling, in which Arm accumulates. Slimb is homologous to βTrCP, a Xenopus protein that regulates Arm in that species. Interestingly, this same E3 protein also regulates ubiquitination of the inhibitory subunit of the NF-κB transcription factor, I-κB (Yaron et al., 1998). The ﬁnding that the slimb mutant phenotype resembles constitutive Wg signaling suggests that despite pleiotropism of function, its effect on Arm regulation is a dominant phenotype. This is also true of Zw3Sgg, which has many other cellular functions outside the realm of Wg, yet when mutated yields a signature Wg-pathway-associated phenotype. Thus, in unstimulated cells Arm is phosphorylated and targeted for rapid degradation. Arm exists in two pools in cells. A signiﬁcant fraction is tightly associated with transmembranous cadherin molecules, which are important in Ca2+ sensitive cell adhesions. A second population is cytoplasmic. In the absence of Wg, only the concentration of this latter pool is maintained at a low level. The cadherin-associated pool of Arm is insensitive to Wg and is not rapidly turned over. The question remained whether there were intermediary proteins between Zw3Sgg and Arm, since the two proteins could not be coerced into an interaction in vitro. The answer came from an unusual source for the Wg pathway—mouse genetics. Analysis of the locus termed “fused” responsible for causing fusion of the embryonic axis in mice revealed the archetypal member of a small family of proteins that act as scaffolds to hold several pieces of the Wg pathway machinery together. Following the isolation of mammalian Axin (and its relative, Axin II or Conductin), a Drosophila ortholog was identiﬁed. This molecule, D-axin, exhibits less than 50% overall identity to mammalian and avian Axins. However, islands of much greater

212

Armen S. Manoukian and James R. Woodgett

similarity are apparent. These map, as might be predicted, to the important functional domains of Axin, as deﬁned biochemically. Beginning at the N terminus, the ﬁrst region of identity shares features present in RGS proteins—regulators of G-protein signaling. This domain mediates an interaction with an important tumor suppressor protein, Adenomatous Polyposis Coli (APC; see Section III.A). This domain is followed by a binding site for Zw3Sgg, which is immediately abutted by a domain that interacts with Arm. A region that interacts with protein phosphatase-2A is followed by the DIX domain, which is also present in Dsh. The DIX domain facilitates Axin dimerization. Axin is thus a clustered collection of interaction domains. The juxtaposition of the Zw3Sgg and Arm binding sites provided insight into at least one of the functions of Axin. Although neither Zw3Sgg nor Arm interact in solution, when both are bound to Axin, Arm becomes a substrate for phosphorylation by Zw3Sgg. Thus, Axin acts as a facilitator to introduce and correctly align these two molecules. Axin is also a substrate for Zw3Sgg, and phosphorylation may affect the afﬁnities of the binding domains. In this respect, the presence of protein phosphatase-2A within the complex is likely relevant. Overexpression of Axin results in increased degradation of Arm, suggesting that it is present at lower levels than Zw3Sgg and can recruit more of the kinase if present at higher concentrations. The biochemical role of APC in the Axin complex, although important, is less clear. APC will be considered in more detail in Section III.B.

1. THE RESTING STATE With all of the (known) components in place, what then is the mechanism by which the Wg pathway operates? The absence of Wg presumably prevents coupling of the receptor, DFz, with Dsh. Inside the cell, the Axin organizer acts to ensnare soluble Arm molecules, presenting them to Zw3Sgg, which adds a couple of phosphate groups, thus initiating their descent into the proteasomal abyss (Fig. 3A). Under these conditions, the Axin complex sweeps the cell, removing Arm and thus preventing formation of the Arm/dTCF transcription factors. The role of Axin phosphorylation is not fully known, but it could possibly act as a ratchet. Phosphorylated Axin has a higher afﬁnity for Arm than unphosphorylated Axin. If phosphorylation of Axin is required for Arm binding, perhaps dephosphorylation of Axin is necessary for Arm release (and degradation). Following dissociation of the phosphoArm, binding of a new Arm molecule would not occur until Axin has been rephosphorylated. This would allow the rate of Arm phosphorylation to be regulatable by the state of Axin phosphorylation. Phosphorylated Axin is more stable that the apo-form. Thus, Zw3Sgg phosphorylation of Axin may provide a means to couple Axin levels to Zw3Sgg activity.

Fig. 3 Regulation of GSK-3 (Zw3/Sgg) by the Wnt signaling pathways. In resting cells GSK-3 is fully active and is present in a ternary complex with β-catenin, Axin, and APC. GSK-3 phosphorylation targets β-catenin for βTrCPmediated ubiquitination and subsequent proteolytic degradation. Steady state β-catenin levels are therefore low. Upon exposure to Wnt, the Dishevelled mediator protein is modiﬁed, and GSK-3 activity associated with the Axin/APC complex is reduced. Phosphorylation of β-catenin is suppressed and the protein accumulates and enhances LEF-1/T cf-mediated transcription.

214

Armen S. Manoukian and James R. Woodgett

2. THE ACTIVATED STATE If there are residual holes in understanding of the resting mode of the Wg pathway, the activated system is almost transparent. Since several of the key molecules have no catalytic activity, the changes that occur upon binding of Wg to DFz2 presumably involve conformational shifts. The structures of the receptor and Dsh have yet to be determined. Simply overexpressing Dsh in cells drives a Wg-like response, suggesting that the ligandless receptor normally represses Dsh function. Given the increase of Wg-induced phosphorylation of Dsh, it is reasonable to suggest that this plays a role: however, it may simply be consequential and not contribute to the actual signal. Arguing against this idea is the ﬁnding that injection of iRNA to casein kinase-1ε (kin-19) in C. elegans phenocopies loss of Wg. There are a few other clues. The presence of a DIX domain in both Dsh and Axin suggests the possibility that the two proteins interact directly, and this has been observed (Kishida et al., 1999; Li et al., 1999b). Thus a Wg-dependent conformational change may promote Dsh–Axin association, which then interferes with the scaffolding functions of Axin. Although many details remain sketchy, Wg signaling acts to break apart or destabilize the Axin-assembled Arm disposal complex (Fig. 3B). The consequential accumulation of cytoplasmic Arm facilitates its association with a family of transcriptional regulators, exempliﬁed by dTCF/pangolin. In the absence of a Wg signal, dTCF is unable to transactivate genes. Indeed, it may act as a repressor. Upon binding of Arm, the activator is restored, resulting in gene transcription. The N- and C-terminal domains of Arm contain transactivation domains, at least as deﬁned by their ability to elevate transcription when fused to an exogenous DNA binding domain (such as GAL4). dTCF also associates with Groucho, a transcriptional repressor molecule. Accumulation of Arm may cause redistribution of Groucho/dTCF complexes, hence alleviating negative gene regulation.

III. MAMMALIAN Wnt SIGNALING AND CANCER As described earlier, understanding of the Wnt pathway has largely evolved from studies in simpler organisms, primarily in Drosophila. Studies on the physiological functions of the mammalian counterparts have beneﬁted from this knowledge. Efforts to understand this signaling system in mammals have been reinforced not only by its role in development but by its subversion in human cancers. Chronic activation of Wnt signaling is the basis for a variety of human malignancies, including colon carcinomas, hepatomas, melanomas, and uterine and ovarian cancers. The archetypal member of the Wnt family, Wnt-1/int-1, was originally identiﬁed as gene ectopically

Regulation of GSK-3

215

expressed via insertion of mouse mammary tumor virus in MMTV-induced mammary adenocarcinomas (Nusse and Varmus, 1982). Other Wnts have also been implicated in human cancer (reviewed in Smalley and Dale, 1999). Two further proteins of the pathway are potential oncogenes: β-catenin and GBP/FRAT (see Section III.A). In addition, mammalian APC and Axin act as classical tumor suppressors (reviewed in Polakis, 2000). Indeed, the ﬁnding that mutations in APC were critical causative elements in familial adenomatous polyposis opened up the study of the Wnt ﬁeld in relation to colon cancer. Although mutations in at least four proteins in the Wnt pathway are associated with human cancer, they all share a common characteristic, namely an increase in the cellular concentration of β-catenin. As discussed earlier, stabilization of β-catenin/Armadillo results in its translocation to the nucleus, binding to transcription factors such as TCF/LEF, and regulation of gene expression. In mammals, stabilized β-catenin causes the induction of an assortment of genes, including several that are known to be associated with cancer: cyclin D1, WISP-1, and c-Myc (Tetsu and McCormick, 1999; Xu et al., 2000; He et al., 1998; reviewed in Hecht and Kemler, 2000). Whether any of these changes is individually sufﬁcient for tumorigenesis is unclear but unlikely. Indeed, transgenic expression of β-catenin in liver generates hepatomegaly without obvious changes in levels of these genes (Cadoret et al., 2001). Several of these target genes are also regulated by other mechanisms, and so it is likely that their dysregulation by β-catenin provides a contributory effect to the phenotype. For example, cyclin D is also regulated at the level of protein stability, by GSK-3 (Diehl et al., 1998; Alt et al., 2000). Although GSK-3 activity is reduced by chronic Wnt signaling (which would lead to stabilization of cyclin D), neoplastic lesions harboring mutations in APC, Axin, or β-catenin should not affect GSK-3 activity per se, at least via Wnt.

A. Role of GBPs Although Drosophila genetics have provided the critical framework for identifying and characterizing the essential components of the Wnt/wg pathway, there are a few important molecules that were ﬁrst identiﬁed in other organisms. One of these is Axin, which was ﬁrst discovered in mice (Zeng et al., 1997) and subsequently revealed to play a similar role in fruit ﬂies (Willert et al., 1999). A second component is somewhat more enigmatic. A small protein was isolated via its ability to induce stabilization of β-catenin in Xenopus and shown to be necessary for establishment of the dorsal axis. This molecule was shown to speciﬁcally bind and inhibit GSK-3 and was termed GSK-3-binding protein (GBP) (Yost et al., 1998) The GSK-3 binding site for GBP appears to be the same as the binding site for Axin, and

216

Armen S. Manoukian and James R. Woodgett

association with these molecules is mutually exclusive. This raises the possibility that GBP could directly regulate Wnt pathway activity by modulating the proportion of GSK-3 bound to Axin. Since GSK-3 can only phosphorylate β-catenin (and hence target it for degradation) when both are bound to Axin, GBP may divert GSK-3 from this task. However, at present it is unclear how GBP might be regulated by Wnt. One possibility is that GBP associates with Dishevelled and is “introduced” to the GSK-3 within the Axin/APC/β-catenin complex via Wnt-induced modiﬁcation of Dishevelled (Li et al., 1999b). One of the wrinkles in this scheme is that a homolog of GBP has yet to be identiﬁed in Drosophila. Since the sequence of the ﬂy genome is known, this suggests that if a ﬂy GBP exists, it displays low sequence identity, is likely redundant (since it has not been identiﬁed by classical genetics), and will only be revealed by functional testing. A role for GBP in Wnt signaling is further reinforced by its relationship to a protein that was ﬁrst identiﬁed via an insertional mutagenesis screen for genes that promoted T-cell lymphomas in mice. This molecule, FRAT1 (Frequently Rearranged in Advanced T-cell lymphoma), is similar to GBP in three domains, including the GSK-3 binding region of GBP (Jonkers et al., 1997). Although FRAT1 overexpression does not induce lymphomagenesis by itself, FRAT is induced upon Moloney murine leukemia virus infection and is often coregulated with Pim1 and Myc. It is currently unclear whether dysregulation of GSK-3 or β-catenin occurs in cells harboring elevated FRAT1, or whether this contributes to the cancer. However, this would appear a reasonable assumption. The common element in Wnt-pathway-associated tumors is the accumulation of β-catenin. Understanding of how this molecule is regulated is therefore at the molecular epicenter of Wnt tumorigenesis research. GSK-3 targets β-catenin on four serines and threonines within the N-terminal region (residues 33, 37, 41, and 45). It is the only identiﬁed protein kinase to date that phosphorylates these residues, and inhibition of GSK-3 selectively blocks their modiﬁcation. Versions of β-catenin that lack these residues are chronically stabilized but are otherwise functionally competent in binding to LEF/TCF and induction of transcription (the transcriptional activation domain is located within the C-terminal region of the molecule). A signiﬁcant fraction of colon cancer cells that have wild-type APC instead harbor mutations in the phosphorylation domain of β-catenin, providing compelling evidence that the critical consequence of APC loss is stabilization of β-catenin. This exclusivity of either an APC or a β-catenin mutation is also observed in aggressive ﬁbromatosis (desmoid tumors) (for discussion, see Polakis, 2000). Other evidence for the critical role of β-catenin stabilization in tumorigenesis has derived from transgenic models in which stabilized mutants of the protein have been expressed in various mouse tissues. Targeted expression of a truncated mutant of human β-catenin to the basal epidermal skin layer induced

Regulation of GSK-3

217

follicle morphogenesis that is usually restricted to embryogenesis (Gat et al., 1998). The animals subsequently developed hair follicular tumors similar to those found in certain familial polyposis syndromes (Chan et al., 1999). Mice expressing Wnt1 in the mammary gland (via the MMTV promoter) develop mammary adenocarcinomas. However, Wnt1 can inﬂuence additional pathways in a cell such as planar polarity via Dishevelled and the JNK/SAPK pathway (Bhanot et al., 1996). Wnts can also induce certain genes independently of β-catenin (Ziemer et al., 2001). It was therefore reassuring to note that MMTV-β-catenin transgenic mice exhibit similar tumorigenic responses to MMTV-Wnt1 mice (Imbert et al., 2001). These mice also demonstrated early differences in mammary gland development and may provide useful clues to changes that occur prior to breast cancer malignancy. A different tack was taken by Harada et al. (1999) in which Cre recombinase was used to selectively excise the phosphorylation-site-containing exon (3) of β-catenin within the epithelial cells lining the intestine. This approach yielded animals that developed adenomatous polyps, a phenotype similar to the well characterized min mouse, which has a germline mutation in APC (reviewed in Heyer et al., 1999).

B. Role of APC Prior to the identiﬁcation of Axin as the scaffolding molecule that assembles β-catenin into a complex with GSK-3 and APC, APC was believed to be the key regulator of the pathway since its loss caused dominant accumulation of β-catenin. Indeed, mutations of Axin in tumors are relatively rare compared with those of APC and/or β-catenin. How, then, does APC modulate the stability of β-catenin? Axin binds to APC via its RGS domain. Deletion of this region decreases the binding of β-catenin to Axin, suggesting that APC association with Axin facilitates catenin binding. In reconstituted systems using puriﬁed components, APC appears to promote incorporation of β-catenin into the Axin complex (and its subsequent phosphorylation). APC can bind both Axin and β-catenin via distinct regions. However, APC mutations in human tumors that do not truncate the molecule tend to be focused in the SAMP repeat domain, which is the region via which APC binds Axin. This suggests that it is the loss of APC binding to Axin that is most relevant to loss of tumor suppression. This, in turn, indicates that the role of APC is to broker the catenin phosphorylation function of the Axin/GSK-3/β-catenin complex. In the absence of APC, the residual complex is inefﬁcient at processing β-catenin. A role for APC in promoting nuclear export of β-catenin has been demonstrated (Henderson, 2000; Neufeld et al., 2000). These data are a reminder of the dynamic nature of β-catenin regulation. Wnt signals are naturally

218

Armen S. Manoukian and James R. Woodgett

transient, and under normal conditions only a pulse of β-catenin stabilization will occur. That population of molecules is not immune to mass action and as the inhibition of the phosphorylation machinery is relieved, the nuclear localized molecules are shipped back to the cytoplasm to be shredded. APC presumably plays an active role in recruiting all available catenin molecules to the disposal chute. It should be stressed that although mutations of APC and β-catenin are well documented in a variety of cancers, in many cases stabilization or enhanced nuclear accumulation of β-catenin is observed in tumors without a known mutation in these genes. Clearly, other mechanisms as yet uncharacterized can lead to activation of this pathway, and the role of β-catenin in human cancer extends beyond the sphere of APC and β-catenin mutations.

IV. REGULATION OF GSK-3 BY OTHER SIGNALING PATHWAYS Genetic analysis of GSK-3 regulation has almost exclusively focused on its role in the Wnt/wg pathway where the activity of the protein kinase is suppressed by the Wnt signal. The molecular details of how GSK-3 activity is suppressed remain unclear, and it is possible that the activity of the enzyme does not change; and that rather, its accessibility to β-catenin is regulated. This scenario does not explain the ﬁnding that GSK-3 activity is catalytically reduced in cells treated with Wg or in cells in which Dishevelled has been induced (Cook et al., 1996; Ruel et al., 1999). However, until the molecular basis of the decrease in β-catenin phosphorylation has been elucidated, indirect effects on the protein kinase cannot be excluded. That is not to say that little is known about the regulation of GSK-3. Indeed, far more molecular details are understood about the control of this enzyme by other pathways outside of its role on the Wnt pathway. In all species examined, GSK-3/Zw3sgg is highly tyrosine phosphorylated at a site analogous to the phosphorylated tyrosine in mitogen-activated protein kinases (Hughes et al., 1993). Phosphorylation of this residue is important but not critical for activity. The solution of the three-dimensional structure of GSK-3β revealed that the unphosphorylated residue does not exclude access to the active site, unlike in the MAPKs where the “T-loop” must be phosphorylated for the T-loop peptide to unencumber access to the active site (Dajani et al., 2001; ter Haar et al., 2001). GSK-3α is tyrosine phosphorylated at residue 279, GSK-3β at residue 216 (Hughes et al., 1993). A variety of mitogenic stimuli (including Wnts, insulin, EGF, and PDGF) result in catalytic inactivation of GSK-3. However, tyrosine phosphorylation of the protein does not change under these conditions and appears to be constitutive

Regulation of GSK-3

219

(Shaw et al., 1997). In Dictyostelium discoideum, a protein kinase termed Zak1 has been shown to be responsible for phosphorylating GSK-A, the slime-mold ortholog of GSK-3 (Kim et al., 1999). Zak1 activity is necessary for function of GSK-A, but a mammalian (or ﬂy) ortholog of Zak1 has yet to be identiﬁed. The catalytic inactivation of GSK-3 induced by most polypeptide mitogens is reversible by treatment with serine/threonine-speciﬁc phosphatases (Cross et al., 1994). The inactivating event has been demonstrated to be due to phosphorylation of serines 21 and 9 of GSK-3α and A-3β, respectively (Sutherland et al., 1993; Stambolic and Woodgett, 1994; Shaw et al., 1997). These residues are speciﬁc targets for several protein-serine kinases, including protein kinase B/Akt, pp90rsk, and cyclic AMP-dependent protein kinase (PKA)(Cross et al., 1995; Fang et al., 2000; Li et al., 2000). The inactivating biochemical consequence of phosphorylation by all three enzymes is identical—what differs is the initiating signal (Fig. 4). Thus, activation of the phosphatidylinositol 3 kinase (PI3 K) pathway (usually via receptor tyrosine kinase activation) results in stimulation of PKB. Inactivation of GSK-3 in response to many mitogens can be inhibited by antagonists of PI3 K such as Wortmannin (Cross et al., 1994). Similarly, engagement of

Fig. 4 Regulation of GSK-3 by other pathways. GSK-3α and GSK-3β are phosphorylated on N-terminal-domain serine residues by protein kinases regulated by the phosphatidylinositol-3 kinase- and GPCR-coupled receptor pathways. The result of phosphorylation is relief of the inhibitory effect of GSK-3-mediated phosphorylation of its substrates (which therefore become activated). Note, that there are other regulatory components of each pathway—the schema is simpliﬁed for clarity.

220

Armen S. Manoukian and James R. Woodgett

G-protein-coupled receptors (GPCRs) that trigger adenylate cyclase results in generation of cyclic AMP, activation of PKA, and inhibition of GSK-3 (Fang et al., 2000). These mechanisms are all independent of Wnt-inducted regulation of GSK-3. Indeed, stimulation of GPCRs or PI3 K does not usually have any effect on β-catenin accumulation, suggesting the possible existence of sub-populations of GSK-3 within cells. Furthermore, Axin-bound GSK-3β is insensitive to insulin-induced phosphorylation of serine 9 (Ding et al., 2000), and GBP-induced inactivation of GSK-3 blocks β-catenin phosphorylation without affecting glycogen synthase phosphorylation (Thomas et al., 1999). In ﬂies, ectopic expression of activated PKB does not induce a naked cuticle phenotype as might be expected if Zw3Sgg is inhibited in the Wingless pathway (A.M. and J.R.W., unpublished). Likewise, injection of activated PKB into Xenopus embryos fails to inﬂuence axis formation. The crystal structure of GSK-3β as well as several biochemical experiments have provided insight into this differential regulation (Dajani et al., 2001; ter Haar et al., 2001; Frame et al., 2001). When phosphorylated at serine-9, the amino terminal peptide of GSK-3β folds back via an interaction with arginine-96. In the active conformation (i.e., when serine-9 is dephosphorylated), this arginine residue plays a critical role in substrate binding. Many of the substrates of GSK-3 are only targeted if they have been “primed” by a previous phosphorylation event. The priming kinase is usually distinct from GSK-3, although the enzyme “ratchets” along and thus is capable of multiply phosphorylating some targets (such as glycogen synthase) by using its own phosphorylation sites as primers for the next site (in the case of GS, the series of phosphorylations is initially primed by phosphorylation of the substrate by casein kinase-II). When arginine-96 is bound to phosphorylated serine-9, the phosphorylation-primed substrate docking site is occluded, inactivating the kinase. Importantly, several targets of GSK-3 do not require prior phosphorylation priming, and these include β-catenin. Thus, it is possible that even when GSK-3β is phosphorylated at serine-9 and catalytically inactive toward most of its substrates, when bound to Axin it is still able to phosphorylate β-catenin because this substrate does not require an interaction with arginine-96. Why might the same protein kinase be coupled to different signaling pathways? The answer probably relates to the relative rarity of a protein kinase being inactivated by a cellular stimulus. Most protein kinases are induced by a cellular stimulus, whereas GSK-3 is shut down. In addition, the enzyme has a broad variety of substrates, most of which are inactivated by phosphorylation by GSK-3. Thus, inhibition of this one case of enzyme will tend to induce the functions of a diverse array of targets. These include transcription factors and other regulatory molecules (Table I). With regard to differential regulation of GSK-3 by the Wnt pathway, it is possible that this pathway evolved to hijack a portion of an existing pathway for its own ends. One of the functions of the Axin/APC destruction

221

Regulation of GSK-3

Table I GSK-3 Substratesa Substrate Wnt pathway related adenomatous polyposis coli protein (APC) Axin β-catenin Cyclin D1 Transcription factors c-Jun JunD c-Myb CREB NF-ATc C/EBPα heat shock factor-1 c-Myc L-myc Metabolic proteins ATP-citrate lyase glycogen synthase eIF-2B translation factor G subunit of phosphatase 1 inhibitor-2 RII subunit of cAMP-dependent protein kinase Other Tau

Phosphorylation site

FXVEXTPXCFSRXSSLSSLS SANDSEQQS330 SDADTLSLT341 SLTDS343 DSGIHSGATTTAPSLSGKGNPEEED EEVDLACTPTDVR EEPQTVPEMPGETPPLSPIDMESQER SPPLSPIDMETQER APVSCLGEHHHCTPSPPVDH KRREILSRRPSYR PYASPQTSPWQSPCVSP LGSPQHSPSTSP PYSPHHSPTPSPHGSPRV TPPPTPVPSP EEPPSPPQSP DIWKKFELLPTPPLSPSRRSG DIWKKFELVPSPPTSPPWGL LLNASGSTSTPAPSRTASFSESR RPASVPPSPSLSRHSSPHQSEDEE DSEELDSRAGSPQLDDIKVF PGFSPQPSRRGSSESSEE DEPSTPYHSMIGDDDDAYSD LREARSRASTPPAAPPS

TPPKSPSAAK SPVVSGDTSPR

a Well characterized GSK-3 targets are grouped along with the sequences of the phosphorylation sites (where identiﬁed). The residues phosphorylated by GSK-3 are in bold type (if known), whereas residues that act as priming sites that are targeted by other kinases are underlined. Note that not all GSK-3 substrates require prephosphorylation by a priming kinase (e.g., β-catenin).

complex is to enlist and sequester a portion of GSK-3 molecules for the speciﬁc task of targeting β-catenin. Although this function is clearly very important (given the consequences of its deregulation), it begs the question of whether the APC/Axin complex regulates the phosphorylation of other targets of GSK-3.

V. OTHER TARGETS OF GSK-3 INVOLVED IN CANCER GSK-3 targets a diverse array of substrate proteins in addition to β-catenin, several of which have been implicated in human disease pathology (see Table I). As noted in Section III, GSK-3 phosphorylates cyclin D and targets

222

Armen S. Manoukian and James R. Woodgett

this molecule for ubiquitin-mediated degradation. GSK-3 also phosphorylates several transcription factors, including c-Jun, CREB, NF-AT, C/EBP, c-Myc, c-Myb, and HSF-1. In addition, GSK-3 regulates several metabolic/ synthetic proteins, including glycogen synthase, protein phosphatase inhibitor-2, eIF2Bε, and ATP-citrate lyase. GSK-3 targets several residues in the neuroﬁlament protein tau that are hyperphosphorylated in Alzheimer’s disease. Transgenic expression of GSK-3β in brain causes an increase in tau phosphorylation (Brownlees et al., 1997; Lucas et al., 2001). Furthermore, small-molecule inhibitors of GSK-3 can protect against neuronal apoptosis (Cross et al., 2001). Together, these data suggest that hyperactivation of GSK-3 may be proapoptotic to neuronally derived cells and may therefore contribute to neurodegenerative diseases.

A. Regulation of Nuclear Factor-κB Disruption of the mouse gene for GSK-3β revealed a surprising effect and a new target for this enzyme (Hoeﬂich et al., 2000). The homozygous knockout mice died during late embryogenesis, around embryonic day 15–16. Prior to death, the embryos showed evidence of liver malfunction characterized by selective apoptosis of the hepatocytes. Within 24 hours essentially all of the hepatocytes had undergone programmed cell death resulting in embryonic death. The liver apoptosis was rescued by maternal injection of anti-tumor necrosis factor-α antibody. The hepatocyte damage was therefore due to elevated levels of this inﬂammatory cytokine, to which the embryonic liver cells showed high sensitivity. The phenotype is remarkably similar to other knockout mice in which disruptions have been engineered in components of the NF-κB transcription factor pathway. These include the DNA-binding domain of NF-κB, c-Rel (Beg et al., 1995), and positive regulators of the pathway, including I-κB kinase-2 (IKKβ)(Li et al., 1999a). NF-κB is sequestered in the cytoplasm of resting cells because of the concealment of a nuclear localization motif by a binding partner termed inhibitor of κB (I-κB). Upon phosphorylation of two serine residues on I-κB by the IKKs, I-κB is targeted for ubiquitin-mediated degradation, much like GSK-3 phosphorylation of β-catenin. This releases NF-κB to translocate to the nucleus to modulate expression of genes harboring κB consensus sequences. Although originally described and characterized in B cells, NF-κB is a widely expressed protein that is activated by many stress-associated stimuli. In general, the transcriptional targets of the factor tend to protect cells from apoptosis (although NF-κB activation can induce death in certain cell types). The similarity of mouse knockout phenotypes was shown to reﬂect a shared defect in NF-κB regulation. Mouse embryo ﬁbroblasts devoid of GSK-3β were defective in activation of the transcription factor and

Regulation of GSK-3

223

hypersensitive to exposure to TNFα (Hoeﬂich et al., 2000). Furthermore, treatment of cells with lithium, a known inhibitor (albeit rather unspeciﬁc) of GSK-3, phenocopied the disruption of GSK-3β. However, unlike the IKK-2 and T2K knockout mice, which are defective in the induction of I-κB phosphorylation, nuclear translocation of NF-κB was found to be normal in the absence of GSK-3β, suggesting that this enzyme exerts its effect downstream of I-κB degradation. Although the requirement of GSK-3β for NF-κB function was unpredicted, its interesting to note that a recently identiﬁed small-molecule inhibitor of GSK-3, termed debromohymenialdisine, was ﬁrst characterized by its antiinﬂammatory properties. This molecule was shown to inhibit NF-κB in an I-κB phosphorylation-independent manner (DiMartino et al., 1995).

VI. GSK-3 AND CANCER In this review, we have focused on aspects of GSK-3 related to cancer. It is notable that this enzyme is negatively regulated by several pathways (Wnt, PI3’K, and cyclic AMP) that have been implicated in human cancer. Several GSK-3 substrates are oncoproteins; and the enzyme is tightly regulated. That said, mutations in GSK-3 genes have not been found in cancer. This may be due to the existence of two closely related gene products in mammals that are largely functionally redundant. Given its importance in development, perhaps mutations in the protein simply yield inviable cells. That is unlikely given the tolerance of cells to inhibitors of GSK-3, although longerterm effects cannot be excluded. A signiﬁcant fraction of the population is also undergoing lithium treatment for bipolar disorder. At therapeutic serum levels of lithium chloride, GSK-3 activity is partially inhibited, yet there are no clinical correlations between chronic lithium therapy and enhanced tumorigenesis. Perhaps tissues adapt and reset the threshold of kinase required to maintain β-catenin and cyclin D at “safe” levels. In view of its association with several diseases, therapeutic inhibition of GSK-3 is an attractive goal for many pharmaceutical companies. However, the utility of such drugs may be limited by the longer term risk of enhanced cancers, especially at acute doses. Longer term examination of mice heterozygous for each of the isoforms of GSK-3 may help in evaluating this risk.

ACKNOWLEDGMENTS Special thanks to current and former laboratory members who have contributed to the work on GSK-3, particularly Adnan Ali, Norm Anthopoulos, Rich Binari, Jing Jin, Liz Rubie, Laurent

224

Armen S. Manoukian and James R. Woodgett

Ruel, Vuk Stambolic, and Ben Wetch. J. R. W. is supported by grants from the CIHR, NCIC, and Howard Hughes Medical Institute and is a CIHR Senior Scientist. A.M. is supported by grants from the CIHR and NCIC.

REFERENCES Alt, J. R., Cleveland, J. L., Hannink, M., and Diehl, J. A. (2000). Phosphorylation-dependent regulation of cyclin D1 nuclear export and cyclin D1-dependent cellular transformation. Genes Dev. 14, 3102–3114. Beg, A. A., Sha, W. C., Bronson, R. T., Ghosh, S., and Baltimore, D. (1995). Embryonic lethality and liver degeneration in mice lacking the RelA component of NF-kappa B. Nature 376, 167– 170. Bhanot, P., Brink, M., Samos, C. H., Hsieh, J. C., Wang, Y., Macke, J. P., Andrew, D., Nathans, J., and Nusse, R. (1996). A new member of the frizzled family from Drosophila functions as a Wingless receptor. Nature 382, 225–230. Bourouis, M., Moore, P., Ruel, L., Grau, Y., Heitzler, P., and Simpson, P. (1990). An early embryonic product of the gene shaggy encodes a serine/threonine protein kinase related to the CDC28/cdc2+ subfamily. EMBO J. 9, 2877–2884. Boutros, M., Paricio, N., Strutt, D. I., and Mlodzik, M. (1998). Dishevelled activates JNK and discriminates between JNK pathways in planar polarity and wingless signaling. Cell 94, 109–118. Brownlees, J., Irving, N. G., Brion, J. P., Gibb, B. J., Wagner, U., Woodgett, J. R., and Miller, C. C. (1997). Tau phosphorylation in transgenic mice expressing glycogen synthase kinase-3β transgenes. Neuroreport 8, 3251–3255. Cadigan, K. M., and Nusse, R. (1997). Wnt signaling: a common theme in animal development. Genes Dev. 11, 3286–3305. Cadoret, A., Ovejero, C., Saadi-Kheddouci, S., Souil, E., Fabre, M., Romagnolo, B., Kahn, A., and Perret, C. (2001). Hepatomegaly in transgenic mice expressing an oncogenic form of beta-catenin. Cancer Res. 61, 3245–3249. Chan, E. F., Gat, U., McNiff, J. M., and Fuchs, E. (1999). A common human skin tumour is caused by activating mutations in beta-catenin. Nat Genet. 21, 410–413. Cook, D., Fry, M. J., Hughes, K., Sumathipala, R., Woodgett, J. R., and Dale, T. C. (1996). Wingless inactivates glycogen synthase kinase-3 via an intracellular signalling pathway which involves a protein kinase C. EMBO J. 15, 4526–4536. Couso, J. P., Bate, M., and Martinez-Arias, A. (1993). A wingless-dependent polar coordinate system in Drosophila imaginal discs. Science 259, 484–489. Couso, J. P., Bishop, S. A., and Martinez Arias, A. (1994). The wingless signalling pathway and the patterning of the wing margin in Drosophila. Development 120, 621–636. Cross, D. A., Alessi, D. R., Cohen, P., Andjelkovich, M., and Hemmings, B. A. (1995). Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 378, 785–789. Cross, D. A., Alessi, D. R., Vandenheede, J. R., McDowell, H. E., Hundal, H. S., and Cohen, P. (1994). The inhibition of glycogen synthase kinase-3 by insulin or insulin-like growth factor 1 in the rat skeletal muscle cell line L6 is blocked by wortmannin, but not by rapamycin: evidence that wortmannin blocks activation of the mitogen-activated protein kinase pathway in L6 cells between Ras and Raf. Biochem J. 303, 21–26. Cross, D. A., Culbert, A. A., Chalmers, K. A., Facci, L., Skaper, S. D., and Reith, A. D. (2001). Selective small-molecule inhibitors of glycogen synthase kinase-3 activity protect primary neurones from death. J. Neurochem. 77, 94–102.

Regulation of GSK-3

225

Cumberledge, S., and Krasnow, M. A. (1993). Intercellular signalling in Drosophila segment formation reconstructed in vitro. Nature 363, 549–552. Dajani, R., Fraser, E., Roe, S. M., Young, N., Good, V., Dale, T. C., and Pearl, L. H. (2001). Crystal structure of glycogen synthase kinase 3 beta: structural basis for phosphate-primed substrate speciﬁcity and autoinhibition. Cell 105, 721–732. Diehl, J. A., Cheng, M., Roussel, M. F., and Sherr, C. J. (1998). Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization. Genes Dev. 12, 3499–3511. DiMartino, M., Wolff, C., Patil, A., and Nambi, P. (1995). Effects of a protein kinase C inhibitor (PKCI) on the development of adjuvant-induced arthritis. Inﬂamm Res. 44 Suppl 2, S123– 124. DiNardo, S., Sher, E., Heemskerk-Jongens, J., Kassis, J. A., and O’Farrell, P. H. (1988). Twotiered regulation of spatially patterned engrailed gene expression during Drosophila embryogenesis. Nature 332, 604–609. Ding, V. W., Chen, R. H., and McCormick, F. (2000). Differential regulation of glycogen synthase kinase 3beta by insulin and Wnt signaling. J. Biol. Chem. 275, 32475–32481. Embi, N., Rylatt, D. B., and Cohen, P. (1980). Glycogen synthase kinase-3 from rabbit skeletal muscle. Separation from cyclic-AMP-dependent protein kinase and phosphorylase kinase. Eur. J. Biochem. 107, 519–527. Fang, X., Lu, Y., Yu, S. X., Lu, Y., Bast, R. C., Woodgett, J. R., and Mills, G. B. (2000). Phosphorylation and inactivation of glycogen synthase kinase-3 by both protein kinase A and protein kinase B. Proc. Natl. Acad. Sci. USA 97, 11960–11965. Frame, S., Cohen, P., and Biondi, R. M. (2001). A common phosphate binding site explains the unique substrate speciﬁcity of GSK3 and its inactivation by phosphorylation. Mol. Cell 7, 1321–1327. Gat, U., DasGupta, R., Degenstein, L., and Fuchs, E. (1998). De novo hair follicle morphogenesis and hair tumors in mice expressing a truncated beta-catenin in skin. Cell 95, 605– 614. Gonzalez, F., Swales, L., Bejsovec, A., Skaer, H., and Martinez Arias, A. (1991). Secretion and movement of wingless protein in the epidermis of the Drosophila embryo. Mech. Dev. 35, 43–54. Hansen, L., Viars, C. S., Rasmussen, S. B., Arden, K. C., Vestergaard, H., Hansen, T., Møller, A. M., Woodgett, J. R., and Pedersen, O. (1997). Chromosomal mapping and mutational analysis of the coding region of the glycogen synthase kinase-3α and β isoforms in patients with non-insulin dependent diabetes mellitus. Diabetologia 40, 940–946. Harada, N., Tamai, Y., Ishikawa, T., Sauer, B., Takaku, K., Oshima, M., and Taketo, M. M. (1999). Intestinal polyposis in mice with a dominant stable mutation of the beta-catenin gene. EMBO J. 18, 5931–5942. He, T. C., Sparks, A. B., Rago, C., Hermeking, H., Zawel, L., da Costa, L. T., Morin, P. J., Vogelstein, B., and Kinzler, K. W. (1998). Identiﬁcation of c-MYC as a target of the APC pathway. Science 281, 1509–1512. He, X., Saint-Jeannet, J.-P., Woodgett, J. R., Varmus, H. E., and Dawid, I. B. (1995). Glycogen synthase kinase 3 and dorsoventral patterning in Xenopus embryos. Nature 374, 617–622. Hecht, A., and Kemler, R. (2000). Curbing the nuclear activities of beta-catenin. Control over Wnt target gene expression. EMBO Rep. 1, 24–28. Hemmings, B. A., Yellowlees, D., Kernohan, J. C., and Cohen, P. (1981). Puriﬁcation of glycogen synthase kinase 3 from rabbit skeletal muscle. Copuriﬁcation with the activating factor (FA) of the (Mg-ATP)-dependent protein phosphatase. Eur. J. Biochem. 119, 443–451. Henderson, B. R. (2000). Nuclear-cytoplasmic shuttling of APC regulates beta-catenin subcellular localization and turnover. Nat. Cell Biol. 2, 653–660. Heyer, J., Yang, K., Lipkin, M., Edelmann, W., and Kucherlapati, R. (1999). Mouse models for colorectal cancer. Oncogene 18, 5325–5333.

226

Armen S. Manoukian and James R. Woodgett

Hoeﬂich, K. P., Luo, J., Rubie, E. A., Tsao, M.-S., Jin, O., and Woodgett, J. R. (2000). Requirement for glycogen synthase kinase-3β in cell survival and NF-κB activation. Nature 406, 86–90. Hughes, K., Nikolakaki, E., Plyte, S E., Totty, N. F., and Woodgett, J. R. (1993). Modulation of the glycogen synthase kinase-3 family by tyrosine phosphorylation. EMBO J. 12, 803– 808. Hughes, K, Pulverer, B., Theocharous, P., and Woodgett, J. R. (1992). Baculovirus-mediated expression and characterisation of glycogen synthase kinase-3β, the mammalian homologue of the Drosophila melanogaster zeste-white3sgg homeotic gene product. Eur. J. Biochem. 203, 305–311. Imbert, A., Eelkema, R., Jordan, S., Feiner, H., and Cowin, P. (2001). Delta N89 beta-catenin induces precocious development, differentiation, and neoplasia in mammary gland. J. Cell Biol. 153, 555–568. Jiang, J., and Struhl, G. (1998). Regulation of the Hedgehog and Wingless signalling pathways by the F-box/WD40-repeat protein Slimb. Nature 391, 493–496. Jonkers, J., Korswagen, H. C., Acton, D., Breuer, M., and Berns, A. (1997). Activation of a novel proto-oncogene, Frat1, contributes to progression of mouse T-cell lymphomas. EMBO J. 16, 441–450. Jue, S. F., Bradley, R. S., Rudnicki, J. A., Varmus, H. E., and Brown, A. M. (1992). The mouse Wnt-1 gene can act via a paracrine mechanism in transformation of mammary epithelial cells. Mol. Cell. Biol. 12, 321–328. Kadowaki, T., Wilder, E., Klingensmith, J., Zachary, K., and Perrimon, N. (1996). The segment polarity gene porcupine encodes a putative multitransmembrane protein involved in Wingless processing. Genes Dev. 10, 3116–3128. Kim, L., Liu, J., and Kimmel, A. R. (1999). The novel tyrosine kinase ZAK1 activates GSK3 to direct cell fate speciﬁcation. Cell 99, 399–408. Kishida, S., Yamamoto, H., Hino, S., Ikeda, S., Kishida, M., and Kikuchi, A. (1999). DIX domains of Dvl and axin are necessary for protein interactions and their ability to regulate beta-catenin stability. Mol. Cell. Biol. 19, 4414–4422. Klingensmith, J., and Nusse, R. (1994). Signaling by wingless in Drosophila. Dev. Biol. 166, 396–414. Klingensmith, J., Nusse, R., and Perrimon, N. (1994). The Drosophila segment polarity gene dishevelled encodes a novel protein required for response to the wingless signal. Genes Dev. 8, 118–130. Kwan, H., Pecenka, V., Tsukamoto, A., Parslow, T. G., Guzman, R., Lin, T. P., Muller, W. J., Lee, F. S., Leder, P., and Varmus, H. E. (1992). Transgenes expressing the Wnt-1 and int-2 proto-oncogenes cooperate during mammary carcinogenesis in doubly transgenic mice. Mol. Cell. Biol. 12, 147–154. Li, M., Wang, X., Meintzer, M. K., Laessig, T., Birnbaum, M. J., and Heidenreich, K. A. (2000). Cyclic AMP promotes neuronal survival by phosphorylation of glycogen synthase kinase 3beta. Mol. Cell Biol. 20, 9356–9363. Li, Q., Van Antwerp, D., Mercurio, F., Lee, K. F., and Verma, I. M. (1999a). Severe liver degeneration in mice lacking the IkappaB kinase 2 gene. Science 284, 321–325. Li, L., Yuan, H., Weaver, C. D., Mao, J., Farr, G. H., 3rd, Sussman, D. J., Jonkers, J., Kimelman, D., and Wu, D. (1999b). Axin and Frat1 interact with dvl and GSK, bridging Dvl to GSK in Wnt-mediated regulation of LEF-1. EMBO J. 18, 4233–4240. Lucas, J. J., Hernandez, F., Gomez-Ramos, P., Moran, M. A., Hen, R., and Avila, J. (2001). Decreased nuclear beta-catenin, tau hyperphosphorylation and neurodegeneration in GSK-3beta conditional transgenic mice. EMBO J. 20, 27–39. McKay, R. M., Peters, J. M., and Graff, J. M. (2001). The casein kinase i family in wnt signaling. Dev Biol. 235, 388–396.

Regulation of GSK-3

227

Morata, G., and Lawrence, P. A. (1977). The development of wingless, a homeotic mutation of Drosophila. Dev. Biol. 56, 227–240. Neufeld, K. L., Zhang, F., Cullen, B. R., and White, R. L. (2000). APC-mediated downregulation of beta-catenin activity involves nuclear sequestration and nuclear export. EMBO Rep. 1, 519–523. Neumann, C. J., and Cohen, S. M. (1996). Distinct mitogenic and cell fate speciﬁcation functions of wingless in different regions of the wing. Development 122, 1781–1789. Noordermeer, J., Johnston, P., Rijsewijk, F., Nusse, R., and Lawrence, P. A. (1992). The consequences of ubiquitous expression of the wingless gene in the Drosophila embryo. Development 116, 711–719. Noordermeer, J., Klingensmith, J., Perrimon, N., and Nusse, R. (1994). dishevelled and armadillo act in the wingless signalling pathway in Drosophila. Nature 367, 80–83. Nusse, R., and Varmus, H. E. (1982). Many tumors induced by the mouse mammary tumor virus contain a provirus integrated in the same region of the host genome. Cell 31, 99–109. Nusse, R., and Varmus, H. E. (1992). Wnt genes. Cell 69, 1073–1087. Nusse, R., Brown, A., Papkoff, J., Scambler, P., Shackleford, G., McMahon, A., Moon, R., and Varmus, H. (1991). A new nomenclature for int-1 and related genes: the Wnt gene family. Cell 64, 231. Perrimon, N., Engstrom, L., and Mahowald, A. P. (1989). Zygotic lethals with speciﬁc maternal effect phenotypes in Drosophila melanogaster. I. Loci on the X chromosome. Genetics 121, 333–352. Peters, J. M., McKay, R. M., McKay, J. P., and Graff, J. M. (1999). Casein kinase I transduces Wnt signals. Nature 401, 345–350. Polakis, P. Wnt signaling and cancer (2000). Genes Dev. 14, 1837–1851. Ramakrishna, N. R., and Brown, A. M. (1993). Wingless, the Drosophila homolog of the protooncogene Wnt-1, can transform mouse mammary epithelial cells. Dev. Suppl., 95–103. Ruel, L., Bourouis, M., Heitzler, P., Pantesco, V., and Simpson, P. (1993a). Drosophila shaggy kinase and rat glycogen synthase kinase-3 have conserved activities and act downstream of Notch. Nature 362, 557–560. Ruel, L., Pantesco, V., Lutz, Y., Simpson, P., and Bourouis, M. (1993b). Functional signiﬁcance of a family of protein kinases encoded at the shaggy locus in Drosophila. EMBO J. 12, 1657–1669. Ruel, L., Stambolic, V., Ali, A., Manoukian, A., and Woodgett, J. R. (1999). Regulation of the protein kinase activity of Shaggy/Zeste-white3 by components of the Wingless in Drosophila cells and embryos. J. Biol. Chem. 274, 21790–21796. Sakanaka, C., Leong, P., Harrison, S. D., and Williams, L. T. (1999). Casein kinase 1 epison in the Wnt pathway: regulation of beta-catenin function. Proc. Natl. Acad. Sci. USA 96, 12548–12552. Sakanaka, C., Sun, T. Q., and Williams, L. T. (2000). New steps in the Wntl beta-catenin signal transduction pathway. Recent Prog. Horm. Res. 55, 225. Shackleford, G. M., MacArthur, C. A., Kwan, H. C., and Varmus, H. E. (1993). Mouse mammary tumor virus infection accelerates mammary carcinogenesis in Wnt-1 transgenic mice by insertional activation of int-2/Fgf-3 and hst/Fgf-4. Proc. Natl. Acad. Sci. USA 90, 740–744. Shaw, M., Cohen, P., and Alessi, D. R. (1997). Further evidence that the inhibition of glycogen synthase kinase-3beta by IGF-1 is mediated by PDK1/PKB-induced phosphorylation of Ser-9 and not by dephosphorylation of Tyr-216. FEBS Lett. 416, 307–311. Siegfried, E., Chou, T. B., and Perrimon, N. (1992). Wingless signaling acts through zestewhite 3, the Drosophila homolog of glycogen synthase kinase-3, to regulate engrailed and establish cell fate. Cell 71, 1167–1179. Siegfried, E., Perkins, L. A., Capaci, T. M., and Perrimon, N. (1990). Putative protein kinase product of the Drosophila segment-polarity gene zeste-white3. Nature 345, 825–829.

228

Armen S. Manoukian and James R. Woodgett

Siegfried, E., Wilder, E. L., and Perrimon, N. (1994). Components of wingless signalling in Drosophila. Nature 367, 76–80. Smalley, M. J., and Dale, T. C. (1999). Wnt signalling in mammalian development and cancer. Cancer Metastasis Rev. 18, 215–230. Stambolic, V., and Woodgett, J. R. (1994). Negative regulation of glycogen synthase kinase 3 in intact cells. Biochem. J. 303, 701–704. Sutherland, C., Leighton, I. A., and Cohen, P. (1993). Inactivation of glycogen synthase kinase-3 beta by phosphorylation: new kinase connections in insulin and growth-factor signalling. Biochem J. 296, 15–19. ter Haar, E., Coll, J. T., Austen, D. A., Hsiao, H.-M., Swenson, L., and Jain, J. (2001). Structure of GSK3beta reveals a primed phosphorylation mechanism. Nat. Struct. Biol. 8, 593– 596. Tetsu, O., and McCormick, F. (1999). Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature 398, 422–426. Thomas, G. M., Frame, S., Goedert, M., Nathke, I., Polakis, P., and Cohen, P. (1999). A GSK3binding peptide from FRAT1 selectively inhibits the GSK3-catalysed phosphorylation of axin and beta-catenin. FEBS Lett. 458, 247–251. Tsukamoto, A. S., Grosschedl, R., Guzman, R. C., Parslow, T., and Varmus, H. E. (1988). Expression of the int-1 gene in transgenic mice is associated with mammary gland hyperplasia and adenocarcinomas in male and female mice. Cell 55, 619–625. Tung, H. Y. L., and Reed, L. J. (1989). Puriﬁcation and characterization of protein phosphatase 1I activating kinase from bovine brain cytosolic and particulate fractions. J. Biol. Chem. 264, 2985–2990. van den Heuvel, M., Harryman-Samos, C., Klingensmith, J., Perrimon, N., and Nusse, R. (1993). Mutations in the segment polarity genes wingless and porcupine impair secretion of the wingless protein. EMBO J. 12, 5293–5302. Vandenheede, J. R., Yang, S. D., Goris, J., and Merlevede, W. (1980). ATP/Mg-dependent protein phosphatase from rabbit skeletal muscle. II. Puriﬁcation of the activating factor and its characterization as a bifunctional protein also displaying synthase kinase activity. J. Biol. Chem. 255, 11768–11774. Wieschaus, E., and Riggleman, R. (1987). Autonomous requirements for the segment polarity gene armadillo during Drosophila embryogenesis. Cell 49, 177–184. Willert, K., Brink, M., Wodarz, A., Varmus, H., and Nusse, R. (1997). Casein kinase 2 associates with and phosphorylates dishevelled. EMBO J. 16, 3089–3096. Willert, K., Logan, C. Y., Arora, A., Fish, M., and Nusse, R. (1999). A Drosophila Axin homolog, Daxin, inhibits Wnt signaling. Development 126, 4165–4173. Wodarz, A., and Nusse, R. (1998). Mechanisms of Wnt signaling in development. Annu. Rev. Cell Dev. Biol. 14, 59–88. Woodgett, J. R. (1990). Molecular cloning and expression of glycogen synthase kinase-3/Factor A. EMBO J. 9, 2431–2438. Xu, L., Corcoran, R. B., Welsh, J. W., Pennica, D., and Levine, A. J. (2000). WISP-1 is a Wnt-1-and beta-catenin-responsive oncogene. Genes Dev. 14, 585–595. Yanagawa, S., van-Leeuwen, F., Wodarz, A., Klingensmith, J., and Nusse, R. (1995). The dishevelled protein is modiﬁed by wingless signaling in Drosophila. Genes Dev. 9, 1087– 1097. Yaron, A., Hatzubai, A., Davis, M., Lavon, I., Amit, S., Manning, A. M., Andersen, J. S., Mann, M., Mercurio, F., and Ben-Neriah, Y. (1998). Identiﬁcation of the receptor component of the IkB-alpha-ubiquitin ligase. Nature 396, 590–594. Yost, C., Farr, G. H., 3rd, Pierce, S. B., Ferkey, D. M., Chen, M. M., and Kimelman, D. (1998). GBP, an inhibitor of GSK-3, is implicated in Xenopus development and oncogenesis. Cell 93, 1031–1041.

Regulation of GSK-3

229

Zecca, M., Basler, K., and Struhl, G. (1996). Direct and long-range action of a wingless morphogen gradient. Cell 87, 833–844. Ziemer, L. T., Pennica, D., and Levine, A. J. (2001). Identiﬁcation of a mouse homolog of the human BTEB2 transcription factor as a beta-catenin-independent Wnt-1-responsive gene. Mol. Cell. Biol. 21, 562–574. Zeng, L., Fagotto, F., Zhang, T., Hsu, W., Vasicek, T. J., Perry, W. L., 3rd, Lee, J. J., Tilghman, S. M., Gumbiner, B. M., and Costantini, F. (1997). The mouse Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway that regulates embryonic axis formation. Cell 90, 181–192.

This Page Intentionally Left Blank

Chronic Immune Activation and Inflammation in the Pathogenesis of AIDS and Cancer Angus G. Dalgleish, and Ken. J. O’Byrne Department of Oncology, St. George’s Hospital Medical School, London, SW17 0RE, United Kingdom and Department of Oncology, Leicester Royal Inﬁrmary, Leicester LE1 5WW, United Kingdom

I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV.

The Pathogenesis of HIV-1 Infection SIV and Simian AIDS Chronic Immune Activation Immunological Imbalance in HIV Infection Cancer and the Immune Response Kaposi’s Sarcoma (KS) and HIV Infection as a Paradigm for Inﬂammatory Cancer Development The Interrelationship between the Immune Response and Angiogenesis Inﬂammation and Apoptosis The Biology of Oncogenic Viruses: Impact on the Immune Response, Angiogenesis, and Apoptosis Features of Tumors Lacking an Obvious Premalignant Inﬂammatory Process Environment Created by Noninﬂammatory Factors That Predispose to Malignant Disease Fractal Mathematics, Carcinogenesis, and the Progression of Malignant Disease Implications for Chemoprevention and Future Treatment of Malignant Disease Conclusions References

Infection with the human immunodeﬁciency virus (HIV) invariably leads to the development of acquired immunodeﬁciency syndrome (AIDS) in most infected humans, yet does so rarely, if at all, in HIV-infected chimpanzees. The differences between the two species are not due to differences in cellular receptors or an inability of the chimpanzee to be infected, but rather to the lack of pan-immune activation in the infected primate. This results in reduced apoptotic death in CD4+ T-helper lymphocytes and a lower viral load. In humans the degree of chronic immune activation correlates with virus load and clinical outcome with high immune activation leading to high viral loads and the more rapid progression to AIDS and death. The type of immune perturbation seen in HIVassociated AIDS is similar to that of chronic graft-versus-host disease (GVHD) where reduced cell-mediated immune (CMI) responses occur early in the course of the disease and where humoral responses (HI) predominate. A reduced CMI response occurs in a

Advances in CANCER RESEARCH 0065-230X/02 $35.00

231

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

232

A. G. Dalgleish and K. J. O’Byrne

number of chronic infectious diseases, including tuberculosis and leishmaniasis. More recently, it has become increasingly apparent that the CMI response is suppressed in virtually all malignant diseases, including melanoma and colorectal and prostate cancer. This raises the possibility that, as the malignant process develops, the cancer cells evolve to subvert the CMI reponse. Moreover, the reduced CMI response seen in colorectal cancer (CRC) patients is completely reversed following curative surgery strongly supporting the hypothesis that CRC can suppress the systemic immune response. Wound healing, ovulation, embyro implantation, and fetal growth are all associated with suppressed CMI and neovascularization (the formation of new blood vessels) or angiogenesis (the formation of new blood vessels from an existing vasculature). If unresolved, wound healing results in chronic inﬂammation, which can give rise to the phenomenon of “scar cancers.” Indeed all the chronic inﬂammatory conditions known to be associated with the subsequent development of malignant disease, including chronic obstructive airway disease (COPD), ulcerative colitis (UC), and asbestosis, give rise to similar proangiogenic, suppressed CMI, and HI-predominant environments. In keeping with this CMI-associated cytokines such as interleukin (IL)-2 and interferon (IFN)-γ tend to be antiangiogenic, whereas HI cytokines such as IL-6 tend to be proangiogenic. Furthermore, chronic immune activation leads to the synthesis and release of factors such as macrophage inﬂammatory protein (MIP)-1 that inhibit apoptosis through suppression of p53 activity. The “Golden Triangle” of suppressed CMI, angiogenesis, and reduced apoptosis would provide the ideal environment for the serial mutations to occur that are required for the development of malignant disease. If the observed association is relevant to carcinogenesis, then treatments aimed at reducing the components of these inﬂammatory conditions may be useful both in the setting of chemoprevention and the therapeutic management of established disease. C 2002 Academic Press.

I. THE PATHOGENESIS OF HIV-1 INFECTION HIV-1 uses the CD4 molecule as its main receptor (Dalgleish et al., 1984; Klatzmann et al., 1984) and the chemokine receptors as secondary receptors, the latter of which determine tropism for different cell types (Berger et al., 1999; Norcross, 1999). The ready ability of HIV-1 to infect and kill CD4 cells in vitro would appear to correlate well with the gradual decline in CD4 cells following HIV-1 infection in vivo, a feature that is also associated with the development of disease in the form of opportunistic infections and malignancies. However, in spite of having identical CD4 and chemokine receptors, progression to disease does not occur in a small minority of HIVinfected humans and the majority of chimpanzees (Candotti et al., 1999; Heeney et al., 1996). Activation at the cellular level has long been known to be important for virus entry and replication (Copeland and Heeney, 1996). However, for disease to develop it would appear that pan-immune activation (i.e., involving all aspects of the immune system including B cells and CD4 and CD8 cells) is also required (Liu et al., 1997; Plaeger et al., 1999). An activated immune system will provide more available activated cells for virus propagation, and it is therefore not surprising to ﬁnd that high activation

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

233

markers are associated with high virus loads, which in turn are associated with rapid progression to disease (Graham, 1996; Schooley, 1995; Ferre et al., 1995). It is important to appreciate that there is a difference between transient immune activation, such as occurs following an immune challenge, and that associated with chronic activation, which appears to require no obvious ongoing extraneous challenge. It is possible to artiﬁcially activate the immune system of an HIV-infected chimpanzee and elevate the virus load. However, upon cessation of the stimulus, the viral load will fall as the immune response diminishes (Fultz et al., 1992). It would therefore appear that the chimpanzee mounts a speciﬁc response, which appropriately ceases when the stimulus stops, and that it does not mount a nonspeciﬁc response to HIV. Nevertheless, relatively high viral loads can be found in the lymph nodes of HIV-1-infected chimpanzees, which does not appear to cause any pathology, in contrast to the situation in HIV-1-positive human lymph nodes (Saksela et al., 1993). It is therefore reasonable to postulate that this ability is genetically controlled. Further support for this comes from the simian immunodeﬁciency virus (SIV) model and its differential clinical outcome in monkey hosts.

II. SIV AND SIMIAN AIDS SIV has many of the properties of HIV, being cytopathic for CD4 cells and causing fatal “AIDS” in infected macaques (Holterman et al., 2000). However, like HIV, it is able to infect some monkeys such as the African green monkey without causing disease (Goldstein et al., 2000). Furthermore, there is a clear link with disease and with the degree of immune activation as well as with the viral load. The importance of immune activation relative to viral load and replication per se is shown by the SIV-infected Sooty Mangabey monkey, which develops marked immune activation, yet maintains a low viral load, and still progresses to “AIDS” (Kaur et al., 1998). Hence, in this model it is the immune activation status and not the viral load that clearly correlates with disease progression.

III. CHRONIC IMMUNE ACTIVATION Pan-activation of the immune system is usually associated with an antigenic hypervariable infectious agent, superantigen stimulation (Held et al., 1994), or allogeneic stimulation (Margolis and Vogelsang, 2000). The most obvious explanation for the pan-immune activation seen in HIV infection is the hypervariability, particularly in the highly antigenic V3 loop

234

A. G. Dalgleish and K. J. O’Byrne

(Lenz et al., 2001), which has so far presented an elusive obstacle to the development of an effective vaccine candidate. However, virus variability occurs in chimpanzees and human long-term nonprogressors (LTNPs), as well as those developing AIDS, and hence this cannot explain the different clinical outcomes. Indeed there would appear to be more variability in LTNPs than in those progressing to AIDS (Shankarappa et al., 1998). Because of the presence of superantigens in other retroviruses such as the mouse mammary tumor virus (MMTV), there was considerable speculation that HIV contained a superantigen (Imberti et al., 1991). Initial reports appeared to conﬁrm this following the discovery of Vβ enhancement and deletion in the variable T-cell receptor (TCR) repertoire. A number of controversial ﬁndings, both for and against, have been published. However, the variability of the Vβ TCR repertoire from person to person is not consistent with a superantigen (Posnett et al., 1993). Indeed, the variability correlates with falling CD4 counts and would hence appear to be an artefact of CD4 loss (Westby et al., 1998b) in that whatever is causing the loss of CD4 cells is working in a random fashion; hence the observed change and variability in the T-cell repertoire as AIDS progresses. Nevertheless there is evidence of a superantigen-like epitope on GP120 that could induce a polyclonal B-cell expression but not CD4 and CD8 activation and therefore cannot be the explanation for the observed pan-immune activation (Goodglick et al., 1995). The other possible explanation is that a chronic allogeneic reaction occurs in HIV-infected people. This is intriguing, as before HIV was identiﬁed, Shearer pointed out the similarities between chronic (not acute!) graft-versushost disease and AIDS and went as far to suggest that allogeneic cells in semen may contribute to AIDS, although transmission via factor 8 excluded this possibility (Shearer, 1983). Both conditions have evidence of immune panactivation, CMI suppression, opportunistic infections, and malignancies, especially Epstein–Barr virus (EBV)-driven lymphomas. Superﬁcially there is no obvious explanation for this similarity of disease, which in the case of chronic GVHD is caused by the graft activating the host’s lymphocytes in an allogeneic-dependent manner (Table I) (Habeshaw et al., 1992). However, there are a number of characteristics of the HIV virus that could explain this similarity. First, HIV gp120 has a number of conserved regions with sequence homology to both Human Lymphocyte Antigen (HLA) class I (A2) and class II (DR1) (Hounsell et al., 1991). Molecular modeling has shown that this could translate into structural similarity, and the ability of gp120 to cross-link peptides that are predicted to bind from three-dimensional modeling in a manner similar to how they bind HLA is compelling (Sheikh et al., 1995). More recently we have shown that gp120 + peptide can stimulate HLA (gp120 homology predicted, e.g., HLA2 + peptide) restricted peptide-speciﬁc T-cell lines (Sheikh et al., 2000). Crystallography of gp120 has proven difﬁcult because of the

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

235

Table I Features Common to HIV Infection Leading to AIDS and GVHD (MHC Class II Mismatched) T cells Increased proliferation Impaired response to mitogens Impaired CD4 function Involving CD8, CTL help B-cell help CTL function correctable with IL-2 Suppressor cells selective for TH Opportunistic infections B cells Polyclonal activation Hypergammaglobulinaemia Autoantibody production Immune complexes (circulating) Impaired response to mitogens Impaired response to T-cell help Lymphomas Non T Non B Increased proliferation Impaired function and responses Increased activation markers

high level of glycosylation, and the only model so far involves co-crystallization with CD4 and a monoclonal antibody (Kwong et al., 1998), which unfortunately results in deletion of the carboxy terminus thought to confer the peptide-binding property of gp120. Consistent with this prediction is the fact that the carboxy-terminus-truncated gp120 used by Sodroski and his team fails to bind a peptide that readily binds the complete gp120 (Fig. 1; see color insert). This structural homolgy between gp120 and HLA may be important at more than one level. First, when gp120 buds from a cell it carries several membrane molecules that are readily ejected (or edited out), with the exception of HLA molecules that form a “copse” around gp120. Indeed, this is so marked in the simian model that anti-HLA antibodies can protect against a challenge virus growing in cells expressing the same HLA as the antibody (Chan et al., 1992; Arthur et al., 1995). HIV gp120 budding in the absence of HLA is unable to infect cells, and this may be due to the ability of a gp120/HLA complex to activate a host cell rather than just bind to CD4 (Cosma et al., 1999). Second, the ability to bind peptides might perturb immune responses, especially when gp120 buds via an antigen-presenting cell (APC) and inappropriately activates T cells. Third, and perhaps very relevant to the ability to induce chronic infection and immune activation, gp120 can furnish peptides for presentation by “self” HLA (Sheikh et al., 2000). Whereas acute GVHD is due to foreign HLA presenting “self” HLA

236

A. G. Dalgleish and K. J. O’Byrne

peptides, chronic GVHD is associated with “self” HLA presenting peptides derived from the HLA of foreign (allo) cells. This could be a potential mechanism to explain the GVHD-like nature of the disease in that the gp120encoded HLA-like sequences induce a response when presented by self HLA similar to that induced by allo-derived peptides. This could therefore explain the HLA restriction that enables some humans and most chimpanzees to coexist with HIV infection without developing disease (Leelayuwat et al., 1993). Based on what sequences are likely to be seen as self and foreign by different HLA types, Habershaw predicted that the carboxy terminus (C5) of gp120 would be a self peptide for HLA B27 and a “foreign” allo peptide in HLA B8 (Habeshaw, 1994). It is of great interest that the only HLA types that were statistically signiﬁcant relative to fast progressors and slow or nonprogressors in an MRC survey of U.K. infected populations were HLA B8 and HLA B27, respectively (McNeil et al., 1996). Nevertheless, it is likely that the activation scenario is more complex in that other HLA classes (e.g., class III) and immune response genes (e.g., TNF allele polymorphisms) may be involved. In addition, the possibility that other HIV genes such as nef and gag could furnish HLAlike sequences such as those described by Young (1988) and contribute to the overall activation cannot be ruled out. Nevertheless, whatever drives the immune activation, it is clearly the cause of disease, and hence reduction of immune activation with steroids should confer beneﬁt and not, as initially supposed, further immunesuppression. One of the ﬁrst AZT-intolerant patients to be given high-dose steroids for his disease, as opposed to opportunistic disease-associated allergic reactions, had a marked clinical improvement with hematological indices allowing the required dose to be titrated to a minimum dose of 40 mg a day (Aitken et al., 1992). Other studies have documented that prednisone can reduce overall virus load (Kilby et al., 1997), underlining the need for immune activation for active viral replication. This association between immune activation and disease has other implications with regard to immunotherapy and vaccine development, as in theory it could be argued that the body needs to be made tolerant of the activating alloepitope component of the virus and not necessarily alerted to it, as it were.

IV. IMMUNOLOGICAL IMBALANCE IN HIV INFECTION The discovery that T cells produce (at least) two major cytokine patterns known as Th-1 cells (which produce IL-2, gamma IFN, and IL-12) and Th-2 cells (which produce IL-4, IL-5, IL-6, and IL-10) has had a major impact on the understanding of the complex changes and interactions between an

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

237

infectious agent and the immune system (Mosmann and Coffman, 1989). Th-1 cytokines are required for a strong CM1 response and are decreased in many infectious diseases, including HIV, TB, and many tropical infections. In what might appear a compensatory response, Th-2 cytokines often increase, as is the case in mid- to late-stage HIV infection. However, there is controversy about the role of IL-4, as it, too, decreases in late-stage HIV infection. Essentially, the decline in IL-2 and gamma IFN production by CD4 cells is associated with the decline in antigen-speciﬁc immune responses resulting in opportunistic infections. This occurs against a background of increasing CD8 activation as measured by the CD38 and CD57 activation markers, which are expressed on nearly every CD8 cell when CD4 cells are barely detectable in late-stage disease (Clerici and Shearer, 1994; Shearer and Clerici, 1992).

V. CANCER AND THE IMMUNE RESPONSE It would appear that most malignancies are associated with suppression of CMI responses (Lee et al., 1997; Maraveyas et al., 1999; Pettit et al., 2000). The number of documented strategies employed by cancers to evade the immune response continues to expand. These include the ability to downregulate HLA and costimulatory molecules, the production of immunosuppressive factors, and the up-regulation of immune cell apoptosis by tumors expressing molecules such as Fas L (Doherty et al., 1994; Ganss and Hanahan, 1998; Garrido et al., 1993; Gorter and Meri, 1999; Melief and Kast, 1991; Strand and Galle, 1998; Pettit et al., 2000). The decline in CMI responses seen in the chronic immune activation of HIV-induced AIDS is also seen in malignant disease. The suppression of local CMI responses has been reported in a number of studies evaluating inﬂammatory cellular inﬁltrates in tumors from patients with malignant tumors, including non-small cell lung cancer (NSCLC) and head and neck, breast, and genitourinary cancers, lymphomas, and sarcomas (Asselin-Paturel et al., 1998; Hildesheim et al., 1997; Lee et al., 1997; Aziz et al., 1998; O’Hara et al., 1998) and carcinoma in situ including Barrett’s esophagus and cervical intraepithelial neoplasia (CIN) (Hildesheim et al., 1997; Oka et al., 1992; Sonnex, 1998). Systemic immunesuppression can be documented by using intracellular cytokine production following stimulation of lymphocytes in vitro and has been documented in several cancer types including melanoma and colorectal cancer (Maraveyas et al., 1999; Heriot et al., 2000). The presence of a dominant Th2 immune response in potentially curable tumors such as lymphomas is associated with a fatal outcome (Lee et al., 1997). Absent or reduced delayed hypersensitivity reactions to common T-cell recall antigens are a manifestation of CMI. These responses are either reduced or absent

238

A. G. Dalgleish and K. J. O’Byrne

in many premalignant and malignant tumors including CIN, Hodgkin’s disease, gastric and small-cell lung cancer, and malignant melanoma (Roses et al., 1979; Lang et al., 1980; Johnston-Early et al., 1983; Richtsmeier and Eisele, 1986; Cainzos, 1989; Hopﬂ et al., 2000). A number of processes may be responsible for the T-cell anergy commonly seen in malignant disease. T cells from cancer patients have abnormalities in their signal transduction pathways. The T-cell receptor (TCR)-αβ or -γ δ chains bind the peptide ligand; in turn the TCR is coupled to intracellular signal transduction components by TCR-ζ subunits. The TCR-associated signaling molecule CD3 is made up of a number of subunits, which stabilizes surface expression of the TCR and is essential for interaction with MHC–antigen complex. The T-cell alterations found in in vivo models of malignant disease include complete absence in CD3-ζ , which is replaced by the Fc εγ -chain, and a reduction in T lymphocyte CD3-γ . Tyrosine kinase p56lck and p59fyn expression is also reduced. These abnormalities are associated with a reduction in the capacity of T-lymphocytes to produce the Th1 cytokines IL-2 and IFN-γ (Mizoguchi et al., 1992; Salvadori et al., 1994; Zea et al., 1995). In malignant mesothelioma the relative CDδ, CDγ , and CDζ mRNA levels expressed by tumor-inﬁltrating lymphocytes (TILs) decrease. Transforming growth factor-β is a potent tumor cell growth and immunosuppressive factor produced by many malignancies, including mesothelioma. In this disease the suppression of the CD3 subunit expression with resultant functional impairment of TILs is reversed in vivo by inducing TGFβ antisense RNA. This indicates that TILs are deactivated by tumor-associated immunosuppressive factors upon inﬁltration of the tumor microenvironment (Jarnicki et al., 1996). Cyclooxygenase (COX) enzymes are responsible for the synthesis of prostaglandins, the precise prostaglandin synthesized depending on the prostaglandin synthase enzyme present in the cell. COX-2, the inducible form of the enzyme, is constitutively expressed in virtually all premalignant and malignant cancers, including colorectal, upper gastrointestinal tract, pancreatic, head and neck, lung, and breast cancers (Murata et al., 1999; Koshiba et al., 1999; Mestre et al., 1999; Molina et al., 1999; Wolff et al., 1998; Huang et al., 1998; Taketo, 1998; Tsujii et al., 1997; Uotila, 1996; Vainio and Morgan, 1998; Vane et al., 1998). COX-2 expression is particularly associated with PGE2. On binding to its receptor on T-cells, PGE2 induces cyclic adenosine monophosphate (cAMP) formation, which inhibits the proliferation of Th1 CMI-associated CD4+ lymphocytes while stimulating the proliferation of Th2 CD4+ lymphocytes, resulting in avoidance of immune surveillance. The importance of the Th1/CMI response in both tumor regression and rejection underscores the importance of these changes. Tumor-speciﬁc cytotoxic T cells represent a major effector arm of Th1/CMI response. Through

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

239

studies of adoptive T-cell transfer, the role of tumor-speciﬁc cytotoxic T cells in mediating tumor regression has been clariﬁed (Kast et al., 1989; Greenberg, 1991; Papadopoulos et al., 1994; Rooney et al., 1995). However, the presence of such effector cells is only seen in a small minority of cases in the setting of tumor progression. Both Th1 and Th2 cytokine gene transduction experiments in animal tumor cell lines have resulted in CMI responses capable of inhibiting a subsequent challenge with parental tumor cells. However, in order to effect established-tumor regression, Th1 cytokine-secreting effector cells are required (Forni and Foa, 1996), and where tumor rejection occurs, the induction of tumor-speciﬁc CMI responses is generally seen. Collectively, these ﬁndings indicate that tumor growth either fails to stimulate an effective CMI response or evades immunesurveillance at least in part through inhibition of TIL CMI functions both locally and systemically (Browning and Bodmer, 1992; Jarnicki et al., 1996). Malignant melanoma is a highly metastatic cancer of the melatonin cells of the skin and is notoriously resistant to classical treatments such as chemotherapy and radiotherapy. Employing the same ﬂow activated cell sorter (FACS) analysis techniques used to detect intracellular cytokines in HIV patients, a signiﬁcant reduction in Th1 cytokine production can be detected in these patients (Maraveyas et al., 1999). For many years, however, it has been recognized that this tumor is sensitive to a variety of immunologically based therapies that act by boosting CMI responses. Skin lesions often disappear following direct intralesional Bacille Calmette-Guerin (BCG), and systemic responses have been reported with IL-2, αIFN, BCG + cell-based vaccines or lysates, and melanoma-speciﬁc peptides with or without IL-2 and/or pulsed onto dendritic cells. Successful treatment with immunotherapy, resulting in either stable disease or an objective tumor response, has been found to be associated with a switch from a Th2/HI-dominant proﬁle to a Th1/CMIdominant one (Grange et al., 1995; Sredni et al., 1996; Hu et al., 1998; Hrouda et al., 1998; Dalgleish, 1999). This occurs even in patients with early small-volume Duke’s A and B tumors where a reduction in systemic Th-1like responses is seen compared to age- and sex-matched controls without cancer (Heriot et al., 2000). In the latter case the observation that these responses return to normal following surgery strongly supports the deduction that it is the cancers themselves that cause the reduction in CMI responses. In NSCLC malignant pleural effusions, the majority of lymphocytes are T cells with a Th2 phenotype, whereas less than 1% are natural killer cells. Following Th1 cytokine therapy with IL-2 and IL-12, the T-helper lymphocytes shift to a Th1 phenotype. The speciﬁc anti-tumor cytotoxic property of these T-lymphocytes can be restored by the use of IL-2 treatment and TCR-CD3 engagement. IL-2 and IL-12 are synergistic in this setting (Chen et al., 1997a, 1999). It has been reported that viral agents may produce altered peptide ligands that antagonize binding of the TCR to antigen-presenting cells. This

240

A. G. Dalgleish and K. J. O’Byrne

results in inhibition of the activation of cytotoxic lymphocytes by blocking CD3-ζ tyrosine phosphorylation (Purbhoo et al., 1998; Sewell et al., 1999). Thus it may be possible for an infection to persist and give rise to an HI-predominant, suppressed-CMI environment that would predispose the tissue to the development of malignancy (see later discussion). Gene knockout experiments have provided more direct evidence for an association between deﬁcient Th1 responses and a predisposition to cancer. An increased incidence of solid tumors is seen in IFN-γ −/−, IFN-γ receptor−/−, or signal transducer and activator of transcription (STAT)1−/−, a component of the IFN signaling pathway (Chen et al., 1998; Kaplan et al., 1998). Therefore, CMI suppression may provide the ideal environment for cancer cells to develop and grow. A single mutation in an oncogene would probably be identiﬁed by a cytotoxic T-cell lymphocyte in an ordinary environment but might survive in a privileged, depressed-CMI immune-response site. As a result the mutation may persist, leaving the cell DNA primed for another stochastic event to occur, such as a p53 mutation. As the neoplastic lesion grows, it becomes progressively hypoxic. Studies indicate that hypoxia is associated with suppression of CMI responses (Lee et al., 1998; SaiRam et al., 1998), which in turn would allow escape of the malignant process from immunosurveillance. Collectively, these ﬁndings indicate that effectively reversing immune tolerance may have a role to play not only in the treatment of established malignant disease but also in chemoprevention.

VI. KAPOSI’S SARCOMA (KS) AND HIV INFECTION AS A PARADIGM FOR INFLAMMATORY CANCER DEVELOPMENT Cancer development is a multistep process leading from the initial benign transformation of cells to overt invasive, metastatic disease (Lengauer et al., 1998; Vogelstein et al., 1988; Raza, 2000). The carcinogenic process takes many years to evolve. The length of time required strongly suggests that cancer arises against a background of rigorous controls that normally maintain cell function and regulate proliferation. The combination of certain environmental factors and a genetic predisposition may have a critical bearing on whether or not an individual exposed to a particular carcinogen develops malignant disease and, if so, the duration of exposure necessary for the tumor to develop. The importance of this combination is underlined by fact that the majority of cigarette smokers never develop lung cancer, whereas many other individuals do so from passive smoking only. Cigarette smoking is associated with chronic airway inﬂammation. However, the nature of the local

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

241

inﬂammatory environment resulting from cigarette smoking is highly variable and is by polymorphic immune response genes in addition to a variety of antioxidant- and DNA-repair-associated genes (Spitz et al., 1999). The development of chronic obstructive airway disease is an independent predictor for the subsequent development of malignant disease (Mayne et al., 1999). HIV research has contributed to our understanding of the biology underlying the long-established association between certain inﬂammatory conditions and the subsequent development of cancer. The ﬁrst concerns the nature of the local and systemic immune responses seen in patients with chronic inﬂammatory conditions known to be associated with the development of malignant disease. HIV-induced AIDS is associated with the development of lymphoma and KS (Weiss, 1999). As discussed earlier, HIV infection is associated with a reduction in CMI (Westby et al., 1998a). In contrast, HI responses are enhanced. The HI predominant chronic immune activation seen in AIDS may be important in providing the necessary environment for EBV to induce lymphoma and HHV-8, KS (Westby et al., 1998a; Clerici and Shearer, 1994). These observations suggest that HI-predominant chronic immune activation associated with suppressed CMI may be a key factor contributing to the ideal environment necessary for virally driven tumors to occur. Indeed, similar immune changes are seen in those chronic infectious as well as noninfectious inﬂammatory conditions that predispose to the subsequent development of cancer. These include Helicobacter pylori– associated gastritis (gastric cancer) (Williams and Pounder, 1999), schistosomiasis associated cystitis (bladder cancer) (Raziuddin et al., 1991), human papilloma virus (HPV) cervicitis (cervical cancer) (al-Saleh et al., 1998; Le Buanec et al., 1999), hepatitis B virus (HBV-) and hepatitis C virus (HCV-) associated chronic hepatitis and cirrhosis (hepatocellular cancer) (Imperial, 1999), asbestos exposure (mesothelioma) (Bielefeldt-Ohmann et al., 1996), extensive inﬂammatory bowel disease (colorectal cancer), and chronic graftversus-host disease (lymphomas) (Habeshaw et al., 1992). Indeed, pathological assessment of early colorectal polyps, the precursor lesions for the development of colorectal cancer, indicate that they are inﬂammatory in nature (Higaki et al., 1999). Experimental evidence suggests that exposure to a foreign antigen results in up-regulation of the nonspeciﬁc proinﬂammatory cytokines IL-1α and -β and the Th1 cytokines in inﬂammatory cells. Cyclooxygenase (COX)-1 and COX-2 are among the most important enzymes in regulation of the immune response and play a key role in angiogenesis, the inhibition of apoptosis, and cell proliferation and motility. COX-1 is constitutively expressed by many cells. In contrast, COX-2 is produced by epithelial, mesenchymal, and inﬂammatory cells following exposure to proinﬂammatory cytokines (Taketo, 1998; Uotila, 1996; Vane et al., 1998), which are induced by infective agents and environmental factors known to be associated with the development of

242

A. G. Dalgleish and K. J. O’Byrne

‘Antigenic’ Carcinogenic Stimuli

Physical

Viral

Chemical

Cigarette smoke Asbestos Sunlight (UV-B)

HIV HHV8 HPV SV40 HBV HCV EBV

Dihydroxy bile salts NNK nicotine

COX-2 PGE2 • Th1 and • Th2 cytokine synthesis

• angiogenic growth factor expression e.g. VEGF, TGF-β

• Apoptosis • Angiogenesis • Cell Mediated Immunity Fig. 2 Exposure to carcinogenic stimuli results in up-regulation of cell survival factors in affected cells, including cyclooxygenase (COX)-2. COX-2 plays a key role in the conversion of arachidonic acid to prostaglandins including PGE2. PGE2 down-regulates the synthesis of Th1 cytokines and up-regulates Th-2 cytokines in inﬂammatory and/or affected epithelial, mesenchymal, or hematopoietic cells, resulting in suppressed cell-mediated immune responses (CMI), increased angiogenesis, and inhibition of apoptosis. In an acute exposure situation, the feedback between the initial proinﬂammatory response and the antiinﬂammatory Th2 cytokines is self-limiting. However, in the case of cancer-associated chronic immune activation conditions, sustained exposure to the antigen/chemical drives the cycle continuously, resulting in a sustained predominant Th2 immune response, angiogenesis, and inhibition of apoptosis, facilitating the development of cancer in a predisposed individual.

malignant disease, including Helicobacter pylori infection (Sawaoka et al., 1998a), nicotine (Schror et al., 1998), and tobacco-speciﬁc nitrosamine 4(methylnitrosamino)-4-(3-pyridyl)-1-butanone (NNK) (El-Bayoumy et al., 1999). Indeed, recent work has indicated that overexpression of COX-2 is sufﬁcient to induce tumorigenesis in the mammary glands of transgenic

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

243

mice derived using the murine mammary tumor virus promoter (Liu et al., 2001). Th2 cytokines such as IL-4 and IL-10, which can inhibit the synthesis of Th1 cytokines by CD4+ T-helper lymphocytes, are produced in COX-2-expressing environments. These Th2 cytokines down-regulate not only both proinﬂammatory/Th1 cytokines, but also COX-2 expression itself (Della Bella et al., 1997; Subbaramaiah et al., 1997; Uotila, 1996; Vane et al., 1998) (Fig. 2). Chronic antigen exposure may drive a continuous cycle in which induced proinﬂammatory and Th1 cytokines up-regulate COX-2, leading to chronic HI/Th2 cytokine production and subsequent impairment of the CMI response. In predisposed individuals this cycle may eventually lead to a predominant HI response environment. The importance of proinﬂammatory cytokines driving the HI response is underpinned by the observation that tumor necrosis factor (TNF)-deﬁcient mice are resistant to skin carcinogenesis (Moore et al., 1999). The results of these studies consistently demonstrate not only that cancer itself is associated with a shift from a Th1 to a Th2 dominant phenotype, but also that conditions predisposing to malignant disease likewise induce similar changes. This suggests that in many cases, the immune response shift precedes the development of the neoplastic process and may play a key role in carcinogenesis.

VII. THE INTERRELATIONSHIP BETWEEN THE IMMUNE RESPONSE AND ANGIOGENESIS After the interaction between inﬂammation and the immune response, the second major relationship between the immune response is with angiogenesis (O’Byrne et al., 2000a). Angiogenesis, the formation of a new blood supply from an existing vasculature, is necessary for the development of early neoplastic lesions and the growth of invasive and metastatic disease. This process occurs in all tumors and is under the regulation of proangiogenic factors, including Th2 cytokines such as IL-6 and vascular endothelial growth factor (VEGF). The intensity of the angiogenic process, as assessed by microvessel counting methods, correlates with primary tumor growth, invasiveness, and metastatic spread of disease (Folkman, 1995; O’Byrne et al., 2000b). Furthermore, there is a strong correlation between tumor cell expression of angiogenic growth factors such as VEGF and angiogenesis and patient outcome (O’Byrne et al., 2000b). Research indicates that normal physiological processes that require angiogenesis, such as ovulation, implantation of the embryo, and wound healing, occur in an HI-predominant environment (Folkman, 1995; Richards et al., 1995; Kodelja et al., 1997; Piccini et al., 1998; Schaffer and Barbul, 1998; Singer and Clark, 1999). HI-stimulated macrophages induce endothelial

244

A. G. Dalgleish and K. J. O’Byrne

cell proliferation 3–3.5 times better than CMI-stimulated macrophages in coculture experiments (Kodelja et al., 1997). These ﬁndings are supported by work in IL-6 knockout mice where the capacity to heal wounds and regenerate normal hepatic tissue, both processes that require angiogenesis, is impaired (Gallucci et al., 2000; Wallenius et al., 2000). In contrast to the up-regulated HI immune response seen, CMI responses are suppressed during ovulation, implantation of the embryo, and wound healing. This would prevent rejection of the male sperm and embryo and presentation of damaged tissues to the immune system as non-self, which might induce an autoimmune response to healing or healed tissues (Richards et al., 1995; Piccini et al., 1998; Schaffer and Barbul, 1998; Singer and Clark, 1999). In contrast to HI immune-response-induced angiogenesis, CMI immune responses tend to inhibit angiogenesis (Watanabe et al., 1997). Unlike normal physiological processes, the factors that suppress CM1 and switch on angiogenesis persist in many established chronic infectious/ inﬂammatory states, particularly conditions associated with the subsequent development of malignant disease (Table II). These include chronic viral infections (see later), asbestos (Bielefeldt-Ohmann et al., 1996), and cigarette smoke (Mayne et al., 1999). Chronic exposure to cigarette smoke leads to chronic obstructive pulmonary disease (COPD) in predisposed individuals. COPD is an independent predictor for the development of lung cancer (Mayne et al., 1999). In keeping with this, inﬂamed lung mucosa has increased vascularity compared with uninﬂamed mucosa (Fisseler-Eckhoff et al., 1996). Furthermore, bronchial dysplasia and carcinoma in situ, precursors to the development of malignant disease, have increased vascularity compared to normal bronchial epithelium (Fontanini et al., 1996; FisselerEckhoff, 1996). Using ﬂuoroscent bronchoscopy, angiogenic squamous dysplastic lesions have been identiﬁed in 34% of high-risk smokers without carcinomas and in 60% of patients with squamous-cell lung carcinoma (Keith et al., 2000). Cigarettes contain a number of factors that may predispose users to the development of malignant disease, including nicotine. Nicotine induces angiogenesis and reduces CMI, which would facilitate the survival and proliferation of a cell transformed by carcinogens such as NNK (Heeschen et al., 2001). If this state occurs for several years, then random mutations in the cells of the affected tissues, caused by carcinogens or unregulated proliferation, would occur not only in an immunologically tolerant, but also a microvessel-rich environment. Phenotypic changes, e.g., proteins resulting from mutations in the ras oncogene, which would normally be detected by cytotoxic lymphocytes, may escape immune surveillance. At the same time these cells would have an adequate supply of oxygen and nutrients and clearance of waste metabolic products, allowing another step in the stochastic progression toward malignancy to occur (Gjertsen et al., 1997). Indeed, it is so important to maintain this environment that developing

245

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

Table II Relationship between Known Causes of Chronic Inﬂammation and Cancer Causative inﬂammatory stimulus EBV

HBV/HCV

HPV

Cancer Burkitts Lymphoma (BL) Nasopharyngeal (NPL) Post transplant lymphoma Immunosupression associated lymphoma NHL Breast cancer Viruses induce hepatitis followed by cirrhosis (angiogenesis) followed by oncogenic transformation—hepatic cancer Cervix/anal/perineum/?upper aerodigestive tract

HHV-8

Kaposi’s Sarcoma

HIV

Lymphoma (EBV driven) Kaposi’s sarcoma (HHV-8 driven) Cervix (HPV driven) (HIV is not oncogenic directly in contrast to the above) Stomach cancer Lyphoma of gut Bladder Lung cancer

Helicobacter pylorii Schistosomiasis Tobacco smoke Nicotine Infections Asbestos Ulcerative colitis Crohn’s ? bile salts Reﬂux +? Prostatitis ?cause Chronic pancreatitis ?cause UV light Chronic tar/soot irritation

Mesothelioma Colorectal cancer

Mechanism of action and precursor states Associated with chromic immune activation such as malaria in BL and smoked food in NPC Immune suppression post transplantation Aﬂatoxin may enhance the degree of inﬂammation

May require extra exogenous causative agent of cervicitis to progress, e.g. chlamydia, and/or immunosuppression Only in presence of immunosupression of either age (mild) or HIV (aggressive) which causes marked immune activation HIV induces immunosuppression as well as chronic immune activation, which appears dependents on the immunogentics of host Gastritis/ulcers Chronic cystitis Chronic bronchitis Inﬂammation of tunica medica

Prostate cancer

Asbestos, ﬁbrotic, plagues Causes of inﬂammatory bowel disease including polyps and adenomas Esophagitis/obesity/tobacco/ nicotine ?infectious cause

Pancreatic cancer

Causative agent unclear

Melanoma

Skin inﬂammation and immunosuppression Common in chimney sweeps in Victorian era

Esophageal cancer

Scrotum

246

A. G. Dalgleish and K. J. O’Byrne

neoplastic cell clones evolve to mimic this state in order to progress and metastasise. This contention is supported by the fact that tumors secrete CMI-immunosuppressive cytokines such as TGFβ and IL-10 (Kiessling et al., 2000). Again induction of COX-2 may be central to the development of an angiogenic environment in many of the conditions, leading to the subsequent development of malignancy. COX-2-expressing tumor cells are associated with the production of a number of angiogenic growth factors and the synthesis and activation of matrix metalloproteinases, both of which favor tumor invasion and angiogenesis (Tsujii et al., 1997, 1998; Takahashi et al., 1999). Cigarette smoke carcinogens include the tobacco-speciﬁc carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) (which reproducibly induces pulmonary adenocarcinomas in laboratory rodents) and nicotine. NNK, a beta-adrenergic receptor agonist that releases arachidonate, and nicotine, acting through nicotinic acetylcholine receptors, upregulate COX-2 expression (Saareks et al., 1998). In keeping with these observations, there is increasing evidence that nicotine itself may induce COX-2 expression. Inhibition of COX-2 activity reduces IL-6 and IL-8 levels secreted by human cell lines, further supporting the need for tumors to produce cytokines that both suppress the CMI response and promote angiogenesis (Luca et al., 1997; Salgado et al., 1999; Hong et al., 2000).

VIII. INFLAMMATION AND APOPTOSIS Experimental evidence indicates that chronic inﬂammation may give rise to the production of growth factors and cytokines, and the activation of intracellular cell survival pathways that would result in inhibition of apoptosis. An example of such a factor released during inﬂammatory states is macrophage inhibitory factor (MIF), which has been shown to repress the transcription activity of p53 and its downstream targets of p21 and bax, thereby having a marked antiapoptotic effect (Cordon-Cardo and Prives, 1999; Hudson et al., 1999). Experimental evidence indicates that p53 plays an important role in the mediation of Th1 cytokine induced cytotoxicity (Yeung and Lau, 1998; Das et al., 1999; Kano et al., 1999; Um et al., 2000; Takagi et al., 2000). p53 induces Transporter Associated with Antigen Processing (TAP) 1 expression through a p53-responsive element. TAP1 is required for the major histocompatibility complex (MHC, HLA in humans), class I antigen presentation pathway. p73, which is homologous to p53, also induces TAP1 and cooperates with p53 to activate TAP1. Through the induction of TAP1, p53 enhances the transport of MHC class I peptides and expression of surface MHC–peptide complexes. p53 cooperates with IFN-γ

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

247

to activate the MHC class I pathway. These results indicate that, as part of their function as tumor suppressors, p53 and its homolog p73 may play in a role in tumor surveillance (Zhu et al., 1999). Therefore, inﬂammation is capable of inactivating one of the most important cell regulatory pathways controlling cancer development, reducing the effectiveness of the body’s own cellular defense reaction to a mutation. p53 also has a key role in the regulation of angiogenesis, in part through induction of the antiangiogenic factor thrombospondin (Dameron, 1994). Furthermore, mutations of p53 may result in induction of the synthesis of the most potent angiogenic agent known, vascular endothelial growth factor, leading to increased angiogenesis (Kieser et al., 1994; Volpert et al., 1997). Therefore, loss of p53 would also result in an impaired CMI response, facilitate angiogenesis, and result in a loss of apoptotic activity. There is increasing evidence that exposure to carcinogens such as ultraviolet B light (Athar et al., 2001), the tobacco speciﬁc carcinogen NNK (El-Bayoumy et al., 1999), nicotine (Saareks et al., 1998), Helicobacter pylori (Konturek et al., 2000; Sawaoka et al., 1998a) and colonic luminal contents, in particular the dihydroxy bile acids deoxycholate and chemodeoxycholate (Zhang et al., 1998; Glinghammar and Rafter, 2001), leads to the up-regulation of COX-2 in the affected tissue. In both nonneoplastic and neoplastic cells, COX-2 is associated with cell proliferation (Tsuji et al., 1996; McGinty et al., 2000) and inhibition of apoptosis at least in part through the induction of bcl-2 (Tsujii and DuBois, 1995). The serine/threonine kinase akt (protein kinase B) is activated in response to a variety of stimuli. This factor provides a survival signal that protects cells from apoptosis induced by growth factor withdrawal. Through the phosphorylation of speciﬁc targets such as Bad (del Peso et al., 1997) and procaspase-9 (Cardone et al., 1998), the akt cell survival signaling pathway inhibits apoptosis. Recent work has indicated that some carcinogens act, at least in part, by inducing oxidative stress in exposed cells. Oxidative stress results in activation of intracellular-survival cell-signaling pathways. In a variety of cell types H2O2, an inducer of oxidative stress, has been shown to induce elevated Akt activity in a time- and dose-dependent manner by a mechanism involving phosphoinositide 3-kinase (PI3K). Inhibitors of PI3K activity, including wortmannin and LY294002, and expression of a dominant negative mutant of p85, a regulatory component of PI3K, inhibited H2O2-induced Akt activation. H2O2 treatment led to EGFR phosphorylation. Inhibition of EGFR activation blocked Akt activation, suggesting that activation of the akt pathway by H2O2 is dependent on EGFR activation. H2O2 induces apoptosis of HeLa cells, which is signiﬁcantly enhanced by the akt inhibitors Wortmannin and LY294002. In contrast, expression of exogenous myristoylated activated Akt, and constitutive expression of v-Akt, inhibits apoptosis of H2O2-treated NIH3T3 cells. These results suggest that

248

A. G. Dalgleish and K. J. O’Byrne

H2O2 activates Akt via an EGFR/PI3-K-dependent pathway and that elevated Akt activity confers protection against oxidative-stress-induced apoptosis (Wang et al., 2000). In keeping with these observations, UV light exposure, which results in EGFR phosphorylation and activation of the akt pathway in a variety of cell types, including keratinocytes, NIH ﬁbroblasts, and HC11 mouse mammary cells, has been demonstrated to induce H2O2 (Huang et al., 1996; Peus et al., 2000). Likewise, carcinogenic asbestos ﬁbres have been demonstrated to induce EGFR phosphorylation, an effect blocked by N-acetylcysteine (Faux et al., 2000). EGFR activation may also result in COX-2 expression (Coffey et al., 1997; Mestre et al., 1999), although the pathways linking EGFR to induction remain to be fully elucidated. However evidence is increasing for an important role for the transcription factor NF-κB. This transcription factor plays an important role in the regulation of a number of genes intrinsic to inﬂammation and cell proliferation (Thanos and Maniatis, 1995). EGFR phosphorylation activates the extracellular regulated kinase 1 and 2 (ERK1/2) and p38 signaling pathways (Peus et al., 2000). PI3K/akt activates the transcription factor NF-κB (Beraud et al., 1999; Madrid et al., 2001). Akt has been shown to target the transactivation function of NF-κB by stimulating the transactivation domain RelA/p65 of the transcription factor. This appears dependent on I-κB kinase-β activity. p38 is required for NF-κB transcriptional function activity. Consistent with this, activated Akt has been shown to induce p38 activity (Madrid et al., 2001). Carcinogenic asbestos ﬁbers induce NF-κB activation in pleural mesothelial cells, and this is linked to cell proliferation (Janssen et al., 1997; Faux et al., 2000). In recent work we have demonstrated that the selective EGFR-tyrosine kinase inhibitor PKI166 (Novartis Pharmaceuticals) inhibits the DNA binding of NF-κB. PKI166 and NF-κB decoy proteins reduced cell viability, showing the importance of this pathway in mesothelial cell survival following asbestos exposure (Faux et al., 2001). A second member of the erb family of type 1 tyrosine kinase receptors, HER-2/neu/c-erbB-2, has also been demonstrated to induce NF-κB. HER-2 overexpression induces the transcription factor by a PI3K/Akt pathway that involves calpein-mediated I-κB-α degradation (Pianetti et al., 2001). The NF-κB binding motif is found in the promoter region of the COX-2 gene (Du Bois et al., 1998). In keeping with the important role of proinﬂammatory cytokines and angiogenic growth factors in the carcinogenic process, IL-1β and bFGF combined with EGF have been shown to enhance the induction of COX-2 (Majima et al., 1997; Yucel-Lindberg et al., 1999). IL-1β induces p38 activity, and this is dependent on Akt and I-κB kinase activation (Madrid et al., 2001). Likewise, akin to HER-2 overexpression in breast cancer, NF-κB activation by IFN-α and -β involves degradation of I-κB-α (Yang et al., 2001). Finally, Akt activation of the NF-kB is involved in cell survival and resistance to apoptosis induced following exposure to TNF-α,

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

249

similar to the situation seen with IL-1β and IFN-α and -β (Zhou et al., 2000). These ﬁndings are in keeping with the observation that proinﬂammatory and Th1 cytokines induce COX-2 expression. Growth factors such as EGFR may protect against apoptosis through a number of other mechanisms. For example, in keratinocytes EGFR upregulates the expression of the antiapoptotic factor bcl-xL through the activation of a MEK-dependent pathway. Furthermore, activation of the PI3K/Akt and phospholipase C-g/protein kinase C-a pathways is required for keratinocyte survival independent of EGFR activation or bcl-xL expression (Jost et al., 2001). In keeping with this are the observations that dihydroxy bile acids, promoters of gastrointestinal tract cancer, induce COX-2 transcription and a 10-fold increase in PGE2 expression in human esophageal adenocarcinoma cells, which is blocked by inhibitors of PKC. Furthermore, increased binding of the transcription factor AP-1, an inducer of COX-2 transcription, to DNA is also seen (Zhang et al., 1999). Therefore, as well as inducing an HI-predominant immune response and angiogenesis, chronic immune activation may lead to inhibition of apoptosis in the affected cells. Carcinogen-induced reactive oxygen species and the formation of carcinogenic metabolites produced by the inﬂammatory process, such as malondialdehyde resulting from the metabolism of arachidonic acid by COX-2 (Subbaramaiah et al., 1997), may lead directly to DNA damage and subsequent mutations. Under these circumstances cells may transform, acquire mutations, proliferate, and, through microenvironmental selection pressures (Pettit et al., 2000), eventually take on a malignant phenotype through the development of cell clones that are themselves capable of resisting apoptosis, suppressing CMI responses and inducing angiogenesis.

IX. THE BIOLOGY OF ONCOGENIC VIRUSES: IMPACT ON THE IMMUNE RESPONSE, ANGIOGENESIS, AND APOPTOSIS The data presented indicate that chronic immune activation may give rise to the prerequisite environment necessary for the development of malignant disease in predisposed individuals. This environment includes the presence of suppressed CMI, angiogenesis, and growth factors that inhibit apoptosis. If such a postulate is true, one would expect that from a biological standpoint oncogenic viruses would have the capacity to induce idenitical environmental features. As mentioned earlier, human papilloma virus (HPV) infection predisposes to cervical carcinoma and to a number of other malignancies, including

250

A. G. Dalgleish and K. J. O’Byrne

anal, head and neck, and possibly oesophageal and lung cancer (Markham, 1996). HPV16 and 18 are the principal HPV subtypes associated with neoplastic disease. These produce the E6 and E7 proteins that inactivate the p53 and retinoblastoma (Rb) tumor suppressor proteins, respectively (Dalgleish, 1991). CMI responses are crucial to the pathogenesis of HPV infection. Regression of genital warts is characterized by a pronounced increase in Th1 cells and macrophages associated with a localized delayed hypersensitivity response. A number of cytokines are secreted, with IL-12 being present in very high levels. However, only recently has clear evidence accumulated that persistent HPV infection is associated with chronic immune activation. Although over 25% of females are infected at ages 19–25, less than 5% remain infected over 35 years of age, suggesting that in a small but signiﬁcant proportion of cases the HPV infection persists. Although differences in methodology and sampling error may have explained these observations, a recent longitudinal study of HPV-infected patients, in which the virus was detected using the hybrid capture II assay, demonstrated that a proportion of patients found to have persistent infection after 2 to 3 assessments developed cervical intraepithelial neoplasia (CIN). In contrast, those individuals found to have cleared the infection did not develop any CIN lesions (Clavel et al., 2000). Failure to clear the viral infection results in persistent inﬂammation with chronic cervicitis and an increased cervical cancer risk (White et al., 1992; Cerqueira et al., 1998; Hsieh et al., 1999). Increased levels of circulating IL-2 soluble receptor, a nonspeciﬁc marker of inﬂammation, are seen in a proportion of otherwise normal infected individuals that rises signiﬁcantly with the development of CIN and subsequently invasive cervical cancer (Hildesheim et al., 1994; Ung et al., 1999). Resolution of HPV infection with clearance of the virus is associated with the development of a CMI response, including an IgA antibody response (Bontkes et al., 1999b). In addition, up-regulation of IFN-γ in exfoliated cervical cells (Scott et al., 1999), IL-2 Th1 responses to the C-terminal domain of the HPV-16 E2 protein (Bontkes et al., 1999a), and a hypersensitivity reaction to the HPV-16 oncoprotein E7 are also seen. This hypersensitivity reaction is itself associated with the subsequent regression of CIN lesions (Hopﬂ et al., 2000). In contrast, active CIN is associated with predominant Th2 immune responses with an increased IL-10/IL-12 ratio seen in wholeblood supernatants (Jacobs et al., 1998). Furthermore, the density of IL-2secreting cells is lower and that of IL-4-positive cells higher in high-grade squamous epithelial lesions (CIN III) than in the transformation zone of healthy women with biopsies showing squamous metaplasia (al-Saleh et al., 1998). Although cytotoxic T-lymphocyte (CTL) responses have been difﬁcult to detect in HPV-infected patients, CTL responses to E6/E7 proteins are more commonly seen in women with cervical HPV 16 infection without evidence

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

251

of CIN than in HPV 16 positive women with CIN (Sonnex, 1998). These results are supported by studies in women with cervical dysplasia who have shown reduced lymphocyte Th1 immune responses to HPV 16 L1 antigen and E6 and E7 peptides compared to healthy adults (Luxton et al., 1997). As CIN progresses to frank carcinoma in situ, the levels of CD4+ and CD3+ DR+ antigen peripheral blood T-cells have been shown to decrease while the level of the CD8+ cells increases (Spivak et al., 1999). More recently, microarray analysis has indicated that infection with HPV16 and 31 may result in down-regulation of the response of IFN-inducible genes to IFN (Chang and Laimins, 2000; Nees et al., 2001). In the case of HPV16 infection, the expression of multiple genes known to be inducible by NFκB and AP-1 was seen. E6 protein increased the secretion of a number of factors, including IL-8 and MIP-1α, which themselves are capable of suppressing CMI responses, inducing angiogenesis and inhibiting apoptosis (Nees et al., 2001). Coinfection with chlamydia may exacerbate the situation, the presence of both organisms being associated with increased proliferation and reduced apoptosis of the ectocervical epithelium (Vaganova, 2000; Anttila et al., 2001). Furthermore, CIN develops in approximately 40% of women with HIV infection. A shift toward a Th2 T-cell cytokine proﬁle is seen in biopsy specimens of normal cervix from HIV-positive women as compared to normal healthy controls, consistent with the overall bias toward Th2 responses in HIV-infected individuals (Olaitan et al., 1998). Epstein–Barr virus (EBV) is implicated in the pathogenesis of lymphomas and solid tumors. In Africa infection EBV-induced Burkitt’s lymphoma (BL) arises in children whose immune systems are chronically activated by malaria (De The, 1993). In keeping with this, EBV-associated nasopharyngeal cancer (NPC) occurs in China among people who are exposed to ﬁsh treated by a smoke-curing process that may also cause local inﬂammation (Zheng et al., 1994). Although the majority of the Western population is infected with EBV, only a minority develop associated malignancies, in particular lymphomas (Yamamoto et al., 1999; Mauray et al., 2000). Other EBV-associated malignancies include squamous esophageal cancer (Wang et al., 1999), gastric cancer (Takada, 2000), and breast cancer where viral DNA is detected in up to 50% of cases studied compared to normal breast tissue (Bonnet et al., 1999). Although preexisting chronic immune activation may not be readily apparent prior to the onset of these conditions, EBV gene products may nonetheless provide the prerequisite environment outlined earlier through the suppression of CMI, induction of angiogenesis, and inhibition of apoptosis. The factors produced by EBV viral DNA and detected in EBV-associated malignancies include the transforming gene product BARF1 (zur Hausen et al., 2000), BHRF1 (Liu et al., 2000), a homolog of the anti-apoptotic factor bcl-2, EBV-encoded poly(A)(−) RNA which induces IL-10 (Kitagawa

252

A. G. Dalgleish and K. J. O’Byrne

et al., 2000), and EBV-encoded latent membrane protein 1 (LMP1), which activates the p38 mitogen-activated protein kinase pathway and has been demonstrated to coregulate IL-6 and IL-8 production (Eliopoulos et al., 1999; Mauray et al., 2000). EBV LMP1 protein expression has been demonstrated to correlate with the expression of the gelatinase MMP-9 and the presence of metastases in nasopharyngeal cancer (Horikawa et al., 2000). Compared to control tissues, increased IL-1α and IL-1β expression is seen in nasopharyngeal cancer, which correlates with EBV-encoded viral IL-10 transcript (Huang et al., 1999). This observation is in keeping with the crucial role induction of proinﬂammatory cytokines is postulated as playing in carcinogenesis and the progression of malignant disease. Simian virus-40 (SV-40) has been implicated as playing an important role in the pathogenesis of a number of malignancies, including ependymomas, choroid plexus tumors, bone tumors, sarcomas, and, in particular, mesothelioma. SV-40 virus oncoprotein, SV-40 large T antigen, binds each of the retinoblastoma family proteins, pRb, p107, and pRb2/p130 and p53 (Carbone et al., 1999; Carbone, 1999; De Luca et al., 1997). Through these effects the oncoprotein would facilitate angiogenesis, reduce Th1 cytokinemediated cytotoxicity, and inhibit apoptosis, thereby predisposing the infected individual to the development of malignant tumors in infected tissues. HBV and HCV do not cause cancer unless chronic inﬂammation occurs, and then only after many years have passed (IIHCSG, 1998). Acute HBV and HCV infections require an effective CMI response in order to resolve. In contrast, chronic infection is associated with weak or undetectable CMI responses and/or a predominant Th2 response. The precore antigen (HBeAg), secreted in chronic hepatitis B, has been shown to deplete antihepatitis-B-speciﬁc Th1 cells that are necessary for viral clearance, while enhancing Th2 cytokine-producing cells (Milich et al., 1998). In hepatitis C the activation of Th2 responses plays a role in the development of chronicity (Tsai et al., 1998). This shift to a Th2 phenotype has been conﬁrmed in vitro in cultured monocytes from patients with chronic HCV. When compared to normal controls, these monocytes respond to antigen stimulation by increasing the synthesis of the Th2 cytokine IL-10, but not the Th1 cytokine IL-12 (Kakumu et al., 1997). Angiogenesis occurs in chronic hepatitis and cirrhosis of the liver induced by hepatitis B and C (Mazzanti et al., 1997; El-Assal et al., 1998). In keeping with this observation, plasma VEGF levels are elevated in chronic hepatitis and cirrhotic liver disease (Jinno et al., 1997) and serum VEGF levels in patients with acute hepatitis as compared to healthy individuals (Akiyoshi et al., 1998). Treatment of chronic HBV and HCV with the therapeutic Th1 cytokines interferon-α and IL-12 results in varying degrees of clearance of viral particles from the liver and clinical remission (Cacciarelli et al., 1996; Carreno and Quiroga, 1997). When

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

253

successful, the viral clearance is associated with a reduction in circulating IL4, IL-6, and IL-10 Th2 cytokine levels (Cacciarelli et al., 1996; Malaguarnera et al., 1997). In conclusion, failure to clear human oncogenic viruses predisposes to the development of malignant disease through the creation of a HI predominant, proangiogenic, anti-apoptotic environment which would favour the survival of transformed cells and be conducive to the development of malignant disease. This applies to those infectious states where chronic immune activation is not readily in which viral oncogenesle of giving rise to a similar environment. The same may also be true for cancers associated with non-infectious carcinogenic insults, including cigarette smoking and asbestos exposure, which themselves often cause inﬂammation.

X. FEATURES OF TUMORS LACKING AN OBVIOUS PREMALIGNANT INFLAMMATORY PROCESS Although initially felt to be relevant to only a minority of cancers, our increased understanding of the environment induced by chronic infective inﬂamatory conditions indicates that virtually all tumors, including those of the lung, esophagus, colon, and rectum, may also ﬁt this model. A similar pattern of suppressed CMI, angiogenesis, and inhibition of apoptosis is seen even in those tumors such as melanoma and renal cell cancer that initially do not appear to readily ﬁt into this model. For example, in melanoma, not only is there a clear genetic susceptibility to the disease (individuals with fair skin, freckles, ginger hair, and a tendency to mole formation), but also a history of recurrent sunburn. Sunburn is associated with activation of the EGFR and the PI3K pathway (Peus et al., 2000), induction of COX-2 expression (Buckman et al., 1998), angiogenesis in the affected tissue, and local and systemic suppression of the Th1 immune response (Pamphilon et al., 1991; Luca et al., 1997). The development of renal-cell cancer is associated, in the majority of cases, with mutations in the Von Hippel Landau tumor suppressor gene. These mutations are associated with up-regulation of the angiogenic growth factor VEGF (Fleming, 1999). Experimental studies clearly demonstrate that VEGF, as well as inducing new microvessel formation, suppresses dendritic cell function, an important mediator of CMI (Gabrilovich et al., 1999). More recently, kidney insults, such as dehydration, have been shown to be associated with induction of COX-2. This may give rise to an HI-predominant, CMI-depressed, proangiogenic, antiapoptotic local environment (Yang et al., 1999). Likewise, chronic immune activation has not been considered a cause for germ-cell, breast and prostate tumors. Testicular cancer, a disease very

254

A. G. Dalgleish and K. J. O’Byrne

sensitive to treatment with cytotoxic chemotherapeutic agents with the majority of cases treated being cured, is increasing rapidly in incidence. The chemosensitivity appears to be related to the low incidence of p53 mutations. Although a speciﬁc carcinogen has not yet been identiﬁed, these germ cell tumors arise at an immunologically privileged site in which immune responses are suppressed (Guillou et al., 1996). The inﬂammatory components in endocrine tumors such as breast and prostate cancer may be more subtle than in nonendocrine tumors. This is because the endocrine system interacts directly with the immune response, with Th1 and DHEA steroids counterregulating Th2 and cortisol steroid pathways (Rook et al., 1994). Nonetheless, epidemiological evidence exists that both breast and prostate cancer occur more frequently in the presence of chronic mastitis and prostatitis, respectively (Prince and Hildreth, 1986; Nakata et al., 1993; Monson et al., 1976). The role of inﬂammation in the pathogenesis of prostate cancer is supported by the observation that there is an increased risk of marker relapse in patients whose disease is found to have high-grade inﬂammation surrounding malignant glands in the resected tumor following prostatectomy (Irani et al., 1999). Therefore, it is somewhat surprising that a possible connection between prostatitis and prostate cancer has only rarely been discussed (De Marzo et al., 1999). If the biological changes associated with chronic immune activation are essential for carcinogenesis (and these changes do seem to occur in noninﬂammatory states that predispose individuals to the subsequent development of malignant disease), it is logical to suppose that the majority of cancers may be amenable to modulation with combination therapies including CMI/Th1 enhancing, antiangiogenic, antiinﬂammatory agents. This approach may be of particular beneﬁt in cancer prevention and have a greater impact on cancer mortality than any other measure other than smoking cessation.

XI. ENVIRONMENT CREATED BY NONINFLAMMATORY FACTORS THAT PREDISPOSE TO MALIGNANT DISEASE Recent observations indicate an important role for insulin-like growth factors (IGF) in the initiation of malignant disease. High normal IGF-I levels are associated with the development of prostate and colorectal cancer, and with the development of breast cancer in premenopausal women (Hankinson et al., 1998; Chan et al., 1998; Ma et al., 1999). Furthermore, there is evidence that the combined effects of IGF-I and mutagen sensitivity contribute signiﬁcantly to lung cancer risk (Guan et al., 1998; Wu et al., 2000). Both IGF-I and IGF-II have been shown to play a role in the induction of COX-2

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

255

and PGE2 in nonmalignant and malignant cells (Di Popolo et al., 2000). In keeping with this is the observation that IGF-mediated muscle cell survival is associated with activation of Akt (Lawlor and Rotwein, 2000). In keeping with this, IGF-I has been shown to be a proangiogenic factor stimulating endothelial cell proliferation and inducing the synthesis of VEGF (Akagi et al., 1998; Dunn, 2000). In adrenocortical tumors, IGF-II has been shown to correlate with VEGF expression and to be inversely associated with expression of the antiangiogenic factor thrombospondin-1 (de Fraipont et al., 2000). Moreover, there is increasing evidence to indicate that IGF-I/IGF-IR system is up-regulated by other angiogenic growth factors such as IL-6 and PDGF (Rubini et al., 1994; Franchimont et al., 1997). Although IGF-I is a proliferative factor for inﬂammatory cells and plays an important role in maintaining normal immune function, under appropriate circumstances the growth factor may act as an antiinﬂammatory agent, suppressing CMI responses to antigens and injury. For example, following experimental renal ischemic injury, IGF-I therapy inhibits TNF-α and major histocompatibility complex (MHC) expression and mild interstitial inﬁltrate normally seen in the affected kidney. IGF-I has also been shown to down-regulate TNF-α serum levels in an in vivo sepsis model, whereas overexpression of the IGF-IR protects cells against TNF-induced apoptosis (Wu et al., 1996; Balteskard et al., 1998). Under normal conditions IGF-I has been shown to up-regulate p53 expression (Wang et al., 1998). In situations where p53 is up-regulated, the proapoptotic protein represses the transcription of IGF-IR, thereby damping down the growth-promoting effects of IGF peptides (Webster et al., 1996). p53 also induces the expression of IGF-binding protein (IGFBP)-3. IGFBP-3 not only antagonizes IGF-I activity, but is also capable of inducing apoptosis by modulating the expression of Bax and Bcl-2 and potentiating p53-independent radiation-induced apoptosis in human breast cancer cells (Butt et al., 2000). However, in stress-inducing conditions there is evidence that IGF-I may induce the expression of mdm2 and down-regulate p53, thereby inhibiting apoptosis (Leri et al., 2000). Furthermore, some mutations in the p53 protein appear to have the capacity to derepress the IGF-IR promoter, leading to growth enhancement by locally produced or systemic IGF-I (Werner et al., 1996). In keeping with these observations, antisense IGF-I/IGF-IR therapies result in reduced tumor cell growth in vitro. They inhibit tumorigenicity and the induction of tumor regression in vivo and result in an increased expression of major histocompatibility complex I antigen and a tumor-speciﬁc immune response (Pass et al., 1996; Lafarge-Frayssinet et al., 1997; Ellouk-achard et al., 1998). Obesity is associated with the development of malignant diseases including breast, colon, and prostate cancer (Moller et al., 1998) and has been demonstrated to suppress lymphocyte function, mitogenesis and natural killer-cell

256

A. G. Dalgleish and K. J. O’Byrne

activity. T-lymphocyte responses to concanavalin A and B-lymphocyte responses to pokeweed mitogen are reduced (Moriguchi et al., 1998). In rats receiving a high-fat diet, the cigarette-smoke carcinogen NNK induces increasingly high levels of COX-2 expression through progressive stages of lung tumorigenesis (El-Bayoumy et al., 1999). Experiments in obese Zucker rats have shown that exercise restores natural killer-cell activity and concanavalin-A-induced splenic lymphocyte responses, reversing the CMI suppression (Moriguchi et al., 1994). Long-term treatment with immunosuppressive therapy, using agents such as cyclosporine A following organ transplantation, is associated with an increased incidence of both virally and non-virally induced tumors (Penn, 1988; Nabel, 1999). Cyclosporin A has been demonstrated to induce malignant cell proliferation, invasion, and metastasis in vitro and/or in vivo. Evidence suggests the immunosuppressive agent acts by inducing the transcription and expression of functional TGF-β. TGF-β antibodies block the observed effects of cyclosporine A on the cell lines tested (Hojo et al., 2000). This is interesting, as TGF-β may also induce apoptosis. Evidence indicates that COX-2 expression converts the apoptotic signaling normally induced by TGF-β to cell proliferation (Saha et al., 1999; Roman et al., 2001). TGF-β is also a potent inhibitor of CMI (Qin et al., 1996; Meert et al., 1996), inducing T-cell anergy (Jarnicki et al., 1996) and directing the immune response toward a Th2/HI phenotype (D’Orazio et al., 1998). Depending on the environment, TGF-β has also been found to be proangiogenic and has been shown to induce the synthesis of VEGF by malignant cells (Pepper et al., 1993). As discussed earlier, tumor-associated macrophages may play a role in tumor angiogenesis. TGF-β is chemotactic for macrophages and induces macrophage production of growth factors and proteases such as IL-6 and urokinase. These factors induce endothelial cell proliferation and break down the extracellular matrix, respectively, thus facilitating new blood-vessel formation (Wahl et al., 1987; Bielefeldt-Ohmann et al., 1996; Hildenbrand et al., 1998). Thus, cyclosporine A induced TGF-β expression may result in all the development of the essential environmental factors that we hypothesize are necessary for carcinogenesis and tumor development.

XII. FRACTAL MATHEMATICS, CARCINOGENESIS, AND THE PROGRESSION OF MALIGNANT DISEASE The immune response and cellular control pathways are a complex system of interacting factors. Based on the chaos theory, the factors involved in the development of cancer represent nonlinear or chaotic processes and are associated with both unpredictability, because there too many forces acting on

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

257

the system, and order, in that major attractors are present that are attempting to maintain the system (Coffey, 1998). The simpliﬁed concept of CMI and HI immune responses represent two attractors that are both self-regulatory with feedback pathways and are affected by certain outside forces such as chronic infections and noninfectious factors (Dalgleish, 1999). The major regulatory pathway factors, including p53, p73, Rb, and the bcl-2 apoptotic family, themselves major cellular policemen in their own right, may also be affected. Once these factors are signiﬁcantly perturbed, further stochastic oncogenic effects can progress. The relevance of this concept is that treatments that return the attractors to normal may be able to have signiﬁcant indirect anticancer activity if applied before the cancer has progressed too far. Therapeutic agents aimed at inducing a predominant CMI response and inhibiting angiogenesis may move the disorder in the system away from tumor cell growth and progression toward inhibition of proliferation and, in some patients, tumor regression. Successful induction of a CMI, antiangiogenic environment may be responsible for the tumor regressions seen in those patients with renal cell cancer and melanoma following nonspeciﬁc vaccination with BCG and cytokine therapy using IL-2 and IFN-α (Dalgleish, 1999; Browning and Dalgleish, 1996; Vile et al., 1996).

XIII. IMPLICATIONS FOR CHEMOPREVENTION AND FUTURE TREATMENT OF MALIGNANT DISEASE Nonsteroidal antiinﬂammatory drugs (NSAIDs) and speciﬁc COX-2 inhibitors inhibit solid malignant-cell proliferation in vitro. Provided the disease overexpresses COX-2, these agents prevent hematogenous spread of malignancy, suppress angiogenesis, and inhibit growth of tumor xenografts (Tsuji et al., 1996; Hida et al., 1998; Sawaoka et al., 1998a, b, 1999; Molina et al., 1999; Tomozawa et al., 1999). However, of greater relevance to the hypothesis that chronic immune activation plays a central role in carcinogenesis is the observation that inhibition of COX-2 activity inhibits carcinogenesis in animal models. Both nonspeciﬁc COX inhibitors such as indomethacin and speciﬁc COX-2 inhibitors such as celocoxib may inhibit malignant transformation in a variety of experimental in vivo models, including those for breast, colorectal, lung, and skin cancer (Lala et al., 1997; Taketo, 1998; Subbarmeieh et al., 1998; Fischer et al., 1999; Mestre et al., 1999; Yao et al., 2000). There is already considerable evidence in the literature that long-term exposure to aspirin and other NSAIDs reduces the incidence of esophageal, gastric, colorectal, bladder, and lung cancer (Thun and Namboodin, 1991; Giovannucci et al., 1995; Study, 1992; Paganini-Hill, 1994; Giovannucci

258

A. G. Dalgleish and K. J. O’Byrne

et al., 1994; Funkhouser and Sharp, 1995; Schreinemachers and Everson, 1994; Castelao et al., 2000; Langman et al., 2000). The effect of aspirin on the inhibition of the COX enzymes and the resultant prostaglandin synthesis may be responsible for the chemoprevention. Inhibition of COX enzymes would prevent susceptible, transformed, and/or dysplastic cells from inducing suppression of CMI, inducing angiogenesis and inhibiting apoptosis (Tsujii et al., 1998; Taketo, 1998; Vane et al., 1998; Subbaramaiah et al., 1997; Grinwich and Plescia, 1977). Randomized placebo-controlled studies in thousands of people conducted over several years would be required to prove the efﬁcacy of NSAIDs and selective COX-2 inhibitors in chemoprevention. There is a need to identify individuals at high risk of developing malignant disease and reduce the “promoter” exposure. Following the success of the tamoxifen trials in reducing the incidence of breast cancer in an at-risk population, this approach should become a high priority in chemoprevention (reviewed in Savanthanan and O’Byrne, 2001). In the not-too-distant future, an increased understanding of the molecular pathogenesis of malignancies such as lung cancer may make it possible to identify those individuals at particular risk of developing the disease. In the absence of obvious factors such as cigarettes, the reduction of inﬂammation, inhibition of angiogenesis, restoration of CMI, and induction of an appropriate proapoptotic state should be primary goals. If successful, chemoprevention should decrease the number of patients requiring one-to-one treatment for malignant disease at a later time. Trials could test the relevant contribution of antiinﬂammatory, antiangiogenic, and CMI immune stimulatory agents alone and in combination as chemopreventive and adjuvant therapies, and in the management of established inoperable/metastatic malignant disease.

XIV. CONCLUSIONS We have shown that chronic infectious diseases are associated with chronic activation of the immune system (AIDS) or local chronic inﬂammation (e.g., Helicobacter pylorii and gastric cancer or lymphoma). This chronic immune activated state is associated with local and/or systemic suppression of CMI, the induction of angiogenesis, and the inhibition of apoptosis. The association between cancer and chronic infection and/or inﬂammation and inherited risk factors such as genetic abnormalities mediates interactive changes in the immune response, angiogenesis, and apoptosis. This “Golden Triangle” of effects predisposes the individual to the development of cancer. Carcinogeninduced or inherited-mutation-induced cellular transformation in a predisposed individual may not be detected by the suppressed immune response,

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

259

thus allowing the mutant cell to survive in the relatively rich blood supply and, despite its altered genetic state, be protected from apoptosis. Thus the cell would be primed to progress down the malignant pathway through the acquisition of further mutations. When the tumor had evolved to the extent that it became capable of producing a similar environment itself, the disease would become invasive and eventually metastasize. Key factors involved in the process include the COX-2 associated pathways as well as apoptotic and cellular regulatory genes such as p53. Although the genetic background greatly affects the degree of the inﬂammatory response and in turn the outcome of disease, this relationship can be exploited to reduce the cancer incidence, particularily in high-risk individuals.

ACKNOWLEDGMENTS Professor A. G. Dalgleish is supported by the Cancer Vaccine Campaign, BBSRC/Onyvax, and Celgene, and Dr. Kenneth O’Byrne by the Institute of Cancer Studies.

REFERENCES Aitken, C., Dalgleish, A., Salama, A., and Forster, G. (1992). Pancytopaenia and hepatosplenomegaly in an AIDS patient responding to high dose steroids [letter]. Int. J. STD AIDS 3, 450–451. Akagi, Y., Liu, W., Zebrowski, B., Xie, K., and Ellis, L. M. (1998). Regulation of vascular endothelial growth factor expression in human colon cancer by insulin-like growth factor-I. Cancer Res. 58, 4008–4014. Akiyoshi, F., Sata, M., Suzuki, H., Uchimura, Y., Mitsuyama, K., and Matsuo Tanikawa, K. (1998). Serum vascular endothelial growth factor levels in various liver diseases. Dig. Dis. Sci. 43, 41–45. al-Saleh, W., Giannini, S. L., Jacobs, N., Moutschen, M., Doyen, J., Boniver, J., and, Delvenne, P. (1998). Correlation of T-helper secretory differentiation and types of antigen-presenting cells in squamous intraepithelial lesions of the uterine cervix. J. Pathol. 184, 283–290. Arthur, L. O., Bess, J. W., Jr., Urban, R. G., Strominger, J. L., Morton, W. R., Mann, D. L., Henderson, L. E., and Benveniste, R. E. (1995). Macaques immunized with HLA-DR are protected from challenge with simian immunodeﬁciency virus. J. Virol. 69, 3117–3124. Asselin-Paturel, C., Echchakir, H., Carayol, G., Gay, F., Opolon, P., Gruenenwald, D., Chouaib, S., and Mami-Chauaib, F. (1998). Quantative analysis of Th1, Th2 and TGFbeta 1 cytokine expression in tumour, TIL and PBL of non-small cell lung cancer patients. Int. J. Cancer 77, 7–12. Anttila, T., Saikku, P., Koskela, P., Bloigu, A., Dillner, J., Ikaheimo, I., Jellum, E., Lehtinen, M., Lenner, P., Hakulinen, T., Narvanen, A., Pukkala, E., Thoresen, S., Youngman, L., and Paavonen, J. (2001). Serotypes of Chlamydia trachomatis and risk for development of cervical squamous cell carcinoma. JAMA 285, (1), 47–51. Aziz, M., Akhtar, S., and Malik, A. (1998). Evaluation of cell-mediated immunity and circulating immune complexes as prognostic indicators in cancer patients. Cancer Detection Prevention 22, 87–99.

260

A. G. Dalgleish and K. J. O’Byrne

Badawi, A. F., Mostafa, M. H., Probert, A., and O’Connor, P. J. (1995). Role of schistosomiasis in human bladder cancer: evidence of association, aetiological factors, and basic mechanisms of carcinogenesis. Eur. J. Cancer Prev. 4, 45–59. Balteskard, L., Unneberg, K., Halvorsen, D., Hansen, J. B., and Revhaug, A. (1998). Effects of insulin-like growth factor on neutrophil and monocyte functions in normal and septic states. J. Parenteral Enteral Nutrition 22, 127–135. Beraud, C., Henzel, W. J., and Baeuerle, P. A. (1999). Involvement of regulatory and catalytic subunits of phosphoinositide 3-kinase in NF-κB activation. Proc. Natl. Acad. Sci. USA 96, 429–434. Berger, E. A., Murphy, P. M., and Farber, J. M. (1999). Chemokine receptors as HIV-1 coreceptors: roles in viral entry, tropism, and disease. Annu. Rev. Immunol. 17, 657–700. Bergers, G., Javaherian, K., Lo, K. M., Folkman, J., and Hanahan, D. (1999). Effects of angiogenesis inhibitors on multistage carcinogenesis in mice. Science 284, 808–812. Bergers, G., Javaherian, K., Lo, K. M., Folkman, J., and Hanahan, D. (1999). Effects of angiogenesis inhibitors on multistage carcinogenesis in mice. Science 284, 808–812. Bielefeldt-Ohmann, H., Jarnicki, A. G., and Fitzpatrick, D. R. (1996). Molecular pathobiology and immunology of malignant mesothelioma. J. Pathol. 178, 369–378. Bonnet, M., Guinebretiere, J. M., Kremmer, E., Grunewald, V., Benhamou, E., Contesso, G., and Joab, I. (1999). Detection of Epstein–Barr virus in invasive breast cancers. J. Natl. Cancer Inst. 91, 1376–1381. Bontkes, H. J., de Gruijl, T. D., Bijl, A., Verheijen, R. H., Meijer, C. J., Scheper, R. J., Stern, P. L., Burns, J. E., Maitland, N. J., and Walboomers, J. M. (1999a). Human papillomavirus type 16 E2-speciﬁc T-helper lymphocyte responses in patients with cervical intraepithelial neoplasia. J. Gen. Virol. 80, 2453–2459. Bontkes, H. J., de Gruijl, T. D., Walboomers, J. M., Schiller, J. T., Dillner, J., Helmerhorst, T. J., Verheijen, R. H., Scheper, R. J., and Meijer, C. J. (1999b). Immune responses against human papillomavirus (HPV) type 16 virus-like particles in a cohort study of women with cervical intraepithelial neoplasia. II. Systemic but not local IgA responses correlate with clearance of HPV-16. J. Gen. Virol. 80, 409–417. Browning, M. J., and Bodmer, W. F. (1992). MHC antigens and cancer: implications for T-cell surveillance. Curr. Opin. Immunol. 4, 613–618. Browning, M., and Dalgleish, A. G. (1996). Introduction and historical perspective. In “Tumour Immunology” (K. Sikora, A. G. Dalgleish, and M. Browning, eds.). Cambridge University Press, London. Buckman, S. Y., Gresham, A., Hale, P., Hruza, G., Anast, J., Masferrer, J., and Pentland, A. P. (1998). COX-2 expression is induced by UVB exposure in human skin: implications for the development of skin cancer. Carcinogenesis 19, 723–729. Butt, A. J., Firth, S. M., King, M. A., and Baxter, R. C. (2000). Insulin-like growth factorbinding protein-3 modulates expression of Bax and Bcl-2 and potentiates p53-independent radiation-induced apoptosis in human breast cancer cells. J. Biol. Chem. 275, 39174–39181. Cacciarelli, T. V., Martinez, O. M., Gish, R. G., Villaneuva, J. C., and Krams, S. M. (1996). Immunoregulatory cytokines in chronic hepatitis C virus infection: pre- and posttreatment with interferon alpha. Hepatology 24, 6–9. Cainzos, M. (1989). Anergy in patients with gastric cancer. Hepatogastroenterology 36, 36–39. Campbell, R. A., Bhat-Nakshatri, P., Patel, N. M., Constantinidou, D., Ali, S., and Nakshatri, H. (2001). Phosphatidylinositol 3-kinase/AKT-mediated activation of estrogen receptor alpha: a new model for estrogen resistance. J. Biol. Chem. 276, 9817–9824. Candotti, D., Costagliola, D., Joberty, C., Bonduelle, O., Rouzioux, C., Autran, B., and Agut, H. (1999). Status of long-term asymptomatic HIV-1 infection correlates with viral load but not with virus replication properties and cell tropism. French ALT Study Group. J. Med. Virol. 58, 256–263.

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

261

Carbone, M. (1999). Simian virus 40 and human tumors: It is time to study mechanisms. J. Cell. Biochem. 76, 189–193. Carbone, M., Fisher, S., Powers, A., Pass, H. I., and Rizzo, P. (1999). New molecular and epidemiological issues in mesothelioma role of SV40. J. Cell. Physiol. 180, 167–172. Cardone, M. H., Roy, N., Stennicke, H. R., Salvesen, G. S., Franke, T. F., Stanbridge, E., Frisch, S., and Reed, J. C. (1998). Regulation of cell death protease caspase-9 by phosphorylation. Science 282, 1318–1321. Carreno, V., and Quiroga, J. A. (1997). Biological properties of interleukin-12 and its therapeutic use in persistent hepatitis B virus and hepatitis C virus infection. J. Viral Hepatitis 4(Suppl. 2), 83–86. Cast, et al. (1989). P. 9. Castelao, J. E., Yuan, J. M., Gago-Dominguez, M., Yu, M. C., and Ross, R. K. (2000). Nonsteroidal anti-inﬂammatory drugs and bladder cancer prevention. Br. J. Cancer 82, 1364– 1369. Cerqueira, E. M., Santoro, C. L., Donozo, N. F., Freitas, B. A., Pereira, C. A., Bevilacqua, R. G., and Machado-Santelli, G. M. (1998). Genetic damage in exfoliated cells of the uterine cervix. Association and interaction between cigarette smoking and progression to malignant transformation? Acta Cytol. 42, 639–649. Chan, J. M., Stampfer, M. J., Giovannucci, E., Gann, P. H., Ma, J., Wilkinson, P., Hennekens, C. H., and Pollak, M. (1998). Plasma insulin-like growth factor-I and prostate cancer risk: a prospective study. Science 279, 563–566. Chan, W. L., Rodgers, A., Hancock, R. D., Taffs, F., Kitchin, P., Farrar, G., and Liew, F. Y. (1992). Protection in simian immunodeﬁciency virus-vaccinated monkeys correlates with anti-HLA class I antibody response. J. Exp. Med. 176, 1203–1207. Chang, Y. E., and Laimins, L. A. (2000). Microarray analysis identiﬁes interferon-inducible genes and Stat-1 as major transcriptional targets of human papillomavirus type 31. J. Virol. 74, 4174–4182. Chen, Y. M., Yang, W. K., Whang-Peng, J., Tsai, W. Y., Hung, Y. M., Yang, D. M., Lin, W. C., Perng, R. P., and Ting, C. C. (1997a). Restoration of the immunocompetence by IL-2 activation and TCR-CD3 engagement of the in vivo anergized tumor-speciﬁc CTL from lung cancer patients. J. Immunother. 20, 354–364. Chen, Y. M., Yang, W. K., Ting, C. C., Yang, D. M., Whang-Peng, J., and Perng, R. P. (1999). Depressed cytolytic activity of peripheral blood mononuclear cells in unusually high paclitaxel concentrations: reversal by IL-2 and IL-12. Chinese Medical Journal (Taipei) 62, 867–874. Chen, E. K., Gadina, M., Galon, J., Chen, M., and O’Shea, J. J. (1998). Not just another meeting: the coming of age of JAKs and STATs. Immunol. Today 19, 338–341. Chen, Y. M., Yang, W. K., Ting, C. C., Tsai, W. Y., Yang, D. M., Whang-Peng, J., and Perng, R. P. (1997b). Cross regulation by IL-10 and IL-2/IL-12 of the helper T cells and the cytolytic activity of lymphocytes from malignant effusions of lung cancer patients. Chest 112, 960– 966. Clavel, C., Masure, M., Levert, M., Putaud, I., Mangeonjean, C., Lorenzato, M., Nazeyrollas, P., Gabriel, R., Quereux, C., and Birembaut, P. (2000). Human papillomavirus detection by the hybrid capture II assay: a reliable test to select women with normal cervical smears at risk for developing cervical lesions. Diagn. Mol. Pathol. 9, 145–150. Clerici, M., and Shearer, G. M. (1994). The Th1-Th2 hypothesis of HIV infection: new insights. Immunol. Today 15, 575–581. Coffey, D. S. (1998). Self-organization, complexity and chaos: the new biology for medicine [see comments]. Nat. Med. 4, 882–885. Coffey, R. J., Hawkey, C. J., Damstrup, L., Graves-Deal, R., Daniel, V. C., Dempsey, P. J., Chinery, R., Kirkland, S. C., DuBois, R. N., Jetton, T. L., and Morrow, J. D. (1997). Epidermal growth factor receptor activation induces nuclear targeting of cyclooxygenase-2, basolateral

262

A. G. Dalgleish and K. J. O’Byrne

release of prostaglandins, and mitogenesis in polarizing colon cancer cells. Proc. Natl. Acad. Sci. USA 94, 657–662. Copeland, K. F., and Heeney, J. L. (1996). T helper cell activation and human retroviral pathogenesis. Microbiol. Rev. 60, 722–742. Cordon-Cardo, C., and Prives, C. (1999). At the crossroads of inﬂammation and tumorigenesis [comment]. J. Exp. Med. 190, 1367–1370. Cosma, A., Blanc, D., Braun, J., Quillent, C., Barassi, C., Moog, C., Klasen, S., Spire, B., Scarlatti, G., Pesenti, E., Siccardi, A. G., and Beretta, A. (1999). Enhanced HIV infectivity and changes in GP120 conformation associated with viral incorporation of human leucocyte antigen class I molecules. AIDS 13, 2033–2042. Dalgleish, A. G. (1991). Viruses and cancer. Br. Med. Bull. 47, 21–46. Dalgleish, A. G. (1992). The pathogenesis of AIDS: classical and alternative views. J. R. Coll. Physicians Lond. 26, 152–158. Dalgleish, A. (1999). The relevance of non-linear mathematics (chaos theory) to the treatment of cancer, the role of the immune response and the potential for vaccines. QJM 92, 347–359. Dalgleish, A. G., Beverley, P. C., Clapham, P. R., Crawford, D. H., Greaves, M. F., and Weiss, R. A. (1984). The CD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature 312, 763–767. Dalgleish, A., and Weiss, R. A. L. (1999). Human immunodeﬁciency viruses, In “Principles and Practice of Clinical Virology” Vol. 44. Wiley, New York. (Eds. Zuckeman, A., Banatvala, J. E., and Pattison, J. R.) Dameron, K. M., Volpert, O. V., Tainsky, M. A., and Bouck, N. (1994). Control of angiogenesis in ﬁbroblasts by p53 regulation of thrombospondin-I. Science 265, 1582–1584. Das, T., Sa, G., Sinha, P., and Ray, P. K. (1999). Induction of cell proliferation and apoptosis: dependence on the dose of the inducer. Biochem. Biophys. Res. Commun. 260, 105–110. de Fraipont, F., El Atiﬁ, M., Gicquel, C., Bertagna, X., Chambaz, E. M., and Feige, J. J. (2000). Expression of the angiogenesis markers vascular endothelial growth factor-A, thrombospondin-1, and platelet-derived endothelial cell growth factor in human sporadic adrenocortical tumors: correlation with genotypic alterations. J. Clin. Endocrinol. Metab. 85, 4734–4741. De Luca, A., Baldi, A., Esposito, V., Howard, C. M., Bagella, L., Rizzo, P., Caputi, M., Pass, H. I., Giordano, G. G., Baldi, F., Carbone, M., and Giordano, A. (1997). The retinoblastoma gene family pRb/p105, p107, pRb2/p130 and simian virus-40 large T-antigen in human mesotheliomas. Nat. Med. 3, 913–916. De Marzo, A. M., Coffey, D. S., and Nelson, W. G. (1999). New concepts in tissue speciﬁcity for prostate cancer and benign prostatic hyperplasia. Urology 53, 29–39; discussion 39–42. De The, G. (1993). The etiology of Burkitt’s lymphoma and the history of shaken dogmas. Blood Cells 19, 667–675. del Peso, L., Gonzalez-Garcia, M., Page, C., Herrera, R., and Nunez, G. (1997). Interleukin-3induced phosphorylation of BAD through the protein kinase Akt. Science 278, 687–689. Della Bella, S., Molteni, M., Compasso, S., Zulian, C., Vanoli, M., and Scorza, R. (1997). Differential effects of cyclo-oxygenase pathway metabolites on cytokine production by T lymphocytes. Prostaglandins Leukot. Essent. Fatty Acids 56, 177–184. Diaz-Flores, L., Gutierrez, R., Valladares, F., Varela, H., and Perez, M. (1994). Intense vascular sprouting from rat femoral vein induced by prostaglandins E1 and E2. Anat. Rec. 238, 68–76. Di Popolo, A., Memoli, A., Apicella, A., Tuccillo, C., di Palma, A., Ricchi, P., Acquaviva, A. M., and Zarrilli, R. (2000). IGF-II/IGF-I receptor pathway up-regulates COX-2 mRNA expression and PGE2 synthesis in Caco-2 human colon carcinoma cells. Oncogene 19, 5517– 5524. Doherty, P. C., Tripp, R. A., and Sixbey, J. W. (1994). Evasion of host immune responses by tumours and viruses. Ciba Found. Symp. 187, 245–256.

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

263

D’Orazio, T. J., and Niederkorn, J. Y. (1998). A novel role for TGF-beta and IL-10 in the induction of immune privilege. J. Immunol. 160, 2089–2098. Donovan, D., Harney, J. H., Toomey, D., Osborne, D. H., Redmond, H. P., and BouchierHayes, D. J. (1997). TGF beta-1 regulation of VEGF production by breast cancer cells. Ann. Surg. Oncol. 4, 621–627. Dubois, R. N., Abramson, S. B., Crofford, L., Gupta, R. A., Simon, L. S., Van D. Putte, L. B., and Lipsky, P. E. (1998). Cycloxygenase in biology and disease. FASEB J. 12, 1063– 1073. Dunn, S. E. (2000). Insulin-like growth factor I stimulates angiogenesis and the production of vascular endothelial growth factor. Growth Horm. IGF Res. 10, Suppl A, S41–S42. Edwards, J. G., Abrams, K. R., Leverment, J. N., Spyt, T. J., Waller, D. A., and O’Byrne, K. J. (2000). Prognostic factors for malignant mesothelioma in 142 patients: validation of CALGB and EORTC prognostic scoring systems. Thorax 55, 731–735. El-Assal, O. N., Yamanoi, A., Soda, Y., et al. (1998). Clinical signiﬁcance of microvessel density and vascular endothelial growth factor expression in hepatocellular carcinoma and surrounding liver: possible involvement of vascular endothelial growth factor in the angiogenesis of cirrhotic liver. Hepatology 27, 1554–1562. El-Bayoumy, K., Iatropoulos, M., Amin, S., Hoffmann, D., and Wynder, E. L. (1999). Increased expression of cyclooxygenase-2 in rat lung tumors induced by the tobacco-speciﬁc nitrosamine 4-(methylnitrosamino)-4-(3-pyridyl)-1-butanone: the impact of a high-fat diet. Cancer Res. 59, 1400–1403. Eliopoulos, A. G., Gallagher, N. J., Blake, S. M., Dawson, C. W., and Young, L. S. (1999). Activation of the p38 mitogen-activated protein kinase pathway by Epstein–Barr virus-encoded latent membrane protein 1 coregulates interleukin-6 and interleukin-8 production. J. Biol. Chem. 274, 16085–16096. Ellouk-Achard, S., Djenabi, S., De Olivieri, G. A., Desauty, G., Duc, H. T., Zohair, M., Trojan, J., Claude, J. R., Sarasin, A., and Lafarge-Frayssinet, C. (1998). Induction of apoptosis in rat hepatocarcinoma cells by expression of IGF-I antisense c-DNA. J. Hepatol. 29, 807–818. Faux, S. P., Houghton, C. E., Hubbard, A., and Patrick, G. (2000). Increased expression of epidermal growth factor receptor in rat pleural mesothelial cells correlates with carcinogenicity of mineral ﬁbres. Carcinogenesis 21, 2275–2280. Faux, S. P., Houghton, C. E., Swain, W. A., Edwards, J. G., Sharma, R. A., Plummer, S. M., and O’Byrne, K. J. (2001). EGFR induced activation of NF-κB in mesothelial cells by asbestos is important in cell survival. Proc. Am. Assoc. Cancer Res. abstract 1315. Ferre, F., Moss, R. B., Daigle, A., Richieri, S. P., Jensen, F., and Carlo, D. J. (1995). Viral load in peripheral blood mononuclear cells as surrogate for clinical progression. J. Acquir. Immune Deﬁc. Syndr. Hum. Retrovirol. 10, S51–S56. Fischer, S. M., Lo, H. H., Gordon, G. B., Seibert, K., Kelloff, G., Lubet, R. A., and Conti, C. J. (1999). Chemopreventive activity of celecoxib, a speciﬁc cyclooxygenase-2 inhibitor, and indomethacin against ultraviolet light-induced skin carcinogenesis. Mol. Carcinog. 25, 231–240. Fisseler-Eckhoff, A., Rothstein, D., and Muller, K. M. (1996). Neovascularization in hyperplastic, metaplastic, and potentially preneoplastic lesions of the bronchial mucosa. Virchows Arch. 429, 95–100. Fleming, S. (1999). Renal cancer genetics: von Hippel Lindau and other syndromes. Int. J. Dev. Biol. 43, 469–471. Folkman, J. (1995). Seminars in Medicine of the Beth Israel Hospital, Boston. Clinical applications of research on angiogenesis [see comments]. N. Engl. J. Med. 333, 1757–1763. Fontanini, G., Vignati, S., Bigini, D., Lucchi, M., Mussi, A., Basolo, F., Angeletti, C. A., and Bevilacqua, G. (1996). Neoangiogenesis: a putative marker of malignancy in non-small cell lung cancer (NSCLC) development. Int. J. Cancer 67, 615–619.

264

A. G. Dalgleish and K. J. O’Byrne

Forni, G., and Foa, R. (1996). The role of cytokines in tumor rejection, in “Tumour Immunology: immunotherapy and cancer vaccines” (A. G. Dalgleish, and M. Browning, eds.), pp. 199– 218. Cambridge University Press. Franchimont, N., Gangji, V., Durant, D., and Canalis, E. (1997). Interleukin-6 with its soluble receptor enhances the expression of insulin-like growth factor-I in osteoblasts. Endocrinology 138, 5248–5255. Fultz, P. N., Gluckman, J. C., Muchmore, E., and Girard, M. (1992). Transient increases in numbers of infectious cells in an HIV-infected chimpanzee following immune stimulation. AIDS Res. Hum. Retroviruses 8, 313–317. Funkhouser, E. M., and Sharp, G. B. (1995). Aspirin and reduced risk of esophageal carcinoma. Cancer 76, 1116–1119. Gabrilovich, D. I., Ishida, T., Nadaf, S., Ohm, J. E., and Carbone, D. P. (1999). Antibodies to vascular endothelial growth factor enhance the efﬁcacy of cancer immunotherapy by improving endogenous dendritic cell function. Clin. Cancer Res. 5, 2963–2970. Gallucci, R. M., Simeonova, P. P., Matheson, J. M., Kommineni, C., Guriel, J. L., Sugawara, T., and Luster, M. I. (2000). Impaired cutaneous wound healing in interleukin-6-deﬁcient and immunosuppressed mice. FASEB J. 14, 2525–2531. Ganss, R., and Hanahan, D. (1998). Tumor microenvironment can restrict the effectiveness of activated antitumor lymphocytes. Cancer Res. 58, 4673–4681. Garrido, F., Cabrera, T., Concha, A., Glew, S., Ruiz-Cabello, F., and Stern, P. L. (1993). Natural history of HLA expression during tumor development. Immunol. Today 14, 491–499. Giovannucci, E., Egan, K. M., Hunter, D. J., Stampfer, M. J., Colditz, G. A., Willet, W. C., and Speizer, F. E. (1995). Aspirin and the risk of colorectal cancer in women. N. Engl. J. Med. 333, 609–614. Giovannucci, E., Rimm, E. B., Stampfer, M. J., Colditz, G. A., Ascherio, A., and Willett, W. C. (1994). Aspirin use and the risk for colorectal cancer and adenoma in male health professionals [see comments]. Ann. Intern. Med. 121, 241–246. Gjertsen, M. K., Bjorheim, J., Saeterdal, I., Myklebust, J., and Gaudernack, G. (1997). Cytotoxic CD4+ and CD8+ T lymphocytes, generated by mutant p21-ras (12Val) peptide vaccination of a patient, recognize 12Val-dependent nested epitopes present within the vaccine peptide and kill autologous tumour cells carrying this mutation. Int. J. Cancer 72, 784–790. Glinghammar, B., and Rafter, J. (2001). Colonic luminal contents induce cyclooxygenase 2 transcription in human colon carcinoma cells. Gastroenterology 120, 401–410. Goldstein, S., Ourmanov, I., Brown, C. R., Beer, B. E., Elkins, W. R., Plishka, R., BucklerWhite, A., and Hirsch, V. M. (2000). Wide range of viral load in healthy African green monkeys naturally infected with simian immunodeﬁciency virus. J. Virol. 74, 11744–11753. Goodglick, L., Zevit, N., Neshat, M. S., and Braun, J. (1995). Mapping the Ig superantigenbinding site of HIV-1 gp120. J. Immunol. 155, 5151–5159. Gorter, A., and Meri, S. (1999). Immune evasion of tumor cells using membrane-bound complement regulatory proteins. Immunol. Today 20, 576–582. Graham, N. M. (1996). The role of immunologic and viral markers in predicting clinical outcome in HIV infection. AIDS 10, Suppl 5, S21–S25. Grange, J. M., Stanford, J. L., and Rook, G. A. (1995). Tuberculosis and cancer: parallels in host responses and therapeutic approaches? Lancet 345, 1350–1352. Greenberg, P. D. (1991). Adoptive T cell therapy of tumors: mechanisms operative in the recognition and elimination of tumor cells. Adv. Immunol. 49, 281–355. Grinwich, K. D., and Plescia, O. J. (1977). Tumor-mediated immunosuppression: prevention by inhibitors of prostaglandin synthesis. Prostaglandins 14, 1175–1182. Guan, Z., Buckman, S. Y., Baier, L. D., and Morrison, A. R. (1998). IGF-I and insulin amplify IL-1 beta-induced nitric oxide and prostaglandin biosynthesis. Am. J. Physiol. 274, F673– F679.

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

265

Guillou, L., Estreicher, A., Chaubert, P., Hurlimann, J., Kurt, A. M., Metthez, G., Iggo, R., Gray, A. C., Jichlinski, P., Leisinger, H. J., and Benhattar, J. (1996). Germ cell tumors of the testis overexpress wild-type p53. Am. J. Pathol. 149, 1221–1228. Habeshaw, J. A. (1994). HLA mimicry by HIV-1 gp120 in the pathogenesis of AIDS. Immunol. Today 15, 39–40. Habeshaw, J., Hounsell, E., and Dalgleish, A. (1992). Does the HIV envelope induce a chronic graft-versus-host-like disease? Immunol. Today 13, 207–210. Hankinson, S. E., Willett, W. C., Colditz, G. A., Hunter, D. J., Michaud, D. S., Deroo, B., Rosner, B., Speizer, F. E., and Pollak, M. (1998). Circulating concentrations of insulin-like growth factor-I and risk of breast cancer. Lancet 351, 1393–1396. Heeney, J., Bogers, W., Buijs, L., Dubbes, R., ten Haaft, P., Koornstra, W., Niphuis, H., Nara, P., and Teeuwsen, V. (1996). Immune strategies utilized by lentivirus-infected chimpanzees to resist progression to AIDS. Immunol Lett. 51, 45–52. Held, W., Acha-Orbea, H., MacDonald, H. R., and Waanders, G. A. (1994). Superantigens and retroviral infection: insights from mouse mammary tumor virus. Immunol. Today 15, 184–190. Heriot, A. G., Marriott, J. B., Cookson, S., Kumar, D., and Dalgleish, A. G. (2000). Reduction in cytokine production in colorectal cancer patients: association with stage and reversal by resection. Br. J. Cancer 82, 1009–1012. Heeschen, C., Jang, J. J., Weis, M., Pathak, A., Kaji, S., Hu, R. S., Tsao, P. S., Johnson, F. L., and Cooke, J. P. (2001). Nicotine stimulates angiogenesis and promotes tumor growth and atherosclerosis. Nat. Med. 7, 775–777. Hida, T., Leyton, J., Makheja, A. N., Ben-Av, P., Hla, T., Martinez, A., Mulshine, J., Malkani, S., Chung, P., and Moody, T. W. (1998). Non-small cell lung cancer cycloxygenase activity and proliferation are inhibited by non-steroidal antiinﬂammatory drugs. Anticancer Res. 18, 775–782. Higaki, S., Akazawa, A., Nakamura, H., Yanai, H., Yoshida, T., and Okita, K. (1999). Metaplastic polyp of the colon develops in response to inﬂammation [published erratum appears in J. Gastroenterol. Hepatol. (1999), 14(8), 832]. J. Gastroenterol. Hepatol. 14, 709–714. Hildenbrand, R., Jansen, C., Wolf, G., Bohme, B., Berger, S., von Minckwitz, G., Horlin, A., Kaufmann, M., and Stutte, H. J. (1998). Transforming growth factor-beta stimulates urokinase expression in tumor-associated macrophages of the breast. Lab. Invest. 78, 59–71. Hildesheim, A., Schiffman, M. H., Tsukui, T., Swanson, C. A., Lucci, J. III, Scott, D. R., Glass, A. G., Rush, B. B., Lorincz, A. T., Corrigan, A., Burk, R. D., Helgesen, K., Houghten, R. A., Sherman, M. E., Kurman, R. J., Berzofsky, J. A., and Kramer, T. R. (1997). Immune activation in cervical neoplasia: cross-sectional association between plasma coluble interleukin 2 receptor levels and disease. Cancer Epidermiol. Biomarkers Prev. 6, 807–813. Hildesheim, A., Schiffman, M. H., Tsukui, T., Swanson, C. A., Lucci, J. III., Scott, D. R., Glass, A. G., Rush, B. B., Lorincz, A. T., Corrigan, A., Burk, R. D., Helgesen, K., Houghten, R. A., Sherman, M. E., Kurman, R. J., Berzofsky, J. A., and Kramer, T. R. (1997). Immune activation in cervical neoplasia: cross-sectional association between plasma soluble interleukin 2 receptor levels and disease. Cancer Epidemiol. Biomarkers Prevention 6, 807–813. Hojo, M., Morimoto, T., Maluccio, M., Asano, T., Morimoto, K., Lagman, M., Shimbo, T., and Suthanthiran, M. (1999). Cyclosporine induces cancer progression by a cell-autonomous mechanism. Nature 397, 530–534. Holterman, L., Niphuis, H., Koornstra, W., Dubbes, R., ten Haaft, P., and Heeney, J. L. (2000). The rate of progression to AIDS is independent of virus dose in simian immunodeﬁciency virus-infected macaques. J. Gen. Virol. 81 Pt 7, 1719–1726. Hong, S. H., Ondrey, F. G., Avis, I. M., Chen, Z., Loukinova, E., Cavanaugh, P. F., Jr., Van Waes, C., and Mulshine, J. L. (2000). Cyclooxygenase regulates human oropharyngeal carcinomas via the proinﬂammatory cytokine IL-6: a general role for inﬂammation? FASEB J. 14, 1499– 1507.

266

A. G. Dalgleish and K. J. O’Byrne

Hopﬂ, R., Heim, K., Christensen, N., Zumbach, K., Wieland, U., Volgger, B., Widschwendter, A., Haimbuchner, S., Muller-Holzner, E., Pawlita, M., Pﬁster, H., and Fritsch, P. (2000). Spontaneous regression of CIN and delayed-type hypersensitivity to HPV- 16 oncoprotein E7. Lancet 356, 1985–1986. Horikawa, T., Yoshizaki, T., Sheen, T. S., Lee, S. Y., and Furukawa, M. (2000). Association of latent membrane protein 1 and matrix metalloproteinase 9 with metastasis in nasopharyngeal carcinoma. Cancer 89, 715–723. Hounsell, E. F., Renouf, D. V., Liney, D., Dalgleish, A. G., and Habeshaw, J. (1991). A proposed molecular model for the carboxy terminus of HIV-1 gp120 showing structural features consistent with the presence of a T-cell alloepitope. Mol. Aspects Med. 12, 283–296. Hrouda, D., Baban, B., Dunsmuir, W. D., Kirby, R. S., and Dalgleish, A. G. (1998). Immunotherapy of advanced prostate cancer: a phase I/II trial using mycobacterium vaccae. Br. J. Urol. 82, 568–573. Hsieh, C. Y., You, S. L., Kao, C. L., and Chen, C. J. (1999). Reproductive and infectious risk factors for invasive cervical cancer in Taiwan. Anticancer Res. 19, 4495–4500. Hu, H. M., Urba, W. J., and Fox, B. A. (1998). Gene-modiﬁed tumor vaccine with therapeutic potential shifts tumor-speciﬁc T cell response from a type 2 to a type 1 cytokine proﬁle. J. Immunol. 16, 3033–3041. Huang, M., Stolina, M., Sharma, S., Mao, J. T., Zhu, L., Miller, P. W., Wollman, J., Herschman, H., and Dubinett, S. M. (1998). Non-small cell lung cancer cyclooxygenase-2-dependent regulation of cytokine balance in lymphocytes and macrophages: up-regulation of interleukin 10 and down-regulation of interleukin 12 production. Cancer Res. 58, 1208–1216. Huang, R. P., Wu, J. X., Fan, Y., and Adamson, E. D. (1996). UV activates growth factor receptors via reactive oxygen intermediates. J. Cell. Biol. 133, 211–220. Huang, Y. T., Sheen, T. S., Chen, C. L., Lu, J., Chang, Y., Chen, J. Y., and Tsai, C. H. (1999). Proﬁle of cytokine expression in nasopharyngeal carcinomas: a distinct expression of interleukin 1 in tumor and CD4+ T cells. Cancer Res. 59, 1599–1605. Hudson, J. D., Shoaibi, M. A., Maestro, R., Carnero, A., Hannon, G. J., and Beach, D. H. (1999). A proinﬂammatory cytokine inhibits p53 tumor suppressor activity. J. Exp. Med. 190, 1375–1382. Imberti, L., Sottini, A., Bettinardi, A., Puoti, M., and Primi, D. (1991). Selective depletion in HIV infection of T cells that bear speciﬁc T cell receptor V beta sequences. Science 254, 860–862. Imperial, J. C. (1999). Natural history of chronic hepatitis B and C. J. Gastroenterol. Hepatol. 14 Suppl, S1–S5. International Interferon Hepatocellular Carcinoma Study Group (1998). Effect of interferon-α on progression of cirrhosis to hepatocellular carcinoma: a retrospective cohort study. Lancet 351, 1535–1539. Irani, J., Goujon, J. M., Ragni, E., Peyrat, L., Hubert, J., Saint, F., and Mottet, N. (1999). Highgrade inﬂammation in prostate cancer as a prognostic factor for biochemical recurrence after radical prostatectomy. Pathologist Multi Center Study Group. Urology 54, 467–472. Jacobs, N., Giannini, S. L., Doyen, J., Baptista, A., Moutschen, M., Boniver, J., and Delvenne, P. (1998). Inverse modulation of IL-10 and IL-12 in the blood of women with preneoplastic lesions of the uterine cervix. Clin. Exp. Immunol. 111, 219–224. Jankowski, J. A., Wright, N. A., Meltzer, S. J., Triadaﬁlopoulos, G., Geboes, K., Casson, A. G., Kerr, D., and Young, L. S. (1999). Molecular evolution of the metaplasia–dysplasia– adenocarcinoma sequence in the esophagus. Am. J. Pathol. 154, 965–973. Janssen, Y. M., Driscoll, K. E., Howard, B., Quinlan, T. R., Treadwell, M., Barchowsky, A., and Mossman, B. T. (1997). Asbestos causes translocation of p65 protein and increases NF-kappa B DNA binding activity in rat lung epithelial and pleural mesothelial cells. Am. J. Pathol. 151, 389–401.

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

267

Janssen, Y. M., Driscoll, K. E., Howard, B., Quinlan, T. R., Treadwell, M., Barchowsky, A., and Mossman, B. T. (2000). Asbestos causes translocation of p65 protein and increases NF-kappaB DNA binding activity in rat lung epithelial and pleural mesothelial cells. Am. J. Pathol. 151, 389–401. Jarnicki, A. G., Fitzpatrick, D. R., Robinson, B. W., and Bielefeldt-Ohmann, H. (1996). Altered CD3 chain and cytokine gene expression in tumor inﬁltrating T lymphocytes during the development of mesothelioma. Cancer Lett. 15, 1–9. Jinno, K., Tanimizu, M., Hyodo, I., Nishikawa, Y., Hosokawa, Y., Doi, T., Endo, H., Yamashita, T., and Okada, Y. (1998). Circulating vascular endothelial growth factor (VEGF) is a possible tumor marker for metastasis in human hepatocellular carcinoma. J. Gastroenterol. 33, 376– 382. Johnston-Early, A., Cohen, M. H., and Fossieck, B. E., Jr. (1983). Delayed hypersensitivity skin testing as a prognostic indicator in patients with small cell lung cancer. Cancer 52, 1395–1400. Jost, M., Huggett, T. M., Kari, C., Boise, L. H., and Rodeck, U. (2001). Epidermal growth factor receptor-dependent control of keratinocyte survival and Bcl-xL expression through a MEK-dependent pathway. J. Biol. Chem. 276, 6320–6326. Kakumu, S., Okumura, A., Ishikawa, T., Iwata, K., Yano, M., and Yoshioka, K. (1997). Production of interleukins 10 and 12 by peripheral blood mononuclear cells (PBMC) in chronic hepatitis C virus (HCV) infection. Clin. Exp. Immunol. 108, 138–143. Kano, A., Haruyama, T., Akaike, T., and Watanabe, Y. (1999). IRF-1 is an essential mediator in IFN-gamma-induced cell cycle arrest and apoptosis of primary cultured hepatocytes. Biochem. Biophys. Res. Commun. 257, 672–677. Kaplan, D. H., Shankaran, V., Dighe, A. S., Stockert, E., Aguet, M., Old, L. J., and Schreiber, R. D. (1998). Demonstration of an interferon gamma-dependent tumor surveillance system in immunocompetent mice. Proc. Natl. Acad. Sci. USA 95, 7556–7561. Kast, W. M., Offringa, R., Peters, P. J., Voordouw, A. C., Meloen, R. H., van de Eb, A. J., and Melief, C. J. (1989). Eradication of adenovirus E1-induced tumors by E1A-speciﬁc cytotoxic T lymphocytes. Cell 59, 603–614. Kaur, A., Grant, R. M., Means, R. E., McClure, H., Feinberg, M., and Johnson, R. P. (1998). Diverse host responses and outcomes following simian immunodeﬁciency virus SIVmac239 infection in sooty mangabeys and rhesus macaques. J. Virol. 72, 9597–9611. Keith, R. L., Miller, Y. E., Gemmill, R. M., Drabkin, H. A., Dempsey, E. C., Kennedy, T. C., Prindiville, S., and Franklin, W. A. (2000). Angiogenic squamous dysplasia in bronchi of individuals at high risk for lung cancer. Clin. Cancer Res. 6, 1616–1625. Kieser, A., Weich, H. A., Brandner, G., Marme, D., and Kolch, W. (1994). Mutant p53 potentiates protein kinase C induction of vascular endothelial growth factor expression. Oncogene 9, 963–969. Kiessling, R., Pawelec, G., Welsh, R. M., Barry, J. D., and Ferrone, S. (2000). Have tumor cells learnt from microorganisms how to fool the immune system? Escape from immune surveillance of tumors and microorganisms: emerging mechanisms and shared strategies. Mol. Med. Today 6, 344–346. Kilby, J. M., Tabereaux, P. B., Mulanovich, V., Shaw, G. M., Bucy, R. P., and Saag, M. S. (1997). Effects of tapering doses of oral prednisone on viral load among HIV-infected patients with unexplained weight loss. AIDS Res. Hum. Retroviruses 13, 1533–1537. Kirk, G. R., and Clements, W. D. (1999). Crohn’s disease and colorectal malignancy. Int. J. Clin. Pract. 53, 314–315. Kitagawa, N., Goto, M., Kurozumi, K., Maruo, S., Fukayama, M., Naoe, T., Yasukawa, M., Hino, K., Suzuki, T., Todo, S., and Takada, K. (2000). Epstein–Barr virus-encoded poly(A)(−) RNA supports Burkitt’s lymphoma growth through interleukin-10 induction. EMBO J. 19, 6742–6750.

268

A. G. Dalgleish and K. J. O’Byrne

Klatzmann, D., Champagne, E., Chamaret, S., Gruest, J., Guetard, D., Hercend, T., Gluckman, J. C., and Montagnier, L. (1984). T-lymphocyte T4 molecule behaves as the receptor for human retrovirus LAV. Nature 312, 767–768. Kodelja, V., Muller, C., Tenorio, S., Schebesch, C., Orfanos, C. E., and Goerdt, S. (1997). Differences in angiogenic potential of classically vs alternatively activated macrophages. Immunobiology 197, 478–493. Konturek, P. C., Hartwich, A., Zuchowicz, M., Labza, H., Pierzchalski, P., Karczewska, E., Bielanski, W., Hahn, E. G., and Konturek, S. J. (2000). Helicobacter pylori, gastrin and cyclooxygenases in gastric cancer. J. Physiol. Pharmacol. 51, 737–749. Koshiba, T., Hosotani, R., Miyamoto, Y., Wada, M., Lee, J. U., Fujimoto, K., Tsuji, S., Nakajima, S., Doi, R., and Imamura, M. (1999). Immunohistochemical analysis of cyclooxygenase-2 expression in pancreatic tumors. Int. J. Pancreatol. 26, 69–76. Kwong, P. D., Wyatt, R., Robinson, J., Sweet, R. W., Sodroski, J., and Hendrickson, W. A. (1998). Structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody. Nature 393, 648–659. Lafarge-Frayssinet, C., Duc, H. T., Frayssinet, C., Sarasin, A., Anthony, D., Guo, Y., and Trojan, J. (1997). Antisense insulin-like growth factor I transferred into a rat hepatoma cell line inhibits tumorigenesis by modulating major histocompatibility complex I cell surface expression. Cancer Gene Ther. 4, 276–285. Lala, P. K., Al-Mutter, N., and Orucevic, A. (1997). Effects of chronic indomethacin therapy on the development and progression of spontaneous mammary tumors in C3H/HEJ mice. Int. J. Cancer 73, 371–380. Lang, J. M., Giron, C., de Leon, J., and Oberling, F. (1980). Delayed cutaneous hypersensitivity testing in untreated Hodgkin’s disease using a standardised new device. Biomedicine 33, 62–64. Langman, M. J., Cheng, K. K., Gilman, E. A., and Lancashire, R. J. (2000). Effect of antiinﬂammatory drugs on overall risk of common cancer: case-control study in general practice research database. BMJ 320, 1642–1646. Lawlor, M. A., and Rotwein, P. (2000). Insulin-like growth factor-mediated muscle cell survival: central roles for Akt and cyclin-dependent kinase inhibitor p21. Mol. Cell. Biol. 20, 8983– 8995. Le Buanec, H., D’Anna, R., Lachgar, A., Zagury, J. F., Bernard, J., Ittele, D., d’Alessio, P., Hallez, S., Giannouli, C., Burny, A., Bizzini, B., Gallo, R. C., and Zagury, D. (1999). HPV-16 E7 but not E6 oncogenic protein triggers both cellular immunosuppression and angiogenic processes. Biomed. Pharmacother. 53, 424–431. Lee, J., Fenton, B. M., Koch, C. J., Frelinger, J. G., and Lord, E. M. (1998). Interleukin 2 expression by tumor cells alters both the immune response and the tumor microenvironment. Cancer Res. 58, 1478–1485. Lee, P. P., Zeng, D., McCaulay, A. E., Chen, Y. F., Geiler, C., Umetsu, D. T., and Chao, N. J. (1997). T helper 2-dominant antilymphoma immune response is associated with fatal outcome. Blood 90, 1611–1617. Leelayuwat, C., Zhang, W. J., Abraham, L. J., Townend, D. C., Gaudieri, S., and Dawkins, R. L. (1993). Differences in the central major histocompatibility complex between humans and chimpanzees. Implications for development of autoimmunity and acquired immune deﬁciency syndrome. Hum. Immunol. 38, 30–41. Lengauer, C., Kinzler, K. W., and Vogelstein, B. (1998). Genetic instabilities in human cancers. Nature 396, 643–649. Lenz, J., Su, M., Mizrachi, Y., Burke, M., and Rubinstein, A. (2001). V3 variation in HIVseropositive patients receiving a V3-targeted vaccine. AIDS 15, 577–581. Leri, A., Liu, Y., Claudio, P. P., Kajstura, J., Wang, X., Wang, S., Kang, P., Malhotra, A., and Anversa, P. (1999). Insulin-like growth factor-1 induces Mdm2 and down-regulates

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

269

p53, attenuating the myocyte renin-angiotensin system and stretch-mediated apoptosis. Am. J. Pathol. 154, 567–580. Lewis, J. D., Deren, J. J., and Lichtenstein, G. R. (1999). Cancer risk in patients with inﬂammatory bowel disease. Gastroenterol. Clin. North Am. 28, 459–477. Liu, C. H., Chang, S. H., Narko, K., Trifan, O. C., Wu, M. T., Smith, E., Haudenschild, C., Lane, T. F., and Hla, T. (2001). Overexpression of cyclooxygenase-2 is sufﬁcient to induce tumorigenesis in transgenic mice. J. Biol. Chem. 276, 18563–18569. Liu, M. Y., Shih, Y. Y., Li, L. Y., Chou, S. P., Sheen, T. S., Chen, C. L., Yang, C. S., and Chen, J. Y. (2000). Expression of the Epstein–Barr virus BHRF1 gene, a homologue of Bcl-2, in nasopharyngeal carcinoma tissue. J. Med. Virol. 61, 241–250. Liu, Z., Cumberland, W. G., Hultin, L. E., Prince, H. E., Detels, R., and Giorgi, J. V. (1997). Elevated CD38 antigen expression on CD8+ T cells is a stronger marker for the risk of chronic HIV disease progression to AIDS and death in the Multicenter AIDS Cohort Study than CD4+ cell count, soluble immune activation markers, or combinations of HLA-DR and CD38 expression. J. Acquir. Immune Deﬁc. Syndr. Hum. Retrovirol. 16, 83–92. Luca, M., Huang, S., Gershenwald, J. E., Singh, R. K., Reich, R., and Bar-Eli, M. (1997). Expression of interleukin-8 by human melanoma cells up-regulates MMP-2 activity and increases tumor growth and metastasis. Am. J. Pathol. 151, 1105–1113. Luxton, J. C., Rose, R. C., Coletart, T., Wilson, P., and Shepherd, P. S. (1997). Serological and T-helper cell responses to human papillomavirus type 16 L1 in women with cervical dysplasia or cervical carcinoma and in healthy controls. J. Gen. Virol. 78, 917–923. Ma, J., Pollak, M. N., Giovannucci, E., Chan, J. M., Tao, Y., Hennekens, C. H., and Stampfer, M. J. (1999). Prospective study of colorectal cancer risk in men and plasma levels of insulin-like growth factor (IGF)-I and IGF-binding protein-3. J. Natl. Cancer Inst. 91, 620– 625. Madrid, L. V., Mayo, M. W., Reuther, J. Y., and Baldwin, A. S., Jr. (2001). Akt stimulates the transactivation potential of the RelA/p65 subunit of NF-{kappa} B through utilization of the I{kappa}B kinase and activation of the mitogen activated protein kinase p38. J. Biol. Chem. 276(22), 18934–40. Majima, M., Isono, M., Ikeda, Y., Hayashi, I., Hatanaka, K., Harada, Y., Katsumata, O., Yamashima, S., Katori, M., and Yamamoto, S. (1997). Jpn. J. Pharmacol. 75, 105–114. Malaguarnera, M., Di Fazio, I., Laurino, A., Ferlito, L., Romano, M., and Trovato, B. A. (1997). Serum interleukin 6 concentrations in chronic hepatitis C patients before and after interferon-alpha treatment. Int. J. Clin. Pharmacol. Ther. 35, 385–388. Maraveyas, A., Baban, B., Kennard, D., Rook, G. A., Westby, M., Grange, J. M., Lydyard, P., Stanford, J. L., Jones, M., Selby, P., and Dalgleish, A. G. (1999). Possible improved survival of patients with stage IV AJCC melanoma receiving SRL 172 immunotherapy: correlation with induction of increased levels of intracellular interleukin-2 in peripheral blood lymphocytes. Ann. Oncol. 10, 817–824. Margolis, J., and Vogelsang, G. (2000). Chronic graft-versus-host disease. J. Hematother. Stem Cell Res. 9, 339–346. Markham, A. F. (1996). Carcinoma of the lung: warts and all. Thorax 51, 878–879. Mauray, S., Fuzzati-Armentero, M. T., Trouillet, P., Ruegg, M., Nicoloso, G., Hart, M., Aarden, L., Schapira, M., and Duchosal, M. A. (2000). Epstein–Barr virus-dependent lymphoproliferative disease: critical role of IL-6. Eur. J. Immunol. 30, 2065–2073. Mayne, S. T., Buenconsejo, J., and Janerich, D. T. (1999). Previous lung disease and risk of lung cancer among men and women nonsmokers. Am. J. Epidemiol. 149, 13–20. Mazzanti, R., Messerini, L., Monsacchi, L., Buzzelli, G., Zignego, A. L., Foschi, M., Monti, M., Lafﬁ, G., Morbidelli, L., Fantappic, O., Bartoloni Saint Omer, F., and Ziche, M. (1997). Chronic viral hepatitis induced by hepatitis C but not hepatitis B virus infection correlates with increased liver angiogenesis. Hepatology 25, 229–234.

270

A. G. Dalgleish and K. J. O’Byrne

McCann, J. (1999). Esophageal cancers: changing character, increasing incidence [news]. J. Natl. Cancer Inst. 91, 497–498. McGinty, A., Chang, Y. W., Sorokin, A., Bokemeyer, D., and Dunn, M. J. (2000). Cyclooxygenase-2 expression inhibits trophic withdrawal apoptosis in nerve growth factordifferentiated PC12 cells. J. Biol. Chem. 275, 12095–12101. McNeil, A. J., Yap, P. L., Gore, S. M., Brettle, R. P., McColl, M., Wyld, R., Davidson, S., Weightman, R., Richardson, A. M., and Robertson, J. R. (1996). Association of HLA types A1-B8-DR3 and B27 with rapid and slow progression of HIV disease. QJM 89, 177–185. Meert, K. L., Ofenstein, J. P., Genyea, C., Sarnaik, A. P., and Kaplan, J. (1996). Elevated transforming growth factor-beta concentration correlates with post-trauma immunosuppression. J. Trauma 40, 901–906. Melief, C. J., and Kast, W. M. (1991). Cytotoxic T lymphocyte therapy of cancer and tumor escape mechanisms. Semin. Cancer Biol. 2, 347–354. ˙ O., Shah, J. P., EdelMestre, J. R., Chan, G., Zhang, F., Yang, E. K., Sacks, P. G., Boyle, J. stein, D., Subbaramaiah, K., and Dannenberg, A. J. (1999). Inhibition of cyclooxygenase-2 expression. An approach to preventing head and neck cancer. Ann. NY Acad. Sci. 889, 62–71. Milich, D. R., Chen, M. K., Hughes, J. L., and Jones, J. E. (1998). The secreted hepatitis B precore antigen can modulate the immune response to the nucleocapsid: a mechanism for persistence. J. Immunol. 160, 2013–2021. Mizoguchi, H., O’Shea, J. J., Longo, D. L., Loefﬂer, C. M., McVicar, D. W., and Ochoa, A. C. (1992). Alterations in signal transduction molecules in T lymphocytes from tumor-bearing mice. Science 258, 1795–1798. Molina, M. A., Sitja-Arnau, M., Lemoine, M. G., Frazier, M. L., and Sinicrope, F. A. (1999). Increased cyclooxygenase-2 expression in human pancreatic carcinomas and cell lines: growth inhibition by nonsteroidal anti-inﬂammatory drugs. Cancer Res. 59, 4356–4362. Moller, H., Mellemgaard, A., Lindvig, K., and Olsen, J. H. (1994). Obesity and cancer risk: a Danish record-linkage study. Eur. J. Cancer 30A, 344–350. Monson, R. R., Yen, S., and MacMahon, B. (1976). Chronic mastitis and carcinoma of the breast. Lancet 2, 224–226. Moore, R. J., Owens, D. M., Stamp, G., Arnott, C., Burke, F., East, N., Holdsworth, H., Turner, L., Rollins, B., Pasparakis, M., Kollias, G., and Balkwill, F. (1999). Mice deﬁcient in tumor necrosis factor-alpha are resistant to skin carcinogenesis. Nat. Med. 5, 828–831. Moriguchi, S., Kato, M., Sakai, K., Yamamoto, S., and Shimizu, E. (1998). Exercise training restores decreased cellular immune functions in obese Zucker rats. J. Appl. Physiol. 84, 311–317. Moriguchi, S., Kato, M., Sakai, K., Yamamoto, S., and Shimizu, E. (1998). Decreased mitogen response of splenic lymphocytes in obese Zucker rats is associated with the decreased expression of glucose transporter 1 (GLUT-1). Am. J. Clin. Nutr. 67, 1124–1129. Mosmann, T. R., and Coffman, R. L. (1989). TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7, 145– 173. Murata, H., Kawano, S., Tsuji, S., Tsuji, M., Sawaoka, H., Kimura, Y., Shiozaki, H., and Hori, M. (1999). Cyclooxygenase-2 overexpression enhances lymphatic invasion and metastasis in human gastric carcinoma. Am. J. Gastroenterol. 94, 451–455. Nabel, G. J. (1999). A transformed view of cyclosporine. Nature 397, 471–472. Nakata, S., Imai, K., and Yamanaka, H. (1993). [Study of risk factors for prostatic cancer]. Hinyokika Kiyo 39, 1017–1024; discussion 1024–1025. Nees, M., Geoghegan, J. M., Hyman, T., Frank, S., Miller, L., and Woodworth, C. D. (2001). Papillomavirus type 16 oncogenes downregulate expression of interferon-responsive genes and upregulate proliferation-associated and NF-kappaB-responsive genes in cervical keratinocytes. J. Virol. 75, 4283–4296.

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

271

Norcross, M. (1999). Chemokine receptors and HIV-1 pathogenesis. In “HIV and the New Viruses” (A. W. Dalgleish, ed.), pp. 133–153. Academic Press, London. O’Byrne, K. J., Dalgleish, A. G., Browning, M. J., Steward, W. P., and Harris, A. L. (2000a). The relationship between angiogenesis and the immune response in carcinogenesis and the progression of malignant disease. Eur. J. Cancer 36, 151–169. O’Byrne, K. J., Koukourakis, M. I., Giatromanolaki, A., Cox, G., Turley, H., Steward, W. P., Gatter, K., and Harris, A. L. (2000b). Vascular endothelial growth factor, platelet-derived endothelial cell growth factor and angiogenesis in non-small cell lung cancer. Br. J. Cancer 82, 1427–1432. O’Hara, R. J., Greenman, J., MacDonald, A. W., Gaskell, K. M., Topping, K. P., Duthie, G. S., Kerin, M. J., Lee, P. W., and Monson, J. R. (1998). Advanced colorectal cancer is associated with impaired interleukin 12 and enhanced interleukin 10 production. Clin. Cancer Res. 4, 1943–1948. Oka, M., Attwood, S. E., Kaul, B., Smyrk, T. C., and De Meester, T. R. (1992). Immunosuppression in patients with Barrett’s esophagus. Surgery 112, 11–17. Olaitan, A., Johnson, M. A., Reid, W. M., and Poulter, L. W. (1998). Changes to the cytokine microenvironment in the genital tract mucosa of HIV+ women. Clin. Exp. Immunol. 112, 100–104. Paganini-Hill, A. (1994). Aspirin and the prevention of colorectal cancer: a review of the evidence. Semin. Surg. Oncol. 10, 158–164. Pamphilon, D. H., Alnaqdy, A. A., and Wallington, T. B. (1991). Immunomodulation by ultraviolet light: clinical studies and biological effects [published erratum appears in Immunol. Today. (1991) 12(7), 227]. Immunol. Today 12, 119–123. Papadopoulos, E. B., Ladanyi, M., Emanuel, D., MacKinnon, S., Boulad, F., Carabasi, M. H., Castro-Malaspina, H., Childs, B. H., Gillio, A. P., and Small, T. N. (1994). Infusions of donor leukocytes to treat Epstein–Barr virus-associated lymphoproliferative disorders after allogeneic bone marrow transplantation. N. Engl. J. Med. 330, 1185–1191. Pass, H. I., Mew, D. J., Carbone, M., Matthews, W. A., Donington, J. S., Baserg, R., Walker, C. L., Resnicoff, M., and Steinberg, S. M. (1996). Inhibition of hamster mesothelioma tumorigenesis by an antisense expression plasmid to the insulin-like growth factor-I receptor. Cancer Res. 56, 4044–4048. Penn, I. (1988). Tumors of the immunocompromised patient. Ann. Rev. Med. 39, 63–73. Pepper, M. S., Vassalli, J. D., Orci, L., and Montesano, R. (1993). Biphasic effect of transforming growth factor factor-beta 1 on in vitro angiogenesis. Exp. Cell Res. 204, 356–363. Pettit, S. J., Seymour, K., O’Flaherty, E., and Kirby, J. A. (2000). Immune selection in neoplasia: towards a microevolutionary model of cancer development. Br. J. Cancer 82, 1900– 1906. Peus, D., Vasa, R. A., Meves, A., Beyerle, A., and Pittelkow, M. R. (2000). UVB-induced epidermal growth factor receptor phosphorylation is critical for downstream signaling and keratinocyte survival. Photochem. Photobiol. 72, 135–140. Pianetti, S., Arsura, M., Romieu-Mourez, R., Coffey, R. J., and Sonenshein, G. E. (2001). Her-2/neu overexpression induces NF-kappaB via a PI3-kinase/Akt pathway involving calpein-mediated degradation of IkappaB-alpha that can be inhibited by the tumour suppressor PTEN. Oncogene 20, 1287–1299. Piccinni, M. P., Beloni, L., Livi, C., Maggi, E., Scarselli, G., and Romagnani, S. (1998). Defective production of both leukemia inhibitory factor and type 2 T-helper cytokines by decidual T cells in unexplained recurrent abortions. Nat. Med. 4, 1020–1024. Plaeger, S., Bass, H. Z., Nishanian, P., Thomas, J., Aziz, N., Detels, R., King, J., Cumberland, W., Kemeny, M., and Fahey, J. L. (1999). The prognostic signiﬁcance in HIV infection of immune activation represented by cell surface antigen and plasma activation marker changes. Clin. Immunol. 90, 238–246.

272

A. G. Dalgleish and K. J. O’Byrne

Posnett, D. N., Kabak, S., Hodtsev, A. S., Goldberg, E. A., and Asch, A. (1993). T-cell antigen receptor V beta subsets are not preferentially deleted in AIDS. AIDS 7, 625– 631. Prince, M. M., and Hildreth, N. G. (1986). The inﬂuence of potential biases on the risk of breast tumors among women who received radiotherapy for acute postpartum mastitis. J. Chronic. Dis. 39, 553–560. Purbhoo, M. A., Sewell, A. K., Klenerman, P., Goulder, P. J., Hilyard, K. L., Bell, J. I., Jakobsen, B. K., and Phillips, R. E. (1998). Co-presentation of natural HIV-agonist and antagonist ligands fails to induce the T cell receptor signalling cascade. Proc. Natl. Acad. Sci. USA 95, 4527–4532. Qin, L., Ding, Y., and Bromberg, J. S. (1996). Gene transfer of transforming growth factorbeta 1 prolongs murine cardiac allograft survival by inhibiting cell mediated immunity. Hum. Gene Ther. 7, 1981–1988. Raza, A. (2000). Consilience across evolving dysplasias affecting myeloid, cervical, esophageal, gastric and liver cells: common themes and emerging patterns. Leuk. Res. 24, 63–72. Raziuddin, S., Shetty, S., and Ibrahim, A. (1991). T-cell abnormality and defective interleukin-2 production in patients with carcinoma of the urinary bladder with schistosomiasis. J. Clin. Immunol. 11, 103–113. Richards, J. S., Fitzpatrick, S. L., Clemens, J. W., Morris, J. K., Alliston, T., and Sirois, J. (1995). Ovarian cell differentiation: a cascade of multiple hormones, cellular signals, and regulated genes. Recent Prog. Horm. Res. 50, 223–254. Richtsmeier, W. J., and Eisele, D. (1986). In vivo anergy reversal with cimetidine in patients with cancer. Arch. Otolaryngol. Head Neck Surg. 112, 1074–1077. Roman, C., Saha, D., and Beauchamp, R. (2001). TGF-beta and colorectal carcinogenesis. Microsc. Res. Tech. 52, 450–457. Rook, G. A., Hernandez-Pando, R., and Lightman, S. L. (1994). Hormones, peripherally activated prohormones and regulation of the Th1/Th2 balance [see comments]. Immunol. Today 15, 301–303. Rooney, C. M., Smith, C. A., Ng, C. Y., Loftin, S., Li, C., Krance, R. A., Brenner, M. K., and Heslop, H. E. (1995). Use of gene-modiﬁed virus-speciﬁc T lymphocytes to control EpsteinBarr-virus-related lymphoproliferation. Lancet 345, 9–13. Roses, D. F., Campion, J. F., Harris, M. N., and Gumport, S. L. (1979). Malignant melanoma. Delayed hypersensitivity skin testing. Arch. Surg. 114, 35–38. Rubini, M., Werner, H., Gandini, E., Roberts, C. T., Jr., LeRoith, D., and Baserga, R. (1994). Platelet-derived growth factor increases the activity of the promoter of the insulin-like growth factor-I (IGF-I) receptor gene. Exp. Cell Res. 211, 374–379. Ryu, H. S., Chang, K. H., Yang, H. W., Kim, M. S., Kwon, H. C., and Oh, K. S. (2000). High cyclooxygenase-2 expression in stage IB cervical cancer with lymph node metastasis or parametrial invasion. Gynecol. Oncol. 76, 320–325. Saareks, V., Mucha, I., Sievi, E., Vapaatalo, H., and Riutta, A. (1998). Nicotine stereoisomers and cotinine stimulate prostaglandin E2 but inhibit thromboxane B2 and leukotriene E4 syntheses in whole blood. Eur. J. Pharmacol. 353, 87–92. Saha, D., Datta, P. K., Sheng, H., Morrow, J. D., Wada, M., Moses, H. L., and Beauchamp, R. D., (1999). Synergistic induction of cyclooxygenase-2 by transforming growth factor-beta1 and epidermal growth factor inhibits apoptosis in epithelial cells. Neoplasia 1, 508–517. SaiRam, M., Sharma, S. K., Dipti, P., Pauline, T., Kain, A. K., Mongia, S. S., Bansal, A., Patra, B. D., Ilavazhagan, G., Devendra, K., and Selvamurthy, W. (1998). Effect of hypobaric hypoxia on immune function in albino rats. Int. J. Biometeorol. 42, 55–59. Saksela, K., Muchmore, E., Girard, M., Fultz, P., and Baltimore, D. (1993). High viral load in lymph nodes and latent human immunodeﬁciency virus (HIV) in peripheral blood cells of HIV-1-infected chimpanzees. J. Virol. 67, 7423–7427.

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

273

Salgado, R., Vermeulen, P. B., Benoy, I., Weytjens, R., Huget, P., Van Marck, E., and Dirix, L. Y. (1999). Platelet number and interleukin-6 correlate with VEGF but not with bFGF serum levels of advanced cancer patients. Br. J. Cancer 80, 892–897. Salvadori, S., Gansbacher, B., Pizzimenti, A. M., and Zier, K. S. (1994). Abnormal signal transduction by T cells of mice with parental tumors is not seen in mice bearing IL-2 secreting tumors. J. Immunol. 153, 5176–5182. Savanthanan, S., and O’Byrne, K. J. (2001). Antiestrogens. J. Br. Meno. Soc. 7, 21–26. Sawaoka, H., Kawano, S., Tsuji, S., Tsuji, M., Sun, W., Gunawan, E. S., and Hori, M. (1998a). Helicobacter pylori infection induces cyclooxygenase-2 expression in human gastric mucosa. Prostaglandins Leukot. Essent. Fatty Acids 59, 313–316. Sawaoka, H., Kawano, S., Tsuji, S., Tsujii, M., Gunawan, E. S., Takei, Y., Nagano, K., and Hori, M. (1998b). Cyclooxygenase-2 inhibitors suppress the growth of gastric cancer xenografts via induction of apoptosis in nude mice. Am. J. Physiol. 274, G1061–G1067. Sawaoka, H., Tsuji, S., Tsujii, M., Gunawan, E. S., Sasaki, Y., Kawano, S., and Hori, M. (1999). Cyclooxygenase inhibitors suppress angiogenesis and reduce tumor growth in vivo. Lab. Invest. 79, 1469–1477. Schaffer, M., and Barbul, A. (1998). Lymphocyte function in wound healing and following injury. Br. J. Surg. 85, 444–460. Schooley, R. T. (1995). Correlation between viral load measurements and outcome in clinical trials of antiviral drugs. AIDS 9 Suppl 2, S15–S19. Schreinemachers, D. M., and Everson, R. B. (1994). Aspirin use and lung, colon, and breast cancer incidence in a prospective study. Epidemiology 5, 138–146. Schror, K., Zimmermann, K. C., and Tannhauser, R. (1998). Augmented myocardial ischaemia by nicotine—mechanisms and their possible signiﬁcance. Br. J. Pharmacol. 125, 79–86. Schuller, H. M., Tithof, P. K., Williams, M., and Plummer, H. 3rd (1999). The tobacco-speciﬁc carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone is a beta-adrenergic agonist and stimulates DNA synthesis in lung adenocarcinoma via beta-adrenergic receptor-mediated release of arachidonic acid. Cancer Res. 59, 4510–4515. Scott, M., Stites, D. P., and Moscicki, A. B. (1999). Th1 cytokine patternsd in cervical human papillomavirus infection. Clin. Diagn. Lab. Immunol. 6, 751–755. Sewell, A. K., Gerth, U. C., Price, D. A., Purbhoo, M. A., Boulter, J. M., Gao, G. F., Bell, J. I., Phillips, R. E., and Jakobsen, B. K. (1999). Antagonism of cytotoxic T-lymphocyte activation by soluble CD8. Nat. Med. 5, 399–404. Shankarappa, R., Gupta, P., Learn, G. H., Jr., Rodrigo, A. G., Rinaldo, C. R., Jr., Gorry, M. C., Mullins, J. I., Nara, P. L., and Ehrlich, G. D. (1998). Evolution of human immunodeﬁciency virus type 1 envelope sequences in infected individuals with differing disease progression proﬁles. Virology 241, 251–259. Shearer, G. M. (1983). Allogeneic leukocytes as a possible factor in induction of AIDS in homosexual men. N. Engl. J. Med. 308, 223–224. Shearer, G. M., and Clerici, M. (1992). T helper cell immune dysfunction in asymptomatic, HIV-1-seropositive individuals: the role of TH1-TH2 cross-regulation. Chem. Immunol. 54, 21–43. Sheikh, J., Souberbielle, B., Westby, M., Austen, B., and Dalgleish, A. G. (2000). HIV gp120 plus speciﬁc peptides are recognized in a similar manner to speciﬁc HLA plus peptide by HLA-restricted antigen-speciﬁc T-cell lines. Viral Immunol. 13, 9–17. Sheikh, M. J., Ongradi, J., Austen, B. M., and Dalgleish, A. G. (1995). The gp120 envelope of HIV-1 binds peptides in a similar manner to human leukocyte antigens. AIDS 9, 1229– 1235. Singer, A. J., and Clark, R. A. (1999). Cutaneous wound healing. N. Engl. J. Med. 341, 738– 746. Sonnex, C. (1998). Human papillomavirus infection with particular reference to genital disease. J. Clin. Pathol. 51, 643–648.

274

A. G. Dalgleish and K. J. O’Byrne

Spitz, M. R., Wei, Q., Li, G., and Wu, X. (1999). Genetic susceptibility to tobacco carcinogenesis [see comments]. Cancer Invest. 17, 645–659. Spivak, M., Lakatosh, V. P., Lazarenko, L. M., Lyanenko, L. M., Azarskova, M. V., Mikhailenko, O. M., Tkacikova, L., and Boroda, A. M. (1999). Interrelation of lymphocyte subpopulations in peripheral blood under cervical papillomavirus infection. Folia Microbiol. 44, 721–725. Sredni, B., Tichler, T., Shani, A., Catane, R., Kaufman, B., Strassmann, G., Albeck, M., and Kalechman, Y. (1996). Predominance of TH1 response in tumor-bearing mice and cancer patients treated with AS101. J. Natl. Cancer Inst. 88, 1276–1284. Strand, S., and Galle, P. R. (1998). Immune evasion by tumours: involvement of the CD95 (APO-1/Fas) system and its clinical implications. Mol. Med. Today 4, 63–68. Study, C. P. (1992). The American Cancer Society Prospective Study. Stat. Bull. Metro. Insur. Co. 73, 21–29. Subbaramaiah, K., Zakim, D., Weksler, B. B., and Dannenberg, A. J. (1997). Inhibition of cyclooxygenase: a novel approach to cancer prevention. Proc. Soc. Exp. Biol. Med. 216, 201–210. Sun, L., Vitolo, M., and Passaniti, A. (2001). Runt-related gene 2 in endothelial cells: inducible expression and speciﬁc regulation of cell migration and invasion. Cancer Res. 61, 4994–5001. Takada, K. (2000). Epstein–Barr virus and gastric carcinoma. Mol. Pathol. 53, 255–261. Takagi, A., Watanabe, S., Igarashi, M., Koike, J., Hasumi, K., Deguchi, R., Koga, Y., and Miwa, T. (2000). The effect of Helicobacter pylori on cell proliferation and apoptosis in gastric epithelial cell lines. Aliment. Pharmacol. Ther. 14 (Suppl 1), 188–192. Takahashi, Y., Kawahara, F., Noguchi, M., Miwa, K., Sato, H., Seiki, M., Inoue, H., Tanabe, T., and Yoshimoto, T. (1999). Activation of matrix metalloproteinase-2 in human breast cancer cells overexpressing cyclooxygenase-1 or -2. FEBS Lett. 460, 145–148. Taketo, M. M. (1998). Cyclooxygenase-2 inhibitors in tumorigenesis (Part II). J. Natl. Cancer Inst. 90, 1609–1620. Thanos, D., and Maniatis, T. (2000). NF-κB: a lesson in family values. Cell 80, 529–532. Thun, M. J., and Namboodin, M. M. (1991). Aspirin use and reduced risk of fatal colon cancer. N. Eng. J. Med. 325, 1593–1596. Tomozawa, S., Nagawa, H., Tsuno, N., Hatano, K., Osada, T., Kitayama, J., Sunami, E., Nita, M. E., Ishihara, S., Yano, H., Tsuruo, T., Shibata, Y., and Muto, T. (1999). Inhibition of haematogenous metastasis of colon cancer in mice by a selective COX-2 inhibitor, JTE-522. Br. J. Cancer 81, 1274–1279. Tsai, S. L., Liaw, Y. F., Chen, M. H., Huang, C. Y., and Kuo, G. C. (1997). Detection of type-2like T-helper cells in hepatitis C virus infection: implications for hepatitis C virus chronicity. Hepatology 25, 449–458. Tsuji, S., Kawano, S., Sawaoka, H., Takei, Y., Kobayashi, I., Nagano, K., Fusamoto, H., and Kamada, T. (1996). Evidences for involvement of cyclooxygenase-2 in proliferation of two gastrointestinal cancer cell lines. Prostaglandins Leukot. Essent. Fatty Acids 55, 179–183. Tsujii, M., and DuBois, R. N. (1995). Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell 83, 493–501. Tsujii, M., Kawano, S., and DuBois, R. N. (1997). Cyclooxygenase-2 expression in human colon cancer cells increases metastatic potential. Proc. Natl. Acad. Sci. USA 94, 3336–3340. Tsujii, M., Kawano, S., Tsuji, S., Sawaoka, H., Hori, M., and DuBois, R. N. (1998). Cyclooxygenase regulates angiogenesis induced by colon cancer cells. Cell 93, 705–716. Um, S. J., Kim, E. J., Hwang, E. S., Kim, S. J., Namkoong, S. E., and Park, J. S. (2000). Antiproliferative effects of retinoic acid/interferon in cervical carcinoma cell lines: cooperative growth suppression of IRF-1 and p53. Int. J. Cancer 85, 416–423. Ung, A., Kramer, T. R., Schiffman, M., Herrero, R., Bratti, M. C., Burk, R. D., Swanson, C. A., Sherman, M. E., Hutchinson, M. L., Alfaro, M., Morales, J., Balmaceda, I., and Hildesheim, A. (1999). Soluble interleukin 2 receptor levels and cervical neoplasia: results

Chronic Immune Activation and Inﬂammation in AIDS and Cancer

275

from a population-based case-control study in Costa Rica. Cancer Epidemiol. Biomarkers Prev. 8, 249–253. Uotila, P. (1996). The role of cyclic AMP and oxygen intermediates in the inhibition of cellular immunity in cancer. Cancer Immunol. Immunother. 43, 1–9. Vaganova, I. G. (2000). [Apoptosis and proliferation of epithelial cells in papillomaviral and chlamydial cervicitis]. Vopr. Onkol. 46, 578–582. Vainio, H., and Morgan, G. (1998). Cyclo-oxygenase 2 and breast cancer prevention. Nonsteroidal anti- inﬂammatory agents are worth testing in breast cancer [editorial]. BMJ 317, 828. Vane, J. R., Bakhle, Y. S., and Botting, R. M. (1998). Cyclooxygenases 1 and 2. Annu. Rev. Pharmacol. Toxicol. 38, 97–120. Vile, R., Souberbielle, B., and Dalgleish, A. G. (1996). Tumour vaccines, in “Immunotherapy in Cancer” (M. Gore, and P. Riches, eds). Wiley, Chichester, U.K. Vogelstein, B., Fearon, E. R., Hamilton, S. R., Kern, S. E., Preisinger, A. C., Leppert, M., Nakamura, Y., White, R., Smits, A. M., and Bos, J. L. (1988). Genetic alterations during colorectal-tumor development. N. Engl. J. Med. 319, 525–532. Volpert, O. V., Dameron, K. M., and Bouck, N. (1997). Sequential development of an angiogenic phenotype by human ﬁbroblasts progressing to tumorigenicity. Oncogene 14, 1495–1502. Wahl, S. M., Hunt, D. A., Wakeﬁeld, L. M., McCartney-Francis, N., Wahl, L. M., Roberts, A. B., and Sporn, M. B. (1987). Transforming growth factor type β induces monocyte chemotaxis and growth factor production. Proc. Natl. Acad. Sci. USA 84, 5788–5792. Wallenius, V., Wallenius, K., and Jansson, J. O. (2000). Normal pharmacologically-induced, but decreased regenerative liver growth in interleukin-6-deﬁcient (IL-6(−/−)) mice. J. Hepatol. 33, 967–974. Wang, L. S., Chow, K. C., Wu, Y. C., Li, W. Y., and Huang, M. H. (1999). Detection of Epstein–Barr virus in esophageal squamous cell carcinoma in Taiwan. Am. J. Gastroenterol. 94, 2834–2839. Wang, P. H., Schaaf, G. J., Chen, W. H., Feng, J., Prins, B. A., Levin, E. R., and Bahl, J. J. (1998). IGF-I induction of p53 requires activation of MAP kinase in cardiac muscle cells. Biochem. Biophys. Res. Commun. 245, 912–917. Wang, X., McCullough, K. D., Franke, T. F., and Holbrook, N. J. (2000). Epidermal growth factor receptor-dependent Akt activation by oxidative stress enhances cell survival. J. Biol. Chem. 275, 14624–14631. Watanabe, M., McCormick, K. L., Volker, K., Ortaldo, J. R., Wigginton, J. M., Brunda, M. J., Wiltrout, R. H., and Fogler, W. E. (1997). Regulation of local host-mediated anti-tumor mechanisms by cytokines: direct and indirect effects on leukocyte recruitment and angiogenesis. Am. J. Pathol. 150, 1869–1880. Webster, N. J., Resnik, J. L., Reichart, D. B., Strauss, B., Haas, M., and Seely, B. L. (1996). Repression of the insulin receptor promoter by the tumor suppressor gene product p53: a possible mechanism for receptor overexpression in breast cancer. Cancer Res. 56, 2781–2788. Werner, H., Karnieli, E., Rauscher, F. J., and LeRoith, D. (1996). Wild-type and mutant p53 differentially regulate transcription of the insulin-like growth factor I receptor gene. Proc. Natl. Acad. Sci. USA 93, 8318–8323. Westby, M., Marriott, J. B., Guckian, M., Cookson, S., Hay, P., and Dalgleish, A. G. (1998a). Abnormal intracellular IL-2 and interferon-gamma (IFN-gamma) production as HIV-1assocated markers of immune dysfunction. Clin. Exp. Immunol. 111, 257–263. Westby, M., Vaughan, A. N., Balotta, C., Galli, M., Clerici, M., and Dalgleish, A. G. (1998b). Low CD4 counts rather than superantigenic-like effects account for differences in expressed T-cell receptor (TCR) repertoires between HIV-1 seropositive long-term non-progressors and individuals with progressive disease. Br. J. Haematol. 102, 1187–1196. Whitby, D., and Boshoff, C. (1998). Kaposi’s sarcoma herpesvirus as a new paradigm for virusinduced oncogenesis. Curr. Opin. Oncol. 10, 405–412.

276

A. G. Dalgleish and K. J. O’Byrne

White, C. D., Macatol, F. R., and DeJosef, A. B. (1992). Inﬂammatory cell inﬁltrate in the cervix as a predictor of residual cervical intraepithelial neoplasia after conization. J. Reprod. Med. 37, 799–802. Williams, M. P., and Pounder, R. E. (1999). Helicobacter pylori: from the benign to the malignant. Am. J. Gastroenterol. 94, S11–S16. Wolff, H., Saukkonen, K., Anttila, S., Karjalainen, A., Vainio, H., and Ristimaki, A. (1998). Expression of cyclooxygenase-2 in human lung carcinoma. Cancer Res. 58, 4997–5001. Wu, X., Yu, H., Amos, C. I., Hong, W. K., and Spitz, M. R. (2000). Joint effect of insulin-like growth factors and mutagen sensitivity in lung cancer risk. J. Natl. Cancer Inst. 92, 737–743. Wu, Y., Tewari, M., Cui, S., and Rubin, R. (1996). Activation of the insulin-like growth factor-I receptor inhibits tumor necrosis factor-induced cell death. J. Cell Physiol. 168, 499–509. Yamamoto, T., Nakamura, Y., Kishimoto, K., Takeuchi, H., Shirakata, M., Mitsuya, T., and Hirai, K. (1999). Epstein–Barr virus (EBV)-infected cells were frequently but dispersely detected in T-cell lymphomas of various types by in situ hybridization with an RNA probe speciﬁc to EBV-speciﬁc nuclear antigen 1. Virus Res. 65, 43–55. Yang, C. H., Murti, A., Pfeffer, S. R., Kim, J. G., Donner, D. B., and Pfeffer, L. M. (2001). IFNalpha/beta promotes cell survival by activating NF-kappaB through phosphatidyl inositol-3 kinase and Akt. J. Biol. Chem. 276(17), 13756–61. Yang, T., Schnermann, J. B., and Briggs, J. P. (1999). Regulation of cyclooxygenase-2 expression in renal medulla by tonicity in vivo and in vitro. Am. J. Physiol. 277, F1–F9. Yao, R., Rioux, N., Castonguay, A., and You, M. (2000). Inhibition of COX-2 and induction of apoptosis: two determinants of nonsteroidal anti-inﬂammatory drugs’ chemopreventive efﬁcacies in mouse lung tumorigenesis. Exp. Lung Res. 26, 731–742. Yeung, M. C., and Lau, A. S. (1998). Tumor suppressor p53 as a component of the tumor necrosis factor-induced, protein kinase PKR-mediated apoptotic pathway in human promonocytic U937 cells. J. Biol. Chem. 273, 25198–25202. Young, J. A. (1988). HIV and HLA similarity. Nature 333, 215. Yucel-Lindberg, T., Ahola, H., Carlstedt-Duke, J., and Modeer, T. (1999). Involvement of tyrosine kinases on cyclooxygenase expression and prostaglandin E2 production in human gingival ﬁbroblasts stimulated with interleukin-1 beta and epidermal growth factor. Biochem. Biophys. Res. Com. 257, 528–532. Zea, A. H., Curti, B. D., Longo, D. L., Alvord, W. G., Strobl, S. L., Mizaguchi, H., Creekmore, S. P., O’Shea, J. J., Powers, G. C., and Urba, W. J. (1995). Alterations in T cell receptor and signal transduction molecules in melanoma patients. Clin. Cancer Res. 1, 1327–1335. Zhang, F., Altorki, N. K., Mestre, J. R., Subbaramaiah, K., and Dannenberg, A. J. (1999). Curcumin inhibits cyclooxygenase-2 transcription in bile acid- and phorbol ester-treated human gastrointestinal epithelial cells. Carcinogenesis 20, 445–451. Zhang, F., Subbaramaiah, K., Altorki, N., and Dannenberg, A. J. (1998). Dihydroxy bile acids activate the transcription of cyclooxygenase-2. J. Biol. Chem. 273, 2424–2428. Zheng, X., Yan, L., Nilson, B., Eklund, G., and Drettner, B. (1994). Epstein–Barr virus infection, salted ﬁsh and nasopharyngeal carcinoma. A case control study in Southern China. Acta Oncol. 33, 867–872. Zhou, B. P., Hu, M., Hu, C. T., Miller, S. A., Yu, Z., Xia, W., Lin, S.-Y., and Hung, M.-C. (2000). HER-2/neu blocks tumour necrosis factor-induced apoptosis via the Akt/NF-κB pathway. J. Biol. Chem. 275, 8027–8031. Zhu, K., Wang, J., Zhu, J., Jiang, J., Shou, J., and Chen, X. (1999). p53 induces TAP1 and enhances the transport of MHC class I peptides. Oncogene 18, 7740–7747. zur Hausen, A., Brink, A. A., Craanen, M. E., Middeldorp, J. M., Meijer, C. J., and van den Brule, A. J. (2000). Unique transcription pattern of Epstein–Barr virus (EBV) in EBVcarrying gastric adenocarcinomas: expression of the transforming BARF1 gene. Cancer Res. 60, 2745–2748.

Molecular Biology of Hodgkin’s Lymphoma Ralf Kuppers ¨ Institute for Genetics and Department of Internal Medicine I University of Cologne 50931 Cologne, Germany

I. Introduction II. Pathology III. Origin of Hodgkin and Reed–Sternberg Cells A. The Germinal Center Reaction B. Origin and Clonality of HRS Cells C. Persistence and Dissemination of HRS Cells D. HRS Cells as Cell Fusions? E. The Relationship of HL and B-cell Non-Hodgkin Lymphomas F. HRS Cells in Diseases Other Than HL G. Searching for Members of the HRS Cell Clone among Small Lymphocytes in the HL Microenvironement H. Rare Occurrence of HL as a T-Cell Lymphoma IV. Pathogenesis A. The Role of EBV in HL Pathogenesis B. Chromosomal Aberrations C. Mutations in Oncogenes and Tumor Suppressor Genes V. Phenotype and Gene Expression Patterns of HRS Cells A. Similarities and Differences Compared with Normal B Cells B. Expression of Apoptosis-Related Genes VI. HRS Cells in Their Microenvironment A. Chemokines and Cytokines B. HRS Cells as Antigen-Presenting Cells VII. Concluding Remarks References

Hodgkin’s lymphoma (HL) is characterized by typical mononucleated Hodgkin and multinucleated Reed–Sternberg cells, which occur at low frequency in a mixed cellular inﬁltrate in the tumor tissue. Because of the rarity of these cells and their unusual immunophenotype, which is strikingly different from those of all normal hematopoietic cell types, the origin of these cells and their clonality have long been unclear. Single-cell studies of rearranged immunoglobulin genes showed that Hodgkin and Reed–Sternberg (HRS) cells represent clonal tumor-cell populations derived from germinal center B cells. In classical HL, the detection of obviously crippling immunoglobulin gene mutations in a fraction of the cases suggests that HRS cells may derive from germinal center B cells that have lost the capacity to be positively selected by antigen and that normally would

Advances in CANCER RESEARCH 0065-230X/02 $35.00

277

C 2002 by Academic Press. Copyright All rights of reproduction in any form reserved.

278

¨ Ralf Kuppers

have undergone apoptosis. In rare cases, HRS cells represent transformed T lymphocytes. The transforming events involved in malignant transformation of HRS cells are still largely unknown. Constitutive activation of the transcription factor NFκB, which can, for example, be induced through Epstein–Barr virus transformation of HRS cells or destructive somatic mutations of the inhibitor of NFκB, is likely to be a key event in HL pathogenesis. Signiﬁcant progress has been made in our understanding of the cellular interactions in HL tissues, which are mainly mediated by a large variety of cytokines and chemokines. C 2002 Academic Press.

I. INTRODUCTION Despite the fact that Hodgkin’s lymphoma (HL) was described 170 years ago (Hodgkin, 1832), the disease remained enigmatic for a long time. Neither the clonality of the putative tumor cells—a hallmark for a malignancy—nor their cellular origin was clariﬁed until recently. The problems associated with the studies of HL were mainly due to the fact that the presumed tumor cells, the mononucleated Hodgkin and the multinucleated Reed–Sternberg cells, are very rare in the tissue, and that these cells show an unusual phenotype, strikingly different from those of all known normal hematopoietic cell types. Moreover, only a few cell lines could be established (Drexler, 1993; Wolf et al., 1996). The scarcity of the Hodgkin and Reed–Sternberg (HRS) cells prevented a reliable application of standard molecular biological techniques, such as Southern-blot hybridization and whole-tissue polymerase chain reaction (PCR), to determine the presence of clonal immunoglobulin (Ig) or T-cell receptor (TCR) gene rearrangements as markers for a B- or T-cell origin of the HRS cells. The low frequency of the tumor cells also hampered cytogenetic studies to analyze the clonality of the cells and to reveal recurrent chromosomal abnormalities. Particularly confusing is the immunophenotype of the HRS cells, which shows coexpression of markers typical for different hematopoietic cell types and no consistent pattern clearly pointing to the cell of origin. It is therefore perhaps not surprising that a large variety of different cell types have been discussed as the cellular origin of these cells. Some of the mysteries surrounding the HRS cells, in particular their clonality and their cellular origin, have been uncovered in the past few years. Besides discussing these ﬁndings, this review will give an overview of the current knowledge regarding the events leading to the malignant transformation of the HRS cells and the role cellular interactions in the complex microenvironment of HL tissues play in the pathogenesis of this lymphoma.

II. PATHOLOGY Characteristic of HL is the occurrence of a small number of typical HRS cells among a major mixed cellular inﬁltrate (Hansmann et al., 1999; Weiss

Hodgkin’s Lymphoma

279

et al., 1999). Usually, HRS cells account for less than 1% of cells in the tissue. The cellular inﬁltrate is mainly composed of T cells, B cells, plasma cells, histiocytes, eosinophils, and neutrophils (Hansmann et al., 1999; Weiss et al., 1999). Based on differences in the cellular composition and the histological picture, ﬁve subsets of HL are deﬁned in the current Revised European American Lymphoma (REAL) and WHO classiﬁcations, namely nodular sclerosis, mixed cellularity, lymphocyte depletion, lymphocyte-rich, and lymphocyte predominant (LP) HL (Weiss et al., 1999). Nodular sclerosis, mixed cellularity, lymphocyte-rich and lymphocyte depletion HL are nowadays grouped together as classical HL and are thus separated from LP HL (Harris, 1999; Weiss et al., 1999). The nodular sclerosis form, which is the most frequent one (ca. 50–60% of cases in the Western world), is characterized by nodular areas that are separated by sclerotic bands (Weiss et al., 1999). Such sclerosis is largely missing in mixed-cellularity HL, which shows a more diffuse distribution of the cells (Weiss et al., 1999). Mixed-cellularity HL accounts for ca. 15% of cases. In nodular sclerosis as well as mixed-cellularity HL, T cells predominate among the inﬁltrating lymphocytes. Lymphocyterich classical HL shows a predominance of B cells among the lymphocytes and a histological picture resembling LP HL, but the phenotype and morphology of the HRS cells are similar to those of nodular sclerosis and mixed-cellularity HL (see below) (Anagnostopoulos et al., 2000). The lymphocyte depletion type is rare (less than 1% of cases) and is characterized by diffuse distribution of cells, little inﬁltration by lymphocytes, and presence of sclerotic as well as necrotic areas (Harris, 1999; Weiss et al., 1999). LP HL accounts for 5% of cases of HL. In this subtype of HL, lymphocytes (in particular B cells) predominate in the cellular inﬁltrate, and the tumor cells are called “lymphocytic and histiocytic” (L&H) cells (Hansmann et al., 1999). The L&H cells usually are not as large as the HRS cells of classical HL, show less pronounced nucleoli, and are only rarely multinucleated. Of diagnostic importance is also the distinct immunophenotype of the HRS cells in classical HL and the L&H cells of LP HL (discussed in more detail later). HRS cells in classical HL express the CD30 surface marker and often CD15, but usually lack B-cell markers such as CD20, whereas L&H cells are CD20-positive and lack CD30 and CD15 expression (Hansmann et al., 1999; Harris, 1999; Weiss et al., 1999).

III. ORIGIN OF HODGKIN AND REED–STERNBERG CELLS A. The Germinal Center Reaction Before the cellular origin of HRS cells is discussed, it is helpful to ﬁrst consider some aspects of normal B-cell development and B-cell differentiation

280

¨ Ralf Kuppers

processes in the germinal center (GC) reaction. B cells are generated in the bone marrow from hematopoietic precursors. In the course of their development, gene segments coding for the variable region of antibody molecules are assembled in somatic recombination processes (Rajewsky, 1996). Since a large number of gene segments (called V, D, and J for the Ig heavy chain and V and J for the κ and λ light chains) are available for the recombination process, and since additional diversity is generated at the joining sites, each B cell is equipped with a unique B-cell receptor (BCR), which can thus serve as a clonal marker. B cells expressing a functional BCR are then released into the periphery (Rajewsky, 1996). If a mature, naive B cell encounters cognate antigen, it is activated and can participate in an immune response against the antigen. In cases where T-cell help is available, activated B cells migrate into B-cell follicles of secondary lymphoid organs such as the lymph nodes or the spleen and in these tissues establish histological structures called GCs (MacLennan et al., 1992; Rajewsky, 1996). In GCs, B cells proliferate vigorously and can give rise to large clones. These proliferating GC B cells are called centroblasts and are concentrated in the dark zone of the GC. In the proliferating GC B cells, the process of somatic hypermutation is activated that modiﬁes the rearranged Ig variable (V) region genes (Berek et al., 1991; ¨ Jacob et al., 1991; Kuppers et al., 1993). Through somatic hypermutation, somatic mutations are introduced at a very high rate into the V-region genes, giving rise to variants of the original BCR. B cells that gain afﬁnity-increasing mutations are positively selected (Fig. 1). The selection process of GC B cells involves two other constituents of the GC, T helper cells and follicular dendritic cells, which are mainly found in the light zone of the GC (MacLennan, 1994). Here, the GC B cells show less proliferative activity and are called centrocytes. It is likely that GC B cells go through repeated rounds of proliferation, mutation, and selection. Finally, selected B cells differentiate to memory B cells or plasma cell precursors and leave the GC (Arpin et al., 1995; MacLennan, 1994). Most somatic mutations, however, will be disadvantagous for a GC B cell (Wiens et al., 1997) (Fig. 1). The majority of amino acid replacements will not lead to an increase in afﬁnity; they may even abolish binding to the respective antigen. Moreover, replacement mutations may interfere with proper folding of the Ig heavy or light chain and thereby prevent surface expression of the receptor. Mutations are also destructive when they generate a stop codon (nonsense mutation) or result in a deletion or duplication of nucleotides that destroy the correct reading frame. Deletions/duplications account for ca. 6% of mutation events in human GC B cells, and nonsense mutations occur at a frequency of about 4–5% of mutation events (Goossens et al., 1998; ¨ Kuppers et al., 1997). It appears that an apoptosis program is activated in GC B cells as a default pathway, thereby allowing a very efﬁcient and quick removal of GC B cells with disadvantageous mutations (Liu et al.,

Hodgkin’s Lymphoma

281

Fig. 1 B-cell selection processes in the germinal center and scenario for HRS cell generation in classical HL. Antigen-activated B cells migrate into B-cell follicles of secondary lymphoid organs and establish GC. GC B cells clonally expand and activate the process of somatic hypermutation that introduces mutations at a high rate into rearranged Ig genes. GC B cells acquiring afﬁnityincreasing mutations are positively selected by GC T helper cells and follicular dendritic cells. Presumably after repeated rounds of proliferation, mutation, and selection, positively selected cells differentiate into either memory B cells or plasma-cell precursors and leave the GC. Most somatic mutations will be disadvantageous for GC B cells, such as amino acid replacement mutations resulting in reduced afﬁnity to the immunizing antigen or deletions that result in loss of the correct reading frame. GC B cells with such disadvantageous mutations are functionally crippled and undergo programmed cell death (apoptosis). The pattern of somatic mutations in rearranged Ig genes of HRS cells from classical HL indicates that these cells are derived from the pool of preapoptotic, crippled GC B cells. The reason for the apoptosis resistance of the HRS cell precursors is largely unclear but may involve several distinct transforming events, such as EBV infection in about half of the cases or somatic mutations of the IκBα gene in some HL.

1989). Only GC B cells that have acquired afﬁnity-improving mutations can appropriately interact with GC T cells and follicular dendritic cells and, by receiving survival signals from these cells, escape the apoptotic pathway.

B. Origin and Clonality of HRS Cells To test HRS cells for a potential origin from B cells, they were isolated by micromanipulation from tissues or cell suspensions and analyzed for Ig gene rearrangments (earlier studies using whole-tissue DNA approaches are not considered here, as they do not allow one to relate the ﬁndings unequivocally

282

¨ Ralf Kuppers

to the HRS cells). The ﬁrst studies revealed contradictory results, reporting ¨ monoclonal Ig gene rearrangements (Kuppers et al., 1994), polyclonal rearrangements (Delabie et al., 1994, 1996), a combination of both (Hummel et al., 1995), or a complete absence of Ig gene rearrangements (Roth et al., 1994). However, these discrepancies were most likely due to technical prob¨ lems associated with the demanding single-cell PCR appraoch (Kuppers et al., ¨ 1996; Kuppers and Rajewsky, 1998). Additional studies by several groups conﬁrmed the presence of clonal Ig gene rearrangements in most cases of HL ¨ (Braeuninger et al., 1997; Brauninger et al., 1999; Irsch et al., 1998; Kanzler ¨ et al., 1996a,b; Kuppers et al., 2001b; Maraﬁoti et al., 1997, 1999, 2000; Ohno et al., 1997; Spieker et al., 2000; Vockerodt et al., 1998). Rearranged Ig loci were detected in HRS cells of classical and LP HL, showing that both types of HL represent B-cell lymphomas. Moreover, the detection of clonal rearrangements establishes the clonality of the HRS cells, conﬁrming that these cells represent the tumor-cell population in this malignancy. In all cases of LP HL, somatic mutations were observed in the rearranged V ¨ region genes (Braeuninger et al., 1997; Kuppers et al., 1994; Maraﬁoti et al., 1997; Ohno et al., 1997). The pattern of mutations indicated that the tumor cells or their precursors were properly selected for BCR expression when ¨ they accumulated the mutations (Braeuninger et al., 1997; Kuppers et al., 1998). In a fraction of the cases, intraclonal sequence diversity was detected, suggesting that the process of somatic hypermutation is still active in the ¨ L&H cells (Braeuninger et al., 1997; Kuppers et al., 1994; Maraﬁoti et al., 1997; Ohno et al., 1997). Since somatic hypermutation is thought to be a GC B-cell-speciﬁc process, the detection of ongoing mutation indicates that LP HL is a GC B-cell lymphoma. This conclusion is also supported by a number of histological and immunophenotypical features of LP HL discussed later. Somatically mutated V genes are also a characteristic feature of the HRS cells in classical HL (Kanzler et al., 1996a,b; Maraﬁoti et al., 2000; Spieker et al., 2000; Vockerodt et al., 1998). In classical HL, however (with rare exceptions), the V gene rearrangements lack intraclonal sequence diversity, indicating that the hypermutation process is no longer active in the HRS cells. Surprisingly, in over 25% (7 of 26) of cases from which we ampliﬁed rearranged V genes, somatic mutations were identiﬁed that rendered originally functional Ig genes nonfunctional (the frequency of cases with obviously destructive mutations may be even higher, as not in all cases were in-frame VH and VL genes and not the entire V region genes ampliﬁed and sequenced) ¨ (Brauninger et al., 1999; Irsch et al., 1998; Jox et al., 1999; Kanzler et al., ¨ ¨ 1996a,b; Kuppers et al., 1994, 2001b; Muschen et al., 2000a, 2001; Spieker et al., 2000; Vockerodt et al., 1998). Such “crippling” mutations included nonsense mutations and deletions that caused frameshifts or removed large parts of the V gene rearrangements. Moreover, in one case a mutation was identifed in the conserved octamer motif of the VH promoter (Jox et al.,

Hodgkin’s Lymphoma

283

1999). Since GC B cells that acquire such destructive mutations are usually efﬁciently eliminated within the GC through apoptosis and are unable to mature into post GC B cells, this observation indicates that the HRS cells in these cases are derived from preapoptotic GC B cells. Importantly, nonsense mutations and deletions account for only a fraction of all disadvantagous mutations (Fig. 1). As discussed earlier, many GC B cells will be driven into apoptosis because of unfavorable replacement mutations. Since such mutations cannot easily be identiﬁed, we speculated that HRS cells as a rule originate from crippled GC B cells, that is, B cells that lost the capacity to be selected by antigen and therefore normally would undergo apoptosis in the ¨ GC (Kanzler et al., 1996b; Kuppers and Rajewsky, 1998). In line with this view, the ratio of replacement to silent mutations in the framework regions of mutated Ig genes of HRS cells in classical HL is in the same range as the one typical for GC centroblasts (1.7–1.8), the mutating GC B cells that are not yet optimally selected, and many of which will undergo apoptosis (Klein ¨ et al., 1998; Kuppers et al., 1997). Memory B cells, GC-derived plasma cells, most B-cell non-Hodgkin lymphomas (e.g., Burkitt’s lymphoma, diffuse large-cell lymphomas, and follicular lymphomas), and the L&H cells of LP HL, on the other hand, show average R/S values of 1.0–1.5, indicating stringent selection against replacement mutations in the framework regions ¨ and hence selection for functionality of the BCR (Klein et al., 1998; Kuppers et al., 1997). The frequent presence of crippling mutations in V genes of HRS cells of classical HL was conﬁrmed by another study, which also detected such mutations in 25% of the cases (the title of the publication is perhaps misleading, as it ignores these ﬁndings) (Maraﬁoti et al., 2000). One case of classical HL with unmutated heavy and light chain V genes has ¨ been described (Muschen et al., 2001). Although this might at ﬁrst glance seem to contradict a uniform GC origin of HRS cells, rare cases of this type are indeed expected in the scenario of an origin of HRS cells from preapoptotic GC B cells. GC founder cells still carry unmutated Ig genes when they enter the GC and start to proliferate. These cells acquire the sensitivity to undergo apoptosis before they accumulate mutations. Hence, some GC B cells may be driven into apoptosis before they managed to accumulate any somatic Ig gene mutations (including afﬁnity-increasing mutations), so that the population of apoptotic GC B cells—the presumed precursor population of HRS cells in classical HL—contains a fraction of cells with unmutated Ig genes (Lebecque et al., 1997). Many B cells not only undergo somatic hypermutation in the GC reaction, but also perform class switch recombination, a process by which the originally expressed Cμ and Cδ genes are replaced by downstream IgH constant region genes. The detection of Cγ transcripts in L&H cells of most cases of LP HL, and the ﬁnding that most HL cell lines established from patients with classical HL show class switch recombination (Irsch et al., 2001; Jox

284

¨ Ralf Kuppers

et al., 1999; Stein et al., 2001), therefore further supports the idea that L&H and HRS cells derive from GC-experienced mature B cells.

C. Persistence and Dissemination of HRS Cells Most of the Ig gene studies of microdissected HRS cells were restricted to a single biopsy specimen. Hence, it was unclear whether the clonality of HRS cells extends to all tissues involved in the disease in a given patient, and whether relapses of the disease usually derive from the original tumor clone or represent independent (therapy-related?) malignancies. Several cases of classical HL were analyzed for the persistence and dissemination of HRS cells. In one case, two lymph-node biopsies taken a year apart harbored the same HRS cell clone, and in a second case, the same HRS cell clone was detected in a lymph node and the spleen biopsied at the same time and another lymph node biopsied 1 year later (Vockerodt et al., 1998). From a relapsed case of mixed cellularity HL, a cell line (L1236) was established from the peripheral blood of the patient in 1994 (Wolf et al., 1996). Molecular analysis of L1236 cells and HRS cells micromanipulated from an HL-involved bone-marrow specimen conﬁrmed the origin of the cell line from the HRS cells in the patient (Kanzler et al., 1996a). Moreover, the clonal IgH gene rearrangement of the HRS cells was detected in a biopsy taken at primary diagnosis in 1991 and at a relapse of the disease in 1993 (Jox et al., 1998). Thus, in this patient, the tumor cells disseminated into the peripheral blood and the HRS cell clone persisted throughout the course of the disease. Persistence and dissemination of HRS cells has also been reported for cases of LP HL (Maraﬁoti et al., 1997; Ohno et al., 1997). Taken together, these data indicate that HRS cells often or perhaps always disseminate and persist in the patient, further supporting the malignant nature of these cells.

D. HRS Cells as Cell Fusions? Based on the unusual immunophenotype of the HRS cells—often showing coexpression of markers typical for different hematopoietic cell types—and the frequent detection of additional copies of various chromosomes even in the mononucleated Hodgkin cells, it has been speculated that HRS cells may represent cell fusions (Bucsky, 1987; Michels, 1995). To address this issue, ﬁve cases of HL with two rearranged IgH loci and one case with two rearranged TCRβ loci (see below) were analyzed for the presence of additional IgH (or TCRβ) alleles in germline conﬁguration as a footprint for a fusion of a B cell with a non-B cell. However, in none of these cases were additional ¨ alleles identiﬁed (Kuppers et al., 2001a). Thus, these ﬁndings argue against a role of cell fusion in HRS cell generation.

Hodgkin’s Lymphoma

285

E. The Relationship of HL and B-cell Non-Hodgkin Lymphomas In rare instances, HL and B-cell non-Hodgkin lymphoma occur in the same patient, either sequentially or simultaneously (Jaffe and Mueller-Hermelink, 1999). Such cases represent interesting models for the study of the relationship between HL and other B-cell malignancies. The clonal relationship of the tumor cells in such cases can be elegantly analyzed using the Ig gene rearrangements carried by these cells as markers of clonality. Six combinations of HL with other B-cell lymphomas have so far been described in detail by ¨ ¨ single-cell analysis (Brauninger et al., 1999; Kuppers et al., 2001b; Maraﬁoti et al., 1999; Ohno et al., 1998b). In one case, in which a small noncleaved cell lymphoma occurred several years after the ﬁrst diagnosis of classical HL, the tumor cells were clonally unrelated (Ohno et al., 1998b). Perhaps the occurrence of the non-Hodgkin lymphoma was related to the chemo- and radiotherapy that the patient received as therapy for the HL. The ﬁve other cases represent three composite follicular lymphomas and classical HL, a combination of B-cell chronic lymphocytic leukemia (B-CLL) with classical HL, and a case in which a classical HL developed 3 years after diagnosis ¨ ¨ of a T-cell-rich B-cell lymphoma (Brauninger et al., 1999; Kuppers et al., 2001b; Maraﬁoti et al., 1999). In each of these cases, the single-cell analysis revealed a common origin for the two lymphoma clones. Moreover, in each of the ﬁve cases, both shared somatic mutations and mutations that were restricted to either of the two lymphoma clones were detected in the rearranged Ig genes. This strongly indicates that the common precursor of the tumors was a GC B cell, and that the lymphomas developed from distinct daughter cells of this shared precursor. This supports a scenario in which a GC B cell that carried one or more transforming events gave rise to descendants that, probably after acquiring additional, separate transforming events, developed ¨ into the two distinct B-cell lymphomas (Brauninger et al., 1999). The ﬁnding of a common clone in a combination of HL and CD5-positive B-CLL may indicate that HRS cells can derive not only from conventional but also from CD5-positive B cells, if B-CLL is indeed a malignancy of CD5 B lym¨ phocytes (Kuppers et al., 2001b). Collectively, these composite lymphomas provide proof of the B-cell nature of the HRS cells in these cases and support the view that decisive events in the pathogenesis of HL (as is true also for ¨ most B-cell non-Hodgkin lymphomas) take place in the GC (Kuppers et al., 1999).

F. HRS Cells in Diseases Other Than HL Although HRS cells are a hallmark of HL, cells with the morphology of HRS cells and expressing CD30 can also be found in other diseases. One of

286

¨ Ralf Kuppers

these is B-CLL, which in rare cases shows HRS-like cells among the B-CLL cells, without the typical cellular microenvironment of HL (and which are thus not classiﬁed as composite lymphomas) (Jaffe and Mueller-Hermelink, 1999). Five cases of this type were analyzed at the single-cell level (Kanzler et al., 2000; Ohno et al., 1998a). In each of the ﬁve cases, the HRS-like cells represent clonal populations, which in three patients were clonally related to the B-CLL (Kanzler et al., 2000; Ohno et al., 1998a). Intriguingly, in one of these cases where a somatic mutation analysis was possible, the related B-CLL and HRS-like cell clones harbored shared as well as distinct mutations in the rearranged Ig genes, indicating derivation of these cells from distinct members of a GC B cell clone (Kanzler et al., 2000). In the other two cases, the HRS-like cells were EBV-infected and clonally unrelated to the B-CLL, showing clonal expansion of EBV-harboring cells in a setting of B-CLL (Kanzler et al., 2000). Also in these instances, the HRS-like cells carried somatically mutated Ig genes, suggesting a derivation from GC-experienced B cells. Moreover, as crippling mutations were observed in the HRS cells of one of the cases, this clone likely originates from a preapoptotic GC B cell (Kanzler et al., 2000). Taking these cases together, it appears that HRS-like cells in B-CLL represent clonal expansions of GC-derived B cells. Based on the clonality of the HRS cells, which are either infected by EBV or related to the B-CLL clone, these (premalignant?) cells may represent potential precursors for HRS cells causing HL. This idea is supported by the identiﬁcation of a composite lymphoma of B-CLL and HL (see earlier discussion). Populations of HRS-like cells, however, do not always represent monoclonal expansions and may sometimes also include naive B cells. In infectious mononucleosis, a self-limiting lymphoproliferative disease caused by primary EBV infection, HRS-like EBV-infected cells are regularly observed. These cells are polyclonal, sometimes carry unmutated Ig genes, and lack obviously crippling mutations (Kurth et al., 2000). Hence, the morphology of HRS cells is not strictly correlated with the derivation from crippled GC B cells. Perhaps the expression of the latent membrane protein 1 of EBV by these cells is involved in the generation of cells with an HRS-like morphology, as discussed later.

G. Searching for Members of the HRS Cell Clone among Small Lymphocytes in the HL Microenvironement Although the presence of the typical HRS cells is the hallmark of HL, the question has arisen as to whether the tumor clone is composed only of HRS cells or also includes members resembling small lymphocytes. In favor of the

Hodgkin’s Lymphoma

287

idea that some of the small lymphocytes in the tumor tissue may belong to the HRS cell clone, Janssen et al. reported that in two cases of HL in which numerical chromosomal abnormalities were observed in the HRS cells by ﬂuorescence in situ hybridization, the same numerical abnormalities (e.g., trisomy of chromosome 1) were also found in some small CD30-negative lymphocytes (Jansen et al., 1999). However, since a trisomy is not a stringent clonal marker, and since it has been reported that lymphocytes of HL patients show an increased frequency of various chromosomal abnormalities (Barrios et al., 1988; Fonatsch et al., 1989), further studies are needed to prove the clonal relationship between the small lymphocytes and the HRS cells showing the same chromosomal abnormalities. We reasoned that in cases of HL in which all HRS cells are infected by Epstein–Barr virus (EBV) (discussed in more detail later), morphologically distinct cells belonging to the HRS clone should also be EBV-positive. Moreover, lymph nodes affected by classical HL often show increased numbers of small EBV-positive cells as compared to lymph nodes without HL involvement (Spieker et al., 2000). On this basis, EBV-positive HRS cells and small, CD30-negative, EBV-positive lymphocytes from three cases of HL were isolated and their clonal relationship was studied by ampliﬁcation of rearranged Ig genes (Spieker et al., 2000). This study revealed that the small EBV-harboring cells are not related to the HRS cells, arguing against the presence of a pool of (morphological) “HRS-cell precursors” among the small B lymphocytes in the patient. In line with this study, it has been shown that in some cases of HL, the HRS cells and small lymphocytes in HL tissues were infected by separate strains of EBV (Faumont et al., 2001). These ﬁndings may also be of clinical relevance, because a population of CD30-negative lymphocytes belonging to the lymphoma clone in HL could severely hamper the success of treatment strategies aiming to cure HL patients by eliminating CD30-positive tumor cells.

H. Rare Occurrence of HL as a T-Cell Lymphoma In a fraction of cases, HRS cells express one or multiple markers that are typical for T cells and partly also natural killer cells. CD3 and the cytotoxic molecules granzyme B, perforin, and T-cell intracellular antigen 1 (TIA-1) have each been detected in HRS cells in about 10 to 15% of cases of classical HL (Felgar et al., 1997; Foss et al., 1996; Krenacs et al., 1997; Oudejans et al., 1996b). Since expression of these markers was taken as an indication for a potential T-cell origin of HRS cells, single HRS cells from 17 such cases ¨ have been analyzed in three studies for TCR gene rearrangements (Muschen et al., 2000a, 2001; Seitz et al., 2000). In three of the lymphomas, clonal TCR gene rearrangements were detected. The HRS cells in these tumors lacked

288

¨ Ralf Kuppers

Ig gene rearrangements. Thus, in rare cases (likely less than 5%), HRS cells derive from T and not B cells. However, the fact that the other 13 cases carried clonal Ig and no TCR gene rearrangements shows that B-cell-derived cases of HL can show aberrant T lineage marker expression. It is interesting to note that the three T-cell-derived cases were the only ones expressing perforin, indicating that this marker perhaps identiﬁes cases of HL with a T-cell origin.

IV. PATHOGENESIS A. The Role of EBV in HL Pathogenesis In ca. 40% of cases of classical HL in the Western World, and in nearly 90% of cases affecting children in Latin America, EBV is present in the HRS cells (Jarrett and MacKenzie, 1999). EBV, a γ herpesvirus, infects more than 90% of the human population worldwide. After primary infection, the virus resides latently in memory B cells at a frequency of about 1 in 105 to 106 infected B cells (Khan et al., 1996). Besides classical HL, EBV is also associated with Burkitt’s lymphoma, B-cell lymphomas in immunodeﬁcient patients, and nasopharyngeal carcinoma. EBV is not found in the L&H cells of LP HL. Patients with infectious mononucleosis have a signiﬁcantly increased risk of developing HL later in their lives (Alexander et al., 2000; Hjalgrim et al., 2000; Munoz et al., 1978; Rosdahl et al., 1974), which is suggestive of a role of EBV in HL pathogenesis. More importantly, the structure of the viral genome in HRS cells indicates a single infection event in the tumor clone, i.e., an infection of the founder cell of the HRS cell population (Anagnostopoulos et al., 1989; Weiss et al., 1989). This directly supports a role of EBV in HL pathogenesis. In HRS cells, the virus is present in latent form. In addition to noncoding RNAs (EBERs), three of the nine latent EBV genes are expressed, namely EBV nuclear antigen 1 (EBNA1) and the latent membrane proteins LMP1 and LMP2a (Young et al., 2000). This pattern of latent gene expression, termed latency II, is peculiar, because in EBV cell lines grown in culture, the expression of LMP1 and LMP2a is regulated by EBNA2, which is, however, not expressed by HRS cells. Interestingly, a study has identiﬁed an alternative promoter of LMP1 and has shown that transcription from this promoter is activated by the transcription factor STAT3 (Chen et al., 2001). Since (EBVpositive and -negative) HRS cells show constitutive activity of STAT3 (Chen et al., 2001; Kube et al., 2001), EBV apparently uses the STAT activity in HRS cells to allow for LMP1 expression in the absence of EBNA2. The function of the EBERs was long unclear. However, studies suggest that these transcripts may play a role in Il-10 induction in EBV-harboring cells

Hodgkin’s Lymphoma

289

(Kitagawa et al., 2000), which may be important for suppression of cytotoxic T cell responses against the EBV-infected HRS cells (see later discussion). EBNA1 is needed for the replication of the episomal viral genome, but may have also some transforming functions (Wilson et al., 1996; Young et al., 2000). LMP1 can transform rodent ﬁbroblasts and is critical for primary B-cell transformation, as is also evident from the fact that transgenic mice expressing this protein in B cells develop B-cell lymphomas (Kaye et al., 1993; Kilger et al., 1998; Kulwichit et al., 1998). LMP1 mimics an activated CD40 receptor, which plays a major role in the differentiation of B cells and the survival of GC B cells (Kilger et al., 1998; Liu et al., 1989). LMP1 signaling is mainly mediated by binding of TRAFs (TNF receptor-associated factors) to the cytoplasmatic portion of LMP1, which ﬁnally results in activation of the transcription factor NFκB (further discussed below) (Devergne et al., 1998; Mosialos et al., 1995). There are some differences between CD40 and LMP1 signaling, which result in a stronger cellular activation through LMP1 (Brown et al., 2001). Additionally, LMP1 signals through the AP-1 transcription factor and uses the JAK/STAT signalling pathway (Gires et al., 1999). LMP2a has a cytoplasmatic motif, the ITAM (immunoreceptor tyrosinebased acivation motif), that is also found in the coreceptors of the BCR and plays an important role in signal transduction of cross-linked BCR (Alber et al., 1993). Via the ITAM, LMP2a can recruit cytoplasmic kinases that normally bind to the BCR and thereby mimic an activated Ig receptor (Longnecker, 2000). Intriguingly, in transgenic mice already expressing LMP2a in B-cell precursors, Ig-negative B cells lacking productive heavy-chain rearrangements develop and colonize peripheral lymphoid organs (Caldwell et al., 1998). Usually, B-cell precursors that do not manage to generate a functional Ig receptor are eliminated within the bone marrow by apoptosis (Rajewsky, 1996). In LMP2a transgenic mice, the viral protein can compensate for the lack of proper Ig signaling. If LMP2a has a similar capacity in GC B cells that fail to receive appropriate survival signals through a high-afﬁnity BCR, this could play an important role in the pathogenesis of EBV-positive HL by rescuing EBV-infected GC B cells with destructive mutations from apoptosis. Taken together, these data indicate that the coexpression of LMP1 and LMP2a could have a decisive role in the development of EBV-positive HL. By mimicking the main survival signals of GC B cells—CD40 signaling and Ig cross-linking—these viral proteins could allow the survival of GC B cells that would otherwise undergo apoptosis. The idea that expression of LMP1 and LMP2a (together with EBNA1) could occur in normal GC B cells is supported by EBV gene expression studies of these cells (Babcock et al., 2000). However, seemingly opposing these latter studies, it has been reported that LMP1 transgenic mice fail to develop GC upon immunization,

290

¨ Ralf Kuppers

suggesting that LMP1 expression suppresses the establishment of these structures (Uchida et al., 1999). Perhaps if EBV infects B cells in already established GCs, or if LMP1 is up-regulated some time after an EBV-infected B cell enters a GC, LMP1 expression transforms the cells and initiates their exit from the GC, replacing the normal CD40-mediated signals that control survivial and differentiation of GC B cells. The fact that EBV is found only in a fraction of cases of classical HL gave rise to speculation on whether the other cases were originally also EBV-positive, but the virus was later lost from the HRS cells because it was no longer needed for survival and expansion of the tumor cells. Since indications of such a “hit-and-run” scenario had indeed been described in some cases of Burkitt’s lymphoma (Razzouk et al., 1996), several cases of classical HL with HRS cells lacking expression of EBV genes were studied for fragments of the EBV genome potentially inserted into the HRS cell genome. However, no such fragments were detected (Staratschek-Jox et al., 2000). HLs were also analyzed for association with other viruses (e.g., herpesviruses 6, 7, and 8 and adenoviruses), but no infection of HRS cells was found (Jarrett and MacKenzie, 1999). A more recent report, however, claims that measles virus can be detected in the HRS cells in more than half of all cases of HL (Gopas et al., 2001).

B. Chromosomal Aberrations Classical cytogenetic analysis of HRS cells was hampered by the rarity of the cells in the tissue, their low proliferation activity, and the problem of assigning metaphase spreads unequivocally to the tumor cells. With these caveats in mind, several studies reported numerical and structural chromosomal aberrations (reviewed in Sarris et al., 1999). The problem of HRS cell identiﬁcation was addressed by combining cytogenetic studies or ﬂuorescense in situ hybridization (FISH) with immunostaining for CD30 (Teerenhovi et al., 1988; Weber-Matthiesen et al., 1995). The latter approach had the additional advantage that the analysis can be perfomed on interphase nuclei, although it is mainly restricted to the identiﬁcation of numerical abnormalities or deﬁned chromosomal translocations. These studies revealed numerical chromosomal abnormalities in all cases of classical HL analyzed (Weber-Matthiesen et al., 1995). In addition to clonal numerical aberrations subclonal gains or losses of chromosomes were observed. Whereas the former observation supports the clonality of the HRS cells, the latter indicates a genomic instability of these cells. Three groups applied comparative genomic hybridization to pools of iso¨ lated HRS or L&H cells (Franke et al., 2001; Joos et al., 2000; Kupper et al., 2001; Ohshima et al., 1999). Whereas one study of HRS cells revealed a very

Hodgkin’s Lymphoma

291

heterogenous picture with many different gains and losses of parts of chromosomes (Ohshima et al., 1999), another study showed a less mixed picture and a predominance of gains (Joos et al., 2000). Among recurrent gains, the chromosomal region 12p14 was identiﬁed, which contains the MDM2 gene ¨ (Kupper et al., 2001). Overexpression of MDM2 could indeed be conﬁrmed and may play a role in inactivating the tumor suppressor gene p53 in the ¨ HRS cells (Kupper et al., 2001). One study of L&H cells of LP HL showed a large number of chromosomal gains and losses affecting nearly all chromosomes, making it difﬁcult to identify genes speciﬁcally ampliﬁed or deleted in LP HL (Franke et al., 2001). Two types of balanced chromosomal translocations well known from nonHodgkin lymphomas have been analyzed in HL, namely the t(14;18), which is a hallmark of follicular lymphoma, and the t(2;5), which is found in many cases of large-cell anaplastic lymphoma. One report suggested a frequent occurrence of t(2;5) translocations resulting in fusion of the NPM and Alk genes (Orscheschek et al., 1995). However, several other studies could not conﬁrm this ﬁnding (Weber-Matthiesen et al., 1996; Weiss et al., 1995; Wellmann et al., 1995). Among more than 120 cases analyzed for t(14;18) involving the antiapoptotic bcl-2 gene and the IgH locus, only two cases with a chromosomal translocation were identiﬁed (Gravel et al., 1998; Miura et al., 2000; Poppema et al., 1992). Whereas bcl-2/IgH translocations appear to be a very rare event in classical HL, translocations involving the IgH locus may well be present in a considerable fraction of HLs. Aberrations at 14q32, the chromosomal region harboring the IgH locus, have been described in more than 10–20% of cases of classical HL (Falzetti et al., 1999; Poppema et al., 1992). However, the genes involved in these chromosomal aberrations have not yet been identiﬁed. The genomic instability that has been described for the HRS cells may not be restricted to the tumor cells in HL patients. Normal lymphocytes in untreated HL patients also show chromosomal aberrations more frequently than lymphocytes of healthy individuals (Barrios et al., 1988; Fonatsch et al., 1989). The reason for this phenomenon is still unclear.

C. Mutations in Oncogenes and Tumor Suppressor Genes Because of the need to analyze isolated HRS cells, only some candidate protooncogenes and tumor suppressor genes have been analyzed for mutations in these cells. No mutations of the N-ras protooncogene were found ¨ in 12 cases analyzed (Table I) (Trumper et al., 1996). Several initial studies suggested an involvement of p53 mutations in HL (Chen et al., 1996; Gupta ¨ et al., 1993; Inghirami et al., 1994; Trumper et al., 1993). However, some

292

¨ Ralf Kuppers

Table I Mutations of Protooncogenes and Tumor Suppressor Genes in HRS Cells Gene

Cases analyzed

Cases with mutations

Reference

bcl-2a

115 28 1 8 4 26 12 8 10 5 10

0 1 1 0 0 3 0 2 1 2 1b

Gravel et al.,1998 Poppema et al., 1992 Miura et al., 2000 Montesinos-Rongen et al., 1999 ¨ Kupper et al., 2001 Maggio et al., 2001 ¨ Trumper et al., 1996 Cabannes et al., 1999 Emmerich et al., 1999 Jungnickel et al., 2000 ¨ Muschen et al., 2000b

p53

N-ras IκBα

CD95

a Analysis for bcl-2/IgH translocations. Since bcl-2/IgH translocation can also be found in some rare normal B cells, only studies that analyzed single HRS cells or conﬁrmed that HRS cells carried the translocation are considered. b Analysis of exon 9, coding for the death domain.

of these studies used enriched populations instead of single HRS cells, so that the origin of the mutated genes is uncertain, and the only “Hodgkin cell line” with p53 mutations (CO) has been identiﬁed as a cell-culture contamination (Drexler et al., 1999). Two single-cell studies of HRS cells did not ¨ identify any mutations of p53 in a total of 12 cases (Kupper et al., 2001; Montesinos-Rongen et al., 1999) (Table I). In another study, monoallelic p53 mutations were detected in 3 of 26 cases (Maggio et al., 2001). Since pools of microdissected HRS cells were used in that analysis, it remains unclear whether the mutations are present in all HRS cells or only in a fraction of the tumor population. These studies together indicate that p53 mutations occur only in a small fraction of HL cases. Based on the observation that HRS cells express constitutive active nuclear factor κB (NFκB) (Bargou et al., 1997), the inhibitor of NFκB, IκBα, was analyzed for mutations as a potential reason for this phenomenon. Usually, the transcription factor NFκB, which plays important roles in the regulation of B cell differentiation and survival (see below), is kept inactive in the cytoplasm by binding to IκBα (reviewed in Baeuerle and Henkel, 1994). Only upon cellular activation is IκBα degraded, thereby releasing NFκB and allowing its transfer into the cell nucleus, where NFκB-responsive genes are activated. Inactivating mutations of IκBα were detected in two HL-derived cell lines, and somatic mutations of IκBα were also identiﬁed in HRS cells isolated or enriched from biopsy specimens in 4 of 23 cases analyzed (Table I) (Cabannes et al., 1999; Emmerich et al., 1999; Jungnickel et al., 2000; Wood et al., 1998). In one case, clonal deleterious mutations of the gene were identiﬁed in both alleles of the gene, suggesting a role for IκBα as a tumor

Hodgkin’s Lymphoma

293

suppressor gene in HL (Jungnickel et al., 2000). Most cases lacked IκBα mutations, or mutations were detected only on one allele, and in the study by Cabannes et al. it remains unclear whether the mutations are present in all HRS cells and hence represent a primary transforming event (Cabannes et al., 1999; Emmerich et al., 1999; Jungnickel et al., 2000). However, these studies may underestimate the frequency of IκBα mutations in HRS cells, since one of the studies was restricted to the analysis of 3 of the 5 IκBα gene exons (Emmerich et al., 1999), and the study by Cabannes et al. focused on the detection of deletions or insertions, so that point mutations would have remained undetected (Cabannes et al., 1999). Taken together, these studies indicate that constitutive active NFκB in HRS cells is related to IκBα mutations in a fraction of the cases, the size of which is not yet known (other factors involved in the activation of NFκB are further discussed later). HRS cells express the Fas (CD95) receptor, which upon cross-linking normally induces cell death of receptor-expressing cells (Metkar et al., 1999; Xerri et al., 1995). As several HL-derived cell lines are resistant to apoptosis mediated by CD95 cross-linking (Metkar et al., 2000; Re et al., 2000), these cell lines were analyzed for somatic mutations in the death domain of the Fas receptor. However, none of seven cell lines analyzed harbored Fas mutations, and in only one of 10 cases of classical HL analyzed were death¨ domain mutations detected in primary HRS cells (Muschen et al., 2000b; Re et al., 2000). Intriguingly, in that case, a missense and a nonsense mutation were identiﬁed, and all HRS cells harbored either the missense or the nonsense mutation, suggesting that these mutations occurred as a late event in tumor development, but were likely of selective advantage to the tumor cells ¨ (Muschen et al., 2000b). A role for defective Fas signaling in HRS cells in the pathogenesis of HL is also indicated from the ﬁnding that patients with a germline mutation of the CD95 gene have a 50-fold increased risk of developing HL (Straus et al., 2001). Thus, although CD95 mutations can occur in a small fraction of cases of classical HL, in most cases the assumed Fas resistance is presumably due to other defects in the Fas signaling pathway or factors inhibiting Fas signaling (Ohshima et al., 2000).

V. PHENOTYPE AND GENE EXPRESSION PATTERNS OF HRS CELLS A. Similarities and Differences Compared with Normal B Cells In LP HL, the B-cell nature of the L&H cells is evident not only from the Ig gene studies discussed earlier, but also from the immunophenotype of

294

¨ Ralf Kuppers

Table II Expression of Immunophenotypic Markers by HRS and L&H Cellsa

Antigen

Normal expression

HRS cells in classical HL

L&H cells in LP HL

CD45 CD19 CD20 CD79a (mb-1) J chain BSAP/Pax-5 Oct-2 Bob-1 Immunoglobulin bcl-6 syndecan-1 Mum-1 CD3 Granzym B TIA-1 Perforin CD30 CD15 EMA TARC Fascin Restin

Leukocytes B cells B cells B cells B cells B cells B cells B cells B cells, plasma cells GC B and T cells Plasma cells Plasma cells, some centrocytes T cells Cytotoxic T cells, NK cells Cytotoxic T cells, NK cells Cytotoxic T cells, NK cells Some activated lymphocytes Granulocytes, monocytes Epithelial cells, some plasma cells Dendritic cells Dendritic cells, EBV+ B cell lines (LCL) Cultivated monocytes, EBV+ B cell lines

− −/+ −/+ −/+ − + − − − −/+ +/− + −/+ −/+ −/+ −/+ + +/− − + + +

+ + + + + + + + + + − − − − − − − − + − − −

a BSAP, B-cell speciﬁc activator protein; EMA, epithelial membrane antigen; LCL, lymphoblastoid cell lines; TARC, thymus and activation regulated cytokine; TIA-1, T-cell intracellular antigen 1.

the cells. L&H cells consistently express the B-cell markers CD20, J chain, CD79a (mb-1), Ig, Pax-5, and the B cell-speciﬁc transcription factors Oct-2 and BOB.1 (Chan, 1999; Foss et al., 1999; Hansmann et al., 1999; Kuzu et al., 1993; Pinkus and Said, 1988; Stein et al., 1986, 2001). Markers of other lineages, such as the T-cell marker CD3 and granzyme B, the dendritic markers fascin and TARC, or the granulocytic/monocytic marker CD15, are not expressed (Table II). A derivation from GC B cells, as suggested by the detection of ongoing Ig gene mutation (see earlier discussion), is also supported by expression of the transcriptional repressor bcl-6, which is a marker for GC B cells (Carbone et al., 1998) and the histological picture of the disease (Hansmann et al., 1999; Timens et al., 1986). HRS cells in classical HL show an immunophenotype that does not relate to any normal cell of the hematopoietic lineage. [A large-scale cDNA sequence analysis has reported a gene expression pattern of single Hodgkin cells resembling B cells (Cossman et al., 1999). However, as the data are based on a mixture of sequences from two L&H and two HRS cells, the

Hodgkin’s Lymphoma

295

“B-cell phenotype” may be partly or largely due to sequences derived from the L&H cells, which are known to resemble phenotypically B cells.] Many B-cell markers (CD19, CD20, J chain, CD79a and b) are either frequently not expressed or expressed only by a small proportion of the HRS cells in a given case (Drexler, 1992; Kuzu et al., 1993; Pinkus and Said, 1988; Stein et al., 1986; Watanabe et al., 2000). On the other hand, HRS cells in classical HL express a variety of markers that are usually not expressed on B cells (Table II). In many cases, HRS cells express CD15, a marker for granulocytes and monocytes, and HRS cells are apparently the only cells besides dendritic cells that secrete the chemokine TARC (Hsu and Jaffe, 1984; van den Berg et al., 1999). Other dendritic cell markers found on HRS cells include fascin and CD83 (Pinkus et al., 1997; Sorg et al., 1997). However, these latter markers are also expressed by EBV-transformed B cells in vitro, showing that they can be expressed by B cells in particular circumstances as well (Pinkus et al., 1997; Sorg et al., 1997). Ten to 15% of cases of classical HL show expression of one or more T-cell markers such as CD3, TIA-1, perforin, and granzyme B (the latter three cytotoxic molecules are also expressed by natural killer cells) (Felgar et al., 1997; Foss et al., 1996; Krenacs et al., 1997; Oudejans et al., 1996b). Although there are rare cases of HL that represent T-cell lymphomas, most of the cases with T-cell marker expression are derived from B cells (see earlier discussion). Although HRS cells in classical HL lack expression of most B lineage markers, some B-cell-associated proteins are nevertheless expressed by a large fraction of the cases. These markers include the transcription factors Pax-5, MUM1, and bcl-6, and the plasma cell marker CD138 (Carbone et al., 1998; Falini et al., 2000; Foss et al., 1999; Krenacs et al., 1998; Tsuboi et al., 2000). The expression of Pax-5 by HRS cells is puzzling, because several of the genes known to be positively regulated by this transcription factor (such as CD19 and CD79a) (Hagman et al., 2000) are usually not expressed by HRS cells. MUM1, bcl-6, and CD138 correspond to late stages of B-cell differentiation, since bcl-6 is normally expressed in GC B cells, MUM1 is found in a subset of centrocytes and plasma cells, and CD138 is a plasma cell marker (Carbone et al., 1998; Cattoretti et al., 1995; Falini et al., 2000; Tsuboi et al., 2000). Since the phenotype bcl-6−/CD138+ resembles normal plasma cells, there has been speculation that the subset of HL showing this expression pattern derives from post-GC B cells (Carbone et al., 1999). In line with this view, one could argue that the down-regulation of several B cell markers (such as CD20) also points to a plasma cell differentiation of HRS cells, since plasma cells lose expression of many B-cell markers. However, given the overall inconsistent expression of lineage markers in classical HL, one should be cautious in classifying HRS cells according to a small set of selected markers. More importantly, although HRS cells resemble post-GC B cells in some aspects phenotypically, the frequent detection of crippled V genes, which

296

¨ Ralf Kuppers

prevent normal GC B cells from differentiating to a post-GC stage of development (see above), argues that key steps in malignant transformation happen in the GC (or earlier, but become relevant in the GC), so that HRS cells in classical HL are likely to derive from transformed GC and not post-GC B cells. The distinction between L&H and HRS cells regarding the expression of B-cell-associated molecules is further confounded by studies of Ig gene expression. Whereas L&H cells show transcripts and protein for Ig, Ig is undetectable in HRS cells of classical HL, both at the RNA and protein level (Hell et al., 1993; Schmid et al., 1991; Stein et al., 2001; Stoler et al., 1995; von Wasielewski et al., 1996). Since BCR expression is the hallmark of a B cell, its absence in HRS cells further illustrates their aberrant phenotype. To further deﬁne the reasons for the lack of Ig transcription, HRS cells were analyzed for expression of transcription factors regulating the activity of Ig promoters. These studies revealed that HRS cells in classical HL (but not L&H cells in LP HL) do not express the B-cell speciﬁc transcription factors Oct-2 and Bob-1 (Re et al., 2001; Stein et al., 2001). Moreover, reintroduction of these factors into two HL-derived cell lines restored transcription of a reporter construct (Stein et al., 2001; Theil et al., 2001). Hence, downregulation of Oct-2 and Bob-1 is likely to be involved in the absence of Ig gene transcription. It has been argued that the identiﬁcation of down-regulated Oct-2 and Bob-1 expression as the likely cause for the lack of Ig transcription in HRS cells dismisses the concept that this lack of Ig transcription is due to crippling Ig mutations (Stein et al., 2001). However, this is a misinterpretation of the scenario that was proposed for the origin of HRS cells based on the presence of obviously crippling Ig mutations in cases of classical HL (see earlier discussion). The unexpected discovery of cases with crippled Ig receptors was the basis for a speculation about the histogenetic origin of HRS ¨ cells and not about issues of Ig transcription (Kanzler et al., 1996b; Kuppers and Rajewsky, 1998). “Crippled GC B cells” were deﬁned as cells that could not be appropriately selected by expression of a high-afﬁnity BCR, including cells with unfavorable replacement mutations. As replacement mutations as such do not abolish Ig transcription, the scenario of a crippled GC B-cell origin of HRS cells did not include any assumptions about Ig transcription ¨ in these tumor cells (Kanzler et al., 1996b; Kuppers and Rajewsky, 1998). Since the accumulation of crippling mutations by HRS cells is a priori not causing down-regulation of Ig transcription, it is an interesting question whether these events are related and which of these processes happens ﬁrst in HL development. The molecular features of somatic hypermutation argue against the idea that (crippling) mutations occur after Ig transcription is shut off: A number of mouse models showed that somatic hypermutation is closely associated with and dependent on Ig transcription, and that in particular initiation of transcription plays a role in hypermutation activity

Hodgkin’s Lymphoma

297

(Fukita et al., 1998; Peters and Storb, 1996; Reynaud et al., 2001). Thus, it is unlikely that the somatic mutations present in rearranged Ig genes of HRS cells—including the crippling mutations—occurred after down-regulation of Ig gene transcription. Moreover, if somatic mutations did continue to accumulate in HRS cells after loss of Ig transcription, one would expect to ﬁnd intraclonal sequence diversity in the V genes of these cells in classical HL, which is, however, not observed (see earlier discussion). Importantly, even if HRS cell precursors ﬁrst down-regulate Ig transcription and then continue to accumulate mutations, these cells would still represent crippled GC B cells, as survival of normal B cells strictly depends on BCR expression (Lam et al., 1997).

B. Expression of Apoptosis-Related Genes On the background of the proposed origin of HRS cells in classical HL from preapoptotic GC B cells, factors preventing apoptotic cell death may be particularly important in the pathogenesis of HL. While the transcription factor NFκB is activated in normal B cells only transiently in response to various stimuli, HRS cells in classical HL show constitutively active NFκB (Bargou et al., 1997). The functional importance of NFκB activity in HRS cells is supported by in vitro studies with HL-derived cell lines, which showed that inactivation of NFκB through dominant negative IκBα induces spontaneous apoptosis of the cells (Hinz et al., 2001; Izban et al., 2001). Hence, NFκB activity may be a key factor for the survival of HRS cells. Among the genes that are positively regulated by NFκB in HL cell lines, several antiapoptotic genes were identiﬁed, such as c-IAP2, Bﬂ-1/A1 and bcl-xL (Hinz et al., 2001). Nearly all cases of classical HL indeed show strong staining for bcl-xL on the HRS cells (Chu et al., 1999), and this molecule appears to be of particular importance for HRS cells, because ectopic expression of bcl-xL in HL cell lines in which NFκB activity is down-regulated rescues the cells from apoptosis (Hinz et al., 2001). It is likely that a number of factors are responsible for constitutive activity of NFκB in HRS cells. As already discussed, in a fraction of cases, somatic deleterious mutations of the IκBα gene may cause NFκB activity. In EBVpositive cases of HL, NFκB activity is likely induced by LMP-1 signaling (see earlier discussion). Indeed, all cases of HL with IκBα mutations reported so far are EBV-negative, supporting the idea that LMP-1 signaling and inactivating IκBα mutations are two alternative pathways for NFκB activation. Moreover, NFκB is induced by signaling through a number of receptors belonging to the TNF receptor family, for example, CD40. HRS cells express this receptor (Carbone et al., 1995a; O’Grady et al., 1994), and since T cells rosetting around HRS cells frequently express CD40 ligand (Carbone et al.,

298

¨ Ralf Kuppers

1995b), signaling through this receptor may further promote NFκB activation. Indeed, CD40 cross-linking on an HL cell line leads to further increase of NFκB activity (Annunziata et al., 2000). Besides bcl-xL, other members of the bcl-2 family are also often expressed by HRS cells, such as the proapoptotic family members Bax and Bad, and the antiapoptotic molecules bcl-2 and mcl-1 (Brousset et al., 1996, 1999). However, the expression level of the molecules can be quite variable between cases, and no functional studies of these factors in HRS cells are available, making it difﬁcult to deﬁne their role in HRS cell survival or apoptosis sensitivity. As discussed previously, HRS cells express the apoptosis-inducing CD95 receptor (Metkar et al., 1999) and are presumably resistant to CD95-mediated apoptosis (Metkar et al., 1999, 2000; Re et al., 2000; Verbeke et al., 2001). In a fraction of cases, somatic mutations in the CD95 gene may account for this resistance (see earlier discussion). HRS cell resistance to Fas-mediated apoptosis in cases lacking CD95 mutations must be due to other factors, such as expression of the decoy receptor 3, which binds to Fas ligand and inhibits Fas-ligand-induced apoptosis (Ohshima et al., 2000), or down-regulation of mediators of Fas signaling (Wrone-Smith et al., 2001).

VI. HRS CELLS IN THEIR MICROENVIRONMENT Usually, about 99% of the cells in HL lymph nodes are composed of various types of inﬂammatory cells. This inﬂammatory response may in part be an (unsuccessful) attempt to eliminate the tumor cells. However, it is likely that HRS cells depend on cellular interactions in the lymph node microenvironment for their survival and proliferation. These cells may therefore have evolved strategies to directly or indirectly attract other cells into the tissue. A dependence of HRS cells on interactions in their microenvironment is also supported by the difﬁculty in establishing in vitro growing HRS cell lines (Drexler, 1993; Wolf et al., 1996), by the failure to grow HRS cells in immunodeﬁcient mice (Kapp et al., 1993; Meggetto et al., 1996), and by the observation that these cells are usually undetectable in peripheral blood in primary cases of the disease (Irsch et al., 1998).

A. Chemokines and Cytokines Many of the effects of HRS cells on their microenvironment are mediated by the secretion of a large variety of chemokines and cytokines. Indeed, HL is often described as a tumor of aberrant cytokine and chemokine expression.

Hodgkin’s Lymphoma

299

The list of cytokines reported to be expressed in HL tissues includes Il-1, Il-3, Il-5, Il-6, Il-7, Il-9, Il-10, IL-13, TNFα, TNFβ, TGFβ, Mig, IP-10, RANTES, MIP1α, and MIP1β (Buri et al., 2001; Chan, 2001; Drexler, 1992; Haluska et al., 1994; Teruya-Feldstein et al., 1999). The following discussion is focused on the proposed effects of a selected number of these cytokines. HRS cells secrete large amounts of the chemokine TARC, which is known to attract CD4+ T cells of the TH2-type (Peh et al., 2001; van den Berg et al., 1999). Hence, the massive tumor inﬁltration by TH2-like CD4 T cells (see later discussion) is likely to be caused mainly by secretion of TARC, which is otherwise speciﬁcally produced by dendritic cells (Lieberam and ¨ Forster, 1999). L&H cells of LP HL do not produce detectable levels of TARC (Peh et al., 2001; van den Berg et al., 1999). Another chemokine that may be involved in the attraction of T helper cells is eotaxin. Whether HRS cells themselves produce this chemokine is controversial, but TNFα, which is secreted by HRS cells, can induce ﬁbroblasts to secrete eotaxin (Jundt et al., 1999; Teruya-Feldstein et al., 1999). Besides the attraction of T cells, eotaxin is also a potent chemoattractant for eosinophils. Therefore, the typical tissue eosinophilia observed in classical HL (Weiss et al., 1999) may be mediated by eotaxin secretion by ﬁbroblasts. Il-13 is a cytokine that is produced mainly by activated T cells and which has multiple effects on B cells, such as promoting survival and inducing Ig class switching to IgG4 and IgE. In HL tissues, Il-13 is strongly expressed by HRS cells (Kapp et al., 1999; Ohshima et al., 2001a; Skinnider et al., 2001). Interestingly, HRS cells express also the Il-13R (Ohshima et al., 2001a; Skinnider et al., 2001), suggesting an autocrine stimulation of these cells. Indeed, in vitro studies of HL-derived cell lines showed that inhibition of Il-13 signaling inhibits cell proliferation (Kapp et al., 1999). Il-13 may also play a role in the ﬁbrosis observed in nodular sclerosis HL through stimulation of Il-13R-expressing ﬁbroblasts. Il-13 is not or only weakly expressed by L&H cells of LP HL (Ohshima et al., 2001a; Skinnider et al., 2001). In EBV-positive cases of HL, the virally encoded latent membrane proteins LMP1 and LMP2a could potentially represent targets for a cytotoxic antitumor T cell response. However, such a response has not been observed. Several factors are presumably involved in preventing a cytotoxic attack on HRS cells (see also the later discussion). HRS cells secrete the cytokines Il-10 and TGFβ, both of which act immunosuppressively on CD8 T cells (Herbst et al., 1996; Newcom and Gu, 1995; Ohshima et al., 1995). Notably, Il-10 expression is particularly prominent in EBV-positive cases of HL, suggesting that EBV up-regulates Il-10 secretion, thereby contributing to the immune evasion of virus-harboring HRS cells. The Met protooncogene is a receptor tyrosine kinase that is activated by the hepatocyte growth factor (HGF). Met signaling in lymphocytes probably plays a role in the regulation of cell adhesion, migration, and survival, and

300

¨ Ralf Kuppers

among B cells, the receptor is speciﬁcally up-regulated in GC B cells (van der Voort et al., 2000). HRS cells were shown to express Met, and HGFpositive dendritic-reticulum cells were found scattered around HRS cells (Teoﬁli et al., 2001). Hence, constitutive activation of Met may represent a further signaling pathway between HRS cells and surrounding cells that is involved in the proliferation and survival of HRS cells.

B. HRS Cells as Antigen-Presenting Cells HRS cells are usually surrounded by rosettes of CD4-positive T helper cells. There is an intimate contact between HRS and these CD4+ T cells, mediated by several pairs of adhesion molecules, such as ICAM1 and LFA-1 on HRS cells and LFA-1 and -2 on T cells (Delabie et al., 1995; Ellis et al., 1992), and by CD40/CD40 ligand interaction (see above). Like professional antigen-presenting cells, HRS cells express high levels of MHC class II and also the costimulatory molecules CD80 and CD86 (Delabie et al., 1993; Munro et al., 1994; Poppema and Visser, 1994; Van Gool et al., 1997). However, the interaction of HRS cells with CD4 T cells is in several aspects unusual. First, it has been reported that a large fraction of the MHC class II molecules on the cell membrane of HRS cells are still associated with the invariant chain peptide (CLIP), which is normally replaced by an antigenic peptide before MHC molecules are presented on the cell surface (Bosshart and Jarrett, 1998). Hence, the MHC class II molecules may be partly nonfunctional as antigen presenters. Second, the CD4 T cells in the HL tissue show an unusual phenotype. The cells are CD45RO+CD45RA−, indicating a memory phenotype, and CD45RBdim, which is compatible with a TH2 phenotype (reviewed in Poppema and van den Berg, 2000). Moreover, the expression of CD38 and CD69 by these cells indicates cellular activation. However, the T cells lack expression of the CD26 activation marker and, as opposed to CD26-negative T cells from healthy humans, cannot be induced in vitro to up-regulate CD26 (Poppema and van den Berg, 2000). Furthermore, whereas the production of Il-4 and Il-5 and the lack of Il-2 production upon in vitro restimulation of T cells from HL tissues is typical for TH2-type T helper cells, the production of IFNγ is unusual for a TH2 T cell. Collectively, these features of the CD4 T cells are suggestive of a TH2like response, although the T-cell activation is incomplete and perhaps partly anergic (Poppema and van den Berg, 2000). Because of the intimate contact between CD4 T cells and HRS cells, the question has arisen whether this represents a speciﬁc interaction. If the rosetting T cells recognize a common antigen presented by the HRS cells, one might expect to ﬁnd a restricted repertoire of TCRs expressed by the T cells. However, neither an immunohistological study using TCR Vβ-family

Hodgkin’s Lymphoma

301

speciﬁc antibodies nor a single-cell analysis of TCRβ gene rearrangements ampliﬁed from micromanipulated T cells revealed any evidence for a restricted TCRβ repertoire of the rosetting T cells or a clonal expansion of these cells (Meggetto et al., 1994; Roers et al., 1998). These ﬁndings might suggest that HRS cells attract CD4 T cells nonspeciﬁcally. The failure of a cytotoxic T cell response to eliminate EBV-harboring HRS cells prompted studies of MHC class I expression in HL. These studies revealed that EBV-positive cases of HL usually show normal expression of class I molecules, whereas MHC class I is often undetectable on the HRS cells in EBV-negative cases (Lee et al., 1998; Murray et al., 1998; Oudejans et al., 1996a; van den Berg et al., 2000). Hence, in EBV-negative cases, a cytotoxic T cell response against the HRS cells is likely to be severely hampered by the low level or absence of MHC class I-associated antigen presentation. EBV-positive cases of HL show normal class I expression, and four of ﬁve HL cell lines were shown to be able to present EBV proteins to HLA class I-restricted cytotoxic T cell clones (Lee et al., 1998). Therefore, the failure of cytotoxic T cells to eliminate EBV-harboring HRS cells cannot be easily explained by insufﬁcient MHC class I-associated antigen presentation. As discussed earlier, the secretion of immunosuppressive cytokines by the HRS cells may play an important role in protecting HRS cells from a T-cell attack. Moreover, expression of the tumor antigen RCAS1 by HRS cells especially in EBV-positive cases of HL may allow the cells to evade immune surveillance, because RCAS1 is suspected to suppress clonal expansion and induce apoptosis of RCAS1-ligand-expressing T and NK cells (Ohshima et al., 2001b).

VII. CONCLUDING REMARKS Our understanding of HL, which has been an enigmatic disease for a long time, has made considerable progress in the past years. We know now that HRS cells represent clonal populations of tumor cells derived from GC B cells (or in rare cases T cells). An involvement of cell fusion in the generation of HRS cells is unlikely, and the identiﬁcation of shared precursors of HRS cells and B-cell non-Hodgkin lymphomas in several composite lymphomas directly demonstrates a close histogenetic relationship of HRS cells to other GC B-cell-derived lymphomas. HRS cells can persist and disseminate in the patient, further supporting their malignant nature. For a comprehensive understanding of the biology of HL, three main issues remain unresolved or only partially clariﬁed: (i) What are the transforming events involved in the pathogenesis of HL? (ii) What causes the strange morphology, immunophenotype, and gene-expression pattern of the HRS cells? (iii) Which cellular interactions are important for HRS cell survival and

302

¨ Ralf Kuppers

Fig. 2 Cellular interactions in classical HL. Shown are some of the direct cellular and cytokinemediated interactions in classical HL. More details and references are given in the text.

proliferation, and how do HRS cells establish the typical cellular inﬁltrate in the tumor tissue? For each of these questions, some partial answers have been found. As discussed earlier, EBV is likely to play an important role in the transformation of HRS cells in about half of cases, presumably through mimicking CD40 and BCR signaling by expression of LMP1 and LMP2a. In other cases, somatic mutations of the tumor suppressor gene IκBα may be causing constitutive NFκB activity, which may be a key event for HRS cell survival. However, other oncogenes and tumor suppressor genes analyzed so far were mutated only in rare cases or not at all, so that additional transforming events involved in HL pathogenesis remain to be identiﬁed. Signiﬁcant progress has been made in identifying pathways of cellular interactions in the HL microenvironment (Fig. 2). For example, the secretion of TARC by HRS cells is likely to play a key role in the attraction of CD4 T cells, eotaxin secretion by ﬁbroblasts may cause tissue eosinophilia, production of Il-10 and TGFβ may rescue HRS cells from cytotoxic attack, and autocrine stimulation through Il-13 may be important for HRS cell survival and proliferation. It is, nevertheless, likely that other cellular or cytokine-mediated interactions remain to be identiﬁed. An identiﬁcation of the major signaling pathways involved in these cellular interactions may also be of therapeutic value. Thus, strategies to abolish NFκB activity or inhibit autocrine Il-13 stimulation may help to eliminate HRS cells and thereby cure HL patients.

Hodgkin’s Lymphoma

303

Largely unknown are the factors that cause the peculiar phenotype of the HRS cells. Although derived from mature B cells, HRS cells were shown to have lost expression of most B-cell markers, and a mixture of markers typical for various other hematopoietic cell types is expressed (Table II). Some aspects of the morphology and immunophenotype of HRS cells may be related to the down-regulation of CD99 expression in HRS cells. CD99 is a glycosylated transmembrane protein that is expressed on many cell types, but the mechanism by which it functions is still unclear (Dworzak et al., 1994; Gelin et al., 1989). Repression of CD99 expression in B cells results in the appearance of multinuclear cells resembling HRS cells and partially expressing CD15 (Kim et al., 1998). Notably, CD99 is down-regulated by LMP1, presumably through NFκB activation, indicating a pathway for the generation of the typical HRS cell morphology in EBV-positive cases of HL (Kim et al., 2000; Lee et al., 2001). The immunophenotypic “re- or dedifferentiation” of HRS cells is in some aspects reminiscent of the situation in mice deﬁcient in the Pax-5 (BSAP) gene, which is a B-cell lineage commitment and maintenance factor (Nutt et al., 2001). In these animals, B lineage cells develop only to the pro-B-cell stage (i.e., cells that carry DHJH joints), as further development along the B-cell developmental pathway is blocked in the absence of Pax-5 (Nutt et al., 1999; Rolink et al., 1999). Intriguingly, these pro-B cells retain the capacity to give rise to T cells, monocytes, and other hematopoietic cell types (Nutt et al., 1999; Rolink et al., 1999). Moreover, Pax-5 deletion in mature B cells results in down-regulation of many B-cell antigens, and a myeloid-speciﬁc gene is induced (Horcher et al., 2001). This demonstrates a surprising plasticity and developmental ﬂexibility of lymphoid cells. In analogy to this model, one may speculate that a similar (incomplete) “redifferentiation” may take ¨ place in the development of HRS cells in classical HL (Kuppers et al., 2001a; Staudt, 2000). Loss of Pax-5 function is likely not involved in this process in HRS cells, as these cells usually express Pax-5 (Foss et al., 1999; Krenacs et al., 1998). If HRS cells indeed derive from preapoptotic GC B cells that are prevented from differentiating along their normal developmental pathway to memory or plasma cells because they lack a high-afﬁnity BCR, but are prevented from undergoing apoptosis because of some transforming event, these cells may perhaps attempt a redifferentiation process, resulting in the peculiar “mixed” phenotype. As B lineage cells are stringently selected for expression of an appropriate BCR throughout their life, loss of the B-cell phenotype of HRS cells may also represent a strategy aimed at escaping the selectional forces acting to eliminate all B cells lacking an appropriate antigen receptor. It is likely that many of the mysteries still surrounding the nature of the HRS cells will be uncovered by large-scale gene-expression proﬁling and the study of differential gene expression between HRS cells and their normal

304

¨ Ralf Kuppers

counterparts. First studies in this direction have already revealed exciting results, such as the detection of TARC by serial analysis of gene expression (SAGE) and the identiﬁcation of up-regulated Il-13 expression on a small low-density gene microarray (Kapp et al., 1999; van den Berg et al., 1999). Although these studies were performed with HL cell lines, an analysis of primary HRS cells—although technically much more demanding—will probably reveal additional features of HRS cells that are lost by the cells growing in vitro. Moreover, similar studies of the various other cell types in the lymphoma tissue will be helpful to further our understanding of the cellular interactions between HRS cells and the cellular inﬁltrate.

ACKNOWLEDGMENTS R.K. is supported by the Deutsche Forschungsgemeinschaft through a Heisenberg Award ¨ and SFB502. I thank Klaus Rajewsky, Andreas Brauninger, and Andrew Davy for valuable comments on the manuscript. I am most grateful to Klaus Rajewsky, Martin-Leo Hansmann, and Volker Diehl for continuous support and many stimulating discussions.

REFERENCES Alber, G., Kim, K. M., Weiser, P., Riesterer, C., Carsetti, R., and Reth, M. (1993). Curr. Biol. 3, 333–339. Alexander, F. E., Jarrett, R. F., Lawrence, D., Armstrong, A. A., Freeland, J., Gokhale, D. A., Kane, E., Taylor, G. M., Wright, D. H., and Cartwright, R. A. (2000). Br. J. Cancer 82, 1117–1121. Anagnostopoulos, I., Herbst, H., Niedobitek, G., and Stein, H. (1989). Blood 74, 810–816. Anagnostopoulos, I., Hansmann, M. L., Franssila, K., Harris, M., Harris, N. L., Jaffe, E. S., Han, J., van Krieken, J. M., Poppema, S., Maraﬁoti, T., Franklin, J., Sextro, M., Diehl, V., and Stein, H. (2000). Blood 96, 1889–1899. Annunziata, C. M., Saﬁran, Y. J., Irving, S. G., Kasid, U. N., and Cossman, J. (2000). Blood 96, 2841–2848. Arpin, C., Dechanet, J., Van Kooten, C., Merville, P., Grouard, G., Briere, F., Banchereau, J., and Liu, Y. J. (1995). Science 268, 720–722. Babcock, G. J., Hochberg, D., and Thorley-Lawson, D. A. (2000). Immunity 13, 497–506. Baeuerle, P. A., and Henkel, T. (1994). Annu. Rev. Immunol. 12, 141–179. Bargou, R. C., Emmerich, F., Krappmann, D., Bommert, K., Mapara, M. Y., Arnold, W., Royer, ¨ H. D., Grinstein, E., Greiner, A., Scheidereit, C., and Dorken, B. (1997). J Clin Invest 100, 2961–2969. Barrios, L., Caballin, M. R., Miro, R., Fuster, C., Berrozpe, G., Subias, A., Batlle, X., and Egozcue, J. (1988). Hum. Genet. 78, 320–324. Berek, C., Berger, A., and Apel, M. (1991). Cell 67, 1121–1129. Bosshart, H., and Jarrett, R. F. (1998). Blood 92, 2252–2259. ¨ Braeuninger, A., Kuppers, R., Strickler, J. G., Wacker, H. H., Rajewsky, K., and Hansmann, M. L. (1997). Proc. Natl. Acad. Sci. USA 94, 9337–9342.

Hodgkin’s Lymphoma

305

¨ Brauninger, A., Hansmann, M. L., Strickler, J. G., Dummer, R., Burg, G., Rajewsky, K., and ¨ Kuppers, R. (1999). N. Engl. J. Med. 340, 1239–1247. Brousset, P., Benharroch, D., Krajewski, S., Laurent, G., Meggetto, F., Rigal-Huguet, F., Gopas, J., Prinsloo, I., Pris, J., Delsol, G., Reed, J. C., and Schlaifer, D. (1996). Blood 87, 2470–2475. Brousset, P., Krajewski, S., Schlaifer, D., Reed, J. C., and Delsol, G. (1999). Leuk. Lymphoma 34, 581–584. Brown, K. D., Hostager, B. S., and Bishop, G. A. (2001). J. Exp. Med. 193, 943–954. Bucsky, P. (1987). Blut 55, 413–420. Buri, C., Korner, M., Scharli, P., Cefai, D., Uguccioni, M., Mueller, C., Laissue, J. A., and Mazzucchelli, L. (2001). Blood 97, 1543–1548. Cabannes, E., Khan, G., Aillet, F., Jarrett, R. F., and Hay, R. T. (1999). Oncogene 18, 3063– 3070. Caldwell, R. G., Wilson, J. B., Anderson, S. J., and Longnecker, R. (1998). Immunity 9, 405– 411. Carbone, A., Gloghini, A., Gattei, V., Aldinucci, D., Degan, M., De Paoli, P., Zagonel, V., and Pinto, A. (1995a). Blood 85, 780–789. Carbone, A., Gloghini, A., Gruss, H. J., and Pinto, A. (1995b). Am. J. Pathol. 147, 912– 922. Carbone, A., Gloghini, A., Gaidano, G., Franceschi, S., Capello, D., Drexler, H. G., Falini, B., and Dalla-Favera, R. (1998). Blood 92, 2220–2228. Carbone, A., Gloghini, A., Larocca, L. M., Antinori, A., Falini, B., Tirelli, U., Dalla-Favera, R., and Gaidano, G. (1999). Blood 93, 2319–2326. Cattoretti, G., Chang, C. C., Cechova, K., Zhang, J., Ye, B. H., Falini, B., Louie, D. C., Ofﬁt, K., Chaganti, R. S. K., and Dalla-Favera, R. (1995). Blood 86, 45–53. Chan, W. C. (1999). Semin. Hematol. 36, 242–252. Chan, W. C. (2001). Hematol. Oncol. 19, 1–17. Chen, H., Lee, J. M., Zong, Y., Borowitz, M., Ng, M. H., Ambinder, R. F., and Hayward, S. D. (2001). J. Virol. 75, 2929–2937. Chen, W. G., Chen, Y. Y., Kamel, O. W., Koo, C. H., and Weiss, L. M. (1996). Lab. Invest. 75, 519–527. Chu, W. S., Aguilera, N. S., Wei, M. Q., and Abbondanzo, S. L. (1999). Hum. Pathol. 30, 1065–1070. Cossman, J., Annunziata, C. M., Barash, S., Staudt, L., Dillon, P., He, W. W., RicciardiCastagnoli, P., Rosen, C. A., and Carter, K. C. (1999). Blood 94, 411–416. Delabie, J., Ceuppens, J. L., Vandenberghe, P., de Boer, M., Coorevits, L., and De Wolf-Peeters, C. (1993). Blood 82, 2845–2852. Delabie, J., Tierens, A., Wu, G., Weisenburger, D. D., and Chan, W. C. (1994). Blood 84, 3291–3298. Delabie, J., Chan, W. C., Weisenburger, D. D., and De Wolf-Peeters, C. (1995). Leuk. Lymphoma 18, 35–40. Delabie, J., Tierens, A., Gavriil, T., Wu, G., Weisenburger, D. D., and Chan, W. C. (1996). Br. J. Haematol. 94, 198–205. Devergne, O., McFarland, E. C., Mosialos, G., Izumi, K. M., Ware, C. F., and Kieff, E. (1998). J. Virol. 72, 7900–7908. Drexler, H. G. (1992). Leuk. Lymphoma 8, 283–313. Drexler, H. G. (1993). Leuk. Lymphoma 9, 1–25. Drexler, H. G., Dirks, W. G., and MacLeod, R. A. (1999). Leukemia 13, 1601–1607. Dworzak, M. N., Fritsch, G., Buchinger, P., Fleischer, C., Printz, D., Zellner, A., Schollhammer, A., Steiner, G., Ambros, P. F., and Gadner, H. (1994). Blood 83, 415–425. Ellis, P. A., Hart, D. N., Colls, B. M., Nimmo, J. C., MacDonald, J. E., and Angus, H. B. (1992). Clin. Exp. Immunol. 90, 117–123.

306

¨ Ralf Kuppers

Emmerich, F., Meiser, M., Hummel, M., Demel, G., Foss, H. D., Jundt, F., Mathas, S., ¨ Krappmann, D., Scheidereit, C., Stein, H., and Dorken, B. (1999). Blood 94, 3129–3134. Falini, B., Fizzotti, M., Pucciarini, A., Bigerna, B., Maraﬁoti, T., Gambacorta, M., Pacini, R., Alunni, C., Natali-Tanci, L., Ugolini, B., Sebastiani, C., Cattoretti, G., Pileri, S., Dalla-Favera, R., and Stein, H. (2000). Blood 95, 2084–2092. Falzetti, D., Crescenzi, B., Matteuci, C., Falini, B., Martelli, M. F., Van Den Berghe, H., and Mecucci, C. (1999). Haematologica 84, 298–305. Faumont, N., Al Saati, T., Brousset, P., Offer, C., Delsol, G., and Meggetto, F. (2001). J. Gen. Virol. 82, 1169–1174. Felgar, R. E., Macon, W. R., Kinney, M. C., Roberts, S., Pasha, T., and Salhany, K. E. (1997). Am. J. Pathol. 150, 1893–1900. Fonatsch, C., Gradl, G., and Rademacher, J. (1989). Recent Results Cancer Res. 117, 35–49. Foss, H. D., Anagnostopoulos, I., Araujo, I., Assaf, C., Demel, G., Kummer, J. A., Hummel, M., and Stein, H. (1996). Blood 88, 4005–4011. Foss, H. D., Reusch, R., Demel, G., Lenz, G., Anagnostopoulos, I., Hummel, M., and Stein, H. (1999). Blood 94, 3108–3113. Franke, S., Wlodarska, I., Maes, B., Vandenberghe, P., Delabie, J., Hagemeijer, A., and De Wolf-Peeters, C. (2001). Blood 97, 1845–1853. Fukita, Y., Jacobs, H., and Rajewsky, K. (1998). Immunity 9, 105–114. Gelin, C., Aubrit, F., Phalipon, A., Raynal, B., Cole, S., Kaczorek, M., and Bernard, A. (1989). EMBO J. 8, 3253–3259. Gires, O., Kohlhuber, F., Kilger, E., Baumann, M., Kieser, A., Kaiser, C., Zeidler, R., Scheffer, B., Uefﬁng, M., and Hammerschmidt, W. (1999). EMBO J. 18, 3064–3073. ¨ Goossens, T., Klein, U., and Kuppers, R. (1998). Proc. Natl. Acad. Sci. USA 95, 2463–2468. Gopas, J., Shemer-Avni, Y., Levy, A., Ariad, S., Bazarsky, E., Barchana, M., Prinsloo, I., RagerZisman, B., Sacks, M., and Benharroch, D. (2001). Proc. Am. Assoc. Cancer Res. 42, 111 (Abstract). Gravel, S., Delsol, G., and Al Saati, T. (1998). Blood 91, 2866–2874. Gupta, R. K., Patel, K., Bodmer, W. F., and Bodmer, J. G. (1993). Proc. Natl. Acad. Sci. USA 90, 2817–2821. Hagman, J., Wheat, W., Fitzsimmons, D., Hodsdon, W., Negri, J., and Dizon, F. (2000). Curr. Top. Microbiol. Immunol. 245, 169–194. Haluska, F. G., Brufsky, A. M., and Canellos, G. P. (1994). Blood 84, 1005–1019. Hansmann, M.-L., Weiss, L. M., Stein, H., Harris, N. L., and Jaffe, E. S. (1999). In “Hodgkin’s Disease” (P. M. Mauch, J. O. Armitage, V. Diehl, R. T. Hoppe, and L. M. Weiss, eds.), pp. 169–180. Lippencott Williams & Wilkins, Philadelphia. Harris, N. L. (1999). Semin. Hematol. 36, 220–232. Hell, K., Pringle, J. H., Hansmann, M. L., Lorenzen, J., Colloby, P., Lauder, I., and Fischer, R. (1993). J. Pathol. 171, 137–143. Herbst, H., Foss, H. D., Samol, J., Araujo, I., Klotzbach, H., Krause, H., Agathanggelou, A., Niedobitek, G., and Stein, H. (1996). Blood 87, 2918–2929. Hinz, M., Loser, P., Mathas, S., Krappmann, D., Dorken, B., and Scheidereit, C. (2001). Blood 97, 2798–2807. Hjalgrim, H., Askling, J., Sorensen, P., Madsen, M., Rosdahl, N., Storm, H. H., HamiltonDutoit, S., Eriksen, L. S., Frisch, M., Ekbom, A., and Melbye, M. (2000). J. Natl. Cancer Inst. 92, 1522–1528. Hodgkin, T. (1832). Med. Chir. Trans. 17, 68–114. Horcher, M., Souabni, A., and Busslinger, M. (2001). Immunity 14, 779–790. Hsu, S. M., and Jaffe, E. S. (1984). Am. J. Clin. Pathol. 82, 29–32. Hummel, M., Ziemann, K., Lammert, H., Pileri, S., Sabattini, E., and Stein, H. (1995). N. Engl. J. Med. 333, 901–906.

Hodgkin’s Lymphoma

307

Inghirami, G., Macri, L., Rosati, S., Zhu, B. Y., Yee, H. T., and Knowles, D. M. (1994). Proc. Natl. Acad. Sci. USA 91, 9842–9846. ¨ Irsch, J., Nitsch, S., Hansmann, M. L., Rajewsky, K., Tesch, H., Diehl, V., Jox, A., Kuppers, R., and Radbruch, A. (1998). Proc. Natl. Acad. Sci. USA 95, 10117–10122. Irsch, J., Wolf, J., Tesch, H., Diehl, V., Radbruch, A., and Staratschek-Jox, A. (2001). Br. J. Haematol. 113, 785–793. Izban, K. F., Ergin, M., Huang, Q., Qin, J., Martinez, R. L., Schnitzer, B., Ni, H., Nickoloff, B. J., and Alkan, S. (2001). Mod. Pathol. 14, 297–310. Jacob, J., Kelsoe, G., Rajewsky, K., and Weiss, U. (1991). Nature 354, 389–392. Jaffe, E. S., and Mueller-Hermelink, K. (1999). In “Hodgkin’s Disease” (P. M. Mauch, J. O. Armitage, V. Diehl, R. T. Hoppe, and L. M. Weiss, eds.) pp. 181–193. Lippencott Williams & Wilkins, Philadelphia. Jansen, M. P., Hopman, A. H., Bot, F. J., Haesevoets, A., Stevens-Kroef, M. J., Arends, J. W., Jox, A., Wolf, J., Ramaekers, F. C., and Schouten, H. C. (1999). J. Pathol. 189, 527–532. Jarrett, R. F., and MacKenzie, J. (1999). Semin. Hematol. 36, 260–269. ¨ Joos, S., Kupper, M., Ohl, S., von Bonin, F., Mechtersheimer, G., Bentz, M., Marynen, P., ¨ Moller, P., Pfreundschuh, M., Trumper, L., and Lichter, P. (2000). Cancer Res. 60, 549–552. ¨ Jox, A., Zander, T., Kornacker, M., Kanzler, H., Kuppers, R., Diehl, V., and Wolf, J. (1998). Ann. Oncol. 9, 283–287. ¨ Jox, A., Zander, T., Kuppers, R., Irsch, J., Kanzler, H., Kornacker, M., Bohlen, H., Diehl, V., and Wolf, J. (1999). Blood 93, 3964–3972. Jundt, F., Anagnostopoulos, I., Bommert, K., Emmerich, F., Muller, G., Foss, H. D., Royer, H. ¨ D., Stein, H., and Dorken, B. (1999). Blood 94, 2065–2071. ¨ Jungnickel, B., Staratschek-Jox, A., Brauninger, A., Spieker, T., Wolf, J., Diehl, V., Hansmann, ¨ M.-L., Rajewsky, K., and Kuppers, R. (2000). J. Exp. Med. 191, 395–401. ¨ Kanzler, H., Hansmann, M. L., Kapp, U., Wolf, J., Diehl, V., Rajewsky, K., and Kuppers, R. (1996a). Blood 87, 3429–3436. ¨ Kanzler, H., Kuppers, R., Hansmann, M. L., and Rajewsky, K. (1996b). J. Exp. Med. 184, 1495–1505. ¨ Kanzler, H., Kuppers, R., Helmes, S., Wacker, H. H., Chott, A., Hansmann, M. L., and Rajewsky, K. (2000). Blood 95, 1023–1031. Kapp, U., Wolf, J., Hummel, M., Pawlita, M., von Kalle, C., Dallenbach, F., Schwonzen, M., ¨ Krueger, G. R. F., Muller-Lantzsch, N., Fonatsch, C., Stein, H., and Diehl, V. (1993). Blood 82, 1247–1256. Kapp, U., Yeh, W. C., Patterson, B., Elia, A. J., Kagi, D., Ho, A., Hessel, A., Tipsword, M., Williams, A., Mirtsos, C., Itie, A., Moyle, M., and Mak, T. W. (1999). J. Exp. Med. 189, 1939–1946. Kaye, K. M., Izumi, K. M., and Kieff, E. (1993). Proc. Natl. Acad. Sci. USA 90, 9150–9154. Khan, G., Miyashita, E. M., Yang, B., Babcock, G. J., and Thorley-Lawson, D. A. (1996). Immunity 5, 173–179. Kilger, E., Kieser, A., Baumann, M., and Hammerschmidt, W. (1998). EMBO J. 17, 1700–9. Kim, S. H., Choi, E. Y., Shin, Y. K., Kim, T. J., Chung, D. H., Chang, S. I., Kim, N. K., and Park, S. H. (1998). Blood 92, 4287–4295. Kim, S. H., Shin, Y. K., Lee, I. S., Bae, Y. M., Sohn, H. W., Suh, Y. H., Ree, H. J., Rowe, M., and Park, S. H. (2000). Blood 95, 294–300. Kitagawa, N., Goto, M., Kurozumi, K., Maruo, S., Fukayama, M., Naoe, T., Yasukawa, M., Hino, K., Suzuki, T., Todo, S., and Takada, K. (2000). EMBO J. 19, 6742–6750. ¨ Klein, U., Goossens, T., Fischer, M., Kanzler, H., Braeuninger, A., Rajewsky, K., and Kuppers, R. (1998). Immunol. Rev. 162, 261–280. Krenacs, L., Wellmann, A., Sorbara, L., Himmelmann, A. W., Bagdi, E., Jaffe, E. S., and Raffeld, M. (1997). Blood 89, 980–989.

308

¨ Ralf Kuppers

Krenacs, L., Himmelmann, A. W., Quintanilla-Martinez, L., Fest, T., Riva, A., Wellmann, A., Bagdi, E., Kehrl, J. H., Jaffe, E. S., and Raffeld, M. (1998). Blood 92, 1308–1316. Kube, D., Holtick, U., Vockerodt, M., Ahmadi, T., Behrmann, I., Heinrich, P. C., Diehl, V., and Tesch, H. (2001). Blood 98, 762–770. Kulwichit, W., Edwards, R. H., Davenport, E. M., Baskar, J. F., Godfrey, V., and Raab-Traub, N. (1998). Proc. Natl. Acad. Sci. USA 95, 11963–11968. ¨ ¨ Kupper, M., Joos, S., Von Bonin, F., Daus, H., Pfreundschuh, M., Lichter, P., and Trumper, L. (2001). Br. J. Haematol. 112, 768–75. ¨ Kuppers, R., and Rajewsky, K. (1998). Annu. Rev. Immunol. 16, 471–493. ¨ Kuppers, R., Zhao, M., Hansmann, M. L., and Rajewsky, K. (1993). EMBO J. 12, 4955–4967. ¨ Kuppers, R., Rajewsky, K., Zhao, M., Simons, G., Laumann, R., Fischer, R., and Hansmann, M. L. (1994). Proc. Natl. Acad. Sci. USA 91, 10962–10966. ¨ Kuppers, R., Kanzler, H., Hansmann, M. L., and Rajewsky, K. (1996). N. Engl. J. Med. 334, 404. ¨ Kuppers, R., Rajewsky, K., and Hansmann, M. L. (1997). Eur. J. Immunol. 27, 1398–1405. ¨ Kuppers, R., Rajewsky, K., Braeuninger, A., and Hansmann, M. L. (1998). N. Engl. J. Med. 338, 763–764. ¨ Kuppers, R., Klein, U., Hansmann, M.-L., and Rajewsky, K. (1999). N. Engl. J. Med. 341, 1520–1529. ¨ ¨ ¨ Kuppers, R., Brauninger, A., Muschen, M., Distler, V., Hansmann, M. L., and Rajewsky, K. (2001a). Blood 97, 818–821. ¨ Kuppers, R., Sousa, A. B., Baur, A. S., Strickler, J. G., Rajewsky, K., and Hansmann, M. L. (2001b). Mol. Med. 7, 285–292. Kurth, J., Spieker, T., Wustrow, J., Strickler, G. J., Hansmann, L. M., Rajewsky, K., and Kuppers, R. (2000). Immunity 13, 485–495. Kuzu, I., Delsol, G., Jones, M., Gatter, K. C., and Mason, D. Y. (1993). Histopathol. 22, 141–144. ¨ Lam, K. P., Kuhn, R., and Rajewsky, K. (1997). Cell 90, 1073–1083. Lebecque, S., de Bouteiller, O., Arpin, C., Banchereau, J., and Liu, Y. J. (1997). J. Exp. Med. 185, 563–571. Lee, I., Kim, M. K., Choi, E. Y., Mehl, A., Jung, K. C., Gil, M. C., Rowe, M., and Park, S. H. (2001). Blood 97, 3596–3604. Lee, S. P., Constandinou, C. M., Thomas, W. A., Croom-Carter, D., Blake, N. W., Murray, P. G., Crocker, J., and Rickinson, A. B. (1998). Blood 92, 1020–1030. ¨ Lieberam, I., and Forster, I. (1999). Eur. J. Immunol. 29, 2684–2694. Liu, Y. J., Joshua, D. E., Williams, G. T., Smith, C. A., Gordon, J., and MacLennan, I. C. (1989). Nature 342, 929–931. Longnecker, R. (2000). Adv. Cancer Res. 79, 175–200. MacLennan, I. C. (1994). Annu. Rev. Immunol. 12, 117–139. MacLennan, I. C., Liu, Y. J., and Johnson, G. D. (1992). Immunol. Rev. 126, 143–161. Maggio, E. M., Stekelenburg, E., Van den Berg, A., and Poppema, S. (2001). Int. J. Cancer, 94, 60–66. Maraﬁoti, T., Hummel, M., Anagnostopoulos, I., Foss, H. D., Falini, B., Delsol, G., Isaacson, P. G., Pileri, S., and Stein, H. (1997). N. Engl. J. Med. 337, 453–458. Maraﬁoti, T., Hummel, M., Anagnostopoulos, I., Foss, H. D., Huhn, D., and Stein, H. (1999). J. Clin. Oncol. 17, 3804–3809. Maraﬁoti, T., Hummel, M., Foss, H.-D., Laumen, H., Korbjuhn, P., Anagnostopoulos, I., Lammert, H., Demel, G., Theil, J., Wirth, T., and Stein, H. (2000). Blood 95, 1443–1450. Meggetto, F., al Saati, T., Rubin, B., and Delsol, G. (1994). Br. J. Haematol. 86, 524–532. Meggetto, F., Muller, C., Henry, S., Selves, J., Mariame, B., Brousset, P., Saati, T. A., and Delsol, G. (1996). Blood 87, 2435–2442.

Hodgkin’s Lymphoma

309

Metkar, S. S., Naresh, K. N., Redkar, A. A., Soman, C. S., Advani, S. H., and Nadkarni, J. J. (1999). Leuk. Lymphoma 33, 521–530. Metkar, S. S., Anand, M., Manna, P. P., Naresh, K. N., and Nadkarni, J. J. (2000). Exp. Cell Res. 255, 18–29. Michels, K. B. (1995). Eur. J. Cancer Prev. 4, 379–388. Miura, I., Tamura, A., Taniwaki, M., Nakamura, S., Nakamine, H., Yoshino, T., Ichinohasama, R., and Miura, A. B. (2000). Cancer Genet. Cytogenet. 123, 97–101. ¨ Montesinos-Rongen, M., Roers, A., Kuppers, R., Rajewsky, K., and Hansmann, M.-L. (1999). Blood 94, 1755–1760. Mosialos, G., Birkenbach, M., Yalamanchili, R., VanArsdale, T., Ware, C., and Kieff, E. (1995). Cell 80, 389–399. Munoz, N., Davidson, R. J., Witthoff, B., Ericsson, J. E., and De-The, G. (1978). Int. J. Cancer 22, 10–13. Munro, J. M., Freedman, A. S., Aster, J. C., Gribben, J. G., Lee, N. C., Rhynhart, K. K., Banchereau, J., and Nadler, L. M. (1994). Blood 83, 793–798. Murray, P. G., Constandinou, C. M., Crocker, J., Young, L. S., and Ambinder, R. F. (1998). Blood 92, 2477–2483. ¨ ¨ Muschen, M., Rajewsky, K., Brauninger, A., Baur, A. S., Oudejans, J. J., Roers, A., Hansmann, ¨ M.-L., and Kuppers, R. (2000a). J. Exp. Med. 191, 387–394. ¨ ¨ ¨ Muschen, M., Re, D., Brauninger, A., Wolf, J., Hansmann, M. L., Diehl, V., Kuppers, R., and Rajewsky, K. (2000b). Cancer Res. 60, 5640–5643. ¨ ¨ ¨ Muschen, M., Kuppers, R., Spieker, T., Brauninger, A., Rajewsky, K., and Hansmann, M. L. (2001). Lab. Invest. 81, 289–295. Newcom, S. R., and Gu, L. (1995). J. Clin. Pathol. 48, 160–163. Nutt, S. L., Heavey, B., Rolink, A., and Busslinger, M. (1999). Nature 401, 556–562. Nutt, S. L., Eberhard, D., Horcher, M., Rolink, A. G., and Busslinger, M. (2001). Int. Rev. Immunol. 20, 65–82. O’Grady, J. T., Stewart, S., Lowrey, J., Howie, S. E., and Krajewski, A. S. (1994). Am. J. Pathol. 144, 21–26. Ohno, T., Stribley, J. A., Wu, G., Hinrichs, S. H., Weisenburger, D. D., and Chan, W. C. (1997). N. Engl. J. Med. 337, 459–465. Ohno, T., Smir, B. N., Weisenburger, D. D., Gascoyne, R. D., Hinrichs, S. D., and Chan, W. C. (1998a). Blood 91, 1757–1761. Ohno, T., Trenn, G., Wu, G., Abou-Elella, A., Reis, H. E., and Chan, W. C. (1998b). Mod. Pathol. 11, 485–490. Ohshima, K., Suzumiya, J., Akamatu, M., Takeshita, M., and Kikuchi, M. (1995). Int. J. Cancer 62, 5–10. Ohshima, K., Ishiguro, M., Ohgami, A., Sugihara, M., Haraoka, S., Suzumiya, J., and Kikuchi, M. (1999). Int. J. Cancer 82, 250–255. Ohshima, K., Haraoka, S., Sugihara, M., Suzumiya, J., Kawasaki, C., Kanda, M., and Kikuchi, M. (2000). Cancer Lett. 160, 89–97. Ohshima, K., Akaiwa, M., Umeshita, R., Suzumiya, J., Izuhara, K., and Kikuchi, M. (2001a). Histopathol. 38, 368–375. Ohshima, K., Muta, K., Nakashima, M., Haraoka, S., Tutiya, T., Suzumiya, J., Kawasaki, C., Watanabe, T., and Kikuchi, M. (2001b). Int. J. Cancer 93, 91–96. Orscheschek, K., Merz, H., Hell, J., Binder, T., Bartels, H., and Feller, A. C. (1995). Lancet 345, 87–90. Oudejans, J. J., Jiwa, N. M., Kummer, J. A., Horstman, A., Vos, W., Baak, J. P., Kluin, P. M., van der Valk, P., Walboomers, J. M., and Meijer, C. J. (1996a). Blood 87, 3844–3851. Oudejans, J. J., Kummer, J. A., Jiwa, M., van der Valk, P., Ossenkoppele, G. J., Kluin, P. M., Kluin-Nelemans, J. C., and Meijer, C. J. (1996b). Am. J. Pathol. 148, 233–240.

310

¨ Ralf Kuppers

Peh, S. C., Kim, L. H., and Poppema, S. (2001). Am. J. Surg. Pathol. 25, 925–929. Peters, A., and Storb, U. (1996). Immunity 4, 57–65. Pinkus, G. S., and Said, J. W. (1988). Am. J. Pathol. 133, 211–217. Pinkus, G. S., Pinkus, J. L., Langhoff, E., Matsumura, F., Yamashiro, S., Mosialos, G., and Said, J. W. (1997). Am. J. Pathol. 150, 543–562. Poppema, S., and van den Berg, A. (2000). Semin. Cancer Biol. 10, 345–350. Poppema, S., and Visser, L. (1994). Am. J. Pathol. 145, 37–41. Poppema, S., Kaleta, J., and Hepperle, B. (1992). J. Natl. Cancer Inst. 84, 1789–1793. Rajewsky, K. (1996). Nature 381, 751–758. Razzouk, B. I., Srinivas, S., Sample, C. E., Singh, V., and Sixbey, J. W. (1996). J. Infect. Dis. 173, 529–535. Re, D., Hofmann, A., Wolf, J., Diehl, V., and Staratschek-Jox, A. (2000). Exp. Hematol. 28, 31–35. ¨ Re, D., Muschen, M., Ahmadi, T., Wickenhauser, C., Staratschek-Jox, A., Holtick, U., Diehl, V., and Wolf, J. (2001). Cancer Res. 61, 2080–2084. Reynaud, C. A., Frey, S., Aoufouchi, S., Faili, A., Bertocci, B., Dahan, A., Flatter, E., Delbos, F., Storck, S., Zober, C., and Weill, J. C. (2001). Phil. Trans. R. Soc. Lond. B Biol. Sci. 356, 91–97. ¨ Roers, A., Montesinos-Rongen, M., Hansmann, M. L., Rajewsky, K., and Kuppers, R. (1998). Eur. J. Immunol. 28, 2424–2431. Rolink, A. G., Nutt, S. L., Melchers, F., and Busslinger, M. (1999). Nature 401, 603–606. Rosdahl, N., Larsen, S. O., and Clemmesen, J. (1974). Br. Med. J. 2, 253–256. ¨ Roth, J., Daus, H., Trumper, L., Gause, A., Salamon-Looijen, M., and Pfreundschuh, M. (1994). Int. J. Cancer 57, 799–804. Sarris, A. H., Jhanwar, S. C., and Cabanillas, F. (1999). In “Hodgkin’s Disease” (P. M. Mauch, J. O. Armitage, V. Diehl, R. T. Hoppe, and L. M. Weiss, eds.), pp. 195–212. Lippencott Williams & Wilkins, Philadelphia. Schmid, C., Sargent, C., and Isaacson, P. G. (1991). Am. J. Pathol. 139, 1281–1289. Seitz, V., Hummel, M., Maraﬁoti, T., Anagnostopoulos, I., Assaf, C., and Stein, H. (2000). Blood 95, 3020–3024. Skinnider, B. F., Elia, A. J., Gascoyne, R. D., Trumper, L. H., von Bonin, F., Kapp, U., Patterson, B., Snow, B. E., and Mak, T. W. (2001). Blood 97, 250–255. Sorg, U. R., Morse, T. M., Patton, W. N., Hock, B. D., Angus, H. B., Robinson, B. A., Colls, B. M., and Hart, D. N. (1997). Pathology 29, 294–299. ¨ ¨ A., and Hansmann, M. L. (2000). Spieker, T., Kurth, J., Kuppers, R., Rajewsky, K., Brauninger, Blood 96, 3133–3138. ¨ Staratschek-Jox, A., Kotkowski, S., Belge, G., Rudiger, T., Bullerdiek, J., Diehl, V., and Wolf, J. (2000). Am. J. Pathol. 156, 209–216. Staudt, L. M. (2000). J. Exp. Med. 191, 207–212. Stein, H., Hansmann, M. L., Lennert, K., Brandtzaeg, P., Gatter, K. C., and Mason, D. Y. (1986). Am. J. Clin. Pathol. 86, 292–297. Stein, H., Maraﬁoti, T., Foss, H. D., Laumen, H., Hummel, M., Anagnostopoulos, I., Wirth, T., Demel, G., and Falini, B. (2001). Blood 97, 496–501. Stoler, M. H., Nichols, G. E., Symbula, M., and Weiss, L. M. (1995). Am. J. Pathol. 146, 812–818. Straus, S. E., Jaffe, E. S., Puck, J. M., Dale, J. K., Elkon, K. B., Rosen-Wolff, A., Peters, A. M., Sneller, M. C., Hallahan, C. W., Wang, J., Fischer, R. E., Jackson, C. M., Lin, A. Y., Baumler, C., Siegert, E., Marx, A., Vaishnaw, A. K., Grodzicky, T., Fleisher, T. A., and Lenardo, M. J. (2001). Blood 98, 194–200. Teerenhovi, L., Lindholm, C., Pakkala, A., Franssila, K., Stein, H., and Knuutila, S. (1988). Cancer Genet. Cytogenet. 34, 305–311.

Hodgkin’s Lymphoma

311

Teoﬁli, L., Di Febo, A. L., Pierconti, F., Maggiano, N., Bendandi, M., Rutella, S., Cingolani, A., Di Renzo, N., Musto, P., Pileri, S., Leone, G., and Larocca, L. M. (2001). Blood 97, 1063–1069. Teruya-Feldstein, J., Jaffe, E., Burd, P. R., Kingma, D. W., Setsuda, J. E., and Tosato, G. (1999). Blood 93, 2463–2470. Theil, J., Laumen, H., Maraﬁoti, T., Hummel, M., Lenz, G., Wirth, T., and Stein, H. (2001). Blood 97, 3191–3196. Timens, W., Visser, L., and Poppema, S. (1986). Lab. Invest. 54, 457–461. ¨ Trumper, L. H., Brady, G., Bagg, A., Gray, D., Loke, S. L., Griesser, H., Wagman, R., Braziel, R., Gascoyne, R., Vicini, S., Iscove, N. N., Cossman, J., and Mak, T. W. (1993). Blood 81, 3097–3115. ¨ Trumper, L., Pfreundschuh, M., Jacobs, G., von Bonin, F., Loftin, U., Moller, P., and Daus, H. (1996). Leukemia 10, 727–730. Tsuboi, K., Iida, S., Inagaki, H., Kato, M., Hayami, Y., Hanamura, I., Miura, K., Harada, S., Kikuchi, M., Komatsu, H., Banno, S., Wakita, A., Nakamura, S., Eimoto, T., and Ueda, R. (2000). Leukemia 14, 449–456. Uchida, J., Yasui, T., Takaoka-Shichijo, Y., Muraoka, M., Kulwichit, W., Raab-Traub, N., and Kikutani, H. (1999). Science 286, 300–303. van den Berg, A., Visser, L., and Poppema, S. (1999). Am. J. Pathol. 154, 1685–1691. van den Berg, A., Visser, L., Eberwine, J., Dadvand, L., and Poppema, S. (2000). Int. J. Cancer 86, 548–552. van der Voort, R., Taher, T. E., Derksen, P. W., Spaargaren, M., van der Neut, R., and Pals, S. T. (2000). Adv. Cancer Res. 79, 39–90. Van Gool, S. W., Delabie, J., Vandenberghe, P., Coorevits, L., De Wolf-Peeters, C., and Ceuppens, J. L. (1997). Leukemia 11, 846–851. Verbeke, C. S., Wenthe, U., Grobholz, R., and Zentgraf, H. (2001). Am. J. Surg. Pathol. 25, 388–394. ¨ Vockerodt, M., Soares, M., Kanzler, H., Kuppers, R., Kube, D., Hansmann, M. L., Diehl, V., and Tesch, H. (1998). Blood 92, 2899–2907. von Wasielewski, R., Wilkens, L., Nolte, M., Werner, M., and Georgii, A. (1996). Mod. Pathol. 9, 334–338. Watanabe, K., Yamashita, Y., Nakayama, A., Hasegawa, Y., Kojima, H., Nagasawa, T., and Mori, N. (2000). Histopathology 36, 353–361. Weber-Matthiesen, K., Deerberg, J., Poetsch, M., Grote, W., and Schlegelberger, B. (1995). Blood 86, 1464–1468. Weber-Matthiesen, K., Deerberg-Wittram, J., Rosenwald, A., Poetsch, M., Grote, W., and Schlegelberger, B. (1996). Am. J. Pathol. 149, 463–468. Weiss, L. M., Movahed, L. A., Warnke, R. A., and Sklar, J. (1989). N. Engl. J. Med. 320, 502–506. Weiss, L. M., Lopategui, J. R., Sun, L. H., Kamel, O. W., Koo, C. H., and Glackin, C. (1995). Blood 85, 2845–7. Weiss, L. M., Chan, J. K. C., MacLennan, K., and Warnke, R. A. (1999). In “Hodgkin’s Disease” (P. M. Mauch, J. O. Armitage, V. Diehl, R. T. Hoppe, and L. M. Weiss, eds.), pp. 101–120. Lippencott Williams & Wilkins, Philadelphia. Wellmann, A., Otsuki, T., Vogelbruch, M., Clark, H. M., Jaffe, E. S., and Raffeld, M. (1995). Blood 86, 2321–2328. Wiens, G. D., Heldwein, K. A., Stenzel-Poore, M. P., and Rittenberg, M. B. (1997). J. Immunol. 159, 1293–1302. Wilson, J. B., Bell, J. L., and Levine, A. J. (1996). EMBO J. 15, 3117–3126. ¨ Wolf, J., Kapp, U., Bohlen, H., Kornacker, M., Schoch, C., Stahl, B., Mucke, S., von Kalle, C., Fonatsch, C., Schaefer, H. E., Hansmann, M. L., and Diehl, V. (1996). Blood 87, 3418–3428.

312

¨ Ralf Kuppers

Wood, K. M., Roff, M., and Hay, R. T. (1998). Oncogene 16, 2131–2139. Wrone-Smith, T., Izban, K. F., Ergin, M., Cosar, E. F., Hsi, E. D., and Alkan, S. (2001). Exp. Hematol. 29, 572–581. Xerri, L., Carbuccia, N., Parc, P., Hassoun, J., and Birg, F. (1995). Histopathology 27, 235–241. Young, L. S., Dawson, C. W., and Eliopoulos, A. G. (2000). Mol. Pathol. 53, 238–247.

Index

A

B

N-Acetyl transferase (NAT), gene mutations in breast cancer, 15 Acquired immunodeﬁciency syndrome, see Human immunodeﬁciency virus AIDS, see Human immunodeﬁciency virus Akt, apoptosis inhibition, 247–248 AMY-1, Myc interactions, 138 Androgen receptor (AR), gene mutations in breast cancer, 15 Angiogenesis cell-mediated immunity suppression, 244 humoral immunite response, 243–244 insulin-like growth factor promotion, 255 metastasis role, 243 Myc role, 104–105 oncogenic viruses and cancer, 249–253 p53 regulation, 247 smoking and cancer, 244, 246 AP-2, Myc interactions, 134 APC, Wnt signaling regulation, 217–218 Apoptosis Akt inhibition, 247–248 chronic immune activation and inhibition, 246–249 Hodgkin and Reed–Sternberg cell expression of genes, 297–298 Myc role Bax activation, 99 cell type differences, 99–100 mitochondrial mechanisms, 98–99 mutation studies, 97 target gene discovery, 97–98 oncogenic viruses and cancer, 249–253 AR, see Androgen receptor Arf, Myc transformation role, 102–103 Arm, wingless signaling, 211 Aspirin, cancer chemoprevention, 257–258 Axin, wingless signaling, 211–212

BACH1, BRCA1 interactions, 14 BAF53, Myc interactions, 121–122 BARD1 BRCA1 interactions, 8 mutations with breast cancer, 13 Bax, Myc activation, 99 B-cell chronic lymphocytic leukemia, Hodgkin and Reed–Sternberg cells, 285–286 Bcl-2 Hodgkin and Reed–Sternberg cell, 298 Myc cooperation in transformation, 103 BIN1, see Bridging Integrator Protein-1 Bob-1, Hodgkin and Reed–Sternberg cell expression, 296 BRAF35 BRCA2 interactions, 9 mutations with breast cancer, 14 BRCA1 DNA microarray analysis, see DNA microarray expression and regulation, 5 locus, 3 mutations and breast cancer incidence, 3 Myc interactions with protein, 136–137 phosphorylation of protein, 7 protein function cell cycle and centrosome regulation, 9 DNA repair, 7–8 transcriptional regulation and chromatin remodeling, 6–7 structure and mutation spectra, 4 BRCA2 discovery, 3 DNA microarray analysis, see DNA microarray expression and regulation, 5 mutations and breast cancer incidence, 3

313

314 BRCA2 (continued ) protein function cell cycle and centrosome regulation, 9 DNA repair, 7–8 transcriptional regulation and chromatin remodeling, 6–7 structure and mutation spectra, 4–5 BRCA3, evidence for existence, 15–16 Breast cancer epidemiology, 2–3 estrogen receptor status and classiﬁcation, 10–11 G1/S regulatory proteins cyclin D1 function, 38 gene ampliﬁcation, 40 protein–protein interactions, 39–40 RNA destabilization, 38 transgenic muose studies, 39 cyclin-dependent kinase inhibitor dysregulation p16, 43–44 p21, 43 p27, 42–43 cyclin E dysregulation, 40–41, 46–49 multiparameter analysis, 44–48 overview, 37–39 p53 mutations, 46–47 retinoblastoma protein dysregulation, 38–39, 44, 47–48 tumor classiﬁcation, 48–50 gene expression analysis DNA microarray BRCA1 versus BRCA2 tumors, 21–24 classiﬁcation of BRCA tumor subgroups, 24 copy number and gene expression analysis, 26 data analysis, 17–22 principles, 16–17 proliferation clusters, 36–37 serial analysis of gene expression, 16 hereditary breast cancer histopathology, 11–12 prospects for study, 26–27 somatic genetic changes, 12–13 immune response, 254 modiﬁer genes, 14–15 proteomics, 36–37 susceptibility genes, see BRCA1; BRCA2 syndromes, 13

Index

Bridging Integrator Protein-1 (BIN1), Myc interactions, 137 Burkitt’s lymphoma, Epstein–Barr virus-speciﬁc cytotoxic T lymphocyte therapy, 191

C

β-Catenin, Wnt signaling role, 215–218 CD95, see Fas CDC6, Myc interactions, 132 Cdc42, see Rho GTPase Cdr2, Myc interactions, 139 C/EBPα, Myc functional interactions in cell differentiation, 96 Cervical cancer, see Human papilloma virus Chaos theory, carcinogenesis, 256–257 Chromatin, Myc remodeling in gene activation ATP-dependent chromatin remodeling, 121–122 histone acetylation, 119–121 promoter clearance, 122 Chronic obstructive pulmonary disease (COPD), cancer association, 240–241, 244 COPD, see Chronic obstructive pulmonary disease COX, see Cyclooxygenase CTL, see Cytotoxic T lymphocyte Cyclin D1 breast cancer dysregulation gene ampliﬁcation, 40 protein–protein interactions, 39–40 RNA destabilization, 38 transgenic muose studies, 39 tumor classiﬁcation, 48–50 function, 38 Myc functional interactions, 94, 106 Cyclin E breast cancer dysregulation, 40–41, 46–49 tumor classiﬁcation, 48–50 Cyclin T1, Myc interactions, 122 Cyclin-dependent kinase inhibitor breast cancer dysregulation p16, 43–44 p21, 43 p27, 42–43 Myc interactionswith p21, 136 Cyclooxygenase (COX) carcinogen upregulation, 247

315

Index

COX-2 tumor expression, 238, 241–243, 246, 253, 259 epidermal growth factor receptor activation and induction, 247–249 inhibition in cancer chemoprevention, 257–258 isoforms, 241 tissue distribution, 241 Cyclosporine A, cancer association, 256 Cytochrome P-450, gene mutations in breast cancer, 14 Cytotoxic T lymphocyte (CTL), adoptive immunotherapy with Epstein–Barr virus-speciﬁc cells Burkitt’s lymphoma treatment, 191 Hodgkin’s disease treatment, 188–190 human immunodeﬁciency virus non-Hodgkin’s lymphoma treatment, 191–192 indications and rationale, 178–179 lymphoproliferative disease treatment fate of cells, 184 generation of lymphocytes, 182–184 hematopoietic stem cell recipient treatment, 184–185 prophylaxis, 184 solid-organ transplant recipient treatment, 185–186 nasopharyngeal carcinoma treatment, 190 prospects, 192–193

D DAP kinase, Myc transformation role, 102–103 Dermo1, Myc transformation role, 103 DNA microarray breast cancer BRCA1 versus BRCA2 tumors, 21–24 classiﬁcation of BRCA tumor subgroups, 24 copy number and gene expression analysis, 26 data analysis artiﬁcial neural networks, 19–20 BRCA1 versus BRCA2 tumors, 19 hierarchical clustering, 18 multidimensional scaling, 18–20 principal components analysis, 21 self organized maps, 18, 20

supervised versus unsupervised methods, 17 principles, 16–17 Myc target gene identiﬁcation, 115, 118 Dsh, wingless signaling, 210

E E2F, Myc functional interactions, 94, 106 EBNA-1 deﬁcient mutant studies, 159 genome maintenance role, 161–162 nuclear localization signal, 159 protein–protein interactions, 162, 170 replication role, 160–161 structure, 157–159 EBV, see Epstein–Barr virus EGFR, see Epidermal growth factor receptor Eotaxin, Hodgkin and Reed–Sternberg cell expression, 299 Epidermal growth factor receptor (EGFR) activation and cyclooxygenase induction, 247–249 apoptosis inhibition mechanisms, 249 Epstein–Barr virus (EBV) adoptive immunotherapy with cytotoxic T lymphocytes Burkitt’s lymphoma treatment, 191 Hodgkin’s disease treatment, 188–190 human immunodeﬁciency virus non-Hodgkin’s lymphoma treatment, 191–192 indications and rationale, 178–179 lymphoproliferative disease treatment fate of cells, 184 generation of lymphocytes, 182–184 hematopoietic stem cell recipient treatment, 184–185 prophylaxis, 184 solid-organ transplant recipient treatment, 185–186 nasopharyngeal carcinoma treatment, 190 prospects, 192–193 associated malignancies and latent gene expression, 176–177, 251 EBNA-1 deﬁcient mutant studies, 159 genome maintenance role, 161–162 nuclear localization signal, 159

316 EBNA-1 (continued ) protein–protein interactions, 162, 170 replication role, 160–161 structure, 157–159 epidemiology, 176 herpesvirus classiﬁcation, 156 Hodgkin and Reed–Sternberg cell infection, 287–290 inﬂammation and malignancy progression, 251–252 latency types, 176–177 latent episomal genome, 157 lymphoproliferative disease diagnosis, 180–181 donor T cell treatment and prophylaxis, 181–182 etiology, 179 incidence following hematopoietic stem cell transplantation, 179–180 lymphokine-activated killer cell therapy, 186 monoclonal antibody therapy, 186–187 OriP elements, 159, 171 initiation of replication, 159–161 prospects for study, 171 replication licensing, 160 vaccination cancer trials, 178 models, 177–178 Estrogen receptor, breast cancer status and classiﬁcation, 10–11

F Fas Hodgkin and Reed–Sternberg cell expression, 298 Hodgkin’s lymphoma mutations, 293 FISH, see Fluorescence in situ hybridization Fluorescence in situ hybridization (FISH), myc activation, 86 Fractal mathematics, carcinogenesis, 256–257 Frz2, wingless signaling, 210

G GAPs, see GTPase-activating proteins GEFs, see Guanine nucleotide exchange factors

Index

Germinal center B cell development and differentiation, 279–281 Hodgkin and Reed–Sternberg cells, 282–284 Glutathione S-transferase (GST), gene mutations in breast cancer, 14–15 Glycogen synthase kinase-3 (GSK-3) discovery, 204 homology between species, 205 isoforms, 204, 223 phosphorylative regulation, 218–221 targets in cancer nuclear factor-κB, 222–223 substrates and phosphorylation sites, 220–222 therapeutic inhibition, 223 Wnt signaling and cancer APC regulation, 217–218 β-catenin role, 215–218 components, 215 differential regulation, 220–221 glycogen synthase kinase-3 binding proteins in signaling, 215–216 Wnt homologs, 206, 214–215 Zw3Sgg in Drosophila gene cloning, 204–205 isoform complementation studies, 205 wingless signaling Arm, 211 Axin, 211–212 components in signaling, 207, 209–212 discovery, 206–207 Dsh, 210 embryogenesis studies, 206–207, 209 Frz2, 210 heparin sulfate proteoglycan binding, 210 resting versus activated state of pathway, 212, 214 GRAF, transformation role, 63 Graft-versus-host disease (GVHD) donor T cell treatment association, 181–182 human immunodeﬁciency virus immune activation similarities, 234–236 GSK-3, see Glycogen synthase kinase-3 GST, see Glutathione S-transferase GTPase-activating proteins (GAPs), transformation role, 63

Index

Guanine nucleotide exchange factors (GEFs) transformation role focus formation activity, 64–65 LARG, 61, 63 Tiam-1, 61 tumor-associated alterations, 61–63 types and Rho GTPase speciﬁcity, 60–61 GVHD, see Graft-versus-host disease

H Hepatitis, viral immune response and carcinogenesis, 252–253 Herpesvirus, see Epstein–Barr virus; Herpesvirus saimiri; Human herpesvirus-8 Herpesvirus saimiri (HVS) genome organization, 165–166 herpesvirus classiﬁcation, 156 LANA1 homolog and function, 164–165, 167 latent episomal genome, 157 terminal repeats and episomal replication, 167–168 HHV-8, see Human herpesvirus-8 Histone acetylation, Myc chromatin remodeling in gene activation, 119–121 Histone H1, LANA1 interactions, 170 HIV, see Human immunodeﬁciency virus HL, see Hodgkin’s lymphoma Hodgkin and Reed–Sternberg (HRS) cells antigen presentation, 300–301 apoptosis-related gene expression, 297–298 B-cell chronic lymphocytic leukemia role, 285–286 B-cell immunophenotype comparison, 293–296, 303 chemokines and cytokines in microenvironment, 298–300 dedifferentiation, 303 detection, 278, 286–287, 304 Epstein–Barr virus infection, 287–290 markers, 279, 293–295 origin B-cell origin tests, 281–282 cell fusion theory, 284 clonality, 282–284 germinal center reaction, 279–281 T-cell origin, 287–288

317 pathology, 278–279 persistence and dissemination, 284 Hodgkin’s lymphoma (HL) cells, see Hodgkin and Reed–Sternberg cells classiﬁcation, 279 Epstein–Barr virus-speciﬁc cytotoxic T lymphocyte therapy, 188–190 histopathology, 278–279 non-Hodgkin’s lymphoma relationship, 285 pathogenesis chromosomal aberrations, 290–291 Epstein–Barr virus, 288–290 gene mutations, 291–293 prognosis and survival, 187–188 prospects for study, 301–304 T-cell lymphoma, 287–288 HPV, see Human papilloma virus HRS cells, see Hodgkin and Reed–Sternberg cells Human herpesvirus-8 (HHV-8) genome organization, 165–166 herpesvirus classiﬁcation, 156 LANA1 dimerization, 164 discovery, 162–163 episomal maintenance, 164–167 locus, 163 mediation of chromosome interactions, 169–170 nuclear localization, 163–164 prospects for study, 171 structure, 163–164 terminal repeat interactions, 169 latent episomal genome, 157 Human immunodeﬁciency virus (HIV) CD4 receptor, 232 chimpanzee infection, 233 chronic immune activation, 233–236 non-Hodgkin’s lymphoma treatment, 191–192 pathogenesis of infection, 232–233 simian immunodeﬁciency virus similarities, 233 T helper cell imbalance, 236–237 Human papilloma virus (HPV) cancer association and serotypes, 249–250 cell-mediated immune response, 250–251 chronic immune activation association, 250–251

318 Human papilloma virus (continued ) coinfection risks for cervical cancer, 251 HVS, see Herpesvirus saimiri

I IFN-γ , see Interferon-γ IGF, see Insulin-like growth factor IL-13, see Interleukin-13 Inﬂammation apoptosis relationship, 246–249 chronic inﬂammation relationships apoptosis, 246–249 cancer, 244–246 oncogenic viruses and cancer, 249–253 INI1, Myc interactions, 132 Insulin-like growth factor (IGF), malignancy role, 254–255 Interferon-γ (IFN-γ ), defects in cancer, 240 Interleukin-13 (IL-13), Hodgkin and Reed–Sternberg cell expression, 299, 302

K Kaposi’s sarcoma, see Human herpesvirus-8

L LANA1 dimerization, 164 discovery, 162–163 episomal maintenance, 164–167 herpesvirus saimiri homolog and function, 164–165, 167 locus, 163 mediation of chromosome interactions, 169–170 nuclear localization, 163–164 prospects for study, 171 structure, 163–164 terminal repeat interactions, 169 Lymphoproliferative disease, see Epstein–Barr virus

M Macrophage inhibitory factor (MIF), release during inﬂammation, 246 Max, Myc interactions, 121, 124–126 Melanoma, immune response, 239, 253

Index

Met, Hodgkin and Reed–Sternberg cell expression, 299–300 Metastasis models, 70, 72 role Rho, 73–75 Tiam-1, 72–73 steps, 70 Microarray, see DNA microarray MIF, see Macrophage inhibitory factor Miz-1, Myc interactions, 124–125, 134 Myc activation detection in human tumors, 85–87 direct versus indirect mechanisms, 85 overview, 84–85, 91 posttranslational modiﬁcation glycosylation, 89 phosphorylation, 88–89 ubiqutination, 88 biological activity angiogenesis role, 104–105 apoptosis role Bax activation, 99 cell type differences, 99–100 mitochondrial mechanisms, 98–99 mutation studies, 97 target gene discovery, 97–98 cell cycle regulation, 92, 94, 107 cell differentiation blocking, 96–97 cell growth, 95 genomic instability promotion, 103–104 overview, 92–93, 105–107 transformation role beta cell targeting studies, 101–102 mouse model studies, 101 mutation studies, 100 oncogene collaboration, 102–103 target genes, 100 isoforms and family members, 82–84 protein–protein interactions, see also speciﬁc proteins coactivators, 127, 131–133 cytoplasmic proteins, 138–139 family and c-Myc isoform speciﬁcity of binding, 139–140 overview, 126–127, 140 prospects for study, 140 table of proteins, 128–129 transcriptional regulators, 133–135 tumor suppressors, 135–138

319

Index

regulation of expression in normal cells, 84 therapeutic targeting for cancer inhibition, 90 rationale, 89–90 tumor activation exploitation for targeting, 90–91 transcriptional regulation activated genes, table, 108–111 activation mechanisms ATP-dependent chromatin remodeling, 121–122 histone acetylation, 119–121 overview, 118–119 promoter clearance, 122 cell cycle regulator genes, 107 essential Myc regions, 118 repressed genes, table, 112–114 repression mechanisms initiator element-dependent genes, 124–125 initiator element-independent genes, 125–126 Max role, 124 overview, 122–123, 126 target genes deﬁning, 115–116 identiﬁcation, 115 strongly supportive evidence, 116–117 suggestive evidence, 117–118

N Nasopharyngeal carcinoma, Epstein–Barr virus-speciﬁc cytotoxic T lymphocyte therapy, 190 NAT, see N-Acetyl transferase NF-κB, see Nuclear factor-κB NF-Y, Myc interactions, 134–135 Nmi, Myc interactions, 133 Nonsense-mediated messenger RNA decay, pathway identiﬁcation in breast cancer, 26 Nonsteroidal anti-inﬂammatory drugs (NSAIDs), cancer chemoprevention, 257–258 NSAIDs, see Nonsteroidal anti-inﬂammatory drugs Nuclear factor-κB (NF-κB) COX-2 expression role, 248–249

glycogen synthase kinase-3 regulation, 222–223 Hodgkin and Reed–Sternberg cell expression, 297–298 inhibitor, Hodgkin’s lymphoma mutations, 292–293

O Obesity, cancer association, 255–256 Oct-2, Hodgkin and Reed–Sternberg cell expression, 296 OriP elements, 159, 171 initiation of replication, 159–161 prospects for study, 171

P p16, see Cyclin-dependent kinase inhibitor p21, see Cyclin-dependent kinase inhibitor p27, see Cyclin-dependent kinase inhibitor p32/TAP, EBNA-1 interactions, 162 p53 breast cancer dysregulation, 46–47 Hodgkin’s lymphoma mutations, 291–292 insulin-like growth factor induction, 255 LANA1 interactions, 170 Th1 cytotoxicity mediation, 246–247 p107, Myc interactions, 135–136 p202a, Myc interactions, 136 PAF400, Myc potential interactions, 131 Pag, Myc interactions, 137–138 Pam, Myc interactions, 132 PDGF, see Platelet-derived growth factor Platelet-derived growth factor (PDGF), Myc functional interactions with receptor, 92, 94, 125 Prostate cancer, immune response, 254 P-TEFb kinase, Myc interactions, 131

R Rac, see Rho GTPase RAD51 BRCA2 interactions, 7–8 mutations with breast cancer, 13 Ras Myc functional interactions, 92, 94 oncogenesis role, 60 Rho GTPase interactions in transformation, 66

320 RB, see Retinoblastoma protein Reed–Sternberg cell, see Hodgkin and Reed–Sternberg cells Renal cancer, immune response, 253 Retinoblastoma protein (RB) breast cancer dysregulation, 38–39, 44, 47–48 Myc functional interactions, 94 Reverse transcription–polymerase chain reaction (RT-PCR), myc activation, 86 Rho GTPase activation cycle, 58–59 cytoskeleton dynamics regulation, 59 effectors, 59 gene expression regulation, 59 metastasis role Rho, 73–75 Tiam-1, 72–73 prospects for cancer studies, 75 transformation role biochemical contributions Cdc42, 69–70 Rac, 68–69 Rho, 67–68 constitutively-activated mutant studies Cdc42, 63, 65–67 Rac, 63, 65–67 Ras and Raf enhancement studies, 66 Rho, 63, 65–67 dominant-negative mutant studies Cdc42, 66 Rac, 66 Ras interactions, 66 Rho, 66 GTPase-activating proteins, 63 guanine nucleotide exchange factors focus formation activity, 64–65 LARG, 61, 63 Tiam-1, 61 tumor-associated alterations, 61–63 types and GTPase speciﬁcity, 60–61 types, 57–58 RT-PCR, see Reverse transcription– polymerase chain reaction

S Simian immunodeﬁciency virus (SIV), human immunodeﬁciency virus similarities, 233 Simian virus-40 (SV-40), immune response, 252

Index

SIV, see Simian immunodeﬁciency virus Sp1, Myc interactions, 135 SV-40, see Simian virus-40 SWI/SNF complex, Myc interactions, 121–122

T TARC, Hodgkin and Reed–Sternberg cell expression, 299, 302, 304 T-cell anergy, malignant disease, 237–238 Testicular cancer, immune response, 253–254 TGF-β, see Transforming growth factor-β T-helper cell imbalance in human immunodeﬁciency virus infection, 236–237 Th1 response in tumor regression and rejection, 238–239 Th2 response and cancer prognosis, 237, 243 Thrombospondin-1, Myc interactions in angiogenesis, 105 Tiam-1 metastasis role, 72–73 transformation role, 61 TIP, Myc interactions, 119, 131 Transforming growth factor-β (TGF-β), cyclosporine A tumor induction role, 256 Translation, Myc regulation of cell growth, 95 TRRAP, Myc interactions, 119, 121, 131 Tubulin, Myc interactions, 138 Twist, Myc transformation role, 103

V Vascular endothelial growth factor (VEGF), Myc interactions in angiogenesis, 105 VEGF, see Vascular endothelial growth factor

W Wnt signaling and cancer APC regulation, 217–218 β-catenin role, 215–218 components, 215 differential regulation, 220–221

321

Index

glycogen synthase kinase-3 binding proteins in signaling, 215–216 homology between species, 205 Wnt homologs, 206, 214–215 wingless signaling Arm, 211 Axin, 211–212 components in signaling, 207, 209–212 discovery, 206–207 Dsh, 210 embryogenesis studies, 206–207, 209 Frz2, 210

heparin sulfate proteoglycan binding, 210 resting versus activated state of pathway, 212, 214

Y Yaf2, Myc interactions, 132–133 YY-1, Myc interactions, 133–134

Z Zw3Sgg, see Glycogen synthase kinase-3

This Page Intentionally Left Blank