No title

Advances in CANCER RESEARCH Volume 77 This Page Intentionally Left Blank Advances in CANCER RESEARCH Volume 77 E...

Author: George F. Vande Woude | George Klein

14 downloads 535 Views 5MB 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 77

This Page Intentionally Left Blank

Advances in

CANCER RESEARCH Volume 77

Edited by

George F. Vande Woude Division of Basic Sciences National Cancer Institute National Institutes of Health Bethesda, Maryland

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

ACADEMIC PRESS San Diego London Boston New York Sydney Tokyo Toronto

This book is printed on acid-free paper.

Copyright © 2000 by ACADEMIC PRESS All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the ﬁrst page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of speciﬁc clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2000 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/00 $30.00

Academic Press A Harcourt Science and Technology Company 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.apnet.com

Academic Press 24-28 Oval Road, London NW1 7DX, UK http://www.hbuk.co.uk/ap/ International Standard Book Number: 0-12-006677-7 PRINTED IN THE UNITED STATES OF AMERICA 99 00 01 02 03 04 EB 9 8 7 6

5

4

3

2

1

Contents

Contributors to Volume 77 vii

The Yin-Yang of TCF/␤-Catenin Signaling Nick Barker, Patrice J. Morin, and Hans Clevers I. II. III. IV. V. VI.

An Introduction to the Tcf Family 2 The HMG Box: DNA Binding and Bending 3 Context-Dependent Transactivation 4 Functions of the Tcfs in Mammalian Development 6 Wingless/Wnt Signaling Pathways 7 Tcf/-Catenin Complexes: Bipartite Transcription Factors Involved in Wingless/Wnt Signaling 8 VII. Regulation of Tcf/-Catenin Signaling 10 VIII. Regulation of Wingless/Wnt Target Genes by Tcfs: Activation versus Repression 15 IX. Tcf/-Catenin Signaling and Cancer 17 References 21

Biochemical and Clinical Implications of the ErbB/HER Signaling Network of Growth Factor Receptors Leah N. Klapper, Mark H. Kirschbaum, Michael Sela, and Yosef Yarden I. II. III. IV.

Introduction 26 Clinical Aspects of ErbB Receptors 29 How Does ErbB-2 Induce Cancer? 42 Evolutionary and Developmental Aspects of the Multiplicity of ErbB Proteins 46 V. The ErbB Signaling Network 48 VI. ErbB-Directed Cancer Therapy 55 VII. Conclusions 61 References 62

p53 and Human Cancer: The First Ten Thousand Mutations Pierre Hainaut and Monica Hollstein I. Introduction 81 II. Biology of the p53 Protein 83

v

vi

Contents

III. Mutations and Variations in the p53 Gene 100 IV. p53 Mutations in Sporadic Cancers: Host–Environment Interactions 104 V. Clinical Relevance of p53 Mutations 119 References 123

Macrophage Stimulating Protein Edward J. Leonard and Alla Danilkovitch I. II. III. IV. V. VI. VII. VIII. IX.

Introduction 139 Structural Aspects of Native and Recombinant pro-MSP 141 Binding of MSP to Its Receptor 142 A Model of MSP-Induced Receptor Dimerization 143 Regulation of MSP Activity: Pathways for pro-MSP Cleavage 144 MSP–Ron Signaling 147 Modes of MSP Receptor Regulation or Activation 152 Target Cells for Macrophage Stimulating Protein 157 Perspective 161 References 162

CD44 Glycoproteins in Colorectal Cancer: Expression, Function, and Prognostic Value Vera J. M. Wielenga, Ronald van der Neut, G. Johan A. Offerhaus, and Steven T. Pals I. II. III. IV.

Introduction 169 Structure and Function of CD44 170 CD44 in Tumor Progression and Metastasis 174 Conclusions 182 References 182

A Simple Model for Carcinogenesis of Colorectal Cancers with Microsatellite Instability Nicolas Janin I. II. III. IV. V.

Introduction 190 A Simple Model of RER Tumor Carcinogenesis 193 Evidence to Support the Model 198 Discussion 203 Conclusion 218 References 218

Index

223

Contributors

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

Nick Barker, Department of Immunology, University Hospital, 85500 GA Utrecht, The Netherlands (1) Hans Clever, Department of Immunology, University Hospital, 85500 GA Utrecht, The Netherlands (1) Alla Danilkovitch, Laboratory of Immunobiology, NCI–Frederick Cancer Research and Development Center, Frederick, Maryland 21702 (139) Pierre Hainaut, IARC, F69372 Lyon, France (81) Monica Hollstein, DKFZ, D69120 Heidelberg, Germany (81) Nicolas Janin, Comité de Pathologie Mammaire, Institut Gustave Roussy, 94805 Villejuif, France (189) Mark H. Kirschbaum, Department of Biological Regulation, The Weizmann Institute of Science, Rehovot 76100, Israel (25) Leah N. Klapper, Department of Immunology, The Weizmann Institute of Science, Rehovot 76100, Israel (25) Edward J. Leonard, Laboratory of Immunobiology, NCI–Frederick Cancer Research and Development Center, Frederick, Maryland 21702 (139) Patrice J. Morin, National Institute for Aging, Baltimore, Maryland 21224 (1) G. Johan A. Offerhaus, Department of Pathology, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands (169) Steven T. Pals, Department of Pathology, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands (169) Michael Sela, Department of Immunology, The Weizmann Institute of Science, Rehovot 76100, Israel (25) Ronald van der Neut, Department of Pathology, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands (169) Vera J. M. Wielenga, Department of Pathology, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands (169) Yosef Yarden, Department of Biological Regulation, The Weizmann Institute of Science, Rehovot 76100, Israel (25) vii

This Page Intentionally Left Blank

The Yin-Yang of TCF/bCatenin Signaling Nick Barker,1 Patrice J. Morin,2 and Hans Clevers1 1

Department of Immunology University Hospital 85500 GA Utrecht, The Netherlands 2 National Institute for Aging Baltimore, Maryland 21224

I. II. III. IV. V. VI.

An Introduction to the Tcf Family The HMG box: DNA Binding and Bending Context-Dependent Transactivation Functions of the Tcfs in Mammalian Development Wingless/Wnt Signaling Pathways Tcf/b-Catenin Complexes: Bipartite Transcription Factors Involved in Wingless/Wnt Signaling VII. Regulation of Tcf/b-Catenin Signaling VIII. Regulation of Wingless/Wnt Target Genes by Tcfs: Activation versus Repression A. Activation of Target Genes by Tcfs B. Repression of Target Genes by Tcfs IX. Tcf/b-Catenin Signaling and Cancer References

Wingless/Wnt signaling directs cell-fate choices during embryonic development. In Drosophila, Wingless signaling mediates endoderm induction and the establishment of segment polarity in the developing embryo. The ﬂy Wingless cascade is strikingly similar to the vertebrate Wnt signaling pathway, which controls a number of key developmental decisions such as dorsal-ventral patterning in Xenopus. Factors of the TCF/LEF HMG domain family (Tcfs) have recently been established as the downstream effectors of the Wingless/Wnt signal transduction pathways. Upon Wingless/Wnt signaling, a cascade is initiated that results in the accumulation of cytoplasmic b-catenin (or its ﬂy homolog, Armadillo). There is also a concomitant translocation of b-catenin/Armadillo to the nucleus, where it interacts with a speciﬁc sequence motif at the N terminus of Tcfs to generate a transcriptionally active complex. This bipartite transcription factor is targeted to the upstream regulatory regions of Tcf target genes including Siamois and Nodal related gene-3 in Xenopus, engrailed and Ultrabithorax in Drosophila via the sequencespeciﬁc HMG box, and mediates their transcriptional activation by virtue of transactivation domains contributed by b-catenin/Armadillo. In the absence of Wingless/Wnt signals, a key negative regulator of the pathway, GSK3b, is activated, which mediates the downregulation of cytoplasmic b-catenin/Armadillo via the ubiquitin-proteasome pathway. In the absence of nuclear b-catenin, the Tcfs recruit the corepressor protein Groucho to the target gene enhancers and actively repress their transcription. An additional corepressor protein, CREB-binding protein (CBP), may also be involved in this repression of Tcf target gene activity. Several other proteins, including adenomatous polyposis coli (APC), GSK3b, and Axin/Conductin, are

Advances in CANCER RESEARCH 0065-230X/00 $30.00

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

2

Nick Barker et al. instrumental in the regulation of b-catenin/Armadillo. In APC-deﬁcient colon carcinoma cell lines, b-catenin accumulates and is constitutively complexed with nuclear Tcf-4. A proportion of APC wild-type colon carcinomas and melanomas also contains constitutive nuclear Tcf-4/b-catenin complexes as a result of dominant mutations in the N terminus of b-catenin that render it insensitive to downregulation by APC, GSK3b, and Axin/Conductin. This results in the unregulated expression of Tcf-4 target genes such as c-myc. Based on the established role for Tcf-4 in maintaining intestinal stem cells it is likely that deregulation of c-myc expression as a result of constitutive Tcf-4/b-catenin activity promotes uncontrolled intestinal cell proliferation. This would readily explain the formation of intestinal polyps during colon carcinogenesis. Similar mechanisms leading to deregulation of Tcf target gene activity are likely to be involved in melanoma and other forms of cancer. © 2000 Academic Press.

I. AN INTRODUCTION TO THE TCF FAMILY The original members of the TCF family, T-cell factor 1 (Tcf-1) and lymphoid-enhancing factor 1 (Lef-1), were discovered during searches for lymphoid-speciﬁc transcription factors. Tcf-1 was cloned as a gene encoding a T-cell-speciﬁc DNA-binding protein with high afﬁnity for the AACAAAG motif in the CD3e enhancer (van de Wetering et al., 1991; Clevers and van de Wetering, 1997). The human Tcf-1 gene, located on chromosome 5, has a complex structure, containing two alternative promoters, four alternative exons, and three alternative splice acceptor sites (van de Wetering et al., 1996). As a consequence, Tcf-1 expression in T cells is characterized by the presence of multiple proteins (Castrop et al., 1995). Transcription from the downstream promoter is predicted to generate Tcf-1 isoforms lacking the Nterminal region important for Tcf function in the Wnt signaling pathway, raising the possibility that these could have a dominant negative function. The closely related gene encoding human Lef-1 (Tcf-1a) was identiﬁed using a classical approach, in which the protein was puriﬁed from the Jurkat T-cell line by column chromatography using a binding site from the HIV long-terminal repeat (Waterman et al., 1991). The peptide sequence was subsequently determined and the information used to clone the corresponding cDNA. Mouse Lef-1 was independently isolated during the subtractive screen for pre-B-cell-speciﬁc factors and subsequently found to be additionally expressed in T cells (Travis et al., 1991). Lef-1 was shown to bind specifically to the TTCAAAGG sequence motif in the TCRa enhancer. The human and mouse Lef-1 genes are located on chromosome 4 and chromosome 3, respectively (Milatovich et al., 1991). Two additional mammalian family members, Tcf-3 and Tcf-4, were identiﬁed via “guessmer” PCR on human genomic DNA using primers designed from the highly conserved DNA-binding domains of Tcf-1 and Lef-1 (Castrop et al., 1992a,b). Full-length hTcf-4 was subsequently isolated from a fetal cDNA library and found to encode a protein that most closely resembles

Yin-Yang of TCF/b-Catenin Signaling

3

hTcf-1 (Fig. 1A) (Korinek et al., 1997). Alternative splicing generates three hTcf-4 isoforms with distinct C termini, which are conserved between hTcf1 and hTcf-4 (Korinek et al., 1997; Sparks et al., 1998). Alternative splicing in the C-terminal region also accounts for the presence of two isoforms of the mouse Tcf-4 homolog (Korinek et al., 1998a). Importantly, all the Tcf-4 isoforms contain the DNA-binding region and the consensus N-terminal bcatenin binding sites, indicative of a similar function in Wnt signaling. Mouse Tcf-3 was cloned simultaneously from an embryonic cDNA library and found to contain the characteristic HMG box and b-catenin binding consensus motif (Korinek et al., 1998b). In accordance with this conservation in their DNA-binding domain, both Tcf-3 and Tcf-4 bind with high afﬁnity to the Tcf/Lef consensus motif A/TA/TCAAAG. No additional human or murine members have been identiﬁed, despite intensive cloning efforts and database searches. The recent explosion of interest of Tcfs following their discovery as Wingless/Wnt signaling effectors has led to the cloning of Tcf genes from the model organisms Xenopus and Drosophila. Four highly related maternally expressed Tcf homologs were isolated from early Xenopus embryos, which were termed XTcf-3 on the basis of their high similarity to the mammalian Tcf-3 gene (Molenaar et al., 1996). The Drosophila homolog, dTCF or Pangolin, which is encoded by a gene on the small chromosome 4, probably represents the only Tcf family member present within the ﬂy genome (van de Wetering et al., 1997; Brunner et al., 1997). Alternative splicing generates two isoforms that differ in the second half of the HMG box (van de Wetering et al., 1997). dTCF, which most closely resembles mammalian Tcf-1, is maternally and ubiquitously expressed throughout the ﬂy embryo. In Caenorhabditis elegans, a mesoderm-specifying gene called Pop-1 was found to encode a distant member of the Tcf/Lef family (Lin et al., 1995). The Tcf family members are characterized by the presence of a single highly conserved DNA-binding high mobility group (HMG) domain. Extensive sequence homology is also present throughout the remainder of the protein, albeit to a lesser extent. However, the C termini of individual members often diverge considerably as a result of alternative splicing (Fig. 1B).

II. THE HMG BOX: DNA BINDING AND BENDING A common feature of the Tcf family members is the presence of a highly conserved sequence-speciﬁc DNA-binding region known as the HMG box. Similar DNA-binding domains have been identiﬁed in SRY/SOX factors and fungal mating type proteins (Laudet et al., 1993). These three subclasses of HMG box all recognize similar 6–8 bp, AT-rich motifs of the canonical sequence [A/T][A/T]CAA[A/T]GG.

4

Nick Barker et al.

Fig. 1A Phylogenetic relationship of the Tcf/Lef gene family based on aligned amino acid sequences. Relative branch lengths are proportional to the number of inferred substitutions per site. (Reproduced with permission of American Society for Micribiology from Korinek, V., et al., Two members of the TCF family implicated in wnt/␤-catenin signaling during embryogenesis in the mouse, Mol. Cell Biol., Vol. 18, No. 3, 1998, pp. 1248–1256.)

Biochemical analyses have determined that the HMG box predominantly binds the minor groove, thereby introducing a sharp bend in the DNA helix (van de Wetering and Clevers, 1992; Giese et al., 1992). Resolution of the crystal structure of non-sequence-speciﬁc HMG boxes such as those present in the high mobility group proteins HMG1 and HMG-D has revealed a structure comprised of three a helices arranged in an L-shape. Although the crystal structure of sequence-speciﬁc HMG boxes has not formally been determined, the results of NMR structural analyses of the Lef-1 HMG box bound to DNA predict an overall similar 3-D structure (Love et al., 1995). The concave side of the L occupies the minor groove of the recognition sequence, and a hydrophobic amino acid located in the ﬁrst a helix intercalates between two AT base pairs. The minor groove of the DNA is consequently widened and a dramatic bend of 70–130⬚ is generated.

III. CONTEXT-DEPENDENT TRANSACTIVATION Tcf-1 and Lef-1 are not “classical” transcription factors in that they are incapable of activating transcription of reporter gene constructs from a synthetic enhancer containing multimerized Tcf/Lef binding sites. Lef-1 acts as

Yin-Yang of TCF/b-Catenin Signaling

5

Fig. 1B Structural organization of the T-cell factor proteins (Tcfs). The DNA-binding HMG box and ␤-catenin/Armadillo binding domains are highly conserved between family members. Lef-1 contains a unique context-dependent activation domain (CAD), which is required for Lef1-mediated TCR␣ enhancer activity. The greatest sequence diversity is present within the C termini, which are often generated via alternative splicing. (Ch, chicken; X, Xenopus; d, Drosophila).

6

Nick Barker et al.

an architectural transcription factor, stimulating the function of the TCR-a enhancer in a manner that is strictly dependent on the presence of other DNA-binding proteins. Two distinct regions of Lef-1 contribute to this context-dependent transactivation. The HMG box binds the minor groove of the TCRa, causing a sharp bend in the DNA helix, which is believed to facilitate interactions between adjacent enhancer-bound ETS1, AML1, and ATF/CREB transcription factors (Giese et al., 1995). Transcriptional activation is mediated by a context-dependent activation domain (CAD), which functions only in the speciﬁc context of the ubiquitously expressed coactivator protein, ally of Lef-1 and AML-1 (ALY) (Carlsson et al., 1993; Giese and Grosschedl, 1992; Bruhn et al., 1997). This architectural function of Lef1 in the regulation of the TCRa enhancer is completely independent of bcatenin, which binds to a region distinct from the CAD in the N terminus of Lef-1 (Carlsson et al., 1993). Tcf-1 differs from Lef-1 in that it does not contain a CAD and cannot bind ALY, thereby providing an explanation as to why it is incapable of activating the TCRa enhancer (van de Wetering et al., 1996).

IV. FUNCTIONS OF THE Tcfs IN MAMMALIAN DEVELOPMENT In adult mammals, Tcf-1 expression is restricted to T lymphocytes, whereas Lef-1 is expressed by both T and pre-B lymphocytes. However, in situ analysis revealed that during mouse embryogenesis both genes exhibit a complex, largely overlapping expression pattern in many tissues (Oosterwegel et al., 1993). It is clear from the analysis of Tcf-1 and Lef-1 knockout mice that these transcription factors have distinct in vivo roles during mouse embryogenesis. Mice lacking Tcf-1 have a severe block in T-cell differentiation, speciﬁcally at the transition from the CD8+ immature single-positive to the CD4+ /CD8+ double-positive stage of thymocyte development (Verbeek et al., 1995). However, the development of Tcf-1⫺/⫺ mice is otherwise normal. Lef-1 knockout mice have a more complex phenotype, lacking hair, teeth, mammary glands, and trigeminal nuclei (Van Genderen et al., 1994). These mice consequently die around birth. Surprisingly, the loss of Lef-1 expression has no major effect on lymphoid development, with the exception of the absence of the B1 B-cell subset, indicating that some redundancy exists within lymphoid transcription factors. Tcf-1/Lef-1 double knockout mice exhibit a complete absence of T cells, indicating that Lef-1 can partially substitute for Tcf-1 during T-cell development (Okamura et al., 1998). Expression of Tcf-3 and Tcf-4 differs markedly within the developing mouse embryo (Korinek et al., 1998a). In situ analysis revealed that Tcf-3 is

Yin-Yang of TCF/b-Catenin Signaling

7

ubiquitously expressed in early embryos (day 6.5), but gradually disappears over the next 3–4 days. The expression pattern of Tcf-3 in adult tissues has not been deﬁned, but a limited histological analysis using a Tcf-3-reactive monoclonal antibody indicates expression within hair follicles and keratinocytes (Barker et al., 1999). In contrast, Tcf-4 expression occurs later in embryonic development (day 10.5) and is restricted to the central nervous system and the intestinal epithelium. Tcf-4 expression remains in the brain and intestinal epithelium throughout life. A detailed histological analysis of Tcf-4 protein in human and mouse intestine revealed a temporal expression gradient along the crypt-villus axis (Barker et al., 1999). Tcf-4 expression is initially high in the actively dividing crypt cells, with little or no expression in the developing villi. However, as development proceeds, Tcf-4 expression on the villus epithelium increases. This increase probably occurs as a result of the acquisition of Tcf-4 expressing crypt cells, which actively migrate along the crypt-villus axis to form villus epithelium. The in vivo function of Tcf-4 was determined by the generation and analysis of mice lacking Tcf-4 protein (Korinek et al., 1998b). The development of the small intestine was severely impaired in these mice as a result of the depletion of the epithelial stem-cell compartments. The Tcf-4 knockout mice died shortly after birth, probably as a consequence of extensive stretching and occasional tearing of the intestinal epithelium. The signiﬁcance of this Tcf-4 function in maintaining the proliferative compartments of the small intestine will become apparent later in this review, when Tcf-4/b-catenin signaling is discussed in the context of colon cancer.

V. WINGLESS/Wnt SIGNALING PATHWAYS Wingless (Wg) and Wnt are secreted signaling molecules that regulate a multitude of key developmental processes in Drosophila and vertebrates, respectively. The linear gene cascade that comprises the Wingless pathway was originally deﬁned during genetic studies on Wingless-controlled segment polarity in Drosophila (Orsulic and Peifer, 1996a). Activation of Wg gene expression leads to secretion of Wg and the subsequent interaction of this morphogen with a seven transmembrane domain receptor, Fz-2, on target cells (Bhanot et al., 1996). This interaction initiates a signal transduction cascade, the ﬁrst event of which is the hyperphosphorylation of a cytoplasmic protein of poorly deﬁned biochemical function, known as Disheveled (Yanagawa et al., 1995). This activated form of Disheveled is recruited to the cell membrane, where it may interact directly with the Fz-2 receptor via a PDZ domain. Activated Disheveled inhibits a serine/threonine kinase termed ZesteWhite-3 (ZW3) (Cook et al., 1996). In the absence of Wingless signaling,

8

Nick Barker et al.

ZW3 targets Armadillo for breakdown, thereby inhibiting the action of this effector molecule. Inactivation of ZW3 via the Wingless signal thus leads to an accumulation of cytoplasmic Armadillo and possibly its nuclear translocation. It was thought that this accumulation of Armadillo would somehow effect genetic reprogramming within the developing embryo, although the molecular mechanism of this action remained a mystery prior to the discovery of Tcfs as mediators of Wingless/Wnt signaling (Orsulic and Peifer, 1996b). A remarkably symmetrical pathway (Wnt) was subsequently identiﬁed in vertebrates using Xenopus as a model organism (Orsulic and Peifer, 1996a; Miller and Moon, 1996; Moon et al., 1997). Embryonic development of this animal was already known to be inﬂuenced by secreted Wnt factors, the vertebrate homologs of the Wingless gene products. Overexpression of these Wnt factors causes the developing embryo to develop a second dorsal body axis (Sokol et al., 1991). Overexpression of b-catenin, the vertebrate homolog of Armadillo, in the developing embryo has a similar effect, resulting in embryos with two heads (Funayama et al., 1995). This Xenopus axis-duplication assay was subsequently used to determine the involvement of other vertebrate homologs of the Drosophila genes in Wnt signaling (Fig. 2).

VI. Tcf/b-CATENIN COMPLEXES: BIPARTITE TRANSCRIPTION FACTORS INVOLVED IN WINGLESS/Wnt SIGNALING A highly unexpected interaction between b-catenin/Armadillo, a cytoplasmic protein with a role in cellular adhesion, and the Tcfs was revealed when Lef-1 was cloned using b-catenin as a bait in a yeast two-hybrid screen (Behrens et al., 1996). This was independently conﬁrmed using a reverse two-hybrid strategy, in which b-catenin was isolated from a brain cDNA library using Tcf-1 as a bait (Molenaar et al., 1996). This interaction was found to be mediated by a highly conserved sequence motif present at the N terminus of the Tcfs. The in vivo relevance of this interaction was subsequently demonstrated when murine Lef-1 and XTcf-3, an embryonically expressed member of the Tcf family in Xenopus, were shown to bind to bcatenin upon microinjection in Xenopus embryos (Behrens et al., 1996; Huber et al., 1996; Molenaar et al., 1996). b-Catenin interacts with Tcfs via a central region comprising 12 imperfect Armadillo repeats, which are putative protein–protein interaction domains present in a range of proteins with diverse functions. Individual repeats comprise three a helices and are tightly packed into a superhelix (Huber et al., 1997). This superhelix features a long, positively charged groove, which is likely to accommodate the negatively charged binding domains of the Tcfs.

Yin-Yang of TCF/b-Catenin Signaling

9

Fig. 2 An alignment of the Wingless/Wnt signaling components. Interactions resulting in stimulation (arrows) or inhibition (bars) of activity are shown. The relationship between APCmediated ␤-catenin/Armadillo regulation and Wingless/Wnt signaling is still somewhat controversial.

The emergence of b-catenin as a potential binding partner of the Tcfs prompted a number of studies into the functional consequences of this complex formation. Using in vitro reporter gene assays it was shown that these Tcf/b-catenin complexes could strongly drive transcription from promoters containing multiple copies of the Tcf/Lef consensus motifs (Molenaar et al., 1996). Deletion of a small region from the N terminus of the Tcfs abolished b-catenin binding and consequently abrogated Tcf/b-catenin transcriptional activity. Biochemical and genetic data indicate that two regions of b-catenin

10

Nick Barker et al.

mediate transcriptional activation. In particular, the carboxy terminus of bcatenin functions as a transcriptional activation domain in a fusion with the GAL4 DNA binding domain (van de Wetering et al., 1996). Genetic analysis has demonstrated that the corresponding region of the Drosophila homolog of b-catenin, Armadillo, is required for Wingless signaling in vivo. More recently, a second transcriptional activation domain has been identiﬁed within the N-terminal region of b-catenin, which augments the activity of the C-terminal domain when associated with Lef-1 (Hsu et al., 1998). Using Xenopus and Drosophila as model systems, the Tcfs were subsequently established as the “missing link” between b-catenin/Armadillo accumulation and genetic reprogramming in response to Wingless/Wnt signaling. In Xenopus, ectopic activation of Wnt signaling can be induced via overexpression of Wnt factors or by b-catenin itself, leading to axis duplication (Funayama et al., 1995). Microinjection of either RNA corresponding to XTcf-3 lacking the N-terminal b-catenin binding site or Lef-1 lacking the DNA-binding HMG box into Xenopus embryos suppresses endogenous axis formation (Molenaar et al., 1996; Behrens et al., 1996; Huber et al., 1996). This indicates that these proteins interfere with the endogenous Wnt signal in a dominant-negative manner. In agreement with these Xenopus data, several studies have demonstrated a strong correlation between coexpression of Wnt-1 and the formation of transcriptionally active nuclear Tcf/b-catenin complexes in cell lines (Korinek et al., 1998a; Hsu et al., 1998). For deﬁnitive proof of the involvement of endogenous Tcfs in Wingless/ Wnt signaling, a genetic approach was adopted using Drosophila as a model organism. The Drosophila Tcf/Lef homolog, dTCF/Pangolin, interacts with Armadillo to generate a bipartite transcription factor capable of activating transcription from multimerized Tcf-binding motifs. Using a genetic approach, several mutant alleles of dTCF have been identiﬁed that exhibit phenotypes similar to those of Wingless and Armadillo loss of function alleles in several tissues, including the embryonic gut, embryonic epidermis, and adult epidermis (van de Wetering et al., 1997; Brunner et al., 1997; Riese et al., 1997). Epistasis analysis determined that the gene encoding dTCF is downstream from Armadillo. Importantly, mutations affecting the N-terminal or DNA-binding regions of dTCF also perturb Wingless signaling (van de Wetering et al., 1997; Brunner et al., 1997).

VII. REGULATION OF Tcf/b-CATENIN SIGNALING Transcriptional activation of target genes in response to Wingless/Wnt signaling is dependent on the nuclear translocation of free cytoplasmic b-

Yin-Yang of TCF/b-Catenin Signaling

11

catenin and complex formation with a member of the Tcf/Lef transcription factor family. The regulation of this transcriptional activity is mainly achieved by strictly controlling the levels of free cytoplasmic b-catenin available for binding to the Tcfs. In the absence of a Wingless/Wnt signal, a quaternary cytoplasmic complex comprising b-catenin, adenomatous polyposis coli (APC), Conductin/Axin, and GSK3b mediates the phosphorylation and, as a consequence, the targeted destruction of b-catenin via the ubiquitin-proteasome pathway. APC was originally identiﬁed as a classical tumor suppressor gene that is mutated in a large proportion of colon cancers (Kinzler and Vogelstein, 1996; Polakis, 1997). The ﬁrst indication that the tumor-suppressive function of APC might be linked to the regulation of cytoplasmic b-catenin was that it speciﬁcally associates with b-catenin via several copies of a 20-aminoacid repeat sequence in the central portion of the protein (Su et al., 1993; Rubinfeld et al., 1993). Colon carcinoma cell lines expressing only truncated APC proteins lacking the 20-amino-acid repeat sequences contain high levels of free cytoplasmic b-catenin and signiﬁcant levels of nuclear b-catenin (Munemitsu et al., 1995). Reintroduction of wild-type (WT) APC into these cells results in a dramatic downregulation of cytoplasmic b-catenin, demonstrating a direct role for APC in the regulation of cytosolic b-catenin (Munemitsu et al., 1995). Consistent with this, b-catenin was found to be distributed throughout the cytoplasm and nucleus of tumor tissue from FAP patients, but was conﬁned to the cell-to-cell contacts of normal intestinal mucosa from the same patients (Inomata et al., 1996). Furthermore, the nuclei of the APC⫺/⫺ colon carcinoma cells were found to contain a constitutively active Tcf-4/b-catenin complex, which dissociates following reintroduction of WT APC (Korinek et al., 1997). Wnt-1 expression counteracts this negative regulation by APC and consequently promotes an increase in cytoplasmic b-catenin levels (Papkoff et al., 1996). In addition, this Wnt-1 expression increases the stability of APC/b-catenin complexes in these cell lines, indicating that APC may also play a positive role within Wnt signaling via complex formation with b-catenin. This is supported by ectopic expression of APC in Xenopus embryos, which generates a phenotype characteristic of Wnt-1 or b-catenin overexpression (Vleminckx et al., 1997). An APC homolog has also been identiﬁed in Drosophila (dAPC), which contains the bcatenin binding domains implicated in the regulation of b-catenin (Hayashi et al., 1997). In line with this, dAPC is able to stimulate the downregulation of b-catenin when introduced into colon carcinoma cell lines. Somewhat puzzling is the observation that mutant embryos lacking dAPC exhibit no defect in Armadillo distribution, and Wingless signaling in the developing ﬂy embryos does not appear to correlate with dAPC expression. These results suggest that APC is not essential for the regulation of Armadillo and is not directly involved in Wingless signaling. However, note that this study could

12

Nick Barker et al.

not exclude the possibility that low levels of maternal APC in these ﬂy embryos are sufﬁcient to mediate Wingless-induced regulation of Armadillo. Additionally, a second Drosophila APC gene has recently been identiﬁed that may compensate for the loss of dAPC function in these mutant ﬂy embryos (Van Es et al., 1999). A second mammalian gene, known as APCL or APC2, has very recently been cloned and partially characterised (Nakagawa et al., 1998; Van Es et al., 1999). The expression pattern of APC2 is more restricted than that of APC, with highest levels present within the brain. APC2 protein has an overall domain structure similar to that of APC, including the 20-amino-acid repeat sequences implicated as being important for b-catenin regulation and the SAMP repeats that mediate interactions with Conductin (Fig. 3). In line with this, the APC2 gene product mediates the downregulation of cytoplasmic b-catenin in colon carcinoma cell lines and interacts with Conductin in the yeast two-hybrid system (Van Es et al., 1999). Additionally, ectopic expression of APC2 in Xenopus embryos inhibits endogenous axis formation, consistent with it being a negative regulator of Wnt signaling (J. Van Es and H. Clevers, unpublished data). The ability of APC to downregulate cytoplasmic b-catenin levels is dependent on complex formation with two additional negative regulators of the Wingless/Wnt pathway: the serine-threonine kinase, glycogen synthase kinase 3b (GSK3b), and Conductin/Axin (Zeng et al., 1997; Behrens et al., 1998). Genetic studies in Drosophila originally identiﬁed Zeste-white 3 (also known as Shaggy), the ﬂy homolog of GSK3b, as a negative regulator of the Wingless signaling pathway (Noordermeer et al., 1994; Peifer et al., 1994). Suppression of endogneous GSK3b activity in the ventral side of Xenopus embryos through ectopic expression of dominant negative (kinase dead) GSK3b results in duplicate axis formation, indicative of a similar negative regulatory function for GSK3b in Wnt signaling (He et al., 1995; Yost et al., 1996). There is also a strong correlation between loss of GSK3b activity and increased b-catenin levels in the ventral side of these embryos, indicating that GSK3b acts to downregulate b-catenin in the absence of a Wnt signal. Biochemical analyses have demonstrated that GSK3b binds to and phosphorylates APC in the presence of excess cytoplasmic b-catenin (Rubinfeld et al., 1996). This phosphorylation of APC results in an increase in its afﬁnity for b-catenin in vitro. GSK3b activity within the complex also mediates the phosphorylation of the b-catenin on several serine-threonine residues at its N terminus (Orford et al., 1997). As a consequence of this modiﬁcation, the bcatenin is targeted for destruction by the ubiquitin-proteasome pathway (Aberle et al., 1997). Conductin was originally identiﬁed during a two-hybrid screen for proteins interacting with b-catenin (Behrens et al., 1998). It was subsequently shown to have additional binding sites for APC and GSK3b and to readily

Fig. 3 A comparison of the domain structures of APC and APC2. Conserved domains are depicted as various shaded boxes. (Adapted with permission of Elsevier Science from Van Es, J., et al., Identiﬁcation of APC2, a homolog of the adenamatous polyposis coli tumor suppressor, Curr. Biol., Vol. 9, No. 2, 1999, p. 105.)

14

Nick Barker et al.

form ternary complexes with these proteins following transient overexpression in cell lines. Ectopic expression of Conductin in Xenopus embryos suppresses endogenous axis formation, and overexpression of dominant-negative mutants lacking the binding sites for GSK3b, APC, and b-catenin in cell lines results in the stabilization of cytoplasmic b-catenin (Behrens et al., 1998). Taken together, these results indicate that Conductin is an essential component of the multiprotein complex that facilitates the degradation of cytoplasmic b-catenin. It was proposed that Conductin acts as a docking platform for the APC, GSK3b, and b-catenin proteins, thereby promoting the formation of a complex between these Wnt signaling components. In response to an appropriate Wnt signal, GSK3b is inactivated and cytoplasmic b-catenin is stabilized (Yost et al., 1996; Papkoff and Aikawa, 1998). The precise mechanism by which GSK3b is inactivated in response to Wnt signaling remains elusive. However, a novel protein, termed GSK3b-binding protein (GBP), has recently been identiﬁed in Xenopus that is essential for body axis formation (Yost et al., 1998). GBP and similar mammalian proteins known as Frats (Jonkers et al., 1997) bind to GSK3b and inhibit its kinase activity, thereby preventing the phosphorylation of target proteins such as b-catenin. It is likely that this GBP inhibition of GSK3b activity facilitates the dorsal accumulation of b-catenin in early Xenopus embryos (Yost et al., 1998). Following the resulting increase in the levels of free cytoplasmic b-catenin there is a concomitant translocation of b-catenin to the nucleus, where it binds Tcfs to generate a transcriptionally active complex. This nuclear translocation of b-catenin may simply be mediated by binding to newly synthesized Tcf protein in the cytoplasm and “hitching a ride” into the nucleus courtesy of the Tcf nuclear localization signal (NLS) (Molenaar et al., 1996; Behrens et al., 1996; Huber et al., 1996). However, this is unlikely to be the main nuclear translocation route since b-catenin/Armadillo mutants that cannot bind Tcfs retain the ability to enter the nucleus (Orsulic and Peifer, 1996b; van de Wetering et al., 1997). Indeed, more recent evidence indicates that this may be an active process, involving binding of b-catenin to the nuclear pore machinery (Fagotto et al., 1998). Regulating the formation of transcriptionally active Tcf/b-catenin complexes is also achieved by post-translational modiﬁcation of the Tcf protein itself. This regulatory mechanism, which has been identiﬁed in Drosophila, involves complex formation between dTCF and the Drosophila CREB-binding protein (dCBP) (Waltzer and Bienz, 1998). dCBP mutants exhibit wingless overactivation phenotypes, consistent with a negative role within the Wingless signaling pathway. Biochemical analyses revealed that dCBP can acetylate dTCF in the Armadillo-binding domain, thereby lowering the afﬁnity of dTCF for Armadillo. It was proposed that in the absence of a Wingless signal the acetylation of dTCF, combined with the low levels of nuclear Armadillo, inhibits the recruitment of Armadillo by dTCF to Wingless tar-

Yin-Yang of TCF/b-Catenin Signaling

15

get genes. Following Wingless signaling, the resulting increase in Armadillo levels is sufﬁcient to overcome this dCBP inhibition, and nuclear dTCF/Armadillo complexes mediate target gene activation.

VIII. REGULATION OF WINGLESS/Wnt TARGET GENES BY Tcfs: ACTIVATION VERSUS REPRESSION A. Activation of Target Genes by Tcfs Several genes whose expression is modiﬁed in response to Wingless signaling have been identiﬁed in Drosophila. The majority of these genes have not been studied in enough detail to determine whether this Wingless responsiveness is linked to Tcf/Armadillo activity. However, notable exceptions include the engrailed and Ultrabithorax genes. Engrailed is a segment polarity gene, which encodes a homeobox-containing protein involved in anterior-posterior compartmentalization of the ﬂy (DiNardo et al., 1988). The minimal enhancer of engrailed contains a potential Tcf binding site and, consistent with this observation, engrailed expression is decreased in dTCF mutant embryos (van de Wetering et al., 1997). Regulation of expression of a second Wingless-responsive gene, Ultrabithorax (Ubx), has been studied in much greater detail (Bienz, 1994). Expression of Ubx in mesodermal cells lining the Drosophila midgut is self-regulating due to a positive feedback loop involving signaling by both Wingless and the product of a Ubx target gene, Decapentaplegic (Dpp). Both Wg and Dpp are required for Ubx expression. The midgut enhancer of Ubx was found to contain a Wingless responsive element adjacent to a Dpp responsive element (Riese et al., 1997). The Wingless responsive element contains an optimal Tcf/Lef binding site that, when occupied with Lef/Armadillo complexes, stimulates transcriptional activation via the Ultrabithorax enhancer. Overexpression of Lef-1 bypasses the requirement for Wingless signaling and mimics the Wingless gainof-function mutations in the midgut, wing, and notum. This Lef-1-induced transactivation of the Ubx enhancer is dependent on concomitant signaling via the Dpp responsive element. Thus, Lef-1 appears to coordinate inputs from two positional signals, Wingless and Dpp. This bears a striking resemblance to the presence of both a Lef-1 binding site and a CRE in the TCRa enhancer (Carlsson et al., 1993; Giese et al., 1995; Riese et al., 1997). It may be that Lef-1 plays similar architectural roles in the Ubx and TCRa enhancer, regulating the assembly of functional multiprotein complexes. The first Wnt responsive target gene to be identified in vertebrates was the Xenopus zygotic gene, Siamois, which plays a crucial role in dorsal axis formation in the developing embryo (Brannon et al., 1997). Siamois was

16

Nick Barker et al.

first identified as a candidate downstream target of the Wnt signaling pathway when ectopic expression was found to cause axis duplication, similar to that seen previously following ectopic expression of Wnt and b-catenin. Additionally, Siamois expression was shown to be induced by Wnt and bcatenin in the absence of transforming growth factor b (TGF-b). The subsequent isolation of the Siamois promoter revealed the presence of a region that was necessary for this response to b-catenin. Importantly, this region contained three XTcf-3 binding sites, which were found to mediate the transactivation of Siamois by XTcf-3/b-catenin complexes during dorsal axis formation. The Xenopus nodal-related 3 (Xnr-3) gene encodes a member of the TGFb family, which is important for the activity of the Spemann organizer during embryonic development. Two distinct enhancer elements confer Wnt inducibility on Xnr-3 (McKendry et al., 1997). The ﬁrst element interacts with a factor that accumulates independently of Wnt signaling during gastrulation. The second speciﬁcally interacts with Lef-1. Importantly, ectopic expression of Lef-1 causes transcriptional activation of Xnr-3, indicating that this gene is a direct target of Tcf/Lef-1 during Xenopus development.

B. Repression of Target Genes by Tcfs The effective regulation of Wingless/Wnt responsive Tcf target genes relies not only on the transactivating potential of Tcfs when complexed with bcatenin/Armadillo, but also on their active repression of target gene expression in unstimulated cells. Tcfs were initially implicated in active repression of Tcf target genes when mutation of the dTCF binding site in the Ubx midgut enhancer was found to promote an increase in expression within cells located distal to the source of the Wingless signal (Riese et al., 1997). In contrast, cells located proximal to the Wingless source showed decreased Ubx expression as the result of loss of dTCF/Armadillo transactivation. During Xenopus development, expression of the Wnt-responsive Tcf target gene Siamois is restricted to the dorsal side of the embryo (Brunner et al., 1997). This correlates well with the distribution of b-catenin, which is present at high levels in the dorsal side of the embryo. However, mutation of the XTcf-3 binding sites in the Siamois enhancer causes a dramatic increase in Siamois expression in the ventral side of the developing embryo. Together, these observations indicate that in the absence of Wingless/Wnt-induced accumulation of b-catenin/Armadillo, Tcfs actively repress transcription of target genes (Brunner et al., 1997). Transcriptional repression by Tcfs involves complex formation with members of the Groucho family of transcriptional repressors (Roose et al., 1998; Cavallo et al., 1998). Groucho (Gro) is a broadly expressed Drosophila corepressor that is important for early segmentation, neurogenesis, and sex

Yin-Yang of TCF/b-Catenin Signaling

17

determination (Hartley et al., 1988; Paroush et al., 1994; Parkhurst, 1998). Enhancer of Split-like (HES) and Hairy factors interact with the non-DNAbinding Gro protein to repress transcription of their target genes (Paroush et al., 1994; Jiminez et al., 1997). Multiple homologs with a similar domain structure exist in mammals. These are known as TLE 1–4 in man, mGrg-1, -3, -4 in mouse, and XGrg-4 and XGrg-5 in Xenopus. An association with Tcf proteins was ﬁrst identiﬁed when the mouse Groucho-related homolog Grg-5 was cloned during a two-hybrid screen for proteins interacting with hTcf-1 (Roose et al., 1998). XGrg-4 inhibits transcriptional activation of synthetic Tcf reporter genes in transient transfection assays and represses transcription of Siamois and Xnr-3 target genes upon injection into Xenopus embryos. Additionally, secondary axis formation induced by a dominant-positive version of XTcf-3, in which the N terminus was replaced by the C-terminal Armadillo transactivation domain, was inhibited by XGrg-4. A genetic analysis of Groucho function in Drosophila demonstrated that mutations in the Gro gene suppress Wingless and Armadillo mutant phenotypes, leading to a derepression of Wingless-responsive dTCF target genes (Cavallo et al., 1998). It is therefore likely that regulation of Tcf target gene expression is a balance between constitutive repression and inducible activation mediated by recruitment of Groucho and b-catenin/Armadillo, respectively. In the absence of Wingless/Wnt signals, constitutive Tcf/Groucho complexes will occupy the enhancers of Tcf target genes, repressing their transcription. Wingless/Wnt activation promotes b-catenin accumulation, facilitating the formation of Tcf/b-catenin complexes and transactivation of target genes. The dual activities of Tcfs may explain the puzzling observation that mutation of the Pop-1 gene (a C. elegans Tcf homolog) produces opposite effects on E versus MS cell speciﬁcation to those resulting from loss of function mutations in Wnt or b-catenin (Rocheleau et al., 1997; Thorpe et al., 1997). It is likely that Pop-1 predominantly functions as a transcriptional repressor in the worm.

IX. Tcf/b-CATENIN SIGNALING AND CANCER The ﬁrst hint of a link between Wnt signaling and cancer was observed when ectopic expression of Wnt-1 was found to promote the formation of mouse mammary tumors (Nusse and Varmus, 1982). However, subsequent research has not been able to establish a direct link between Wnt signaling and human breast cancer. During recent years, much attention has focused on the molecular events contributing to the transformation of colon epithelial cells following loss of function of the tumor suppressor protein APC. The vast majority of germline and somatic mutations in APC result in the for-

18

Nick Barker et al.

Table I RT-PCR Assay for Tcf Expression in Human Cancers Tcf-1 Tumor origin Colon Lung Stomach Esophagus Head/neck Prostate Ovary

Lef-1

Tcf-3

Tcf-4

Samples

%a

Samples

%

Samples

%

Samples

%

43 16 4 4 8 4 6

53 75 100 25 38 25 50

42 15 3 4 8 4 6

88 87 33 100 88 100 83

43 16 4 4 8 4 6

50 81 25 100 100 75 83

43 16 4 4 8 4 6

100 94 100 100 100 100 100

a% denotes the percentage of samples positive for Tcf expression.

mation of extensively truncated proteins. The most common route to colon carcinoma is inheritance of a mutant copy of APC, followed by somatic mutation in the second allele (LOH, loss of heterozygosity). Colonic epithelial cells that display this LOH express only truncated forms of APC, which lack a central domain essential for the downregulation of cytoplasmic b-catenin (Munemitsu et al., 1995). The resulting elevation in the levels of cytoplasmic b-catenin in these cells promotes its nuclear translocation, where it associates with Tcf-4, a member of the Tcf family expressed in normal colonic epithelium (Korinek et al., 1997) and colon carcinoma cell lines (Table I). The resulting bipartite transcription factor is capable of activating expres-

Fig. 4 A model for Tcf/␤-catenin signaling in the intestinal epithelium. I Intestinal stem cell: An appropriate Wnt signal is delivered by the surrounding mesenchyme, which activates Dsh and leads to inhibition of GSK3␤ activity. ␤-catenin is thus stabilized and is concomittantly translocated to the nucleus, where it binds to Tcf-4. This active Tcf-4/␤-catenin complex mediates transcriptional activation of c-myc and presumably of other Tcf target genes required for stem cell activity. This signaling activity also induces de novo Lef-1 expression, which leads to the formation of transcriptionally active Lef-1/␤-catenin complexes. II. Normal intestinal epithelial cell: In the absence of a Wnt signal, complexes comprising APC, GSK3␤, Conductin/ Axin, and ␤-catenin are present in the cytoplasm. The activity of GSK3␤ within this complex mediates the N-terminal phosphorylation of ␤-catenin, which targets the ␤-catenin for degradation by the ubiquitin-proteasome pathway. As a consequence, ␤-catenin levels are dramatically reduced and the formation of nuclear Tcf-4/␤-catenin complexes is inhibited. In the absence of nuclear ␤-catenin, Tcf-4 recruits the corepressor protein Groucho to the enhancers of Tcf target genes and actively represses their transcription. III Colon carcinoma cells: Mutation of APC or ␤-catenin itself severely impairs the downregulation of ␤-catenin in the absence of a Wnt signal. This leads to accumulation of ␤-catenin in the cytoplasm and nucleus, which promotes the formation of transcriptionally active Tcf-4/␤-catenin and Lef-1/␤-catenin complexes. Tcf target genes, including c-myc are thus constitutively activated, which results in uncontrolled cellular proliferation and polyp formation.

20

Nick Barker et al.

sion of Tcf target genes. The high levels of b-catenin in these cell lines also mediate the de novo transcription of Lef-1, indicating that Lef-1 may itself be a Tcf-4 target gene (Porﬁri et al., 1997). Reintroduction of wild-type APC into these cells downregulates the cytoplasmic b-catenin and abolishes the Tcf/b-catenin-mediated transcription in their nuclei (Munemitsu et al., 1995; Korinek et al., 1997). However, APC is not the sole target for mutation in this pathway. In a proportion of colon carcinoma cell lines expressing wild-type APC, b-catenin itself acquires oncogenic activity as the result of dominant mutations within its N terminus (Morin et al., 1997). Many of these mutations lead to loss of phosphorylation consensus sites implicated in the negative regulation of b-catenin by the GSK3b/APC/Conductin complex (Peifer et al., 1994; Yost et al., 1996; Ilyas et al., 1997). As a consequence of this deregulation, b-catenin levels are high in these cells and Tcf/ b-catenin complexes are constitutively present in their nuclei. Similar mutations were found in malignant melanoma, inducing a prolonged half-life of b-catenin and the formation of constitutive nuclear complexes with Lef-1 (Rubinfeld et al., 1997). To date, no mutations have been identiﬁed in other components of the Wnt pathway in colon carcinoma cells (Sparks et al., 1998). Thus, activating mutations in b-catenin or loss of function mutations in APC lead to the same transcriptional deregulation: the constitutive transcription of Tcf target genes (Fig. 4). This deregulation of Tcf target gene expression is considered likely to be a critical event in the malignant transformation of colonic epithelium. Two recent studies have provided evidence to support this hypothesis. Analysis of Tcf-4 knockout mice demonstrated a unique role for Tcf-4 in maintaining the epithelial stem cells within the nascent crypts of the small intestine (Korinek et al., 1998b). Maintenance of these proliferative compartments of the small intestine is mediated via Tcf-4/b-catenin signaling. Wnt factors secreted by the gut mesenchyme are prime candidates for initiating this signaling pathway in the stem cells. Constitutive activation of Tcf-4 signaling in APCdeﬁcient cells may therefore be predicted to promote the maintenance of stem cell characteristics, such as cycling and longevity, in cells that would normally terminally differentiate into one of the epithelial cell types. Support for this model was provided by the recent identiﬁcation of c-myc as a Tcf-4 target gene in colon cells (He et al., 1998). By employing a technique known as serial analysis of gene expression (SAGE), expression of c-myc was found to be directly regulated by the APC pathway. c-myc expression is repressed by wild-type APC and activated by b-catenin. Signiﬁcantly, these effects are mediated via Tcf-4-binding motifs within the c-myc promoter. Overexpression of c-myc, which is consistently observed in colon carcinoma, is therefore likely to be a direct consequence of deregulation of Tcf-4/b-catenin signaling through mutations in APC or b-catenin. The oncogenic activities of c-myc are well documented and its overexpression in response to constitu-

Yin-Yang of TCF/b-Catenin Signaling

21

tive Tcf-4/b-catenin signaling is likely to promote an unchecked expansion of epithelial stem cells of the small intestine. Tumorigenesis in APCMin mice is known to affect the multipotent stem cell in the intestinal crypt and the resulting expansion of these crypt cells is proposed to result in the formation of an intestinal microadenoma, which then expands into a neighboring villus (Moser et al., 1992; Oshima et al., 1995). This mechanism accounts for the formation and characteristics of polyps in the APCMin mice and the obligatory activation of Tcf-4 transcriptional activity via mutations in APC or bcatenin, as found in human cancer (Fig. 4). Based on its critical role in so many key developmental processes during embryogenesis it is perhaps not unreasonable to consider that deregulation of Wnt signaling activity may be linked to other forms of cancer. Indeed, a recent study of Wnt signaling in murine hair development demonstrated that deregulation of b-catenin levels through N-terminal mutation resulted in the formation of hair tumors (Gat et al., 1998). Interestingly, this provides a possible explanation for the incidence of hair tumors in some patients suffering from familial colon cancer. The rapid rate at which our understanding of these developmental processes is progressing will undoubtedly be of great beneﬁt for the diagnosis and treatment of human cancers in years to come.

ACKNOWLEDGMENTS We are grateful to Johan Van Es for his constructive comments on the manuscript.

REFERENCES Aberle, H., Bauer, A., Stappert, J., Kispert, A., and Kemler, R. (1997). EMBO J. 13, 3797–3804. Barker, N., Huls, G., and Clevers, H. (1999). Am. J. Pathol. 154, 29–35. Behrens, J., von Kries, J. P., Kuehl, M., Bruhn, L., Wedlich, D., Grosschedl, R., and Birchmeier, W. (1996). Nature (London) 382, 638–642. Behrens, J., Jerchow, B.-A., Würtele, M., Grimm, J., Asbrand, C., Wirtz, R., Kühl, M., Wedlich, D., and Birchmeier, W. (1998). Science 280, 596–599. Bhanot, P., Brink, M., Samos, C., Hsieh, J.-C., Wang, Y., Macke, J., Andrew, D., Nathans, J., and Nusse, R. (1996). Nature (London) 382, 225–230. Bienz, M. (1994). Trends Genet. 10, 22–26. Brannon, M., Gomperts, M., Sumoy, L., Moon, R., and Kimelman, D. (1997). Genes Dev. 11, 2359–2370. Bruhn, L., Munnerlyn, A., and Grosschedl, R. (1997). Genes Dev. 11, 640–653. Brunner, E., Peter, O., Schweizer, L., and Basler, K. (1997). Nature (London) 385, 829–833. Carlsson, P., Waterman, P., and Jones, K. (1993). Genes Dev. 7, 2418–2430.

22

Nick Barker et al.

Castrop, J., van Norren, K., and Clevers, H. C. (1992a). Nucleic Acids Res. 20, 611. Castrop, J., Hoevenagel, R., Young, J., and Clevers, H. (1992b). Eur. J. Immunol. 22, 1327– 1330. Castrop, J., van Wichen, D., Koomans-Bitter, M., van de Wetering, M., de Weger, R., van Dongen, J., and Clevers, H. (1995). Blood 85, 3050–3059. Cavallo, R., Cox, R., Moline, M., Roose, J., Polevoy, G., Clevers, H., Peifer, M., and Bejsovec, A. (1998). Nature (London) 395, 604–608. Clevers, H., and van de Wetering, M. (1997). Trends Genet. 13, 485–489. Cook, D., Fry, M., Hughes, K., Sumathipala, R., Woodgett, J. R., and Dale, T. C. (1996). EMBO J. 15, 4526–4536. DiNardo, S., Sher, E., Heemskerk-Jongens, J., Kassis, J., and O’Farrell, P. (1988). Nature (London) 332, 604–609. Fagotto, F., Glück, U., and Gumbiner, B. (1998). Curr. Biol. 8, 181–190. Funayama, N., Fagotto, F., McCrea, P., and Gumbiner, B. (1995). J. Cell Biol. 128, 959–968. Gat, U., DasGupta, R., Degenstein, L., and Fuchs, E. (1998). Cell (Cambridge, Mass.) 95, 1–20. Giese, K., and Grosschedl, R. (1992). EMBO J. 12, 4667–4676. Giese, K., Cox, J., and Grosschedl, R. (1992). Cell (Cambridge, Mass.) 69, 185–195. Giese, K., Kingsley, C., Kirshner, J., and Grosschedl, R. (1995). Genes Dev. 9, 995–1008. Hartley, D., Preiss, A., and Tsakonas, S. (1988). Cell (Cambridge, Mass.) 55, 789–795. Hayashi, S., Rubinfeld, B., Souza, B., Polakis, P., Wieschaus, E., and Levine, A. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 242–247. He, X., Saint-Jeanett, J.-P., Woodgett, J., Varmus, H., and Dawid, I. (1995). Nature (London) 374, 617–622. He, T-C., Sparks, A., Rago, C., Hermeking, H., Zawel, L., da Costa, L., Morin, P., Vogelstein, B., and Kinzler, K. (1998). Science 281, 1509–1512. Hsu, S.-C., Galceran, J., and Grosschedl, R. (1998). Mol. Cell. Biol. 18, 4807–4818. Huber, A., Nelson, J., and Weis, W. (1997). Cell (Cambridge, Mass.) 90, 871–882. Huber, O., Korn, R., McLaughlin, J., Ohsugi, M., Hermann, B. G., and Kemler, R. (1996). Mech. Dev. 59, 3–10. Ilyas, M., Tomlinson, P., Rowan, A., Pignatelli, M., and Bodmer, W. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 10330–10334. Inomata, M., Ochiai, A., Akimoto, S., Kitano, S., and Hirohashi, S. (1996). Cancer Res. 56, 2213–2217. Jiminez, G., Paroush, Z., and Harowicz, D. (1997). Genes Dev. 11, 3072–3082. Jonkers, J., Korswagen, H., Acton, D., Bruer, M., and Berns, A. (1997). EMBO J. 16, 441–450. Kinzler, K., and Vogelstein, B. (1996). Cell (Cambridge, Mass.) 87, 159–170. Korinek, V., Barker, N., Morin, P., van Wichen, D., de Weger, R., Kinzler, K. W., Vogelstein, B., and Clevers, H. (1997). Science 275, 1784–1787. Korinek, V., Barker, N., Willert, K., Molenaar, M., Roose, J., Wagenaar, G., Markman, M., Lamers, W., Destree, O., and Clevers, H. (1998a). Mol. Cell. Biol. 18, 1248–1256. Korinek, V., Barker, N., Moerer, P., van Donselaar, E., Huls, G., Peters, P., and Clevers, H. (1998b). Nat. Genet. 19, 279–283. Laudet, V., Stehelin, D., and Clevers, H. (1993). Nucleic Acids Res. 21, 2493–2501. Lin, R., Thompson, S., and Priess, J. (1995). Cell (Cambridge, Mass.) 83, 599–609. Love, J., Li, X., Case, D., Giese, K., Grosschedl, R., and Wright, P. (1995). Nature (London) 376, 791–795. McKendry, R., Hsu, S.-C., Harland, R., and Grosschedl, R. (1997). Dev. Biol. 192, 420– 431. Milatovich, A., Travis, A., Grosschedl, R., and Francke, U. (1991). Genomics 11, 1040–1048. Miller, J., and Moon, R. (1996). Genes Dev. 10, 2527–2539. Molenaar, M., van de Wetering, M., Oosterwegel, M., Peterson-Maduro, J., Godsave, S., Ko-

Yin-Yang of TCF/b-Catenin Signaling

23

rinek, V., Roose, J., Destree, O., and Clevers, H. (1996). Cell (Cambridge, Mass.) 86, 391–399. Moon, R., Brown, J., and Torres, M. (1997). Trends Genet. 13, 157–162. Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B., and Kinzler, K. (1997). Science 275, 1787–1790. Moser, A. R., Dove, W. F., Roth, K. A., and Gordon, J. I. (1992). J. Cell Biol. 116, 1517–1526. Munemitsu, S., Albert, I., Souza, B., Rubinfeld, B., and Polakis, P. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 3046–3050. Nakagawa, H., Murata, Y., Koyama, K., Fujiyama, A., Miyoshi, Y., Monden, M., Akiyama, T., and Nakamura, Y. (1998). Cancer Res. 58, 5176–5181. Noordermeer, J., Klingensmith, J., Perrimon, N., and Nusse, R. (1994). Nature (London) 367, 80–83. Nusse, R., and Varmus, H. (1982). Cell (Cambridge, Mass.) 31, 99–109. Okamura, R., Sigvardsson, M., Galceron, J., Verbeek, S., Clevers, H., and Grosschedl, R. (1998). Immunity 8, 11–20. Oosterwegel, M., van de Wetering, M., Timmerman, J., Kruisbeek, A., Destree, O., Meijlink, F., and Clevers, H. (1993). Development (Cambridge, UK) 118, 439–448. Orford, K., Crockett, C., Jensen, J., Weismann, A., and Byers, S. (1997). J. Biol. Chem. 272, 24735–24738. Orsulic, S., and Peifer, M. (1996a). Curr. Biol. 6, 1363–1367. Orsulic, S., and Peifer, M. (1996b). J. Cell Biol. 134, 1283–1301. Oshima, M., Oshima, H., Kitagawa, K., Kobayashi, M., Itakura, C., and Takito, M. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 4482–4486. Papkoff, J., and Aikawa, M. (1998). Biochim. Biophys. Acta 247, 851–858. Papkoff, J., Rubinfeld, B., Schryver, B., and Polakis, P. (1996). Mol. Cell. Biol. 16, 2128–2134. Parkhurst, S. (1998). Trends Genet. 14, 130–132. Paroush, Z., Finley, R., Kidd, T., Wainwright, S., Ingham, P., Brent, R., and Ish-Horowicz, D. (1994). Cell (Cambridge, Mass.) 79, 805–815. Peifer, M., Sweeton, D., Casey, M., and Wieschaus, E. (1994). Development (Cambridge, UK) 120, 369–380. Polakis, P. (1997). Biochim. Biophys. Acta 1332, 127–147. Porﬁri, E., Rubinfeld, B., Albert, I., Hovanes, K., Waterman, M., and Polakis, P. (1997). Oncogene 15, 2833–2839. Riese, J., Yu, X., Munnerlyn, A., Eresh, S., Hsu, S.-C., Grosschedl, R., and Bienz, M. (1997). Cell (Cambridge, Mass.) 88, 777–787. Rocheleau, C., Downs, W., Lin, R., Wittmann, C., Bei, Y., Cha, Y.-H., Ali, M., Priess, J., and Mello, C. (1997). Cell (Cambridge, Mass.) 90, 707–716. Roose, J., Molenaar, M., Peterson, P., Hurenkamp, J., Brantjes, H., Moerer, P., van de Wetering, M., Destree, O., and Clevers, H. (1998). Nature (London) 395, 608–612. Rubinfeld, B., Souza, B., Albert, I., Muller, O., Chamberlain, S. H., Masiarz, F. R., Munemitsu, S., and Polakis, P. (1993). Science 262, 1731–1734. Rubinfeld, B., Albert, I., Porﬁri, E., Fiol, C., Munemitsu, S., and Polakis, P. (1996). Science 272, 1023–1026. Rubinfeld, B., Robbins, P., El-Gamil, M., Albert, I., Porfori, E., and Polakis, P. (1997). Science 275, 17990–1792. Sokol, S., Christian, J., Moon, R., and Melton, D. (1991). Cell (Cambridge, Mass.) 67, 741–752. Sparks, A., Morin, P., Vogelstein, B., and Kinzler, K. (1998). Cancer Res. 58, 1130–1134. Su, L. K., Vogelstein, B., and Kinzler, K. (1993). Science 262, 1734–1736. Thorpe, C., Schlesinger, A., Carter, J., and Bowerman, B. (1997). Cell (Cambridge, Mass.) 90, 695–705.

24

Nick Barker et al.

Travis, A., Amsterdam, A., Bélanger, C., and Grosschedl, R. (1991). Genes Dev. 5, 880–894. van de Wetering, M., and Clevers, H. (1992). EMBO J. 11, 3039–3044. van de Wetering, M., Oosterwegel, M., Dooijes, D., and Clevers, H. (1991). EMBO J. 10, 123–132. van de Wetering, M., Castrop, J., Korinek, V., and Clevers, H. (1996). Mol. Cell. Biol. 16, 745–752. van de Wetering, M., Cavallo, R., Dooijes, D., van Beest, M., van Es, J., Loureiro, J., Ypma, A., Hursh, D., Jones, T., Bejsovec, A., Peifer, M., Mortin, M., and Clevers, H. (1997). Cell (Cambridge, Mass.) 88, 789–799. Van Es, J., Kirkpatrick, C., van de Wetering, M., Molenaar, M., Miles, A., Kuipers, J., Destreé, O., Peifer, M., and Clevers, H. (1998). Curr. Biol. 9, 105–108. van Genderen, C., Okamura, R. M., Farinas, I., Quo, R. G., Parslow, T. G., Bruhn, L., and Grosschedl, R. (1994). Genes Dev. 8, 2691–2703. Verbeek, S., Izon, D., Hofhuis, F., Robanus-Mandaag, E., te Riele, H., van de Wetering, M., Oosterwegel, M., Wilson, A., MacDonald, R., and Clevers, H. (1995). Nature (London) 374, 70–74. Vleminckx, K., Wang, E., Guger, K., Rubinfeld, B., Polakis, P., and Gumbiner, B. (1997). J. Cell Biol. 136, 411–420. Waltzer, L., and Bienz, M. (1998). Nature (London) 395, 521–525. Waterman, M., Fischer, W., and Jones, K. (1991). Genes Dev. 5, 656–669. Yanagawa, S., van Leeuwen, F., Wodarz, A., Klingensmith, J., and Nusse, R. (1995). Genes Dev. 9, 1087–1097. Yost, C., Torres, M., Miller, J., Huang, E., Kimelman, D., and Moon, R. (1996). Genes Dev. 10, 1443–1454. Yost, C., Farr, G., III, Pierce, S., Ferkey, D., Chen, M., and Kimelman, D. (1998). Cell (Cambridge, Mass.) 93, 1031–1041. Zeng, L., Fagotto, F., Zhang, T., Hsu, W., Vasicek, T., Perry, W., III, Lee, J., Tilghman, S., Gumbiner, B., and Constantini, F. (1997). Cell (Cambridge, Mass.) 90, 181–192.

Biochemical and Clinical Implications of the ErbB/HER Signaling Network of Growth Factor Receptors Leah N. Klapper,1 Mark H. Kirschbaum,2 Michael Sela,1 and Yosef Yarden2 2

1 Department of Immunology Department of Biological Regulation The Weizmann Institute of Science Rehovot 76100, Israel

I. Introduction II. Clinical Aspects of ErbB Receptors A. Breast Cancer B. Gynecological Cancers C. Prostate Cancer D. Gastrointestinal Cancer E. Lung Cancer F. Head and Neck Cancer G. Kidney Cancer H. Bladder Cancer I. Brain Tumors III. How Does ErbB-2 Induce Cancer? A. In Vitro Transforming Potential of ErbB-2 B. ErbB Activating Ligands C. Does ErbB-2 Have a Ligand of Its Own? D. Ligand-Independent Receptor Dimerization IV. Evolutionary and Developmental Aspects of the Multiplicity of ErbB Proteins V. The ErbB Signaling Network A. Evidence for Inter-Receptor Interactions B. Transforming Ability of Heterodimers C. Ligand-Bivalency Selects Dimer Participants D. Extending the Variation of Signaling Complexes by Diversiﬁcation of Ligand Recognition E. Intracellular Signaling F. Tuning of ErbB Signaling by Receptor Endocytosis VI. ErbB-Directed Cancer Therapy A. Immunotherapy B. Gene Therapy C. Other Modes of Therapy VII. Conclusions References

Advances in CANCER RESEARCH 0065-230X/00 $30.00

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

26

Leah N. Klapper et al.

Carcinoma, cancer of epithelial cells, is a major cause of morbidity and mortality in Western societies. Clonal ﬁxation and propagation of oncogenic genetic changes, sporadically accumulating in epithelial cells, depend on growth factors and their surface receptors. One of the large families of receptors is that of the ErbB tyrosine kinases, which bind multiple neuregulins and other epidermal growth factor-like molecules. Certain ErbB members and their ligands are involved in human cancers of various origins. However, most of the clinical data relate to ErbB-2, a protein whose overexpression in subsets of carcinomas can predict poor prognosis. Although no ligand has so far been assigned to ErbB-2, recent biochemical evidence implies that this oncoprotein operates as a shared receptor subunit of other ErbBs. Several biochemical attributes enable ErbB-2 to act as an epithelial cell ampliﬁer of stroma-derived growth factor signals: It delays ligand dissociation, enhances coupling to the mitogen-activated protein kinase pathway, and impedes the rate of receptor downregulation. The realization that ErbB-2 is a master regulator of a signaling network that drives epithelial cell proliferation identiﬁes this protein as a target for cancer therapy. Indeed, various ErbB-2-directed therapeutic approaches, including immunological and genetic therapies, demonstrate promising clinical potential. © 2000 Academic Press.

I. INTRODUCTION Cellular transformation, underlying the promotion and progression of human tumors, is, to date, acknowledged as the result of cumulative independent mutations (Fearon and Vogelstein, 1990; Kinzler and Vogelstein, 1996; Nowell, 1976). Numerous players of a cellular network designed to convey proliferative signals are plausible and often established candidates for such transformation-driving mutations (Bishop, 1991). Among these, polypeptide growth factors, as well as their receptors, comprise a group of major activators of such cellular signaling pathways, enabling the inﬂux of information and mediating its frequency and intensity, making these molecules prime suspects in the promotion of pathophysiology. Activation of a growth factor receptor was ﬁrst linked to human cancers via identiﬁcation of the homology between the epidermal growth factor receptor (EGFR) and the viral oncogene v-erbB, encoded by an avian erythroblastosis retrovirus (Downward et al., 1984). Cancers of the rat nervous system, induced by a chemical carcinogen, led to the discovery of an EGFRrelated 185-kDa phosphoprotein, designated the Neu oncoprotein (Padhy et al., 1982; Schechter et al., 1984). A single point mutation, replacing a valine with a glutamic acid in the transmembranal region of the Neu oncoprotein, provided its transforming capability (Bargmann et al., 1986). Independently, sequence similarity to the erbB-1 gene, encoding for the EGFR, resulted in the isolation of the human ortholog of neu, HER2 (for human EGF receptor 2) or erbB-2 (Coussens et al., 1985; King et al., 1985; Yamamoto et al., 1986). Screening of genomic DNA and messenger RNAs with probes de-

ErbB/HER Signaling Network of Growth Factor Receptors

27

rived from ErbBs allowed isolation of two additional relatives of the human erbB-1 gene. These were named erbB-3 (or HER3) and erbB-4 (or HER4) (Kraus et al., 1989; Plowman et al., 1990, 1993a,b). Early on, the similarity between ErbB-2 and ErbB-1 (the four ErbB proteins are compared in Fig. 1) suggested its activation by a direct ligand, and led to several attempts to isolate an ErbB-2 binding protein (Lupu et al., 1990; Yarden and Peles, 1991; Yarden and Weinberg, 1989). These ventures yielded an ErbB-2 activating molecule termed neu differentiation factor (NDF) (Peles et al., 1992; Wen et al., 1992) or heregulin (Holmes et al., 1992) only to be revealed later as the direct binding ligand of the two closely ho-

Fig. 1 The ErbB family of receptors. The prototypic protein for the four members of the family is represented as a bilobular membrane-spanning structure. An extracellular portion, stabilized by two cysteine-rich domains (CRD-1 and CRD-2), possesses the ligand binding capacity of the receptor. Its four zones show a high diversity between the family members as presented in percentage of homology to the ﬁrst identiﬁed receptor, ErbB-1. This diversity confers specific primary recognition of multiple ligands as listed for each of the receptors. Speciﬁcity is extended by cross-reactivity of several of the ligands; NRG1 and NRG2 are shared by ErbB-3 and ErbB-4; betacellulin, epiregulin, and HB-EGF are recognized by both ErbB-1 and ErbB-4. ErbB2 is ligandless. High homology of the cytoplasmic tyrosine kinase domain is altered in the case of ErbB-3 that is catalytically impaired. The C-terminal tail (CT), harboring docking sites for effector proteins of signaling pathways, exhibits a low similarity, implying a high diversity of interacting molecules.

28

Leah N. Klapper et al.

mologous receptors, ErbB-3 (Carraway et al., 1994; Kita et al., 1994; Sliwkowski et al., 1994; Tzahar et al., 1994) and ErbB-4 (Plowman et al., 1993a; Tzahar et al., 1994), thus rendering ErbB-2 orphan of a known high afﬁnity binding molecule. This contrasts with its family members capable of binding multiple ligands with some overlapping speciﬁcities (Fig. 1). Together, the four ErbB proteins comprise the ﬁrst subtype of the family of receptor tyrosine kinases (RTKs) (van der Geer et al., 1994). Typical to proteins of RTK afﬁliation, ErbB receptors bear an extracellular ligand binding domain, characterized by a relatively high diversity between family members and capable of transmitting a signal that results in the activation of the intracellular portion of the protein. Central to this activation is an intrinsic tyrosine kinase catalytic core showing close resemblance between the receptors, but wide variety in the ﬂanking sequences enabling diversity of interactions with receptor-speciﬁc effector proteins. The latter are adapter molecules, some endowed with an enzymatic activity, that share one of several phosphotyrosine binding domains (e.g., a Src homology 2 domain). Receptor activation results in phosphorylation of speciﬁc tyrosine residues located within the receptor’s cytoplasmic region, which leads to the recruitment of phosphotyrosine-binding effector proteins, and subsequent simultaneous stimulation of multiple signaling pathways. As with other allosteric enzymes, the monomeric RTK molecule is inactive but a dimeric form is fully active. The EGF-like ligands act as allosteric modiﬁers by promoting rapid receptor dimerization (Yarden and Schlessinger, 1987). Despite their overall structural similarity, the four ErbBs display differences in structures of autophosphorylation docking sites, in substrate speciﬁcity, and in potency of the kinase activity. Most remarkable is the defective kinase function of ErbB-3 and the absence of a known ligand for ErbB-2, although the transforming ability of this receptor is higher than that of other ErbBs. The prerequisite of receptor oligomerization for the activation of ErbB kinases paved the way for the understanding of the capability of ErbB-2 to transform cells regardless of its orphanhood. It suggests that receptor dimerization, between identical or sibling molecules, can induce the activation of ErbB-2, independent of a speciﬁcally binding ligand. This network of receptor interactions serving to increase the diversity and robustness of signaling induced by the activation of ErbB receptors (Alroy and Yarden, 1997) operates as a major proliferation engine of several types of cells, of which most critical for cancer are the epithelial cells. This function places the additional three members of the family, alongside ErbB-2, in the center of cancer research interests. As expected from this model, concomitant expression of these receptors or their ligands is a common feature of many adenocarcinomas (Gullick, 1990; Hynes and Stern, 1994; Salomon et al., 1995a; Stancovski et al., 1994). It is, however, most dominantly characterized by a high incidence of the ErbB-2 protein, suggesting further that it is this protein that confers a pro-

ErbB/HER Signaling Network of Growth Factor Receptors

29

liferative advantage to cells (Tzahar and Yarden, 1998). The mode by which ErbB-2 exerts this effect, driving cells toward a more transformed phenotype, is still an open question and will reside in the center of our discussion. Commencing with the clinical aspects of ErbB proteins and their ligands, we proceed to discuss the mechanisms underlying their possible role in transformation. Speciﬁcally, the clinical signiﬁcance of ErbB-2 is discussed in the context of a model that attributes receptor interactions to bivalent ligand binding, enabling receptor cooperativity. Moreover, the high transforming potential of ErbB-2 in epithelial tumors is considered in view of its ability to act as a shared low afﬁnity receptor for the many stroma-derived EGF-like growth factors. Last, we debate the therapeutic opportunities developing through our better understanding of the role of ErbB receptors and their ligands in human cancer.

II. CLINICAL ASPECTS OF ErbB RECEPTORS Under normal conditions the ErbB signaling module acts as a network that mediates the interactions between different cell types, such as the crosstalk between neurons and muscle ﬁbers at the neuromuscular synapse, Schwann cell–neuron interactions in the peripheral nervous system, and endocardium to myocardium crosstalk in the developing heart (Burden and Yarden, 1997). These interactions are unleashed in tissues undergoing malignant transformation. The best example is the mesenchyme–epithelial interaction, which allows ErbB-expressing epithelial sheets to receive morphogenic cues from the stroma in the form of EGF- and neuregulin-like growth factor ligands (Threadgill et al., 1995; Yang et al., 1995). Several animal models of cancer display activated forms of the ErbB signaling network, such as a truncated ErbB-1, which is encoded by the viral erbB oncogene of the avian erythroblastosis virus (Downward et al., 1984), a carcinogen-induced mutant of ErbB-2 that can promote Schwannomas in rodents (Bargmann et al., 1986), and transcriptionally active autocrine loops involving TGF-␣ in several retrovirally transformed cells (Salomon et al., 1995a). Apparently, such autocrine loops are operational in some types of human tumors, as descried later, and aberrant forms of ErbB-1 are found in human neuroblastomas (see later discussion), but most relevant to human cancer is overexpression, often as a result of gene ampliﬁcation, of ErbB-2. In normal epithelial cells, ErbB-2 is expressed at low levels, especially in embryonic stages (Kokai et al., 1987), but a variable fraction of several types of epithelial cancers exhibits remarkably increased expression, up to 100-fold. The clinical implications of ErbB-2 overexpression are well described in breast and ovarian cancers, two tumor types that are the major focus of our discussion, but oth-

30

Leah N. Klapper et al.

er types of tumors also exhibit high ErbB-2 expression, including tumors of the lung, salivary gland, kidney, and bladder. Here we discuss the most recent data related to ErbB-1, ErbB-3 and ErbB-4, and then concentrate on ErbB-2. The reader is referred to earlier reviews that cover speciﬁc aspects of the ﬁeld (Salomon et al., 1995a; Hynes and Stern, 1994; Gullick, 1990; Stancovski et al., 1994). ErbB-1 ErbB-1 expression is found more frequently in squamous cell carcinomas of the lung as opposed to other lung cancer histologies (Veale et al., 1987), and it correlates signiﬁcantly with high metastatic rate, poor differentiation, short patient survival time (Pavelic et al., 1993), and poor prognosis (Volm et al., 1992). Neoplasms such as renal, ovarian, and breast carcinoma display elevated expression of ErbB-1, but correlation with clinical outcome is inconclusive. For example, approximately 37% of invasive ductal carcinomas of the breast express ErbB-1, but correlation with prognosis is unclear: In a study of 309 breast cancer patients, no relation of ErbB-1 to disease-free or overall survival was noted (Charpin et al., 1993). However, in node negative disease, ErbB-1 expression was a signiﬁcant negative prognostic factor for relapse and survival (Harris et al., 1992; Nicholson et al., 1991). In addition, ErbB-1 expression signiﬁcantly predicts for relapse in operable breast cancer patients (Gasparini et al., 1994a), and is associated with shorter disease-free and overall survival in advanced breast cancer (Archer et al., 1995). In a study assessing ﬁne-needle aspirations on women at high risk for breast cancer, ErbB-1 was signiﬁcantly more highly expressed in high-risk than in low-risk women, suggesting that it may be a useful marker in this subset of women (Fabian et al., 1993, 1996). In bladder cancer, ErbB-1 expression correlates with stage and poor prognosis in patients with a transitional cell type (Berger et al., 1987; Chow et al., 1997; Korkolopoulou et al., 1997; Lonn et al., 1993; Nguyen et al., 1994). It is also overexpressed in head and neck tumors (Irish and Bernstein, 1993), which are also squamous cell cancers, and tend to be prevalent within the same patient population as lung cancer. ErbB-1 mRNA overexpression, but not gene ampliﬁcation, was noted in about 40% of benign prostatic hypertrophy cases (Schwartz et al., 1998), while in prostatic carcinoma, overexpression of ErbB-1 was associated with poorer prognosis (Visakorpi et al., 1992). In renal cell carcinoma, coexpression of ErbB-1 and ErbB-2 was signiﬁcantly correlated with metastatic disease (Stumm et al., 1996). Taken together, only non-small cell lung cancer (NSCLC) is presently considered a disease in which ErbB-1 can serve as a prognostic marker, but other types of carcinomas are likely candidates. The situation is different in brain malignancies because several types of clinically relevant alterations of the erbB-1 gene were reported. Both EGF and its receptor are overexpressed in glial tumors (Libermann et al., 1985). In fact, overexpression, often due to ampliﬁcation of the erbB-1 gene, is the most common genetic alteration in glial tumors, and it

ErbB/HER Signaling Network of Growth Factor Receptors

31

correlates with poor prognosis (Wong et al., 1992). Overexpression is more frequent in higher grade neoplasms and correlates with higher proliferation and reduced survival (Jaros et al., 1992). Further, in a signiﬁcant fraction of tumors showing overexpression the erbB-1 gene displays rearrangements, resulting in in-frame deletions of portions of the extracellular domain of ErbB1. The most common mutation (type III) deletes amino acids 6–273, yielding a protein whose ligand binding is defective, but it is constitutively phosphorylated and its tumorigenicity in vivo is enhanced (Nishikawa et al., 1994). ErbB-3 The third member of the ErbB family, ErbB-3, differs from the others in that it contains a nonfunctional tyrosine kinase domain, and thus it signals only as a heterodimer partner. ErbB-3 is normally expressed in many tissues other than the hematopoetic system, in a pattern differing from that of ErbB-1 or ErbB-2 (Prigent et al., 1992). This pattern in normal tissue is paralleled in neoplastic processes as well. In breast cancer, several studies have shown that more than 50% of tumors will have some ErbB-3 positivity by immunohistochemistry, and about 20–30% will show strong membrane staining (Gasparini et al., 1994b; Lemoine et al., 1992a; Quinn et al., 1994; Travis et al., 1996). As opposed to the case of ErbB-2, no signiﬁcant correlation between ErbB-3 overexpression in breast cancer and patient survival, tumor size, and recurrence was found. In gastrointestinal cancers, extensive overexpression of ErbB-3 seems to occur. For example, in gastric cancer, ErbB-3 was found to be more highly expressed than ErbB-2 and it was detectable in most tumors (Sanidas et al., 1993). Similarly, a large number of pancreatic tumors express ErbB-3 (Lemoine et al., 1992b) and 55% of colon tumors, as opposed to 22% of normal colon tissue, express the erbB-3 mRNA (Ciardiello et al., 1991). In oral squamous cell cancers, ErbB-3 overexpression was linked to lymph node involvement, invasion, and patient survival (Shintani et al., 1995). ErbB-3 is overexpressed in prostate cancer, with high levels of membranous staining, which persists in metastatic disease (Myers et al., 1994). Another difference between ErbB-2 and ErbB-3 as prognostic parameters is seen in ovarian cancer, where 85% of tumors stain positively, but with stronger staining seen in borderline and early stage tumors (Simpson et al., 1995). In melanoma as well, ErbB-3 overexpression was found more frequently in nevi than in malignant states and was not seen in cases of metastatic disease (Korabiowska et al., 1996). Taken together, these initial clinical correlates of ErbB-3 attribute to it a role in carcinoma development, which may differ from that of ErbB-2. Given the necessity of ErbB-3 heterodimerization, the true role of ErbB-3 overexpression as a prognostic factor might be better clariﬁed if coexpression with ErbB-2 were analyzed. Indeed, in one study of papillary thyroid cancer 64% of tumors showed coexpression of ErbB-2, -3, and -4 (Haugen et al., 1996). ErbB-4 Only a few clinical studies of ErbB-4 are available. Consistent

32

Leah N. Klapper et al.

with the abundant expression of this neuregulin receptor in the nervous system, a large fraction of pediatric neuroblastomas expresses ErbB-4 (Gilbertson et al., 1997). Even more important, coexpression with ErbB-2 was found in 54% of tumors, but neuregulin expression was limited to 31% of cancers, and no correlation with clinicopathological disease features was observed. By contrast, in a survey of 24 prostatectomy specimens no tumor expressed neuregulin, but 100% of the stroma contained the ligand, and 23% of prostate cancer specimens expressed ErbB-4 (Lyne et al., 1997). More studies on ErbB-4 involvement in cancer, and in particular its cooperation with ErbB-2, are indicated.

A. Breast Cancer Unlike ductal carcinoma in situ (DCIS) and its probable outcome, inﬁltrating ductal cancers (IDC), which show relatively high amounts of the protein, ErbB-2 overexpression has not been found in benign breast disease (Allred et al., 1992a; Gusterson et al., 1988; Regidor et al., 1995; Schimmelpenning et al., 1992). However, metastatic lesions arising from ErbB-2overexpressing tumors maintain overexpression (Iglehart et al., 1995; Niehans et al., 1993), suggesting that ErbB-2 is not involved in premalignant stages, but its function may be essential for progression and metastasis.

1. DCIS Ductal carcinoma in situ is deﬁned as a ductal proliferation of malignant cells that have not invaded the basement membrane. Much attention has been given recently to attempting to improve classiﬁcation and stratiﬁcation of DCIS, because a great deal of uncertainty remains and a wide range of approaches is available for treatment and prognosis of this condition. The incidence has risen dramatically, from 5,000 cases in 1983 to more than 23,000 in 1992, according to the NCI SEER study (Ernster et al., 1996), and it represents about 12% of all breast cancer (Kerlikowske et al., 1997). Despite the recognition that at least 30% of these cases may progress to advanced disease, there is continued debate about which cases require aggressive intervention. Early on, it was recognized that the comedo form of DCIS often shows a greater degree of membranous staining for ErbB-2 than does the non-comedo form (Allred et al., 1992a; Lodato et al., 1990). In fact, it is estimated that up to 90% of comedo DCIS overexpress ErbB-2 (Barnes et al., 1992; van de Vijver et al., 1988). To better stratify risk among DCIS cases, different grading systems were developed. Moreno, breaking down DCIS into high grade versus low grade or intermediate grade, showed a signiﬁcant difference in ErbB-2 expression between the high-grade DCIS and all others,

ErbB/HER Signaling Network of Growth Factor Receptors

33

but not between low and intermediate grade (Moreno et al., 1997). In a study of 127 cases of DCIS, 57% were found to be ErbB-2 positive, with overexpression found more frequently in less differentiated DCIS according to the grading system of Holland (Zafrani et al., 1994). Mack et al. (1997) analyzed ErbB-2 levels in DCIS cases stratiﬁed according to the new histological classiﬁcations as described by Scott (Scott et al., 1997) and found that the differences in ErbB-2 staining were signiﬁcant when related to subclass (Mack et al., 1997). It was found that even in analysis of invasive disease, the level of ErbB-2 expression depended on whether the in situ component was comedo or non-comedo (Brower et al., 1995). Along the same line, 21% of axillary metastases were shown to have a revertant phenotype, to a DCIS identical to the original presentation, and in these cases ErbB-2 levels were identical to those predevelopment of IDC (Barsky et al., 1997). Paget’s disease, which is essentially an aggressive form of DCIS, overexpresses ErbB-2 (Bose et al., 1996; Lodato et al., 1990), while lobular carcinoma in situ does not (Fisher et al., 1996; Lodato et al., 1990; Porter et al., 1991; Ramachandra et al., 1990).

2. MALE BREAST CANCER In two studies, one of 30 male breast cancer specimens and the other of 41, overexpression of ErbB-2 was similar to that detected in females (Bruce et al., 1996; Willsher et al., 1997). This differs from earlier results on 21 patients published by Fox et al. (1991) where no male breast cancer specimens were positive. These variant results highlight the need for standardization of techniques and reagents in order to document deﬁnitively the role of molecular markers in the clinical setting, because the current impression is that although male breast cancer is much less common, there are few molecular differences between male and female breast cancer (for example, male breast cancer is frequently estrogen and progesterone receptor positive), even if the etiology in the male is still enigmatic (Memon and Donohue, 1997; Wagner et al., 1995).

3. CARCINOMAS Studies show that overall about 30% of invasive ductal carcinomas manifest ampliﬁcation of ErbB-2 (Lipponen et al., 1993; Slamon et al., 1987). No difference was seen in ErbB-2 expression between different ethnic groups in the United States (Elledge et al., 1994; Weiss et al., 1995). While DCIS of the comedo type more frequently shows higher expression than does invasive disease, tumors greater than 1 cm in size tend to show ErbB-2 overexpression more frequently than do smaller tumors (Schimmelpenning et al., 1992), and higher grade tumors more frequently overexpress than do lower

34

Leah N. Klapper et al.

grade ones (Tervahauta et al., 1991). Lobular carconoma, similar to lobular carcinoma in situ, does not overexpress ErbB-2 (Gusterson et al., 1992). The signiﬁcance of the lower incidence of ErbB-2 overexpression in IDC compared with DCIS remains unknown. While some evidence supports the notion that comedo DCIS more often than other types progresses to IDC, the probable precursor–product relationships are still unclear. Nevertheless, it appears likely that invasion through the basement membrane selects malignant cells that do not overexpress ErbB-2, but later on those ErbB-2-overexpressing tumors beneﬁt a proliferative advantage. Indeed, some recent observations in animal model systems and in breast cancer patients suggest that once a tumor is established in a distant organ, it becomes more engaged in proliferation and even reverts to its premetastatic phenotype (Barsky et al., 1997).

4. MOLECULAR MARKERS, LYMPH NODE STATUS, AND OVERALL PATIENT SURVIVAL ErbB-2 expression is related to the absence of the two steroid hormone receptors, ER and PR, in breast cancers (Gusterson et al., 1992; Tandon et al., 1989). Multiple studies have demonstrated a positive correlation between p53 expression and ErbB-2 expression in tumors ranging from DCIS to invasive ductal carcinoma (Lipponen et al., 1993; Naidu et al., 1998). In estrogen receptor positive tumors, the combination of p53 and ErbB-2 positivity predicts more undifferentiated carcinoma, the presence of axillary nodes, and shorter disease-free survival (Bebenek et al., 1998; Wiltschke et al., 1994). E-cadherin, the reduction of which is associated with increased invasiveness, was inversely related to ErbB-2 expression (Charpin et al., 1997). The status of another marker of tumor invasion and metastasis, the acidic lysosomal proteinase cathepsin D, is less clear. Whereas one study found no relation between ErbB-2 expression and tumor cell levels of cathepsin D (Scorilas et al., 1993; Tetu et al., 1993), a correlation was noted in node positive patients (Seshadri et al., 1994). Intriguingly, high levels of ErbB-2 in breast tumors correlate with increased cathepsin D in the surrounding stromal cells, which was predictive of shorter metastasis-free survival in chemotherapy-treated patients. Lastly, chromosome 1 aneusomy, as detected by FISH analysis, was signiﬁcantly related to increased ErbB-2 expression (Farabegoli et al., 1996). One of the problems in clinical oncology is the determination of nodal metastases in breast cancer patients; if there were a marker that could predict this accurately, then more conservative surgery could be performed and more aggressive therapy reserved for those patients with more extensive disease. The identiﬁcation of increased levels of ErbB-2 mRNA in ﬁne-needle biopsy specimens was predictive of lymph node involvement preoperatively (Anan et al., 1998). Several studies have shown positive correlation between

ErbB/HER Signaling Network of Growth Factor Receptors

35

ErbB-2 overexpression and nodal metastases (Eissa et al., 1997; Midulla et al., 1995; Noguchi et al., 1993; Tiwari et al., 1992), but this was not conﬁrmed by other studies (Rilke et al., 1991). Although lymph node status per se is the major prognostic factor indicating future relapse, even in lymph node negative breast cancer 30% of patients will relapse. Thus, it is imperative to determine other prognostic indicators that could identify which patients may have a more aggressive disease that requires more intensive treatment. The amount of ErbB-2, especially when analyzed in relatively large groups of patients (⬎300 specimens), appears to predict clinical outcome in both node negative and node positive patients, and thus may serve as a useful indicator. For example, in node negative breast cancer, overexpression of ErbB-2 or p53 was predictive of a decreased overall survival as well as disease-free survival (Albanell et al., 1996; Han et al., 1997; O’Malley et al., 1996; Paterson et al., 1991; Sauer et al., 1992). In some studies, the decreased overall survival was seen particularly in node positive patients (Rilke et al., 1991; Slamon et al., 1987; Tandon et al., 1989; Toikkanen et al., 1992; Tsuda et al., 1998). Bertheau et al. (1998) showed that the status of ErbB-2 was prognostic by the Cox model analysis in the younger patient cohort, under age 35, whereas p53 was prognostic in the age 36–50 group. Others have shown that prognostic information was more reliable in the postmenopausal patient (Tervahauta et al., 1991). In early stage breast cancer (stages I and early stage II) use of p53 and ErbB-2 in an artiﬁcial neural network analysis gave better predictive information than the TNM staging system, and was useful in predicting response to adjuvant chemo- and radiotherapy (Burke et al., 1998). Some studies show that ErbB-2 overexpression is an independent prognostic factor for high-grade disease (Lipponen et al., 1993), patient survival (in either univariate or multivariate analysis) (Eissa et al., 1997; Lonn et al., 1994, 1995; Noguchi et al., 1993), regardless of nodal status (Gullick et al., 1991). Other studies did not conﬁrm these correlations (Clark and McGuire, 1991; Kury et al., 1990). These and other conﬂicting results may relate to the size of the population analyzed and to some technical considerations. Essentially, determination of positivity may vary, especially in earlier studies, with the type of analysis or the identity of antibody used for ErbB2 immunostaining. In addition, because ErbB-2 function may not be autonomous (see later discussion), the extent of coexpression of neuregulins and EGF-like stromal growth factors, as well as the expression of other ErbB proteins (especially ErbB-3), the status of ErbB-2 phosphorylation, and its membrane localization are all factors that may be critical for determining the signiﬁcance of an overexpressed ErbB-2.

5. CIRCULATING ErbB-2 AND ANTIBODIES Despite ErbB-2 being a normal self-antigen and not yet shown to be mutated in human disease, signiﬁcant antibody responses have been seen in the

36

Leah N. Klapper et al.

presence of breast cancer. Disis showed titers greater than 1:100 in 20% of ErbB-2-expressing breast carcinoma patients (Disis et al., 1997), with several of the patients having titers greater than 1:5,000. Tumors expressing ErbB2 were found to release a soluble factor that corresponds to the extracellular domain of ErbB-2 (Langton et al., 1991; Lin and Clinton, 1991). Circulating ErbB-2 itself, at serum levels grater than 120 fmol/ml, has been shown to be a negative prognostic factor for disease-free survival (Fehm et al., 1997), poor prognosis in breast or ovarian cancer (Kandl et al., 1994; Leitzel et al., 1995; Mansour et al., 1997; Meden et al., 1997; Molina et al., 1996; Willsher et al., 1996), larger tumor size and nodal status (Fontana et al., 1994; Krainer et al., 1997), and decreased response to hormonal treatment (Leitzel et al., 1995), but in many of the studies cited, was independent of the detection of ErbB-2 on the tumor specimens themselves by immunohistochemistry.

6. RESPONSE TO TREATMENT The more aggressive phenotype of ErbB-2-overexpressing cells, both in clinical and laboratory settings, implied that the level of ErbB-2 is a predictive factor. Indeed, the reports reviewed next suggest some correlation with response to adjuvant hormonal and chemotherapy.

a. Hormonal Therapy ErbB-2 overexpression has been linked to shorter disease-free survival and overall survival in ER positive patients treated with tamoxifen (Berns et al., 1995; Borg et al., 1994; Carlomagno et al., 1996), but not in all studies (Elledge et al., 1998). This concurs with basic research ﬁndings, in that tamoxifen induces ErbB-2 expression and enhances its signaling (Warri et al., 1991, 1996). Recently, it has been shown that withdrawal of estrogen or treatment with tamoxifen leads to an increase in ErbB-2 expression, potentially due to intron 1 transcription factor binding sites, which mediate the transcriptional response to estrogens (Bates and Hurst, 1997).

b. Chemotherapy As mentioned earlier, approximately 30% of patients with node negative disease are at risk of relapsing, thus there is a very practical need to identify tumors that might be more aggressive, suggesting the need for more intensive treatment protocols. Patients with node negative disease who were ErbB-2 positive were found to have a signiﬁcantly decreased disease-free survival after chemotherapy (Allred et al., 1992b). In a large study from the International (Ludwig) Breast Cancer Study Group Trial V, disease-free survival was greater for ErbB-2 negative patients, both node positive and node negative, who were treated with CMF (cyclophosphamide, methotrexate, 5-

ErbB/HER Signaling Network of Growth Factor Receptors

37

ﬂuorouracil) chemotherapy (Gusterson et al., 1992). A prospective study from the Toronto Breast Cancer Study Group involving 580 patients identiﬁed ErbB-2 positivity as a signiﬁcant negative prognostic factor in node negative patients, with the difference in disease-free survival accentuated among those receiving chemotherapy (Andrulis et al., 1998). The poorer response to standard chemotherapy in patients overexpressing ErbB-2 is documented in node positive patients as well (Tetu and Brisson, 1994; Tsuda et al., 1998). A case control study showed ErbB-2 positivity to predict poor disease-free survival after conservative surgery and radiotherapy without chemotherapy (Haffty et al., 1996). The CALGB study comparing high-dose CAF [5-ﬂuorouracil, doxorubicin (adriamycin), cyclophosphamide] chemotherapy with standard dose chemotherapy in women with node positive disease found that patients with ErbB-2 overexpression actually responded better than other patients after high-dose CAF treatment (Muss et al., 1994). It is possible that the general improvement seen with the introduction of anthracyclines such as doxorubicin into standard and high-dose regimens such as CAF or AC may be explained by the ﬁnding that ErbB-2 is occasionally co-overexpressed with toposiomerase IIa, and experiments in vitro show that the ErbB-2-topo II-overexpressing cells are relatively sensitive to inhibitors of the enzyme (Smith et al., 1993). ErbB-2 as a prognostic factor has been studied in neoadjuvant therapy as well. In a study of neoadjuvant chemotherapy ErbB-2 overexpression was identiﬁed as the major prognostic factor correlated with disease-free survival and overall survival (MacGrogan et al., 1996), although in a recent study of neoadjuvant CAF chemotherapy or radiotherapy, neither ErbB-2 nor p53 detection was of prognostic value (Rozan et al., 1998). Last, in a study of patients with metastatic disease, ErbB-2 overexpression did not predict survival or response to chemotherapy (Niskanen et al., 1997). These results suggest that it may be of value to document ErbB-2 status in the clinical trial setting, as an additional parameter that may inﬂuence choice of treatment protocol, particularly in early stage disease, where ErbB-2 overexpression may suggest including chemotherapy in the care of the node negative patient.

B. Gynecological Cancers 1. OVARIAN CANCER Breast and ovarian cancer share responsiveness to steroid hormones, and incidence of the vast majority of tumors in the epithelial rather than the stromal component. In addition, these diseases appear to share etiologic factors: Women with one kind of tumor have an increased risk of developing the other type of tumor. Whereas normal ovarian tissue does not overexpress ErbB-

38

Leah N. Klapper et al.

2 (Huettner et al., 1992; Wong et al., 1995), 20–35% of borderline tumors (Eltabbakh et al., 1997; Harlozinska et al., 1997) and 30–50% of ovarian cancers express ErbB-2. About 60% of ovarian tumors may overexpress erbB-2 by mRNA analysis (Huettner et al., 1992). In some, but not all studies (Medl et al., 1995; Rubin et al., 1993, 1994; Tanner et al., 1996), ErbB2 has been associated with advanced FIGO stage, worse prognosis, or decreased response to therapy (Berchuck et al., 1990a; Harlozinska et al., 1997; Meden et al., 1994; Natali et al., 1990; Slamon et al., 1989). ErbB-2 expression by immunocytochemistry is higher in sporadic ovarian cancer than in the familial variety (Auranen et al., 1997). Expression may also be higher in extra ovarian mullerian adenocarcinomas (Kowalski et al., 1997). Unlike breast cancer, some ovarian tumor cell lines display rearranged erbB2 gene or variant transcripts (Hung et al., 1992; King et al., 1992), and more often than in breast cancer, overexpression may not be linked to gene ampliﬁcation. According to one study, coexpression of ErbB-2 and p21/WAF1 correlates with shorter overall survival and disease-free survival (Katsaros et al., 1995). High serum titers of an ErbB-2 fragment have been associated with shorter survival in ovarian cancer (Meden et al., 1997).

2. VULVAR AND ENDOMETRIAL CANCER Patients with stage I or II vulvar carcinoma are more likely to have nodal metastases if they overexpress ErbB-2 (Gordinier et al., 1997). erbB-2 Ampliﬁcation is also associated with a more aggressive course in gestational trophoblastic disease (Bauer et al., 1997). On the other hand, although overexpression was found in up to 48% of endometrial tumors, no overall prognostic signiﬁcance was noted by some researchers (Backe et al., 1997; Bell et al., 1997; Gassel et al., 1998). Nevertheless, others correlated more intense ErbB-2 staining with metastatic disease (Berchuck et al., 1991), decreased survival (Hamel et al., 1996; Kohlberger et al., 1996), shorter disease-free survival by univariate analysis (Lukes et al., 1994), and relation to deep myometrial invasion (Seki et al., 1998). In stage I tumors treated by hysterectomy, overexpression of ErbB-2 has been correlated with decreased survival (Nazeer et al., 1995). As in the case of ovarian and breast cancer, these results may be technique dependent. Gene ampliﬁcation determined by using ﬂuorescence in situ hybridization (FISH) did not correspond to protein expression in a series of endometrial tumors, but nevertheless ampliﬁcation predicted poorer survival (Riben et al., 1997).

3. CERVICAL CANCER ErbB-2 ampliﬁcation was detected in 12–22% of stage II/III squamous cell cancers of the cervix (Kristensen et al., 1996; Mitra et al., 1994; Ndubisi

ErbB/HER Signaling Network of Growth Factor Receptors

39

et al., 1997; Wong et al., 1996). Increased ErbB-2 staining is seen in squamous metaplasia, raised condyloma, and carcinoma in situ (Berchuck et al., 1990b). Staining for ErbB-1 was high, while staining for ErbB-2 decreased in cervical squamous neoplasms (Berchuck et al., 1990b). Ampliﬁcation of ErbB-2, H-ras, and c-myc was seen in high-grade cervical intraepithelial neoplasia (CIN3) but not in earlier stages (Pinion et al., 1991). Cervical cancer patients with ErbB-2 overexpression showed signiﬁcantly decreased 5-year survival in cohorts treated with radiation therapy (Nakano et al., 1997, 1998), but decreased survival was not seen in all subgroups (Ndubisi et al., 1997). In cervical glandular carcinoma in situ and adenocarcinoma, ErbB-2 overexpression is associated more frequently with human papilloma virus type 16 versus type 18 (Roland et al., 1997).

C. Prostate Cancer About 30% of prostate cancer stains positive for ErbB-2 (Kuhn et al., 1993; Sadasivan et al., 1993), whereas benign prostatic hypertrophy (BPH) is generally reported as being negative (Kuhn et al., 1993; Sadasivan et al., 1993). Some studies that showed strong positive staining for ErbB-2 in malignant tissue have demonstrated positive immunostaining for ErbB-2 in BPH as well (Giri et al., 1993; Gu et al., 1996). In fact, discrimination between membrane or cytoplasmic staining may yield an even stronger difference between benign and malignant tissue. A retrospective study showed ErbB-2 staining by immunocytochemistry to correlate with disease progression in node negative patients (Veltri et al., 1994). As with endometrial tumors, the recent use of FISH for analysis of prostate carcinoma has proven itself to be more sensitive, showing about 40% positivity versus 29% by IHC, and positivity was signiﬁcantly associated with higher Gleason grade, DNA ploidy, and disease-free survival (Ross et al., 1997a,b).

D. Gastrointestinal Cancer 1. COLON CANCER While normal colonic mucosa is mostly negative for ErbB-2, the levels of this antigen increase with Dukes stage in colon carcinoma and show significant correlation with relapse-free and postoperative survival period (Kapitanovic et al., 1997). Consistent with a correlation with poor prognosis, colon cancers that metastasized to the liver have higher levels of ErbB-2 than those that do not (Yang et al., 1997). ErbB-2 overexpression was found in about 50% of early colon carcinomas, particularly those with-

40

Leah N. Klapper et al.

out an adenomatous component (Caruso and Valentini, 1996; Shirai et al., 1995), and colon cancer patients with lymph node metastases were reported to have higher ErbB-2 levels (Saeki et al., 1995). Interestingly, significant trends for coexpression of ErbB-1 and either TGF-␣ or amphiregulin were detected in this latter study, suggesting that the two ligands of ErbB-1 may play an important role in the development of colorectal carcinomas through an autocrine mechanism. Lastly, circulating serum levels of ErbB2 correlated with poor prognosis in colorectal carcinoma (Vogel et al., 1996).

2. ESOPHAGEAL CANCER One study showed 43% ErbB-2 positivity in esophageal tumors, and association with a better prognosis to neoadjuvant therapy (Duhaylongsod et al., 1995). In Barrett’s adenocarcinoma expression correlates with decreased patient survival (Flejou et al., 1994).

3. GASTRIC CANCER Overexpression of ErbB-2 in gastric cancer is common (Kim et al., 1993; Ooi et al., 1998), particularly in advanced stage and intestinal subtype (Wu et al., 1998) and may be associated with worse prognosis (Amadori et al., 1997; Yonemura et al., 1998).

4. PANCREATIC CANCER ErbB-2 is frequently overexpressed in pancreatic adenocarcinomas, specifically in well- or moderately differentiated glandular areas of the tumor, and is decreased in the poorly differentiated areas (Dugan et al., 1997; Yamanaka et al., 1993b). Overexpression was linked with worse prognosis in ampulla of Vater tumors (Vaidya et al., 1996a,b).

E. Lung Cancer A signiﬁcant additive effect between p53 and ErbB-2 in predicting poor prognosis was found in stage I NSCLC (Harpole et al., 1995), but not in a larger study (Pastorino et al., 1997). ErbB-2 expression has an inverse correlation with angiogenesis in NSCLC (Giatromanolaki et al., 1996). The combination of K-ras mutation by PCR with ErbB-2 expression by immunocytochemistry was a poor prognostic factor in lung NSCLC (Nemunaitis et al., 1998).

ErbB/HER Signaling Network of Growth Factor Receptors

41

F. Head and Neck Cancer Overexpression of ErbB-2 is associated with a decreased disease-free survival in intestinal-type adenocarcinoma of the paranasal sinuses (Gallo et al., 1998), palatal salivary gland neoplasms (Giannoni et al., 1995), and mucoepidermoid carcinoma of the salivary gland (Press et al., 1994).

G. Kidney Cancer An inverse relationship between ErbB-1 and ErbB-2 levels in renal cell carcinoma has been shown (Weidner et al., 1990). ErbB-2 overexpression was noted with increased frequency in certain renal cystic disorders as well as in neoplasms (Herrera, 1991). Combined high expression of ErbB-1 and ErbB2 was correlated with metastatic disease (Stumm et al., 1996).

H. Bladder Cancer A relationship between transitional cell cancer of the bladder and ErbB2 expression was ﬁrst noted by Zhau et al. (1990) who found that about 70% of tumors, and none of the normal tissue specimens, overexpress ErbB2 by immunohistochemistry and Western blotting. Overexpression of erbB2 has been correlated with grade and survival (Korkolopoulou et al., 1997; Lipponen et al., 1991; X. H. Zhang et al., 1997), and is infrequent in superﬁcial tumors (Tetu et al., 1996). The presence of disease is frequently diagnosed through bladder washings, and the presence of ErbB-2 in the cytological specimen is found only in tumors of higher grade (Lonn et al., 1993). Interestingly, in a large recent prospective study, presence of ErbB-1 or ErbB-2 in grade 3 tumors predicted for less invasive disease (Vollmer et al., 1998).

I. Brain Tumors In astrocytomas, ErbB-2 tends to correlate with higher grade histology (Bernstein et al., 1993; Schwechheimer et al., 1994). On the other hand, in meningiomas, expression of ErbB-2 is higher in the typical forms, whereas expression is lost in progression to atypical forms; the inverse is true for p53 (Chozick et al., 1996). PCR analysis of CNS ﬂuid can detect erbB-2 sequences from breast cancer metastatic to brain (Rhodes et al., 1994). In neuroblastoma, ErbB-2, but not ErbB-1, predicted signiﬁcantly shorter patient

42

Leah N. Klapper et al.

survival, and the combination of ErbB-2 with p53 positivity was especially strong as a prognostic indicator (Layﬁeld et al., 1995).

III. HOW DOES ErbB-2 INDUCE CANCER? A. In Vitro Transforming Potential of ErbB-2 How ErbB-2 causes cancer is a question that has been repeatedly asked— and the answers reevaluated—during the past 15 years. Its role in cellular transformation, insinuated by its abundance in a wide range of human tumors, is supported by a few but solid lines of evidence that therapies directed against this receptor can indeed impede tumor growth. Although no analogous point mutation, to that of the transforming rat neu, was found in the human gene, site-directed mutagenesis conﬁrmed that a similar change can activate human erbB-2 as an oncogene (Akiyama et al., 1991). When overexpressed in mouse ﬁbroblasts, the human gene conferred a transformed phenotype in vitro and tumorigenesis in vivo (Di Fiore et al., 1987; Hudziak et al., 1987), indicating that a threshold level of overexpression is crucial for its oncogenic potential. In accordance, manipulating ectopic expression of ErbB-2 by tetracycline-induced abrogation reversed the transformed phenotype of the cells and their ability to establish tumors (Baasner et al., 1996). Moreover, tissue-speciﬁc induction of the human, unmutated transgene promoted the appearance of mouse mammary adenocarcinomas with a tendency to metastasize (Suda et al., 1990). Tumor progression by ErbB-2 overexpression was attributed to aberrant activation of the tyrosine kinase (Lonardo et al., 1990; Pierce et al., 1991) and indeed it has been demonstrated that this activity is essential for transformation (Weiner et al., 1989a). Consequently the removal of ErbB-2 from the cell surface of breast cancer cells (Beerli et al., 1994) reduces mitogenicity and correlates this effect with a downstream decrease in growth factor-induced signaling (Graus-Porta et al., 1995; Karunagaran et al., 1996). Cellular transformation by ErbB-2 is, therefore, most probably linked to its elevated expression at the cell surface where it can initiate a hypermitogenic intracellular signal.

B. ErbB Activating Ligands Residing in cells of epithelial origin, ErbB receptors can be nourished with binding proteins either via an autocrine secretory loop or with molecules produced by adjacent tissues. Stromal cells in the vicinity of propagating tumors serve as the prime source for ErbB-activating ligands, whereas the ex-

ErbB/HER Signaling Network of Growth Factor Receptors

43

tracellular matrix surrounding them is a reservoir that can increase local ligand concentrations (see Fig. 2). EGF was the ﬁrst identiﬁed and is the prototype for ligands capable of binding ErbB proteins. These have been shown to harbor the capacity to activate the receptors within a common structural motif of 45–55 amino acids, called the EGF domain. Three covalently held loops, formed by six cysteine residues that are typically spaced within this motif, are crucial for receptor recognition and, together with a critical arginine residue and several glycines, are shared by all ligands (Groenen et al., 1994). All mammalian ErbB ligands are derived from transmembrane precursors and give rise to the mature soluble protein by speciﬁc cleavage. Similar to EGF, the transforming growth factor ␣ (TGF-␣) (Marquardt et al., 1984), amphiregulin (Shoyab et al., 1988), betacellulin (Sasada et al., 1993), epiregulin (Toyoda et al., 1995), and the heparin-binding EGF (HB-EGF) (Higashiyama et al., 1991) can all induce activation and phosphorylation of ErbB-1, but none are able to stimulate homodimeric complexes of ErbB-2. Correlation between these ligands and

Fig. 2 Epithelial–mesenchymal interactions in the promotion of cancer. Cells of mesenchymal origin, embedded within the stroma, synthesize and secrete EGF-like growth factors. Diffusing through the basal lamina, surrounding glandular structures, the polypeptide ligands encounter binding receptors exposed on the surface of epithelial cells. The receptor repertoire of the cell can determine the biological response achieved by such interactions. Accordingly, high expression of ErbB-2 can contribute to the ampliﬁcation of growth factor signaling, driving the cells toward a malignant phenotype. Moreover, extracellular matrix can serve as a ligand reservoir, increasing its local concentrations available for receptor activation.

44

Leah N. Klapper et al.

cancer has been most convincingly demonstrated for TGF-␣. The expression of this ligand, accompanied by its binding receptor ErbB-1, correlates with poor prognosis for several gastrointestinal (Ihara et al., 1993), lung (Tateshi et al., 1990), and ovarian (Kohler et al., 1989) malignancies. In a similar manner, the coexpression of ErbB-1 with either EGF or TGF-␣ was reported in 38% of pancreatic tumors and correlated with both tumor size and a reduction in patient survival (Yamanaka et al., 1993a). Enhanced synthesis of EGF has been reported in several tumors, including lung and ovarian tumors and the levels of both growth factors were found to be elevated in the urine of glioma patients. Another ligand, HB-EGF, may act as a stromal mediator of prostate cancer, being synthesized by interstitial and vascular smooth muscle cells and capable of efﬁciently stimulating cell growth (Freeman et al., 1998). Neuregulins (NRGs) comprise an additional group of ErbB activating proteins, demonstrating a speciﬁcity of binding toward ErbB-3 and ErbB-4 (Burden and Yarden, 1997). NRG1, which was initially identiﬁed as an ErbB-2 phosphorylating activity (Falls et al., 1993; Holmes et al., 1992; Mabrouk et al., 1996; Marchionni et al., 1993; Peles et al., 1992; Wen et al., 1992), inhibits terminal differentiation and induces adenocarcinomas when targeted to the mammary gland of transgenic mice (Krane and Leder, 1996), suggesting involvement in tumor-propagating processes mediated by ErbB receptors. The exact mode of action by which NRG exerts its effect on epithelia is not simple, as exempliﬁed by its counteracting effects on cells of the mammary gland. Essentially, synthesis of the ligand by the connective tissue, primarily the fat pad, surrounding ductal and alveolar structures, is upregulated at pregnancy and is able to promote lobular-alveolar budding and milk production (Yang et al., 1995). Several breast cancer cell lines differentiate, in accordance with this phenomenon, to produce milk proteins and lipids (Bacus et al., 1992a, 1993). However, others undergo enhanced proliferation in response to NRG stimulation (Lewis et al., 1996). A variety of biological observations led to the identiﬁcation of splice variants of the NRG1 gene revealing their pleiotropic mode of action. Among these are the glial growth factor (GGF), which induces Schwann cell proliferation (Marchionni et al., 1993), and acetylcholine receptor inducing activity (ARIA) functioning at the neuromuscular junction (Falls et al., 1993). A related group of molecules, termed NRG2, binds both ErbB-3 and ErbB-4 (Carraway et al., 1997; H. Chang et al., 1997), and a third molecule, NRG3, exclusively binds to ErbB-4 (D. Zhang et al., 1997).

C. Does ErbB-2 Have a Ligand of Its Own? Intuitively, a transforming activity transmitted by overexpressed levels of ErbB-2 could be evoked by a direct ligand. Such a binding protein would in-

ErbB/HER Signaling Network of Growth Factor Receptors

45

duce receptor homodimerization, cumulatively resulting in a high mitogenic signal. Several candidate ligands suggested to fulﬁll this function have been isolated. Among these are NAF (neu activating factor) isolated from a transformed human T-cell line (Samanta et al., 1994), p75 puriﬁed from human mammary carcinoma cells (Lupu et al., 1992), and a surface proteoglycan from a rat adenocarcinoma (Wu et al., 1994). However, although capable of stimulating ErbB-2 phosphorylation, the speciﬁcity and function of these putative ligand molecules are not well established, maintaining ErbB-2’s orphanhood. Reinforcing the nonexistence of a direct ErbB-2 ligand is the identiﬁcation of three EGF-like ligands encodd by poxviruses: the vaccinia virus growth factor (VGF) (Brown et al., 1985; Stroobant et al., 1985), the Shope sarcoma virus growth factor (SGF) (Chang et al., 1987), and the Myxoma virus growth factor (MGF) (Upton et al., 1987). Apparently, these ligands harness the proliferation-inducing activity of the receptors for the enhancement of their virulence (McFadden et al., 1996). Synthetic analogs of these three viral ligands revealed a speciﬁc pattern of binding to ErbB proteins: Whereas VGF is similar to EGF in its receptor speciﬁcity, and SFGF is a pan-ErbB ligand, MGF emerges as a factor that binds exclusively to heterodimeric ErbB2/ErbB-3 complexes, but not to either of the corresponding homodimers (Tzahar et al., 1998). Likewise, neither VGF nor SFGF can recognize ErbB2 alone, but both can recruit it into heterodimers, suggesting that no highafﬁnity ligand of ErbB-2, of either viral or mammalian origin, may exist (Tzahar et al., 1998; Klapper et al., 1999).

D. Ligand-Independent Receptor Dimerization An alternative to ligand-induced activation of the ErbB-2 protein is a conformational change driven by dimer formation in a ligand-independent manner. Conformational energy analysis of the transmembrane domain of the receptor predicted that the mutated transforming Neu, presenting a helical form of the sequence, should have a higher tendency toward dimer formation than that of the wild-type protein (Brandt Rauf et al., 1990). Indeed, covalent cross-linking could detect dimers of the transforming protein, whereas the normal receptor appeared only as a monomer (Stern et al., 1988; Weiner et al., 1989b). Apparently the glutamic acid residue, substituting the normal valine at position 664 in the transmembrane portion of the wild-type Neu, is strongly hydrogen bonded, suggesting that direct interactions of this residue underlie the propensity to form dimers (Smith et al., 1996). Dimer formation between molecules of transforming Neu is indeed essential as demonstrated by the ability of a cytoplasmic portion deleted oncoprotein to inhibit transformation and tumorigenicity of Neu-expressing ﬁbroblasts and

46

Leah N. Klapper et al.

rat-derived neuroglioblastomas (Qian et al., 1996). This dominant-negative effect of the dimerizing partner is attributed to the lack of kinase activity and can be reproduced by mutating the ATP-binding site of the receptor, indicating that the proliferative effect is dependent on an intact signaling cascade triggered by receptor transphosphorylation (Schlegel et al., 1997). The quality and not only the level of receptor phosphorylation seems to be important because receptor dimerization induced by a foreign transmembranal sequence elevated phosphorylation on tyrosine residues, but lacked the ability to cause transformation in vitro (Burke et al., 1997). Interestingly, mammary tumors induced by a transgenic wild-type Neu arise due to somatic activating mutations within the extracellular part of the receptors. These mutations, occurring at conserved cysteine residues, promote constitutive dimerization via disulﬁde bonds, resulting in a transforming activity (Siegel and Muller, 1996). The question of transforming mechanisms remains, however, unanswered, because none of the suggested dimerization-driving mutations has been identiﬁed in human cancer. Moreover, homodimers of the wild-type ErbB-2, induced by monoclonal antibodies (Pinkas-Kramarski et al., 1996b) or a point mutation (Campion et al., 1993), imply that the activity of the homodimeric ErbB-2 complexes is relatively weak, suggesting that constitutive homodimerization arising due to high receptor expression would not necessarily lead to transformation. In support of this scenario, transfection of oncogenic Neu conferred ligand-independent transformation only in the presence of ErbB-1, ErbB-3, or ErbB-4 (Cohen et al., 1996b). Possibly, heterodimer formation, more than the formation of ErbB-2 homodimers, is linked to a transforming signal.

IV. EVOLUTIONARY AND DEVELOPMENTAL ASPECTS OF THE MULTIPLICITY OF ErbB PROTEINS Signaling via ErbB receptor tyrosine kinases matured to the module known in mammals following a paradigm of increased complexity, enabling interactions in a tissue-speciﬁc manner as well as a high degree of ﬁne tuning and regulation. ErbB signaling was originally handled by a single family member as represented in Caenorhabditis elegans by the let-23 gene encoding a primordial EGF receptor (Aroian et al., 1990). ErbB structural motifs, as well as the amino acid sequence of the cytoplasmic tyrosine kinase domain, were inherited through evolution as was the structure of a single Let-23 binding protein, the Lin-3 ligand, bearing a typical EGF-like domain (Hill and Sternberg, 1992). The relative simplicity of ErbB signaling in C. elegans enabled the characterization of its central role in development as reﬂected in vulval

ErbB/HER Signaling Network of Growth Factor Receptors

47

induction essential for nematode reproduction. Gonadal secretion of the Lin3 protein exposes the adjacent vulva precursor Let-23-expressing cells to a ligand gradient, resulting in a series of divisions and morphogenetic changes leading to vulval generation (Sternberg and Horvitz, 1991). Cell fate determination by Let-23 activation was also demonstrated in neuroectoblasts of C. elegans, however, unlike vulval induction, differentiation is uncoupled to cell division (Sternberg et al., 1995), indicating a versatility of action overriding the expected limitations of a single-receptor-ligand interaction. Proceeding several steps higher in evolution, to the fruit ﬂy Drosophila, reveals a complexity achieved by ligand multiplicity represented by four members: gurken, spitz, vein, and argos (Neuman-Silberberg and Schupbach, 1993; Rutledge et al., 1992; Schnepp et al., 1996; Schweitzer et al., 1995), all apparently capable of binding a single receptor homolog, the Drosophila EGF receptor (DER) protein (Livneh et al., 1985). Isolation of mutant alleles of the Drosophila receptor revealed the central role of the DER signaling pathway in the development of multiple tissues and organs spanning a temporal range of processes (Clifford and Schuppbach, 1994; Shilo and Raz, 1991). As in lower and higher organisms, DER is widely expressed in embryos, assigning its regional activation to spatial restriction of the ligands (Stein and Stevens, 1991). Expanding the complexity even further, evolutionary progress applied repeated duplications of genes encoding the ErbB receptors, as well as their EGF-like ligands, achieving the wide repertoire known in mammals. ErbBdependent developmental patterns supplied valuable evidence that the availability of several ligand-receptor combinations evolved to serve an interplay of receptor interactions rather than to expand parallel nonoverlapping pathways. The ﬁrst family member that was successfully targeted and thereby expression eliminated was the erbB-1 gene. Absence of ErbB-1 expression is lethal, death occurring at a variety of developmental phases depending on animal genetic background, as a result of major defects in the assembly of epithelia (Miettinen et al., 1995; Sibilia and Wagner, 1995; Threadgill et al., 1995). Ligand multiplicity predicted that disruption of a single gene would result in a milder phenotype than that arising from receptor knockout. Indeed, mice homozygous for a disrupted TGF-␣ gene displayed only part of the abnormalities of erbB-1⫺/⫺ mice, the most eminent being aberrant eye development and waviness of whiskers (Luetteke et al., 1993; Mann et al., 1993). A different ligand, HB-EGF, seems to govern yet another role of ErbB1 in development, enabling blastocyte attachment to the uterus (Das et al., 1994). Signaling by ErbB-3 and ErbB-4 inﬂicts an additional level of complexity by applying a wide but shared variety of direct ligands to both receptors. Genetic evidence implies that signaling by neuregulins is essential for epithelial organs, for effectors of the nervous system such as Schwann cells and neu-

48

Leah N. Klapper et al.

romuscular junctions, and for muscle cells of the heart’s trabeculae (Meyer and Birchmeier, 1995). Mouse embryos lacking ErbB-4 die of heart malfunction caused by an undifferentiated ventricle, manifested in the absence of myocardial extensions (Gassmann et al., 1995). This phenotype is shared by NRG mutants speciﬁcally targeted at immunoglobulin-like (Ig-like) domain containing isoforms, indicating that ligand abundance does not confer redundancy in biological effects (Kramer et al., 1996). This important feature of multiplicity is further emphasized by the requirement for Ig-NRGs that have a glycosylation domain (type I), as opposed to NRGs without a glycosylation domain (type II), playing an important role in the early development of Schwann cells, but not in heart differentiation (Meyer et al., 1997) Remarkably, the phenotype of erbB-2⫺/⫺ mice shares characteristics with both NRG and ErbB-4 knockout animals (Lee et al., 1995). This phenotypic similarity to both NRG⫺/⫺ and erbB-4⫺/⫺ mice implies the three genes to be of close relevance in development and essential for the activation of an overlapping pathway. Likewise, targeted mutations to the ErbB-3 receptor cause severe neuropathies (Riethmacher et al., 1997), the earliest observed, namely, absence of neurons deriving from hindbrain, being similar to an erbB-2⫺/⫺ phenotype. Such an overlap in activities suggests that ErbB-2 cooperates with ErbB-3 and ErbB-4 in the mediation of signals induced by NRGs. As discussed later, this cooperation appears essential for ErbB-2-mediated transformation of epithelial and other cell types.

V. THE ErbB SIGNALING NETWORK A. Evidence for Inter-Receptor Interactions Functional interactions between ErbB proteins were ﬁrst suggested by the observation that ErbB-2 is a substrate of the ligand-activated ErbB-1 (King et al., 1988; Stern et al., 1986). This was later shown to correlate with a physical association of the two proteins leading to the formation of heterodimeric complexes and strictly dependent on the binding of a ligand (Goldman et al., 1990; Wada et al., 1990b). Molecular modeling of the kinase domains of both receptors implied that heterodimers are energetically favored over homodimers (Murali et al., 1996). In line with this prediction, all six possible heterodimeric complexes of ErbB receptors can be observed (Tzahar et al., 1996). The most remarkable variation achieved by this phenomenon is due to the impaired catalytic activity of ErbB-3 requiring a partner with an active kinase to promote its signaling (Guy et al., 1994). Assessing the incidence of dimer formation reveals that an ErbB receptor bound by its direct

ErbB/HER Signaling Network of Growth Factor Receptors

49

ligand will preferentially recruit ErbB-2 as a heterodimerizing partner (Graus Porta et al., 1997; Tzahar et al., 1996). Moreover, EGF and NRG receptors compete with each other for the interaction with the ErbB-2 receptor (Karunagaran et al., 1995), implying an advantage of ErbB-2 as a signaling subunit. Complying with its emerging role as a favored surrogate receptor, the expression of ErbB-2 is the most expanded amongst all four family members (Pinkas-Kramarski et al., 1997).

B. Transforming Ability of Heterodimers The bias toward the formation of ErbB-2-containing heterodimers could have evolved via superiority of their signaling. Indeed, signals generated by activated ErbB-1/ErbB-2 heterodimers lead to an enhanced proliferative response to EGF (Wada et al., 1990a), resulting in greater normal and tumorigenic cell growth in comparison to ErbB-1 homodimers (Kokai et al., 1989). Accordingly, transgenic expression of both Neu and TGF-␣ in the mammary epithelium appears synergistic in the promotion of multifocal mammary tumors arising after a signiﬁcantly shorter latency period than either parental strain alone (Muller et al., 1996). Importantly, none of the ErbB receptors alone could cause growth in soft agar or tumorigenicity in animals, transformation occurring only when ErbB-2 was expressed with one or more of its sibling receptors (Cohen et al., 1996b). This heterodimer-dependent transformability of the complexes was further demonstrated by its inhibition with antibodies against either participating receptor (Wada et al., 1990a), as well as by reducing cell proliferation with heterodimer-destabilizing monoclonal antibodies (Klapper et al., 1997). Both ErbB-3 and ErbB-4 show a mitogenic superiority and promote cellular transformation (Alimandi et al., 1995; Wallasch et al., 1995; Zhang et al., 1996) in a ligand-dependent manner, when co-introduced with ErbB2. Most signiﬁcant is the reconstitution of the ErbB2/ErbB-3 heterodimer. Induced by NRGs, this complex accommodates the highest signaling activity among all receptor combinations (Pinkas-Kramarski et al., 1996b; Riese et al., 1995) as well as being the predominant ligand-binding moiety in several human adenocarcinomas (Chen et al., 1996). The functional linkage between the increased mitogenicity of ErbB-2-containing dimers and its high expression in cancer could reside, on the one hand, in its high basal activity (Lonardo et al., 1990) and, on the other, in a propensity to form dimers (Weiner et al., 1989a). The latter, reminiscent of the transforming rodent Neu, appears crucial to understanding the role of ErbB-2 in signal transduction and in cancer: Although not a direct receptor for EGF or NRG,

50

Leah N. Klapper et al.

ErbB-2 can decrease the rate of dissociation of these ligands from their direct receptors (Karunagaran et al., 1996). This results in signiﬁcant prolongation of downstream signaling to the mitogen-activated protein kinase (MAPK) pathways and remarkable signal ampliﬁcation. Thus, by forming heterodimers with other ErbB proteins, ErbB-2 may function as an ampliﬁer of stroma-derived mitogenic signals (Fig. 3). As discussed later, this capability is enhanced by two factors: First, the propensity of ErbB-2 to undergo endocytosis is low compared to that of ErbB-1, and second, ErbB-2-containing heterodimers are endowed with an unexpectedly wide speciﬁcity to various ErbB ligands.

Fig. 3 Signaling superiority of ErbB-2-containing heterodimers. ErbB receptor dimers comprise the ligand-binding and signal-promoting complexes. ErbB-2 (“2”) is a superior heterodimerizing partner in comparison to sibling receptors. Superiority is attributed to a hierarchy of recognition by which ErbB-2 is preferred to other members. Furthermore, complexes comprised of ErbB-2 molecules show high stability in comparison to non-ErbB-2-containing dimers (as expressed by the balance between arrows). Dimer stability is reﬂected in the afﬁnity of EGFlike ligands to their binding moieties, being higher in ErbB-2 containing heterodimers than in homodimers. High afﬁnity is the result of a decrease in ligand dissociation rate and the outcome is a strong and prolonged signal.

ErbB/HER Signaling Network of Growth Factor Receptors

51

C. Ligand-Bivalency Selects Dimer Participants The ability of ErbB-2 to increase ErbB signaling in a ligand-dependent fashion, despite its lack of a direct binding protein of its own, addresses a fundamental question of dimerization-driving forces. The mechanism of receptor dimerization is best understood in the case of the human growth hormone receptor (Wells, 1996), in which the ligand uses two different sites to sequentially bind two receptor molecules. A similar model is suggested for ErbB receptors, originally relying on the duplicated structure of ErbB-1’s extracellular domain (Gullick, 1994), and recently supported by evidence of bivalent recognition of EGF-like ligands (Tzahar et al., 1997). According to this model, encouraged by NMR studies (Jacobsen et al., 1996), NRG1 bears two distinct receptor binding sites, an N-terminal high-afﬁnity site and a low-afﬁnity/broad-speciﬁcity site located at the C terminus (Barbacci et al., 1995; Tzahar et al., 1997). High-afﬁnity binding of the ligand to its primary receptor demands a stringent ﬁt, whereas the low-afﬁnity site allows more ﬂexible paring. Furthermore, derivatives of EGF indicated that the N-terminal tail of this ligand binds to the N-terminal subdomain of its receptor, whereas the C terminus of EGF juxtaposes to subdomain III of ErbB-1 (Summerﬁeld et al., 1996), pinpointing this domain as a possible low-afﬁnity broad-speciﬁcity binding pocket. It is suggested that a functionally analogous portion exists within the ErbB-2 receptor, selecting it as a preferred coreceptor. Supporting this is the cooperative binding of both EGF (Wada et al., 1990b) and NDF (Peles et al., 1993; Sliwkowski et al., 1994; Tzahar et al., 1996) to cells co-overexpressing a primary receptor together with ErbB2, implying its direct low-afﬁnity binding to the ligands. In accordance, biophysical assays reported that ErbB-2 binds EGF-like ligands at an afﬁnity ranging in micromolars as opposed to the nanomolar range afﬁnity of the ligands to their primary respective receptor (Horan et al., 1995; Tzahar et al., 1997). In addition, monoclonal antibodies directed against the putative ligand binding site of ErbB-2 can accelerate dissociation of EGF-like ligands, thereby blocking the formation of ErbB-2-containing dimers (Kalpper et al., 1997). Cumulatively these independent studies reason that ErbB-2, capable of directly recognizing bivalent ligands, has no ligand of its own and has evolved to function as a shared signaling subunit of ErbB receptor complexes, analogous to the gp120 shared subunit of lymphokine receptors.

D. Extending the Variation of Signaling Complexes by Diversiﬁcation of Ligand Recognition According to the model implying bivalence of ErbB binding ligands, each EGF-like domain can select its own unique set of preferred receptor dimers,

52

Leah N. Klapper et al.

suggesting that ligand multiplicity is segregated not only by differential expression but also by distinct recognition. Indeed, two ErbB-1 activating ligands, EGF and TGF-␣, have been shown to bind their receptor at nonoverlapping sites (Katsuura and Tanaka, 1989; Richter et al., 1992). This is extended to variations in the recruitment of dimeric partners as demonstrated for two isoforms of NRG1, both able to stabilize ErbB-2/ErbB-3 heterdimers, whereas only NRG1-␤ can promote ErbB-1/ErbB-3 heterodimers (Pinkas-Karmarski et al., 1996a). A similar diversity in complex formation is suggested for ErbB-1 ligands different in transactivation of the various ErbB receptors (Beerli and Hynes, 1996). Complex variation is further increased by the dual speciﬁcity of betacellulin (Riese et al., 1996) and HBEGF (Elenius et al., 1997), which bind both ErbB-1 and ErbB-4. In addition to its ability to stabilize ligand-bound complexes, and thereby prolong signal transduction, the superiority of ErbB-2 as a dimerizing partner is enhanced by its ability to expand speciﬁcity of ligand-receptor recognition. Epiregulin, a broad speciﬁcity ligand that preferentially binds ErbB-4 (Komurasaki et al., 1997), gains ErbB-1 activation and augments ErbB-4 activation in the presence of ErbB-2 (Riese et al., 1998; Shelly et al., 1998). Cross-speciﬁcity, conferred by ErbB-2, is further demonstrated by the ability of high concentrations of EGF and betacellulin to promote ErbB-2/ErbB3 dimers (Alimandi et al., 1997; Pinkas-Kramarski et al., 1998) and the ability of EGF, TGF-␣, and HB-EGF to signal through ErbB-2/ErbB-4 complexes (Shelly et al., 1998; Wang et al., 1998).

E. Intracellular Signaling The diversity of dimer formation could be attractively rationalized by the possibility that every ligand-bireceptor complex recruits a unique set of signaling proteins activating a distinct pathway. Indeed, a large number of cytoplasmic proteins, containing phosphotyrosine binding motifs, engage the activated ErbB dimers. However, in contrast to the dogma underlying diversity, many of these proteins overlap, interacting with most if not all dimeric species (Alroy and Yarden, 1997). These include effectors such as Shc (Pelicci et al., 1992; Segatto et al., 1993), Grb-2 (Buday and Downward, 1993), and Src (Anderson et al., 1990; Luttrell et al., 1994). Other substrates showing some speciﬁcity are Cbl, a protooncogenic adaptor that is recruited by all ErbB-1 containing dimers (Levkowitz et al., 1996); phospholipase C␥ (PLC␥), which associates with ErbB-1 and ErbB-2 (Cohen et al., 1996a; Peles et al., 1991), but not with ErbB-3 and ErbB-4; and phosphatidylinositol 3⬘-kinase (P13K), which shows a preference toward ErbB-3 (Stoltoff et al., 1996). Additionally, certain phosphorylated docking sites have permissive recognition enabling the binding of more than one adaptor via a hierarchy

ErbB/HER Signaling Network of Growth Factor Receptors

53

that is characteristic of each receptor. Such is the preferential selectivity of ErbB-3 toward Grb-7 upon Grb-2 (Fiddes et al., 1998), suggesting the interaction to be dependent on the cellular milieu. Despite the wide availability of transmitting molecules, mammalian ErbBs conserve the signaling pattern observed in invertebrates that utilizes the Ras-Raf-MAP-kinase cascade as its major route (Brunner et al., 1994; Hsu and Perrimon, 1994). It seems that receptor identity determines both the intensity and the kinetics of MAPK activation, increasing from homodimers to heterodimers and more so in heterodimers containing ErbB-2 (Graus-Porta et al., 1995; Karunagaran et al., 1996; Pinkas-Kramarski et al., 1996b). The ﬁne tuning of this cascade most probably governs the decision of differentiation or proliferation on growth factor stimulation (Marshall, 1995). Accordingly, NRG1 stimulation of cancer cells can cause proliferation or differentiation, depending on receptor expression (Bacus et al., 1992b; Daly et al., 1997), both effects correlating with the activation of MAPK (Grasso et al., 1997; Sepp Lorenzino et al., 1996). Likewise, dominant-negative inhibition of the MAPK-activating pathway abrogated transcriptional stimulation by NRG in differentiating myotubes (Altiok et al., 1997; Si et al., 1996). Regardless of the biological outcome, it seems that the response triggered by ErbB-speciﬁc growth factors follows a pattern of signals that spreads outward from the activated receptor and passes down a number of parallel pathways before converging onto a speciﬁc set of signaling molecules. Theoretically this can stabilize cell signaling pathways against transient ﬂuctuations in the concentrations of cell signaling molecules. The emergence of ErbB-2 as a favored signaling subunit of ErbB receptors highlights this concept even further due to its apparent strong coupling to MAPK (Ben-Levy et al., 1994). ErbB-2 couples a deleted ErbB-1, lacking its phosphotyrosine residues, to the Ras signaling pathway, recovering the DNA synthesis characteristic of the wild-type ErbB-1 (Sasaoka et al., 1996). Furthermore, a kinase-defective Neu abrogates motigenicity and transformation by ErbB-1, without causing its uncoupling to typical signaling molecules such as PLC␥, the ras-GTPase activating protein (Ras-GAP), and Shc, suggesting additional pathways underlying ErbB-2 superiority (Dougal et al., 1996). In such a case the ErbB-2dominated network will respond correctly even if changes to the network occur or the input is incomplete. Parallel network-pathways, originated to ensure correct signaling, could underlie pathological hypersignaling, leading to unregulated cell proliferation. Interestingly and in agreement with this possibility, two ErbB-2 binding proteins, Grb-2 and Grb-7, although known to interact with the same receptor residue, do not compete with each other in breast cancer cells. Apparently, a common co-overexpression of Grb-7 and ErbB-2 is correlated with increased ErbB-2/Grb-2 interactions and enhanced MAPK activation (Janes et al., 1997).

54

Leah N. Klapper et al.

F. Tuning of ErbB Signaling by Receptor Endocytosis The funneling of ErbB signaling into common pathways suggests that the variability in biological outcome would depend, at least partially, on kinetic regulation. A key determinant of signaling duration is the endocytic ligand-induced removal of receptors from the cell surface, a process called downregulation. Dependent on ligand binding, endocytosis and subsequent sorting to recycling back to the cell surface or degradation in lysosomes critically preserve a ﬁne balance mediating activation by growth factors (Wells et al., 1990). Indeed, alternative intracellular routing of ErbB proteins contributes to the diversiﬁcation of signal transduction (Baulida et al., 1996; Pinkas-Kramarski et al., 1996b). Whereas ErbB-3 undergoes slow endocytosis followed by recycling, most EGF-stimulated ErbB-1 molecules are destined to lysosomal degradation, the difference being expressed in the mitogenic potency of the complexes (Waterman et al., 1998). The importance of receptor-surface expression is further assigned by the identity of the binding ligand as demonstrated in certain cells by TGF-␣, a stronger activator than EGF, with an impaired ability to induce receptor downregulation (Ebner and Derynck, 1991). Interestingly, coexpression of ErbB-2 potentiates EGF signaling to the level achieved by TGF-␣, following heterodimer disintegration in the early endosome, and receptor recycling to the cell surface (Lenferink et al., 1998). Apparently, two processes determine intracellular sorting of endocytosed ErbB molecules (Fig. 4): First, recruitment of the c-Cbl adaptor protein positively shunts ligand–receptor complexes to degradation in lysosomes by elevating ubiquitination of the tyrosine-phosphorylated receptor (Levkowitz et al., 1998). Second, once in the sorting endosome, some ligand–receptor complexes dissociate under the mildly acidic pH of this vesicular compartment, leading to uncoupling from c-Cbl and destination to the default pathway, namely, recycling. By contrast, complexes that resist low pH, such as the EGF/ErbB-1 complex, do recruit c-Cbl, undergo ubiquitination, and thereby they are destined to degradation in lysosomes. In summary, signal transduction by the many EGF-like ligands is best described in terms of a neural-like network (see Fig. 5). This layered organization of multiple ligands, 10 dimeric receptors, and many downstream effector molecules funnels extracellular signals primarily into the Ras-Raf-MAPK pathway. However, despite its uniformity, the output of the network is variable: Prolonged activation is mediated by ErbB-2-containing heterodimers and may lead to transformation. Homodimers elicit only weak signaling because their endocytosis is relatively rapid, or because their selection of signaling molecules determines only transient coupling to the MAPK pathway. The network can be constitutively active if high ligand concentration is maintained through an autocrine or paracrine loop, or one of the receptors is mu-

ErbB/HER Signaling Network of Growth Factor Receptors

55

Fig. 4 Cell-surface receptor determination by endocytic routing. The identity of receptors pared by ligand binding determines the fate of the complex by intracellular routing. Following dimerization, receptors are aggregated within coated pits and internalized. The decision between degradation and recycling back to the cell surface is dependent on the stability of the complexes within cellular compartment. The acidic environment of the early endosome challenges the volatile interaction between the two receptors and the ligand causing the disintegration of unstable interactions. A heterodimer of ErbB-2 and ErbB-1 is readily dissociated at pH 5.5, enabling the rapid reappearance of the receptors on the cell surface. Such recycling reintroduces these molecules into the signaling milieu, thus conferring a sustained activation. ErbB-1 homodimers exhibit high stability under acidic conditions; remaining intact they are directed to subsequent compartments in which they undergo protein degradation. This results in a signiﬁcant reduction in cell-surface expression inﬂuencing the kinetics of signaling.

tated or overexpressed. The latter mode of activation is most relevant to ErbB-2 because it allows biased formation of the most active receptor combinations—those containing ErbB-2.

VI. ErbB-DIRECTED CANCER THERAPY The apparent correlation between ErbB expression and human cancer has attracted attention to these molecules as potential targets for the development of therapeutic modalities. Being mostly correlated to aggressiveness

56

Leah N. Klapper et al.

Fig. 5 The ErbB signaling network. Growth factor induced signaling is illustrated as protruding several consecutive layers. Multiple EGF-like ligands (⬎30) bind ErbB receptors recruited by differential and partially overlapping speciﬁcities, transmitting the signal through the membranal barrier. Ligand-receptor binding gives rise to 10 possible dimeric complexes, comprising the second level of signaling complexity. The identity of dimer-receptors is dependent on receptor expression and ligand afﬁnity and results in the activation of a variety of downstream pathways. Effector molecules containing SH2 or PTB domains serve as adaptors by binding activated receptors and conveying the signal further, activating parallel, interacting, or overlapping cascades. The most frequently evoked is the activation of MAPK followed by its translocation to the nucleus and subsequent stimulation of transcriptional activity. Stimuli funneling through this pathway can lead to both differentiation and proliferation, promoting normal or pathological cell growth. This network pattern suggests that the absorbance of a wide variety of growth-stimulating signals results in the activation of a multitude of intracellular molecules, only to converge into few mainstream pathways. Complexity of such a system is contemplated to confer ﬂexibility against ﬂuctuations in extracellular conditions, ensuring an intact cellular response.

and poor prognosis of epithelial cancers, ErbB-2 has been assigned most of the attempts, utilizing strategies directed to inhibit its activity as well as promote speciﬁc immunity (Disis and Cheever, 1998a,b).

A. Immunotherapy Antibodies that can block the biological activities of growth factor receptors are expected to alter the autocrine and paracrine loops. The rationale that speciﬁc ligand binding inhibition can decrease the mitogenic signaling

ErbB/HER Signaling Network of Growth Factor Receptors

57

of ErbB receptors elicited the development of several antibodies, adequate and beneﬁcial for human treatment (Fan and Mendelsohn, 1998). Antibody 225, a mouse monoclonal antibody that has been chimerized for use in humans, has a signiﬁcant antitumor activity on a variety of cultured and xenografted cancer cell lines (Prewett et al., 1996). It has been shown to successfully target primary lung cancers and metastasis (Divgi et al., 1991), and is apparently well tolerated in patients receiving repeated administrations. Other ErbB-1-speciﬁc monoclonal antibodies have been assessed in phase I clinical studies for their safety and efﬁcient binding in patients suffering from malignant gliiomas (Faillot et al., 1996), non-small cell cancer of the lung (Perez Soler et al., 1994), and head and neck cancers (Modjahedi et al., 1996). Likewise, successful inhibition of tumor growth has been accomplished by the use of monoclonal antibodies that speciﬁcally recognize ErbB2, in either a conventional athymic murine system or in a transgenic animal model of breast cancer (Katsumata et al., 1995). The immunological approach has recently been extended to patients: A phase II clinical trial revealed that a humanized antibody was clinically active in patients with ErbB2-overexpressing metastatic breast cancers (Baselga et al., 1996). Potentially, soluble ErbB-2 may interfere with the antitumorigenic effect of certain monoclonal antibodies directed to this receptor, and thus may limit the effect of immunotherapy (Brodowicz et al., 1997). In addition, possible mechanisms underlying the antitumorigenic effect are constantly challenged. Different antibodies directed against the extracellular domain of the receptor have been shown to both decrease and increase receptor phosphorylation (Stancovski et al., 1991), implying that simple inhibition of an evoked signaling cascade cannot explain the outcome. More appealing is the capability of inhibitory antibodies to downregulate the receptor from the cell surface (Hurwitz et al., 1995) or, alternatively, to destabilize heterodimeric complexes (Klapper et al., 1997), both mechanisms leading to a decreased signaling capacity. Anti-ErbB-2 antibodies have also been shown to affect the progression of the cell cycle either by inducing differentiation (Bacus et al., 1992b) or by driving cells toward apoptosis (Kita et al., 1996). Patients with ErbB-2-positive cancers have been occasionally shown to develop an immune response against the protein (Disis et al., 1994, 1997; Fisk et al., 1995) predicting that antireceptor vaccines could be successful in evoking an anticancer response. The high expression of ErbB-2 on cancer cells, in comparison to normal tissues, suggests that such a response should be preferentially directed against malignancy with no or residual autoimmune toxicity. Originally, murine tumors overexpressing the rat oncogenic neu were successfully targeted and treated by immunization with a vaccinia virus recombinant of the protein’s extracellular domain (Bernards et al., 1987). Peptides from both intracellular (Ioannides et al., 1992; Peiper et al., 1997) and extracellular (Fisk et al., 1995) portions of the receptor can elicit a spe-

58

Leah N. Klapper et al.

ciﬁc response of cytotoxic T lymphocytes (CTL) originating from cancer patients. Tolerance to self-proteins has been suggested to depend on dominant epitopes allowing the promotion of an immune response to such molecules by the exposure of subdominant epitopes (Sercarz et al., 1993). Accordingly, immunizing rats with peptides derived from the self-rat Neu, but not with the whole protein, can promote antibody and T-cell responses against the native protein (Disis et al., 1996). Similar peptides, derived from the murine ErbB-2, can induce CTL activity, resulting in the suppression of growth of receptor-overexposing cells in syngeneic hosts (Nagata et al., 1997). To confer ErbB-2 recognition to T cells, without the need for antigen processing while circumventing MHC restriction, expression of chimeric antibodies against ErbB-2 fused to the signaling subunit of the T-cell receptor was designed (Stancovski et al., 1993). Adoptive transfer of such CTLs can markedly inhibit the growth of ErbB-2-transformed cells in nude mice (Moritz et al., 1994) and in a syngeneic immunocompetent model (Altenschmidt et al., 1997a). ErbB-2-speciﬁc targeting and activation of T cells can also be achieved by fusing an antibody speciﬁc for ErbB-2 to a sequence encoding the extracellular domain of the B7-1 (Challita Eid et al., 1998) or B7-2 (Gerstmayer et al., 1997) T-cell costimulatory proteins. A similar methodology is utilized to attract and activate an additional arm of the immunological response including monocytes and macrophages. A bispeciﬁc antibody, directed against ErbB-2 and the Fc␥ RIII, systemically administered to SCID mice bearing ovarian cancer signiﬁcantly improved survival while associated with no observed toxicity (Weiner et al., 1993). This encouraged a phase I clinical trial and contemplated future studies (Weiner et al., 1995). Similarly, antibodies directed against Fc␥ RI and ErbB-2 or ErbB1 were evaluated in phase II clinical trials for treatment of a variety of neoplasms (Curnow, 1997; van-Ojik et al., 1997) showing a promising range of responses as expressed by a reduction in metastasis and serum markers.

B. Gene Therapy Transcriptional upregulation signiﬁcantly contributes to ErbB-2 overexpression in cancers (Kraus et al., 1987; Miller et al., 1994; Pasleau et al., 1993), suggesting that manipulation of the promoter activity can be utilized for therapy. Such is the selective expression of suicide genes driven by regulatory regions of the erbB-2 promoter rendering cells sensitive to gancyclovir (Ring et al., 1997). A different approach is the use of adenovirus type 5 early region 1A gene product (E1A) to repress ErbB-2 expression, suppressing the tumorigenic potential of overexpressing cells (Yu and Hung, 1991). The growth of human cancer cells of breast (J. Y. Chang et al., 1997), ovary, and lung (Chang et al., 1996) origin in nude mice is efﬁciently inhibited by the

ErbB/HER Signaling Network of Growth Factor Receptors

59

viral product when delivered by vector or liposomes. This can also be effectively used to sensitize cells toward chemotherapy as demonstrated for breast cancer cells exposed to taxol (Ueno et al., 1997). Alternatively, mRNA levels can be manipulated by conditional depletion; anti-ErbB-2 targeted hammerhead ribozymes, expressed under the control of a tetracycline-regulated promoter, can almost completely abrogate expression of the protein at the cell surface, resulting in the inhibition of tumor growth in nude mice, as well as in tumor regression upon tetracycline withdrawal (Juhl et al., 1997). Similarly, antisense cDNA constructs encompassing different regions of the erbB-2 gene inhibit the tumorigenicity of lung adenocarcinoma cells (Casalini et al., 1997). Last, DNA delivery by adenoviral vectors has also been utilized for the introduction of an anti-ErbB-2 single chain antibody capable of retaining the protein within the cell. Intraperitoneal injection of the vector resulted in the reduction of tumor burden in SCID mice (Deshane et al., 1997), encouraging a phase I clinical trial with ovarian cancer patients (Alvarez and Curiel, 1997).

C. Other Modes of Therapy 1. ANTIBODY-DRUG COMBINATION As discussed earlier, tumors overexpressing ErbB-2 show lower responsiveness to adjuvant therapy that includes cyclophosphamide, methotrexate, and 5⬘-ﬂuorouracil (CMF) (Allred et al., 1992b; Gusterson et al., 1992). Furthermore, ErbB-2 seems to synergize with the multidrug-resistant protein, p170mdr-1, rendering breast cancer cells more resistant to taxol (Yu et al., 1998). One possible explanation for this sensitization could be the enhancement of cellular proliferation enabling cells surviving a therapeutic course to rapidly propagate (Pegram et al., 1997). The increase in resistance to therapy conferred by receptor overexpression suggests that interference with ErbB-2 expression at the cell surface could lead to a better response (Torre et al., 1997). Several studies have examined the possible advantage of combining anti-ErbB-2 effects with chemotherapeutic treatment. A synergistic inhibitory effect between mAbs to the EGF receptor and the DNA-damaging drug cisplatin has previously been reported (Aboud Pirak et al., 1988). Similarly, an enhanced cytotoxicity of cisplatin, in breast and ovarian cells overexpressing ErbB-2, has been observed when cells were concomitantly exposed to an anti-ErbB-2 antibody (Hancock et al., 1991; Pietras et al., 1994). Further analysis of this phenomenon showed a reduction in both DNA synthesis and repair of cisplatin-DNA adducts in the presence of the antibody (Arteaga et al., 1994; Pietras et al., 1994), suggesting an elevated chemosensitivity as a result of antibody treatment. Enhanced cisplatin sensitivity in

60

Leah N. Klapper et al.

the presence of anti-ErbB-2 mAbs, has been shown to depend on agonistic properties of the antibody (Arteaga et al., 1994). Tyrphostin 50864-2, a low molecular weight tyrosine kinase inhibitor, can abrogate the elevated drugmediated cell killing induced by an anti-ErbB-2 antibody. Moreover, the enhancement was not observed with an anti-ErbB-2 mAb that does not induce cell signaling (Arteaga et al., 1994). A similar sensitization was achieved for the treatment with the anti-estrogen drug, tamoxifen (Witters et al., 1997), as well as with TNF (Hudziak et al., 1989), showing an enhanced inhibitory effect in vitro in the presence of an anti-ErbB-2 antibody.

2. IMMUNOTOXINS Antibodies directed against ErbB-2 have been suggested as useful vehicles for the targeting of therapeutic agents to tumors. This approach is attractive due to both the correlation of receptor expression with cancer as well as the ability of antibodies to internalize with the receptor and introduce the toxic agent into the cell (Hudziak et al., 1989; Hurwitz et al., 1995; van Leeuwen et al., 1990). Conjugates of mAbs and toxins have been used in preclinical trials as antitumor agents (Pastan and FitzGerald, 1991). Several immunotoxins have been constructed using various anti-ErbB-2 antibodies that have been coupled to Lys-PE40, a recombinant form of Pseudomonas exotoxin lacking its cell-binding domain (Batra et al., 1992). Anti-ErbB-2-exotoxin successfully inhibits the growth of Schwannoma cells in nude mice (Altenschmidt et al., 1997b). Several other agents have been similarly targeted, including ricin (Rodriguez et al., 1993), doxorubicin (Park et al., 1995), and enzyme prodrugs (Eccles et al., 1994), all presenting speciﬁc cell inhibitory effects. Ligands directed against ErbB proteins have also been examined as beneﬁcial carriers, utilizing their high binding afﬁnity to respective receptors. A fusion toxin of NRG1 with exotoxin-a induced complete regression of human breast cancer xenografts in nude mice (Groner et al., 1997). Betacellulin-pseudomonas toxin fusion is effective against cells expressing ErbB-1 but not cells expressing ErbB-4, probably due to a limited internalizing capacity (Mixan et al., 1998). A bispeciﬁc toxin combining the recognition ability of an anti-ErbB-2 antibody with that of TGF-␣ inhibits the growth of breast cancer cells in vivo (Schmidt et al., 1996), probably by the induction of heterodimeric complexes and their subsequent internalization. Drug delivery has also been attempted by antibody-targeting of drugloaded liposomes. Immunoliposomes efﬁciently bind to cancer cells, delivering cytotoxic doses of doxorubicin in a targeted manner (Park et al., 1995) as a function of their ability to internalize (Goren et al., 1996). A prolonged tumor-localized supply of an ErbB-2 speciﬁc toxin has been elegantly achieved by the development of a new class of tumor-speciﬁc killer lymphocytes. These cells produce and secrete an antibody-targeted toxin in the vicinity of

ErbB/HER Signaling Network of Growth Factor Receptors

61

the tumor, overcoming depletion by clearance, and result in high cytotoxicity toward tumors in an athymic murine model (Chen et al., 1997).

3. TYROSINE KINASE INHIBITORS In an attempt to inhibit the mitogenic signaling of receptor tyrosine kinases, several chemical compounds have been designed and synthesized to interfere with enzymatic activity (Klohs et al., 1997). Two groups of molecules, termed tryphostins, have been developed to bear selective speciﬁcities toward the ATP-binding sites of ErbB-1 or ErbB-2, resulting in an inhibition of proliferation of cells expressing the respective receptors (Osherov et al., 1993). Tyrphostins speciﬁc for the ErbB-1 receptor inhibit primary glioblastoma cells from invading brain aggregates (Penar et al., 1997) and prostate cancer from proliferation (Kondapaka and Reddy, 1996). A similar compound, capable of inhibiting activation of ErbB receptors, is a potent in vivo inhibitor of various human xenografts expressing ErbBs (Rewcastle et al., 1998). AG825, a speciﬁc inhibitor of the ErbB-2 tyrosine kinase, sensitizes receptor-overexpressing cells to chemotherapy including doxorubicin, etoposide, and cisplatin (Tsai et al., 1996), suggesting the involvement of ErbB-2 signaling in resistance toward chemotherapy. Thus, low molecular weight compounds capable of selective inhibition of the catalytic activity of specific ErbB proteins, either alone or in combination with other drugs, are potential future cancer therapeutic agents.

VII. CONCLUSIONS Mounting experimental evidence now supports the notion that signaling by the ErbB receptors and their ligand growth factors may be explained in terms of a protein network. The signaling module evolved to function in inductive morphogenesis, but it is opportunistically exploited by malignant processes. This may not be limited to ErbBs; other growth factors, such as the hepatocyte growth factor and its sibling factors affecting cell migration and metastasis, and the vascular growth factors, which control angiogenesis, probably operate through similar signaling networks. It is currently unclear to which extent ErbB signaling is involved in the control of cell migration, angiogenesis, or apoptosis. For example, the possibility that ErbB proteins confer resistance to apoptosis induced by certain chemotherapeutic drugs is attractive but needs additional experimental support. Nevertheless, in vitro, as well as in vivo and clinical lines of evidence indicate that the major function of the ErbB network is to control the decision to proliferate or differentiate. Biochemically, this decision is executed by a linear cascade that

62

Leah N. Klapper et al.

includes Ras, Raf, and MAPK. However, ErbB signaling recruits several other pathways, for example, the phosphatydilinositol 3⬘-kinase pathway, whose physiological role is less understood. Crucial to understanding ErbB signaling is the unsolved biochemical role of the ligand-less oncoprotein, ErbB-2. The emerging notion arguing that ErbB-2 acts solely as a low-afﬁnity/broad-speciﬁcity subunit of the three other ErbB receptors is appealing, because it attributes the oncogenic superiority of ErbB-2 to its probable ability to augment the proliferative actions of multiple stroma-derived growth factors. Accordingly, overexpression of ErbB-2 in tumor cells biases the formation of the respective heterodimer, thereby favoring formation of signaling complexes whose activity is more potent and prolonged. While this model explains the ﬂexibility and robustness of the ErbB network, it also offers a scenario for therapy directed at blocking ErbB function in cancer cells: The lack of signaling autonomy of ErbB-2 means that not only antibodies to this protein, but also antagonists of speciﬁc growth factors, and blockers of other receptors and their enzymatic activities, will inhibit ErbB-mediated transformation. It is likely that speciﬁc ligands and certain heterodimeric ErbBs are more critical than others in tumors of different origins. Characterization of this speciﬁcity and its pharmacological targeting remain future challenges.

ACKNOWLEDGMENTS Our laboratories are supported by grants from the National Institutes of Health, the Department of the Army, the Israel Basic Research Fund, and by a Bristol-Myers Squibb Foundation Cancer Grant Award (to M.S.). M.H.K. is the recipient of a postdoctoral fellowship from the Susan G. Komen Breast Cancer Foundation.

REFERENCES Aboud Pirak, E., Hurwitz, E., Pirak, M. E., Bellot, F., Schlessinger, J., and Sela, M. (1988). J. Natl. Cancer Inst. 80, 1605–1611. Akiyama, T., Matsuda, S., Namba, Y., Saito, T., Toyoshima, K., and Yamamoto, T. (1991). Mol. Cell. Biol. 11, 833–842. Albanell, J., Bellmunt, J., Molina, R., Garcia, M., Caragol, I., Bermejo, B., Ribas, A., Carulla, J., Gallego, O. S., Espanol, T., and Sole Calvo, L. A. (1996). Anticancer Res. 16, 1027– 1032. Alimandi, M., Romano, A., Curia, M. C., Muraro, R., Fedi, P., Aaronson, S. A., Di Fiore, P. P., and Kraus, M. H. (1995). Oncogene 10, 1813–1821. Alimandi, M., Wang, L.-M., Bottaro, D., Lee, C.-C., Angera, K., Frankel, M., Fedi, P., Tang, F., Tang, C., Lippman, M., and Pierce, J. H. (1997). EMBO J. 16, 5608–5617.

ErbB/HER Signaling Network of Growth Factor Receptors

63

Allred, D. C., Clark, G. M., Molina, R., Tandon, A. K., Schnitt, S. J., Gilchrist, K. W., Osborne, C. K., Tormey, D. C., and McGuire, W. L. (1992a). Hum. Pathol. 23, 974–979. Allred, D. C., Clark, G. M., Tandon, A. K., Molina, R., Tormey, D. C., Osborne, C. K., Gilchrist, K. W., Mansour, E. G., Abeloff, M., Eudey, L., et al. (1992b). J. Clin. Oncol. 10, 599–605. Alroy, I., and Yarden, Y. (1997). FEBS Lett. 410, 83–86. Altenschmidt, U., Klundt, E., and Groner, B. (1997a). J. Immunol. 159, 5509–5515. Altenschmidt, U., Schmidt, M., Groner, B., and Wels, W. (1997b). Int. J. Cancer 73, 117–124. Altiok, N., Altiok, X., and Changeux, J. P. (1997). EMBO J. 16, 717–725. Alvarez, R. D., and Curiel, D. T. (1997). Hum. Gene Ther. 8, 229–242. Amadori, D., Maltoni, M., Volpi, A., Nanni, O., Scarpi, E., Renault, B., Pellegata, N. S., Gaudio, M., Magni, E., and Ranzani, G. N. (1997). Cancer (Philadelphia) 79, 226–232. Anan, K., Morisaki, T., Katano, M., Ikubo, A., Tsukahara, Y., Kojima, M., Uchiyama, A., Kuroki, S., Torisu, M., and Tanaka, M. (1998). Eur. J. Surg. Oncol. 24, 28–33. Anderson, D., Koch, C. A., Grey, L., Ellis, C., Moran, M. F., and Pawson, T. (1990). Science 250, 979–982. Andrulis, I. L., Bull, S. B., Blackstein, M. E., Sutherland, D., Mak, C., Sidlofsky, S., Pritzker, K. P., Hartwick, R. W., Hanna, W., Lickley, L., Wilkinson, R., Qizilbash, A., Ambus, U., Lipa, M., Weizel, H., Katz, A., Baida, M., Mariz, S., Stoik, G., Dacamara, P., Strongitharm, D., Geddie, W., and McCready, D. (1998). J. Clin. Oncol. 16, 1340–1349. Archer, S. G., Eliopoulos, A. Spandidos, D., Barnes, D., Ellis, I. O., Blamey, R. W., Nicholson, R. I., and Robertson, J. F. (1995). Br. J. Cancer 72, 1259–1266. Aroian, R. V., Koga, M., Mendel, J. E., Ohshima, Y., and Sternberg, P. W. (1990). Nature (London) 348, 693–699. Arteaga, C. L., Winnier, A. R., Poirier, M. C., Lopez Larraza, D. M., Shawver, L. K., Hurd, S. D., and Stewart, S. J. (1994). Cancer Res. 54, 3758–3765. Auranen, A., Grenman, S., and Kleml, P. J. (1997). Cancer (Philadelphia) 79, 2147–2153. Baasner, S., von Melchner, H., Klenner, T., Hilgard, P., and Beckers, T. (1996). Oncogene 13, 901–911. Backe, J., Gassel, A. M., Krebs, S., Muller, T., and Cafﬁer, H. (1997). Arch. Gynecol. Obstet. 259, 189–195. Bacus, S. S., Huberman, E., Chin, D., Kiguchi, K., Simpson, S., Lippman, M., and Lupu, R. (1992a). Cell Growth Differ. 3, 401–411. Bacus, S. S., Stancovski, I., Huberman, E., Chin, D., Hurwitz, E., Mills, G. B., Ullrich, A., Sela, M., and Yarden, Y. (1992b). Cancer Res. 52, 2580–2589. Bacus, S. S., Gudkov, A. V., Zelnick, C. R., Chin, D., Stern, R., Stancovski, I., Peles, E., Ben Baruch, N., Farbstein, H., Lupu, R., Wen, D., Sela, M., and Yarden, Y. (1993). Cancer Res. 53, 5251–5261. Barbacci, E. G., Guarino, B. C., Stroh, J. G., Singleton, D. H., Rosnack, K. J., Moyer, J. D., and Andrews, G. C. (1995). J. Biol. Chem. 270, 9585–9589. Bargmann, C. I., Hung, M. C., and Weinberg, R. A. (1986). Nature (London) 319, 226– 230. Barnes, D. M., Bartkova, J., Camplejohn, R. S., Gullick, W. J., Smith, P. J., and Millis, R. R. (1992). Eur. J. Cancer 28, 644–648. Barsky, S. H., Doberneck, S. A., Sternlicht, M. D., Grossman, D. A., and Love, S. M. (1997). J. Pathol. 183, 188–194. Baselga, J., Tripathy, D., Mendelsohn, J., Baughman, S., Benz, C. C., Dantis, L., Sklarin, N. T., Seidman, A. D., Hudis, C. A., Moore, J., Rosen, P. P., Twaddell, T., Henderson, I. C., and Norton, L. (1996). J. Clin. Oncol. 14, 737–744. Bates, N. P., and Hurst, H. C. (1997). Oncogene 15, 473–481. Batra, J. K., Kasprzyk, P. G., Bird, R. E., Pastan, I., and King, C. R. (1992). Proc. Natl. Acad. Sci. U.S.A. 89, 5867–5871.

64

Leah N. Klapper et al.

Bauer, M., Horn, L. C., Kowalzik, J., Mair, W., and Czerwenka, K. (1997). Gen. Diagn. Pathol. 143, 185–190. Baulida, J., Kraus, M. H., Alimandi, M., Di Fiore, P. P., and Carpenter, G. (1996). J. Biol. Chem. 271, 5251–5257. Bebenek, M., Bar, J. K., Harlozinska, A., and Sedlaczek, P. (1998). Anticancer Res. 18, 619–623. Beerli, R. R., and Hynes, N. E. (1996). J. Biol. Chem. 271, 6071–6076. Beerli, R. R., Wels, W., and Hynes, N. E. (1994). J. Biol. Chem. 269, 23931–23936. Bell, J. G., Minnick, A., Reid, G. C., Judis, J., and Brownell, M. (1997). Gynecol. Oncol. 66, 388–392. Ben-Levy, R., Paterson, H. F., Marshall, C. J., and Yarden, Y. (1994). EMBO J. 13, 3302– 3311. Berchuck, A., Kamel, A., Whitaker, R., Kerns, B., Olt, G., Kinney, R., Soper, J. T., Dodge, R., Clarke Pearson, D. L., Marks, P., Mckenzie, S., Yin, S., and Bast, C. R., Jr. (1990a). Cancer Res. 50, 4087–4089. Berchuck, A., Rodriguez, G., Kamel, A., Soper, J. T., Clarke Pearson, D. L., and Bast, R. C., Jr. (1990b). Obstet. Gynecol. 76, 381–387. Berchuck, A., Rodriguez, G., Kinney, R. B., Soper, J. T., Dodge, R. K., Clarke Pearson, D. L., and Bast, C. R., Jr. (1991). Am. J. Obstet. Gynecol. 164, 15–21. Berger, M. S., Greenﬁeld, C., Gullick, W. J., Haley, J., Downward, J., Neal, D. E., Harris, A. L., and Waterﬁeld, M. D. (1987). Br. J. Cancer 56, 533–537. Bernards, R., Destree, A., McKenzie, S., Gordon, E., Weinberg, R. A., and Panicali, D. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 6854–6858. Berns, E. M., Foekens, J. A., van Staveren, I. L., van Putten, W. L., de Koning, H. Y., Portengen, H., and Klijn, J. G. (1995). Gene 159, 11–18. Bernstein, J. J., Anagnostopoulos, A. V., Hattwick, E. A., and Laws, E. R., Jr. (1993). J. Neurosurg. 78, 240–251. Bertheau, P., Steinberg, S. M., and Merino, M. J. (1998). Hum. Pathol. 29, 323–329. Bishop, J. M. (1991). Cell (Cambridge, Mass.) 64, 235–248. Borg, A., Baldetorp, B., Ferno, M., Killander, D., Olsson, H., Ryden, S., and Sigurdsson, H. (1994). Cancer Lett. 81, 137–134. Bose, S., Lesser, M. L., Norton, L., and Rosen, P. P. (1996). Arch. Pathol. Lab. Med. 120, 81–85. Brandt Rauf, P. W., Rackovsky, S., and Pincus, M. R. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 8660–8664. Brodowicz, T., Wiltschke, C., Budinsky, A. C., Krainer, M., Steger, G. G., and Zielinski, C. C. (1997). Int. J. Cancer 73, 875–879. Brower, S. T., Ahmed, S., Tartter, P. I., Bleiweiss, I., and Amberson, J. B. (1995). Ann. Surg. Oncol. 2, 440–444. Brown, J. P., Twardzik, D. R., Marquardt, H., and Todaro, G. J. (1985). Nature (London) 313, 491–492. Bruce, D. M., Heys, S. D., Payne, S., Miller, I. D., and Eremin, O. (1996). Eur. J. Surg. Oncol. 22, 42–46. Brunner, D., Oellers, N., Szabad, J., Biggs, W. R., Zipursky, S. L., and Hafen, E. (1994). Cell (Cambridge, Mass.) 76, 875–888. Buday, L., and Downward, J. (1993). Cell (Cambridge, Mass.) 73, 611–620. Burden, S., and Yarden, Y. (1997). Neuron 18, 847–855. Burke, C., Lemmon, M., Coren, B., Engelman, D., and Stern, D. (1997). Oncogene 14, 687–696. Burke, H. B., Hoang, A., Iglehart, J. D., and Marks, J. R. (1998). Cancer (Philadelphia) 82, 874–877. Campion, S. R., Geck, M. K., and Niyogi, S. K. (1993). J. Biol. Chem. 268, 1742–1748. Carlomagno, C., Perrone, F., Gallo, C., De Laurentiis, M., Lauria, R., Morabito, A., Pettinato,

ErbB/HER Signaling Network of Growth Factor Receptors

65

G., Panico, L., D’Antonio, A., Bianco, A. R., and De Placido, S. (1996). J. Clin. Oncol. 14, 2702–2708. Carraway, K. L., Weber, J., Unger, M., Ledesma, J., Yu, N., Gassmann, M., and Lai, C. (1997). Nature (London) 387, 512–516. Carraway, K. L., Sliwkowski, M. X., Akita, R., Platko, J. V., Guy, P. M., Nuijens, A., Diamonti, A. J., Vandlen, R. L., Cantley, L. C., and Cerione, R. A. (1994). J. Biol. Chem. 269, 14303– 14306. Caruso, M. L., and Valentini, A. M. (1996). Anticancer Res. 16, 3813–3818. Casalini, P., Menard, S. M., Malandrin, S. M., Rigo, C. M., Colnaghi, M. I., Cultraro, C. M., and Segal, S. (1997). Int. J. Cancer 72, 631–636. Challita Eid, P. M., Penichet, M. L., Shin, S. U., Poles, T., Mosammaparast, N., Mahmood, K., Slamon, D. J., Morrison, S. L., and Rosenblatt, J. D. (1998). J. Immunol. 160, 3419–3426. Chang, H., Riese, D., Gilbert, W., Stern, D. F., and McMahan, U. J. (1997). Nature (London) 387, 509–512. Chang, J. Y., Xia, W., Shao, R., and Hung, M. C. (1996). Oncogene 13, 1405–1412. Chang, J. Y., Xia, W., Shao, R., Sorgi, F., Hortobagyi, G. N., Huang, L., and Hung, M. C. (1997). Oncogene 14, 561–568. Chang, W., Upton, C., Hu, S.-L., Purchio, A. F., and McFadden, G. (1987). Mol. Cell Biol. 7, 535–540. Charpin, C., Devictor, B., Bonnier, P., Andrac, L., Lavaut, M. N., Allasia, C., and Piana, L. (1993). Breast Cancer Res. Treat. 25, 203–210. Charpin, C., Garcia, S., Bouvier, C., Devictor, B., Andrac, L., Choux, R., and Lavaut, M. (1997). J. Pathol. 181, 294–300. Chen, S. Y., Yang, A. G., Chen, J. D., Kute, T., King, C. R., Collier, J., Cong, Y., Yao, C., and Huang, X. F. (1997). Nature (London) 385, 78–80. Chen, X., Levkowitz, G., Tzahar, E., Karunagaran, D., Lavi, S., Ben-Baruch, N., Leitner, O., Ratzkin, B. J., Bacus, S. S., and Yarden, Y. (1996). J. Biol. Chem. 271, 7620–7629. Chow, N. H., Liu, H. S., Yang, H. B., Chan, S. H., and Su, I. J. (1997). Virchows Arch. 430, 461–466. Chozick, B. S., Benzil, D. L., Stopa, E. G., Pezzullo, J. C., Knuckey, N. W., Epstein, M. H. Finkelstein, S. D., and Finch, P. W. (1996). J. Neuro-Oncol. 27, 117–126. Ciardiello, F., Kim, N., Saeki, T., Dono, R., Persico, M. G., Plowman, G. D., Garrigues, J., Radke, S., Todaro, G. J., and Salomon, D. S. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 7792–7796. Clark, G. M., and McGuire, W. L. (1991). Cancer Res. 51, 944–948. Clifford, R., and Schuppbach, T. (1994). Genetics 137, 531–550. Cohen, B. D., Green, J. M., Foy, L., and Fell, H. P. (1996a). J. Biol. Chem. 271, 4813–4818. Cohen, B. D., Kiener, P. K., Green, J. M., Foy, L., Fell, H. P., and Zhang, K. (1996b). J. Biol. Chem. 271, 30897–30903. Coussens, L., Yang Feng, T. l., Liao, Y. C., Chen, E., Gray, A., McGrath, J., Seeburg, P. H., Libermann, T. A., Schlessinger, J., Francke, U., Levinson, A., and Ullrich, A. (1985). Science 230, 1132–1139. Curnow, R. T. (1997). Cancer Immunol. Immunother. 45, 210–215. Daly, J., Jannot, C., Beerli, R., Graus-Porta, D., Maurer, F., and Hynes, N. (1997). Cancer Res. 57, 3804–3811. Das, S., Wang, X.-N., Paria, B., Damm, D., Abraham, J., Klagsbrun, M., Andrews, G., and Dey, S. (1994). Development (Cambridge, UK) 120, 1071–1083. Deshane, J., Siegal, G. P., Wang, M., Wright, M., Bucy, R. P., Alvarez, R. D., and Curiel, D. T. (1997). Gynecol. Oncol. 64, 378–385. Di Fiore, P. P., Pierce, J. H., Kraus, M. H., Segatto, O., King, C. R., and Aaronson, S. A. (1987). Science 237, 178–182.

66

Leah N. Klapper et al.

Disis, M. L., and Cheever, M. A. (1998a). Crit. Rev. Immunol. 18, 37–45. Disis, M. L., and Cheever, M. A. (1998b). Adv. Cancer Res. 71, 344–371. Disis, M. L., Bernhard, H., Gralow, J. R., Hand, S. L., Emery, S. R., Calenoff, E., and Cheever, M. A. (1994). Ciba Found. Symp. 187, 198–207. Disis, M. L., Gralow, J. R., Bernhard, H., Hand, S. L., Rubin, W. D., and Cheever, M. A. (1996). J. Immunol. 156, 3151–3158. Disis, M. L., Pupa, S. M., Gralow, J. R., Dittadi, R., Menard, S., and Cheever, M. A. (1997). J. Clin. Oncol. 15, 3363–3367. Divgi, C. R., Welt, S., Kris, M., Real, F. X., Yeh, S. D., Gralla, R., Merchant, B., Schweighart, S., Unger, M., Larson, S. M., and Mendelsohn, J. (1991). J. Natl. Cancer Inst. 83, 97–104. Dougal, W. C., Quian, X., Miller, M. J., and Greene, M. I. (1996). DNA Cell Biol. 15, 31–40. Downward, J., Yarden, Y., Mayes, E., Scrace, G., Totty, N., Stockwell, P., Ullrich, A., Schlessinger, J., and Waterﬁeld, M. D. (1984). Nature (London) 307, 521–527. Dugan, M. C., Dergham, S. T., Kucway, R., Singh, K., Biernat, L., Du, W., Vaitkevicius, V. K., Crissman, J. D., and Sarkar, F. H. (1997). Pancreas 14, 229–236. Duhaylongsod, F. G., Gottfried, M. R., Iglehart, J. D., Vaughn, A. L., and Wolfe, W. G. (1995). Ann. Surg. 221, 677–683; discussion: 683–684. Ebner, R., and Derynck, R. (1991). Cell Regul. 2, 599–612. Eccles, S. A., Court, W. J., Box, G. A., Dean, C. J., Melton, R. G., and Springer, C. J. (1994). Cancer Res. 54, 5171–5177. Eissa, S., Khalifa, A., el-Gharib, A., Salah, N., and Mohamed, M. K. (1997). Anticancer Res. 17, 3091–3097. Elenius, K., Paul, S., Allison, G., Sun, J., and Klagsbrun, M. (1997). EMBO J. 16, 1268–1278. Elledge, R. M., Clark, G. M., Chamness, G. C., and Osborne, C. K. (1994). J. Natl. Cancer Inst. 86, 705–712. Elledge, R. M., Green, S., Ciocca, D., Pugh, R., Allred, D. C., Clark, G. M., Hill, J., Ravdin, P., O’Sullivan, J., Martino, S., and Osborne, C. K. (1998). Clin. Cancer Res. 4, 7–12. Eltabbakh, G. H., Belinson, J. L., Kennedy, A. W., Biscotti, C. V., Casey, G., and Tubbs, R. R. (1997). Gynecol. Oncol. 65, 218–224. Ernster, V. L., Barclay, J., Kerlikowske, K., Grady, D., and Henderson, C. (1996). JAMA, J. Am. Med. Assoc. 275, 913–918. Fabian, C. J., Zalles, C., Kamel, S., McKittrick, R., Moore, W. P., Zeiger, S., Simon, C., Kimler, B., Cramer, A., Garcia, F., and Jewell, W. (1993). J. Cell Biochem. Suppl. 17G, 153–160. Fabian, C. J., Kamel, S., Zalles, C., and Kimler, B. F. (1996). J. Cell Biochem., Suppl. 25, 112–122. Faillot, T., Magdelenat, H., Mady, E., Stasiecki, P., Fohanno, D., Gropp, P., Poisson, M., and Delattre, J. Y. (1996). Neurosurgery 39, 478–483. Falls, D. L., Rosen, K. M., Corfas, G., Lane, W. S., and Fischbach, G. D. (1993). Cell (Cambridge, Mass.) 72, 801–815. Fan, Z., and Mendelsohn, J. (1998). Curr. Opin. Oncol. 10, 67–73. Farabegoli, F., Ceccarelli, C., Santini, D., Trere, D., Baldini, N., Taffurelli, M., and Derenzini, M. (1996). Int. J. Cancer 69, 381–385. Fearon, E. F., and Vogelstein, B. (1990). Cell (Cambridge, Mass.) 61, 759–767. Fehm, T., Maimonis, P., Weitz, S., Teramoto, Y., Katalinic, A., and Jager, W. (1997). Breast Cancer Res. Treat. 43, 87–95. Fiddes, R. J., Campbell, D. H., Janes, P. W., Sivertsen, S. P., Sasaki, H., Wallasch, C., and Daly, R. J. (1998). J. Biol. Chem. 273, 7717–7724. Fisher, E. R., Costantino, J., Fisher, B., Palekar, A. S., Paik, S. M., Suarez, C. M., and Wolmark, N. (1996). Cancer (Philadelphia) 78, 1403–1416. Fisk, B., Blevins, T. L., Wharton, J. T., and Ionnides, C. G. (1995). J. Exp. Med. 181, 2109–2117.

ErbB/HER Signaling Network of Growth Factor Receptors

67

Flejou, J. F., Paraf, F., Muzeau, F. Fekete, F., Henin, D., Jothy, S., and Potet, F. (1994). J. Clin. Pathol. 47, 23–26. Fontana, X., Ferrari, P., Namer, M., Peysson, R., Salanon, C., and Bussiere, F. (1994). Anticancer Res. 14, 2099–2104. Fox, S. B., Day, C. A., and Rogers, S. (1991). J. Clin. Pathol. 44, 960–961. Freeman, M., Paul, S., Kaefer, M., Ishikawa, M., Adam, R., Renshaw, A., Elenius, K., and Klagsbrun, M. (1998). J. Cell. Biochem. 68, 328–338. Gallo, O., Franchi, A., Fini-Storchi, I., Cilento, G., Boddi, V., Boccuzzi, S., and Urso, C. (1998). Head Neck 20, 224–231. Gasparini, G., Boracchi, P., Bevilacqua, P., Mezzetti, M., Pozza, F., and Weidner, N. (1994a). Breast Cancer Res. Treat. 29, 59–71. Gasparini, G., Gullick, W. J., Maluta, S., Palma, P. D., Caffo, O., Leonardi, E., Boracchi, P., Pozza, F., Lemoine, N. R., and Bevilacqua, P. (1994b). Eur. J. Cancer 30a, V2. Gassel, A. M., Backe, J., Krebs, S., Schon, S., Cafﬁer, H., and Muller-Hermelink, H. K. (1998). J. Clin. Pathol. 51, 25–29. Gassmann, M., Casangranda, F., Orioli, D., Simon, H., Lai, C., Klein, R., and Lemke, G. (1995). Nature (London) 378, 390–394. Gerstmayer, B., Altenschmidt, U., Hoffmann, M., and Wels, W. (1997). J. Immunol. 158, 4584–4590. Giannoni, C., el Naggar, A. K., Ordonez, N. G., Tu, Z. N., Austin, J., Luna, M. A., and Batsakis, J. G. (1995). Otolaryngol. Head Neck Surg. 112, 391–398. Giatromanolaki, A., Koukourakis, M. I., O’Byrne, K., Kaklamanis, L., Dicoglou, C., Trichia, E., Whitehouse, R., Harris, A. L., and Gatter, K. C. (1996). Anticancer Res. 16, 3819–3825. Gilbertson, R. J., Perry, R. H., Kelly, P. J., Pearson, A. D., and Lunec, J. (1997). Cancer Res. 57, 3272–3280. Giri, D. K., Wadhwa, S. N., Upadhaya, S. N., and Talwar, G. P. (1993). Prostate 23, 329–336. Goldman, R., Ben-Levy, R., Peles, E., and Yarden, Y. (1990). Biochemistry 29, 11024–11028. Gordinier, M. E., Steinhoff, M. M., Hogan, J. W., Peipert, J. F., Gajewski, W. H., Falkenberry, S. S., and Granai, C. O. (1997). Gynecol. Oncol. 67, 200–202. Goren, D., Horowitz, A. T., Zalipsky, S., Woodle, M. C., Yarden, Y., and Gabizon, A. (1996). Br. J. Cancer 74, 1749–1756. Grasso, A. W., Wen, D., Miller, C. M., Rhim, J. S., Pretlow, T. G., and Kung, H. J. (1997). Oncogene 15, 2705–2716. Graus-Porta, D., Beerli, R. R., and Hynes, N. E. (1995). Mol. Cell. Biol. 15, 1182–1191. Graus Porta, D., Beerli, R. R., Daly, J. M., and Hynes, N. E. (1997). EMBO J. 16, 1647–1655. Groenen, L. C., Nice, E. C., and Burgess, A. W. (1994). Growth Factors 11, 235–257. Groner, B., Wick, B., Jeschke, M., Fiebig, H. H., Dengler, W., Runau, T., Mihatsch, M., Kahl, R., Schmidt, M., Wels, W., and Stocklin, E. (1997). Int. J. Cancer 70, 682–687. Gu, K., Mes Masson, A. M., Gauthier, J., and Saad, F. (1996). Cancer Lett. 99, 185–189. Gullick, W. J. (1990). Int. J. Cancer, Suppl. 5, 55–61. Gullick, W. J. (1994). Eur. J. Cancer 30A, 2186. Gullick, W. J., Love, S. B., Wright, C., Barnes, D. M., Gusterson, B., Harris, A. L., and Altman, D. G. (1991). Br. J. Cancer 63, 434–438. Gusterson, B. A., Machin, L. G., Gullick, W. J., Gibbs, N. M., Powles, T. J., Elliott, C., Ashley, S., Monaghan, P., and Harrison, S. (1988). Br. J. Cancer 58, 453–457. Gusterson, B. A., Gelber, R. D., Goldhirsch, A., Price, K. N., Save Soderborgh, J., Anbazhagan, R., Styles, J., Rudenstam, C. M., Golouh, R., Reed, R., Martinez-Tello, F., Tiltman, A., Torhorst, J., Grigolato, P., Bettelheim, R., Neville, A. M., Burki, K., Castiglione, M., Collins, J., and Senn, H. J. (1992). J. Clin. Oncol. 10, 1049–1056. Guy, P. M., Platko, J. V., Cantley, L. C., Cerione, R. A., and Carraway, K. L. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 8132–8136.

68

Leah N. Klapper et al.

Haffty, B. G., Brown, F., Carter, D., and Flynn, S. (1996). Int. J. Radiat. Oncol. Biol. Phys. 35, 751–757. Hamel, N. W., Sebo, T. J., Wilson, T. O., Keeney, G. L., Roche, P. C., Suman, V. J., Hu, T. C., and Podratz, K. C. (1996). Gynecol. Oncol. 62, 192–198. Han, S. Yun, I. J., Noh, D. Y., Choe, K. J., Song, S. Y., and Chi, J. G. (1997). J. Surg. Oncol. 65, 22–27. Hancock, M. C., Langton, B. C., Chan, T., Toy, P., Monahan, J. J., Mischak, R. P., and Shawver, L. K. (1991). Cancer Res. 51, 4575–4580. Harlozinska, A., Bar, J. K., Sobanska, E., and Goluda, M. (1997). Anticancer Res. 17, 3545–3552. Harpole, D. H., Jr., Herndon, J. E. N., Wolfe, W. G., Iglehart, J. D., and Marks, J. R. (1995). Cancer Res. 55, 51–56. Harris, A. L., Nicholson, S., Sainsbury, R., Wright, C., and Farndon, J. (1992). J. Natl. Cancer Inst. Monogr. 11, 181–187. Haugen, D. R., Akslen, L. A., Varhaug, J. E., and Lillehaug, J. R. (1996). Cancer Res. 56, 1184–1188. Herrera, G. A. (1991). Kidney Int. 40, 509–513. Higashiyama, S., Abraham, J. A., Miller, J., Fiddes, J. C., and Klagsbrun, M. (1991). Science 251, 936–939. Hill, R. J., and Sternberg, P. W. (1992). Nature (London) 358, 470–476. Holmes, W. E., Sliwkowski, M. X., Akita, R. W., Henzel, W. J., Lee, J., Park, J. W., Yansura, D., Abadi, N., Raab, H., Lewis, G. D., Shepard, M., Wood, W. I., Goeddel, D. V., and Vandlen, R. L. (1992). Science 256, 1205–1210. Horan, T., Wen, J., Arakawa, T., Liu, N., Brankow, D., Hu, S., Ratzkin, B., and Philo, J. S. (1995). J. Biol. Chem. 270, 24604–24608. Hsu, J. C., and Perrimon, N. 1994). Genes Dev. 8, 2176–2187. Hudziak, R. M., Schlessinger, J., and Ullrich, A. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 7159–7163. Hudziak, R. M., Lewis, G. D., Winget, M., Fendly, B. M., Shepard, H. M., and Ullrich, A. (1989). Mol. Cell. Biol. 9, 1165–1172. Huettner, P. C., Carney, W. P., Naber, S. P., DeLellis, R. A., Membrino, W., and Wolfe, H. J. (1992). Mod. Pathol. 5, 250–256. Hung, M. C., Zhang, X., Yan, D. H., Zhang, H. Z., He, G. P., Zhang, T. Q., and Shi, D. R. (1992). Cancer Lett. 61, 95–103. Hurwitz, E., Stancovski, I., Sela, M., and Yarden, Y. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 3353–3357. Hynes, N. E., and Stern, D. F. (1994). Biochim. Biophys. Acta 1198, 165–184. Iglehart, J. D., Kerns, B. J., Huper, G., and Marks, J. R. (1995). Breast Cancer Res. Treat. 34, 253–263. Ihara, K., Shiozaki, H., Tahata, K., Koyabashi, M., Inoue, S., Tamura, M., Miyata, H., Oka, Y., Doki, T., and Mori, T. (1993). Cancer (Philadelphia) 71, 2902–2909. Ioannides, C., Ioannides, M., and O’Brian, C. (1992). Mol. Carcinog. 6, 77–82. Irish, J. C., and Bernstein, A. (1993). Laryngoscope 103, 42–52. Jacobsen, N. E., Abadi, N., Sliwkowski, M. X., Reilly, D., Skelton, N. J., and Fairbrother, W. J. (1996). Biochemistry 35, 3402–3417. Janes, P. W., Lackmann, M., Church, W. B., Sanderson, G. M., Sutherland, R. L., and Daly, R. J. (1997). J. Biol. Chem. 272, 8490–8497. Jaros, E., Perry, R. H., Adam, L., Kelly, P. J., Crawford, P. J., Kalbag, R. M., Mendelow, A. D., Sengupta, R. P., and Pearson, A. D. (1992). Br. J. Cancer 66, 373–385. Juhl, H., Downing, S. G., Wellstein, A., and Czubayko, F. (1997). J. Biol. Chem. 272, 29482–29486.

ErbB/HER Signaling Network of Growth Factor Receptors

69

Kandl, H., Seymour, L., and Bezwoda, W. R. (1994). Br. J. Cancer 70, 739–742. Kapitanovic, S., Radosevic, S., Kapitanovic, M., Andelinovic, S., Ferencic, Z., Tavassoli, M., Primorac, D., Sonicki, Z., Spaventi, S., Pavelic, K., and Spaventi, R. (1997). Gastroenterology 112, 1103–1113. Karunagaran, D., Tzahar, E., Liu, N., Wen, D., and Yarden, Y. (1995). J. Biol. Chem. 270, 9982–9990. Karunagaran, D., Tzahar, E., Beerli, R. R., Chen, X., Graus-Porta, D., Ratzkin, B. J., Seger, R., Hynes, N. E., and Yarden, Y. (1996). EMBO J. 15, 254–264. Katsaros, D., Theillet, C., Zola, P., Louason, G., Sanﬁlippo, B., Isaia, E., Arisio, R., Giardina, G., and Sismondi, P. (1995). Anticancer Res. 15, 1501–1510. Katsuura, M., and Tanaka, S. (1989). J. Biochem. (Tokyo) 106, 87–92. Katsumata, M., Okudaira, T., Samanta, A., Clark, D. P., Drebin, J. A., Jolicoeur, P., and Greene, M. I. (1995). Nat. Med. 1, 644–648. Kerlikowske, K., Barclay, J., Grady, D., Sickles, E. A., and Ernster, V. (1997). J. Natl. Cancer Inst. 89, 76–82. Kim, Y. J., Ghu, H. D., Kim, D. Y., Kim, H. J., Kim, S. K., and Park, C. S. (1993). J. Surg. Oncol. 54, 167–170. King, B. L., Carter, D., Foellmer, H. G., and Kacinski, B. M. (1992). Am. J. Pathol. 140, 23–31. King, C. R., Kraus, M. H., and Aaronson, S. A. (1985). Science 229, 974–976. King, C. R., Borrello, I., Bellot, F., Comoglio, P., and Schlessinger, J. (1988). EMBO J. 7, 1647–1651. Kinzler, W. K., and Vogelstein, B. (1996). Cell (Cambridge, Mass.) 87, 159–170. Kita, Y., Tseng, J., Horan, T., Wen, J., Philo, J., Chang, D., Ratzkin, B., Paciﬁci, R., Brankow, D., Hu, S., Luo, Y., Wen, D., Arakawa, T., and Nicolson, M. (1996). Biochem. Biophys. Res. Commun. 226, 59–69. Kita, Y. A., Barff, J., Luo, Y., Wen, D., Brankow, D., Hu, S., Liu, N., Prigent, S. A., Gullick, W. J., and Nicolson, M. (1994). FEBS Lett. 349, 139–143. Klapper, L. N., Vaisman, N., Hurwitz, E., Pinkas Kramarski, R., Yarden, Y., and Sela, M. (1997). Oncogene 14, 2099–2109. Klapper, L. N., Glathe, S., Vaisman, N., Hynes, N. E., Andrews, G. C., Sela, M., and Yarden, Y. (1999). Proc. Natl. Acad. Sci. U.S.A. 96, 4995–5000. Klohs, W. D., Fry, D. W., and Kraker, A. J. (1997). Curr. Opin. Oncol. 9, 562–568. Kohlberger, P., Loesch, A., Koelbl, H., Breitenecker, G., Kainz, C., and Gitsch, G. (1996). Cancer Lett. 98, 151–155. Kohler, M., Janz, I., Wintzer, H.-O., Wagner, E., and Bauknecht, T. (1989). Anticancer Res. 9, 1537–1548. Kokai, Y., Cohen, J. A., Drebin, J. A., and Greene, M. I. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 8498–8501. Kokai, Y., Myers, J. N., Wada, T., Brown, V. I., LeVea, C. M., Davis, J. G., Dobashi, K., and Greene, M. I. (1989). Cell (Cambridge, Mass.) 58, 287–292. Komurasaki, T., Toyoda, H., Uchida, D., and Morimoto, S. (1997). Oncogene 15, 2841–2848. Kondapaka, B. S., and Reddy, K. B. (1996). Mol. Cell. Endocrinol. 117, 53–58. Korabiowska, M., Mirecka, J., Brinck, U., Hoefer, K., Marx, D., and Schauer, A. (1996). Anticancer Res. 16, 471–474. Korkolopoulou, P., Christodoulou, P., Kapralos, P. ,Exarchakos, M., Bisbiroula, A., Hadjiyannakis, M., Georgountzos, C., and Thomas-Tsagli, E. (1997). Pathol. Res. Pract. 193, 767–775. Kowalski, L. D., Kanbour, A. I., Price, F. V., Finkelstein, S. D., Christopherson, W. A., Seski, J. C., Naus, G. J., Burhnam, J. A., Kanbour Shakir, A., and Edwards, R. P. (1997). Cancer (Philadelphia) 79, 1587–1594.

70

Leah N. Klapper et al.

Krainer, M., Brodowicz, T., Zeillinger, R., Wiltschke, C., Scholten, C., Seifert, M., Kubista, E., and Zielinski, C. C. (1997). Oncology 54, 475–481. Kramer, R., Bucay, N., Kane, D. J., Martin, L. E., Tarpley, J. E., and Theill, L. E. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 4833–4838. Krane, I. M., and Leder, P. (1996). Oncogene 12, 1781–1788. Kraus, M. H., Popescu, N. C., Amsbaugh, S. C., and King, R. C. (1987). EMBO J. 6, 605–610. Kraus, M. H., Issing, M., Popescu, N. C., and Aaronson, S. A. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 9193–9197. Kristensen, G. B., Holm, R., Abeler, V. M., and Trope, C. G. (1996). Cancer (Philadelphia) 78, 433–440. Kuhn, E. J., Kurnot, R. A., Sesterhenn, I. A., Chang, E. H., and Moul, J. W. (1993). J. Urol. 150, 1427–1433. Kury, F., Sliutz, G., Schemper, M., Reiner, G., Reiner, A., Jakesz, R., Wrba, F., Zeillinger, R., Knogler, W., Huber, J., Holzner, H., and Spona, J. (1990). Eur. J. Cancer 26, 946–949. Langton, B. C., Crenshaw, M. C., Chao, L. A., Stuart, S. G., Akita, R. W., and Jackson, J. E. (1991). Cancer Res. 51, 2593–2598. Layﬁeld, L. J., Thompson, J. K., Dodge, R. K., and Kerns, B. J. (1995). J. Surg. Oncol. 59, 21–27. Lee, K. F., Simon, H., Chen, H., Bates, B., Hung, M. C., and Hauser, C. (1995). Nature (London) 378, 394–398. Leitzel, K., Teramoto, Y., Konrad, K., Chinchilli, V. M., Volas, G., Grossberg, H., Harvey, H., Demers, L., and Lipton, A. (1995). J. Clin. Oncol. 13, 1129–1135. Lemoine, N. R., Barnes, D. M., Hollywood, D. P., Hughes, C. M., Smith, P., Dublin, E., Prigent, S. A., Gullick, W. J., and Hurst, H. C. (1992a). Br. J. Cancer 66, 1116–1121. Lemoine, N. R., Lobresco, M., Leung, H., Barton, C., Hughes, C. M., Prigent, S. A., Gullick, W. J., and Kloppel, G. (1992b). J. Pathol. 168, 269–273. Lenferink, A. E. G., Pinkas-Kramarski, R., van de Poll, M. L. M., van Vugt, M. J. H., Klapper, L. N., Tzahar, E., Waterman, H., Sela, M., van Zoelen, E. J. J., and Yarden, Y. (1998). EMBO J. 17, 3385–3397. Levkowitz, G., Klapper, L. N., Tzahar, E. Freywald, A., Sela, M., and Yarden, Y. (1996). Oncogene 12, 1117–1125. Levkowitz, G., Waterman, H., Zamir, E., Kam, Z., Oved, S., Langdon, W. Y., Beguinot, L., Geiger, B., and Yarden, Y. (1998). Genes Dev. 12, 3663–3674. Lewis, G. D., Lofgren, J. A., McMurtrey, A. E., Nuijens, A., Fendly, B. M., Bauer, K. D., and Sliwkowski, M. X. (1996). Cancer Res. 56, 1457–1465. Libermann, T. A., Nusbaum, H. R., Razon, N., Kris, R., Lax, I., Soreq, H., Whittle, N., Waterﬁeld, M. D., Ullrich, A., and Schlessinger, J. (1985). Nature (London) 313, 144–147. Lin, Y. Z., and Clinton, G. M. (1991). Oncogene 6, 639–643. Lipponen, H. J., Aaltomaa, S. Syrjanen, S., and Syrjanen, K. (1993). Anticancer Res. 13, 1147–1152. Lipponen, P., Eskelinen, M., Syrjanen, S., Tervahauta, A., and Syrjanen, K. (1991). Eur. Urol. 20, 238–242. Livneh, E., Glazer, L., Segal, D., Schlessinger, J., and Shilo, B. Z. (1985). Cell (Cambridge, Mass.) 40, 599–607. Lodato, R. F., Maguire, H. C., Jr., Greene, M. I., Weiner, D. B., and LiVolsi, V. A. (1990). Mod. Pathol. 3, 449–454. Lonardo, F., Di Marco, E., King, C. R., Pierce, J. H., Segatto, O., Aaronson, S. A., and Di Fiore, P. P. (1990). New Biol. 2, 992–1003. Lonn, U., Lonn, S., Nylen, U., Friberg, S., and Stenkvist, B. (1993). Cancer (Philadelphia) 71, 3605–3610. Lonn, U., Lonn, S., Nilsson, B., and Stenkvist, B. (1994). Breast Cancer Res. Treat. 29, 237–245.

ErbB/HER Signaling Network of Growth Factor Receptors

71

Lonn, U., Lonn, S., Nilsson, B., and Stenkvist, B. (1995). Cancer (Philadelphia) 75, 2681–2687. Luetteke, N. C., Qiu, T. H., Peiffer, R. L., Oliver, P., Smithies, O., and Lee, D. C. (1993). Cell (Cambridge, Mass.) 73, 263–278. Lukes, A. S., Kohler, M. F., Pieper, C. F., Kerns, B. J., Bentley, R., Rodriguez, G. C., Soper, J. T., Clarke Pearson, D. L., Bast, R. C., Jr., and Berchuck, A. (1994). Cancer (Philadelphia) 73, 2380–2385. Lupu, R., Colomer, R., Zugmaier, G., Sarup, J., Shepard, M., Slamon, D., and Lippman, M. (1990). Science 249, 1552–1555. Lupu, R., Colomer, R., Kannan, B., and Lippman, M. E. (1992). Proc. Natl. Acad. Sci. U.S.A. 89, 2287–2291. Luttrell, D. K., Lee, A., Lansing, T. J., Crosby, R. M., Jung, K. D., Willard, D., Luther, M., Rodriguez, M., Berman, J., and Gilmer, T. M. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 83–87. Lyne, J. C., Melhem, M. F., Finley, G. G., Wen, D., Liu, N., Deng, D. H., and Salup, R. (1997). Cancer J. Sci. Am. 3, 21–30. Mabrouk, G. M., Helal, S. A., El-Lamie, K. I., and Khalifa, A. (1996). Clin. Chem. (WinstonSalem, N.C.) 42, 981–982. MacGrogan, G., Mauriac, L., Durand, M., Bonichon, F., Trojani, M., de Mascarel, I., and Coindre, J. M. (1996). Br. J. Cancer 74, 1458–1465. Mack, L., Kerkvliet, N., Doig, G., and O’Malley, F. P. (1997). Hum. Pathol. 28, 974–979. Mann, G., Fowler, K., Gabriel, A., Nice, E., Williams, R., and Dunn, A. (1993). Cell (Cambridge, Mass.) 73, 249–261. Mansour, O. A., Zekri, A. R., Harvey, J., Teramoto, Y., and el-Ahmady, O. (1997). Anticancer Res. 17, 3101–3106. Marchionni, M. A., Goodearl, A. D. J., Chen, M. S., Bermingham-McDonogh, O., Kirk, C., Hendricks, M., Denehy, F., Misumi, D., Sudhalter, J., Kobayashi, K., Wroblewski, D., Lynch, C., Baldassare, M., Hiles, I., Davis, J. B., Hsuan, J. J., Totty, N. F., Otsu, M., McBury, R. N., Waterﬁeld, M. D., Stroobant, P., and Gwynne, D. (1993). Nature (London) 362, 312–318. Marquardt, H., Hunkapiller, M. W., Hood, L. E., and Todaro, G. J. (1984). Science 223, 1079–1082. Marshall, C. J. (1995). Cell (Cambridge, Mass.) 80, 179–185. McFadden, G., Graham, K., and Barry, M. (1996). Tranplant. Proc. 28, 2085–2088. Meden, H., Marx, D., Rath, W., Kron, M., Fattahi Meibodi, A., Hinney, B., Kuhn, W., and Schauer, A. (1994). Int. J. Gynecol. Pathol. 13, 45–53. Meden, H., Marx, D., Schauer, A., Wuttke, W., and Kuhn, W. (1997). Anticancer Res. 17, 757–760. Medl, M., Sevelda, P., Czerwenka, K., Dobianer, K., Hanak, H., Hruza, C., Klein, M., Leodolter, S., Mullauer Ertl, S., Rosen, A., Salzer, H., Vaura, N., and Spona, J. (1995). Gynecol. Oncol. 59, 321–326. Memon, M. A., and Donohue, J. H. (1997). Br. J. Surg. 84, 433–435. Meyer, D., and Birchmeier, C. (1995). Nature (London) 378, 386–390. Meyer, D., Yamaai, T., Garratt, A., Reithmacher-Sonnenberg, E., Kane, D., Theill, L., and Birchmeier, C. (1997). Development (Cambridge, U.K.) 124, 3575–3586. Midulla, C., Giovagnoli, M. R., Valli, C., and Vecchione, A. (1995). Anal. Quant. Cytol. Histol. 17, 157–162. Miettinen, P., Berger, J., Meneses, J., Phung, Y., Pederson, R., Werb, Z., and Derynck, R. (1995). Nature (London) 376, 337–341. Miller, S. J., Suen, T.-C., Sexton, T. B., and Hung, M. C. (1994). Int. J. Oncol. 4, 599–608. Mitra, A. B., Murty, V. V., Pratap, M. Sodhani, P., and Chaganti, R. S. (1994). Cancer Res. 54, 637–639. Mixan, B., Cohen, B. D., Bacus, S. S. Fell, H. P., and Siegall, C. B. (1998). Oncogene 16, 1209–1215.

72

Leah N. Klapper et al.

Modjtahedi, H., Hickish, T., Nicolson, M., Moore, J., Styles, J., Eccles, S., Jackson, E., Salter, J., Sloane, J., Spencer, L., Priest, K., Smith, I., Dean, C., and Gore, M. (1996). Br. J. Cancer 73, 228–235. Molina, R., Jo, J., Filella, X., Zanon, G., Pahisa, J., Munoz, M., Farrus, B., Latre, M. L., Gimenez, N., Hage, M., Estape, J., and Ballesta, A., M. (1996). Anticancer Res. 16, 2295–2300. Moreno, A., Lloveras, B., Figueras, A., Escobedo, A., Ramon, J. M., Sierra, A., and Fabra, A. (1997). Mod. Pathol. 10, 1088–1092. Moritz, D., Wels, W., Mattern, J., and Groner, B. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 4318–4322. Muller, W., Artega, C., Muthuswamy, S., Siegel, P., Webster, M., Cardiff, R., Meise, K., Li, F., Halter, S., and Coffey, R. (1996). Mol. Cell. Biol. 16, 5726–5736. Murali, R., Brennan, P., Keiber-Emmons, T., and Green, M. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 6252–6257. Muss, H. B., Thor, A. D., Berry, D. A., Kute, T., Liu, E. T., Koerner, F., Cirrincione, C. T., Budman, D. R., Wood, W. C., Barcos, M., and Henderson, I. C. (1994). N. Engl. J. Med. 330, 1260–1266. Myers, R. B. Srivastava, S., Oelschlager, D. K., and Grizzle, W. E. (1994). J. Natl. Cancer Inst. 86, 1140–1145. Nagata, Y., Furugen, R., Hiasa, A., Ikeda, H., Ohta, N., Furukawa, K., Nakamura, H., Furukawa, K., Kanematsu, T., and Shiku, H. (1997). J. Immunol. 159, 1336–1343. Naidu, R., Yadav, M., Nair, S., and Kutty, K. K., (1998). Anticancer Res. 18, 65–70. Nakano, T., Oka, K., Ishikawa, A., and Morita, S. (1997). Cancer (Philadelphia) 79, 513–520. Nakano, T., Oka, K., Ishikawa, A., and Morita, S. (1998). Cancer Detect. Prev. 22, 120–128. Natali, P. G., Nicotra, M. R., Bigotti, A., Venturo, I., Slamon, D. J., Fendly, B. M., and Ullrich, A. (1990). Int. J. Cancer 45, 457–461. Nazeer, T., Ballouk, F., Malfetano, J. H., Figge, H., and Ambros, R. A. (1995). Am. J. Obstet. Gynecol. 173, 1829–1834. Ndubisi, B., Sanz, S., Lu, L., Podczaski, E. Benrubi, G., and Masood, S. (1997). Ann. Clin. Lab. Sci. 27, 396–401. Nemunaitis, J., Klemow, S., Tong, A., Courtney, A., Johnston, W., Mack, M., Taylor, W., Solano, M., Stone, M., Mallams, J., and Mues, G. (1998). Am. J. Clin. Oncol. 21, 155–160. Neuman-Silberberg, F. S., and Schupbach, T. (1993). Cell (Cambridge, Mass.) 75, 165–174. Nguyen, P. L., Swanson, P. E., Jaszcz, W., Aeppli, D. M., Zhang, G., Singleton, T. P., Ward, S., Dykoski, D., Harvey, J., and Niehans, G. A. (1994). Am. J. Clin. Pathol. 101, 166–176. Nicholson, S., Richard, J., Sainsbury, C., Halcrow, P., Kelly, P., Angus, B., Wright, C., Henry, J., Farndon, J. R., and Harris, A. L. (1991). Br. J. Cancer 63, 146–150. Niehans, G. A., Singleton, T. P. Dykoski, D., and Kiang, D. T. (1993). J. Natl. Cancer Inst. 85, 1230–1235. Nishikawa, R., Ji, X. D., Harmon, R. C., Lazar, C. S., Gill, G. N., Cavenee, W. K., and Huang, H. J. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 7727–7731. Niskanen, E. Blomqvist, C., Franssila, K., Hietanen, P., and Wasenius, V. M. (1997). Br. J. Cancer 76, 917–922. Noguchi, M., Koyasaki, N., Ohta, N., Kitagawa, H. Earashi, M., Thomas, M., Miyazaki, I., and Mizukami, Y. (1993). arch. Surg. (Chicago) 128, 242–246. Nowell, P. (1976). Science 194, 23–28. O’Malley, F. P., Saad, Z., Kerkvliet, N., Doig, G., Stitt, L., Ainsworth, P., Hundal, H., Chambers, A. F., Turnbull, D. I., and Bramwell, V. (1996). Hum. Pathol. 27, 955–963. Ooi, A., Kobayashi, M., Mai, M., and Nakanishi, I. (1998). Lab Invest. 78, 345–351. Osherov, N., Gazit, A., Gilon, C., and Levitzki, A. (1993). J. Biol. Chem. 268, 11134– 11142.

ErbB/HER Signaling Network of Growth Factor Receptors

73

Padhy, L., Shih, C., Cowing, D., Finkelstein, R., and Weinberg, R. (1982). Cell (Cambridge, Mass.) 28, 865–871. Park, J. W., Hong, K., Carter, P., Asgari, H., Guo, L. Y., Keller, G. A., Wirth, C., Shalaby, R., Kotts, C., Wood, W. I., Papahadjopoulos, D., and Benz, C. C. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 1327–1331. Pasleau, F., Grooteclaes, M., and Gol-Winkler, R. (1993). Oncogene 8, 849–854. Pastan, I., and FitzGerald, D. (1991). Science 254, 1173–1177. Pastorino, U., Andreola, S., Tagliabue, E., Pezzella, F., Incarbone, M., Sozzi, G., Buyse, M., Menard, S., Pierotti, M., and Rilke, F. (1997). J. Clin. Oncol. 15, 2858–2865. Paterson, M. C., Dietrich, K. D., Danyluk, J., Paterson, A. H., Lees, A. W., Jamil, N., Hanson, J., Jenkins, H., Krause, B. E., McBlain, W. A., et al. (1991). Cancer Res. 51, 556– 567. Pavelic, K., Banjac, Z., Pavelic, J., and Spaventi, S. (1993). Anticancer Res. 13, 1133–1137. Pegram, M. D., Finn, R. S., Arzoo, K., Beryt, M., Pietras, R. J., and Slamon, D. J. (1997). Oncogene 15, 537–547. Peiper, M., Goedegebuure, P. S., Linehan, D. C., Ganguly, E., Douville, C. C., and Eberlein, T. J. (1997). Eur. J. Immunol. 27, 1115–1123. Peles, E., Ben-Levy, R., Or, E., Ullrich, A., and Yarden, Y. (1991). EMBO J. 10, 2077–2086. Peles, E., Bacus, S. S., Koski, R. A., Lu, H. S. Wen, D., Ogden, S. G., Levy, R. B., and Yarden, Y. (1992). Cell (Cambridge, Mass.) 69, 205–216. Peles, E., Ben Levy, R., Tzahar, E., Liu, N., Wen, D., and Yarden, Y. (1993). EMBO J. 12, 961–971. Pelicci, G. Lanfrancone, L., Grignani, F., McGlade, J., Cavallo, F., Forni, G., Nicoletti, I., Grignani, F., Pawson, T., and Pelicci, P. G. (1992). Cell (Cambridge, Mass.) 70, 93–104. Penar, P. L., Khoshyomn, S., Bhushan, A., andTritton, T. R. (1997). Neurosurgery 40, 141–151. Perez Soler, R., Donato, N. J., Shin, D. M., Rosenblum, M. G., Zhang, H. Z., Tornos, C., Brewer, H., Chan, J. C., Lee, J. S., Hong, W. K., and Murray, J. L. (1994). J. Clin. Oncol. 12, 730–739. Pierce, J. H., Arnstein, P., Di Marco, E., Artrip, J., Kraus, M. H., Lonardo, F., Di Fiore, P. P., and Aaronson, S. A. (1991). Oncogene 6, 1189–1194. Pietras, R. J., Fendly, B. M., Chazin, V. R., Pegram, M. D., Howell, S. B., and Slamon, D. J. (1994). Oncogene 9, 1829–1838. Pinion, S. B., Kennedy, J. H., Miller, R. W., and MacLean, A. B. (1991). Lancet 337, 819–820. Pinkas-Kramarski, R., Shelly, M., Glathe, S., Ratzkin, B. J., and Yarden, Y. (1996a). J. Biol. Chem. 271, 19029–19032. Pinkas-Kramarski, R., Soussan, L., Waterman, H., Levkowitz, G., Alroy, I., Klapper, L., Lavi, S., Seger, R., Ratzkin, B. J., Sela, M., and Yarden, Y. (1996b). EMBO J. 15, 2452–2467. Pinkas-Kramarski, R., Eilam, R., Alroy, I., Levkowitz, G., Lonai, P., and Yarden, Y. (1997). Oncogene 15, 2803–2815. Pinkas-Kramarski, R., Lenferink, A. E., Bacus, S. S., Lyass, L., van de Poll, M. L., Klapper, L. N., Tzahar, E., Sela, M., van Zoelen, E. J., and Yarden, Y. (1998). Oncogene 16, 1249–1258. Plowman, G. D., Whitney, G. S., Neubauer, M. G., Green, J. M., McDonald, V. I., Todaro, G. J., and Shoyab, M. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 4905–4909. Plowman, G. D., Culouscou, J. M., Whitney, G. S., Green, J. M., Carlton, G. W., Foy, L., Neubauer, M. G., and shoyab, M. (1993a). Proc. Natl. Acad. Sci. U.S.A. 90, 1746–1750. Plowman, G. D., Green, J. M., Culouscou, J. M., Carlton, G. W., Rothwell, V. M., and Buckley, S. (1993b). Nature (London) 366, 473–475. Porter, P. L., Garcia, R., Moe, R., Corwin, D. J., and Gown, A. M. (1991). Cancer (Philadelphia) 68, 331–334.

74

Leah N. Klapper et al.

Press, M. F., Pike, M. C., Hung, G., Zhou, J. Y., Ma, Y., George, J., Dietz Band, J., James, W., Slamon, D. J., Batsakis, J. G., and El-Naggar, A. K. (1994). Cancer Res. 54, 5675–5682. Prewett, M., Rockwell, P., Rockwell, R. F., Giorgio, N. A., Mendelsohn, J., Scher, H. I., and Goldstein, N. I. (1996). J. Immunother. Emphasis Tumor Immunol. 19, 419–427. Prigent, S. A. Lemoine, N. R., Hughes, C. M., Plowman, G. D., Selden, C., and Gullick, W. J. (1992). Oncogene 7, 1273–1278. Qian, X., O’Rourke, D. M., Zhao, H., and Greene, M. I. (1996). Oncogene 13, 2149 2157. Quinn, C. M. Ostrowski, J. L., Lane, S. A., Loney, D. P., Teasdale, J., and Benson, F. A. (1994). Histopathology 25, 247–252. Ramachandra, S., Machin, L., Ashley, S., Monaghan, P., and Gusterson, B. A. (1990). J. Pathol. 161, 7–14. Regidor, P. A., Callies, R., and Schindler, A. E. (1995). Eur. J. Gynaecol. Oncol. 16, 130– 137. Rewcastle, G. W., Murray, D. K., Elliott, W. L., Fry, D. W., Howard, C. T., Nelson, J. M., Roberts, B. J., Vincent, P. W., Showalter, H. D., Winters, R. T., and Denny, W. A. (1998). J. Med. Chem. 41, 742–751. Rhodes, C. H., Honsinger, C., and Sorenson, G. D. (1994). J. Neuropathol. Exp. Neurol. 53, 364–368. Riben, M. W., Malfetano, J. H., Nazeer, T., Muraca, P. J., Ambros, R. A., and Ross, J. S. (1997). Mod. Pathol. 10, 823–831. Richter, A., Conlan, J. W., Ward, M. E., Chamberlin, S. G., Alexander, P, Richards, N. G. J., and Davies, D. E. (1992). Biochemistry 31, 9546–9554. Riese, D. J., van Raaij, T. M., Plowman, G. D., Andrews, G. C., and Stern, D. F. (1995). Mol. Cell. Biol. 15, 5770–5776. Riese, D. J., Bermingham, Y., van Raaij, T. M., Buckley, S., Plowman, G. D., and Stern, D. F. (1996). Oncogene 12, 345–353. Riese, D. J., Komurasaki, T., Plowman, G. D., and Stern, D. F. (1998). J. Biol. Chem. 273, 11288–11294. Riethmacher, D., Sonnenberg-Riethmacher, E., Brinkmann, V., Yamaai, T., Lewin, G. R., and Birchmeier, C. (1997). Nature (London) 389, 725–730. Rilke, F., Colnaghi, M. I., Cascinelli, N., Andreola, S. Baldini, M. T., Bufalino, R., Della Porta, G., Menard, S., Pierotti, M. A., and Testori, A. (1991). Int. J. Cancer 49, 44–49. Ring, C. J., Blouin, P., Martin, L. A., Hurst, H. C., and Lemoine, N. R. (1997). Gene Ther. 4, 1045–1052. Rodriguez, G. C., Boente, M. P., Berchuck, A., Whitaker, R. S., O’Briant, K. C., Xu, F., and Bast, R. C., Jr. (1993). Am. J. Obstet. Gynecol. 168, 228–232. Roland, P. Y., Stoler, M. H., Broker, T. R., and Chow, L. T. (1997). Am. J. Obstet. Gynecol. 177, 133–138. Ross, J. S., Sheehan, C., Hayner Buchan, A. M., Ambros, R. A., Kallakury, B. V., Kaufman, R., Fisher, H. A., and Muraca, P. J. (1997a). Hum. Pathol. 28, 827–833. Ross, J. S., Sheehan, C. E., Hayner Buchan, A. M., Ambros, R. A., Kallakury, B. V., Kaufman, R. P., Jr., Fisher, H. A., Rifkin, M. D., and Muraca, P. J. (1997b). Cancer (Philadelphia) 79, 2162–2170. Rozan, S., Vincent Salomon, A., Zafrani, B., Validire, P., De Cremoux, P., Benoux, A., Nieruchalski, M., Fourquet, A., Clough, K., Dieras, V., Pouillart, P., and Sastre Garau, X. (1998). Int. J. Cancer 79, 27–33. Rubin, S. C., Finstad, C. L., Wong, G. Y., Almandrones, L., Plante, M., and Lloyd, K. O. (1993). Am. J. Obstet. Gynecol. 168, 162–169. Rubin, S. C., Finstad, C. L., Federici, M. G., Scheiner, L., Lloyd, K. O., and Hoskins, W. J. (1994). Cancer (Philadelphia) 73, 1456–1459.

ErbB/HER Signaling Network of Growth Factor Receptors

75

Rutledge, B. J., Zhang, K., Bier, E., Jan, Y. N., and Perrimon, N. (1992). Genes Dev. 6, 1503–1517. Sadasivan, R., Morgan, R., Jennings, S., Austenfeld, M., Van Veldhuizen, P., Stephens, R., and Noble, M. (1993). J. Urol. 150, 126–131. Saeki, T., Salomon, D. S., Johnson, G. R., Gullick, W. J., Mandai, K., Yamagami, K., Moriwaki, S. Tanada, M., Takashima, S., and Tahara, E. (1995). Jpn. J. Clin Oncol. 25, 240–249. Salomon, D. S., Brandt, R. Ciardiello, F., and Normanno, N. (1995a). Crit. Rev. Oncol. Hematol. 19, 183–232. Salomon, D. S., Normanno, N., Cirdiello, F., Brandt, R., Shoyab, M., and Todaro, G. J. (1995b). Breast Cancer Res. Treat. 33, 103–114. Samanta, A. LeVea, C. M., Dougall, W. C., Qian, X., and Greene, M. I. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 1711–1715. Sanidas, E. E., Filipe, M. I., Linehan, J., Lemoine, N. R., Gullick, W. J., Rajkumar, T., and Levison, D. A. (1993). Int. J. Cancer 54, 935–940. Sasada, R., Ono, Y., Taniyama, Y., Sing, Y., Folkman, J., and Igrashi, K. (1993). Biochem. Biophys. Res. Commun. 190, 1173–1179. Sasaoka, T., Langlois, W. J., Bai, F., Rose, D. W., Leitner, J. W., Decker, S. J., Saltiel, A. R., Gill, G. N., Koyabashi, M., Draznin, B., and Olefsky, J. M. (1996). J. Biol. Chem. 271, 8338–8344. Sauer, R., Schauer, A., Rauschecker, H. F., Schumacher, M., Gatzemeier, W., Schmoor, C., DUnst, J. Seegenschmiedt, M. H., and Marx, D. (1992). Int. J. Radiat. Oncol. Biol. Phys. 23, 907–914. Schechter, A. L., Stern, D. F., Vaidyanathan, L., Decker, S. J., Drebin, J. A., Greene, M. I., and Weinberg, R. A. (1984). Nature (London) 312, 513–516. Schimmelpenning, H., Eriksson, E. T., Falkmer, U. G., Azavedo, E., Svane, G., and Auer, G. U. (1992). Virchows Arch. A: Pathol. Anat. Histopathol. 420, 433–440. Schlegel, J., Trenkle, T., Stumm, G., and Kiessling, M. (1997). Int. J. Cancer 70, 78–83. Schmidt, M., Hynes, N. E., Groner, B., and Wels, W. (1996). Int. J. Cancer 65, 538–546. Schnepp, B. Grumbling, G., Donaldson, T., and Simcox, A. (1996). Genes Dev. 15, 2302– 2313. Schwartz, S., Jr., Caceres, C., Morote, J., De Torres, I., Rodriguez-Vallejo, J. M., Gonzalez, J., and Reventos, J. (1998). Int. J. Cancer 76, 464–467. Schwechheimer, K., Lauﬂe, R. M. Schmahl, W., Knodlseder, M., Fischer, H., and Hoﬂer, H. (1994). Hum. Pathol. 25, 772–780. Schweitzer, R., Howes, R., Smith, R., Shilo, B.-Z., and Freeman, M. (1995). Nature (London) 376, 699–702. Scorilas, A., Yotis, J., Gouriotis, D., Keramopoulos, A., Ampela, K., Trangas, T., and Talieri, M. (1993). Anticancer Res. 13, 1895–1900. Scott, M. A., Lagios, M. D., Axelsson, K., Rogers, L. W., Anderson, T. J., and Page, D. L. (1997). Hum. Pathol. 28, 967–973. Segatto, O., Pelicci, G., Giuli, S., Digiesi, G., Di Fiore, P. P., McGlade, J., Pawson, T., and Pelicci, P. G. (1993). Oncogene 8, 2105–2112. Seki, A., Nakamura, K., Kodama, J., Miyagi, Y., Yoshinouchi, M., and Kudo, T. (1998). Eur. J. Gynaecol. Oncol. 19, 90–92. Sepp Lorenzino, L., Eberhard, I., Ma, Z., Cho, C., Serve, H., Liu, F., Rosen, N., and Lupu, R. (1996). Oncogene 12, 1679–1687. Secarz, E. E., Lehmann, P. V., Ametani, A., Benichou, G., Miller, A., and Moudgil, K. (1993). Annu. Rev. Immunol. 11, 729–766. Seshadri, R., Horsfall, D. J., Firgaira, F., McCaul, K., Setlur, V., Chalmers, A. H., Yeo, R. Ingram, D., Dawkins, H., and Hahnel, R. (1994). Int. J. Cancer 56, 61–65. Shelly, M., Pinkas-Kramarski, R., Guarino, B. C., Waterman, H., Wang, L.-M., Lyass, L., Ali-

76

Leah N. Klapper et al.

mandi, M., Juo, A., Bacus, S. S. Pierce, J. H., Andrews, G, C., and Yarden, Y. (1998). J. Biol. Chem. 273, 10496–10505. Shilo, B.-Z., and Raz, E. (1991). Trends Genet. 7, 388–392. Shintani, S. Funayama, T., Yoshihama, Y., Alcalde, R. E., and Matsumura, T. (1995). Cancer Lett 95, 79–83. Shirai, H., Ueno, E., Osaki, M., Tatebe, S., Ito, H., and Kaibara, N. (1995). Anticancer Res. 15, 2889–2894. Shoyab, M., McDonald, V. L., Bradley, J. G., and Todaro, G. J. (1988). Proc. Natl. Acad. Sci. U.S.A. 85, 6528–6532. Si, J., Luo, Z., and Mei, L. (1996). J. Biol. Chem. 271, 19752–19759. Sibilia, M., and Wagner, E. F. (1995). Science 269, 234–238. Siegel, P., and Muller, W. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 8878–8883. Simpson, B. J., Weatherill, J., Miller, E. P., Lessells, A. M., Langdon, S. P., and MIller, W. R. (1995). Br. J. Cancer 71, 758–762. Slamon, D. J., Clark, G. M., Wong, S. G., Levin, W. J., Ullrich, A., and McGuire, W. L. (1987). Science 235, 177–182. Slamon, D. J., Godolphin, W., Jones, L. A., Holt, J. A., Wong, S. G., Keith, D. E., levin, W. J., Stuart, S. G., Udove, J., Ullrich, A., and Press, M. F. (1989). Science 244, 707–712. Sliwkowski, M. X., Schaefer, G., Akita, R. W., Lofgren, J. A. Fitzpatrick, V. D., Nuijens, A., Fendly, B. M., Cerione, R. A., Vandlen, R. L., and Carraway, K. L. (1994). J. Biol. Chem. 269, 14661–14665. Smith, K., Houlbrook, S., Greenall, M., Carmichael, J., and Harris, A. L. (1993). Oncogene 8, 933–938. Smith, S., Smith, C., and Bormann, B. (1996). Nat. Struct. Biol. 3, 252–258. Stancovski, I., Hurwitz, E., Leitner, O., Ullrich, A., Yarden, Y., and Sela, M. (1991). Proc. Natl. Acad. Sci. U.S.A. 88, 8691–8695. Stancovski, I., Schindler, D. G., Waks, T., Yarden, Y., Sela, M., and Eshhar, Z. (1993). J. Immunol. 151, 6577–6582. Stancovski, I., Sela, M., and Yarden, Y. (1994). Cancer Treat. Res. 71, 161–191. Stein, D. S., and Stevens, L. M. (1991). Curr. Opin. Genet. Dev. 1, 247–254. Stern, D. F., Heffernan, P. A., and Weinberg, R. A. (1986). Mol. Cell. Biol. 6, 1729–1740. Stern, D. F., Kamps, M. P., and Cao, H. (1988). Mol. Cell. Biol. 8, 3969–3973. Sternberg, P. W., and Horvitz, R. (1991). Trends Genet. 7, 366–371. Sternberg, P. W., Lesa, G., Lee, J., Katz, W. S. Yoon, C. Clandinin, T. R., Huang, L. S., Chamberlin, H. M., and Jongeward, G. (1995). Mol. Reprod. Dev. 42, 523–528. Stoltoff, S. P., Carraway, K. L., Prigent, S. A., Gullick, W. G., and Cantley, L. C. (1996). Mol. Cell. Biol. 14, 3550–3558. Stroobant, P., Rice, A. P., Gullick, W. J., Cheng, D. J., Kerr, I. M., and Waterﬁeld, M. D. (1985). Cell (Cambridge, Mass.) 42, 383–393. Stumm, G., Eberwein, S., Rostock Wolf, S., Stein, H., Pomer, S., Schlegel, J., and Waldherr, R. (1996). Int. J. Cancer 69, 17–22. Suda, Y., Aizawa, S., Furuta, Y., Yagi, T., Ikawa, Y., Saitoh, K., Yamada, Y., Toyoshima, K., and Yamamoto, T. (1990). EMBO J. 9, 181–190. Summerﬁeld, A. E., Hudnall, A. K., Lukas, T. J., Guyer, C. A., and Staros, J. V. (1996). J. Biol. Chem. 271, 19656–19659. Tandon, A. K., Clark, G. M., Chamness, G. C., Ullrich, A., and McGuire, W. L. (1989). J. Clin. Oncol. 7, 1120–1128. Tanner, B., Kreutz, E., Weikel, W., Meinert, R., Oesch, F., Knapstein, P. G., and Becker, R. (1996). Gynecol. Oncol. 62, 268–277. Tateshi, M., Ishida, T., Mitsudomi, T., Kaneko, S., and Sugimachi, K. (1990). Cancer Res. 50, 7077–7080.

ErbB/HER Signaling Network of Growth Factor Receptors

77

Tervahauta, A., Eskelinen, M., Syrjanen, S., Lipponen, P., Pajarinen, P., and Syrjanen, K. (1991). Anticancer Res. 11, 1677–1681. Tetu, B., and Brisson, J. (1994). Cancer (Philadelphia) 73, 2359–2365. Tetu, B., Brisson, J., Cote, C., Brisson, S., Potvin, D., and Roberge, N. (1993). Int. J. Cancer. 55, 429–435. Tetu, B., Fradet, Y., Allard, P., Veilleux, C., Roberge, N., and Bernard, P. (1996). J. Urol. 155, 1784–1788. Threadgill, D., Dlugosz, A., Hansen, L., Tennenbaum, T., Lichti, U., Yee, D., LaMantia, C., Mourton, T., Herrup, K., and Harris, R. (1995). Science 269, 230–234. Tiwari, R. K., Borgen, P. I., Wong, G. Y., Cordon Cardo, C., and Osborne, M. P. (1992). Anticancer Res. 12, 419–425. Toikkanen, S., Helin, H., Isola, J., and Joensuu, H. (1992). J. Clin. Oncol. 10, 1044–1048. Torre, E. A., Salimbeni, V., and Fulco, R. A. (1997). J. Chemother. 9, 51–55. Toyoda, H., Komursaki, T., Uchida, D. Takayama, Y., Isobe, T., Okuyama, T., and Hanada, K. (1995). J. Biol. Chem. 270, 7495–7500. Travis, A., Pinder, S. E., Robertson, J. F., Bell, J. A., Wencyk, P., Gullick, W. J., Nicholson, R. I., Poller, D. N., Blamey, R. W., Elston, C. W., and Ellis, I. O. (1996). Br. J. Cancer 74, 229–233. Tsai, C. M., Levitzki, A., Wu, L. H., Chang, K. T., Cheng, C. C., Gazit, A., and Perng, R. P. (1996). Cancer Res. 56, 1068–1074. Tsuda, H., Sakamaki, C., Tsugane, S., Fukutomi, T., and Hirohashi, S. (1998). Breast Cancer Res. Treat. 48, 21–32. Tzahar, E., and Yarden, Y. (1998). Biochim. Biophys. Acta 1377, M25–M37. Tzahar, E., Levkowitz, G., Karunagaran, D., Yi, L., Peles, E., Lavi, S., Chang, D., Liu, N., Yayon, A., Wen, D., and Yarden, Y. (1994). J. Biol. Chem. 269, 25226–25233. Tzahar, E., Waterman, H., Chen, X., Levkowitz, G., Karunagaran, D., Lavi, S., Ratzkin, B. J., and Yarden, Y. (1996). Mol. Cell. Biol. 16, 5276–5287. Tzahar, E., Pinkas Kramarski, R., Moyer, J. D., Klapper, L. N., Alroy, I., Levkowitz, G., Shelly, M., Henis, S., Eisenstein, M., Ratzkin, B. J., Sela, M., Andrews, G. C., and Yarden, Y. (1997). EMBO J. 16, 4938–4950. Tzahar, E., Guarino, B. C., Waterman, H., Levkowitz, G., Shelly, M., Pinkas-Kramarski, R., Wang, L.-M., Alimandi, M., Kuo, A., Moyer, J. D., Pierce, J. H., Andrews, G. C., and Yarden, Y. (1998). EMBO J. 17, 5948–5963. Ueno, N. T., Yu, D., and Hung, M. C. (1997). Oncogene 15, 953–960. Upton, C., Macen, J. L., and McFadden, G. (1987). J. Virol. 61, 1271–1275. Vaidya, P., Kawarada, Y., Higashiguchi, T., Yoshida, T., Sakakura, T., and Yatani, R. (1996a). J. Pathol. 178, 140–145. Vaidya, P., Yoshida, T., Sakakura, T., Yatani, R., Noguchi, T., and Kawarada, Y. (1996b). Pancreas 12, 196–201. van der Geer, P., Hunter, T., and Lindberg, R. A. (1994). Annu. Rev. Cell Biol. 10, 251–337. van de Vijver, M., Peterse, J. L., Mooi, W. J., Wisman, P., Lomans, J., Dalesio, O., and Nusse, R. (1988). N. Engl. J. Med. 319, 1239–1245. van Leeuwen, F., van de Vijver, M. J., Lomans, J., van Deemter, L., Jenster, G., Akiyama, T., Yamamoti, T., and Nusse, R. (1990). Oncogene 5, 497–503. van-Ojik, H. H., Repp, R., Groenewegen, G., Valerius, T., and van-de-Winkel, J. G. (1997). Cancer Immunol. Immunother. 45, 207–209. Veale, D., Ashcroft, T., Marsh, C., Gibson, G. J., and Harris, A. L. (1987). Br. J. Cancer 55, 513–516. Veltri, R. W., Partin, A. W., Epstein, J. E., Marley, G. M., Miller, C. M., Singer, D. S., Patton, K. P., Criley, S. R., and Coffey, D. S. (1994). J. Cell. Biochem., Suppl. 19, 249– 258.

78

Leah N. Klapper et al.

Visakorpi, T., Kallioniemi, O. P., Koivula, T., Harvey, J., and Isola, J. (1992). Mod. Pathol. 5, 643–648. Vogel, W., Kath, R., Kosmehl, H., Olschowsky, E., and Hoffken, K. (1996). J. Cancer Res. Clin. Oncol. 122, 118–121. Vollmer, R. T., Humphrey, P. A., Swanson, P. E., Wick, M. R., and Hudson, M. L. (1998). Cancer (Philadelphia) 82, 715–723. Volm, M., Efferth, T., and Mattern, J. (1992). Anticancer Res. 12, 11–20. Wada, T., Myers, J. N., Kokai, Y., Brown, V. I., Hamuro, J., LeVea, C. M., and Greene, M. I. (1990a). Oncogene 5, 489–495. Wada, T., Qian, X., and Greene, M. I. (1990b). Cell (Cambridge, Mass.) 61, 1339–1347. Wagner, J. L., Thomas, C. R., Jr., Koh, W. J., and Rudolph, R. H. (1995). Med. Pediatr. Oncol. 24, 123–132. Wallasch, C., Weiss, F. U., Niederfellner, G., Jallal, B., Issing, W., and Ullrich, A. (1995). EMBO J. 14, 4267–4275. Wang, L. M., Kuo, A., Alimandi, M., Very, M. C., Lee, C. C., Kapoor, V., Ellmore, N., Chen, X. H., and Pierce, J. H. (1998). Proc. Natl. Acad. Sci. U.S.A. 95, 6809–6814. Warri, A. M., Laine, A. M., Majasuo, K. E., Alitalo, K. K., and Harkonen, P. L. (1991). Int. J. Cancer 49, 616–623. Warri, A. M., Isola, J. J., and Harkonen, P. L. (1996). Eur. J. Cancer 32A, 134–140. Waterman, H., Sabanai, I., Geiger, B., and Yarden, Y. (1998). J. Biol. Chem. 273, 13819–13827. Weidner, U., Peter, S., Strohmeyer, T., Hussnatter, R., Ackermann, R., and Sies, H. (1990). Cancer Res. 50, 4504–4509. Weiner, D. B., Kokai, Y., Wada, T., Cohen, J. A., Williams, W. V., and Greene, M. I. (1989a). Oncogene 4, 1175–1183. Weiner, D. B., Liu, J., Cohen, J. A., Williams, W. V., and Greene, M. I. (1989b). Nature (London) 339, 230–231. Weiner, L. M., Holmes, M., Adams, G. P., LaCreta, F., Watts, P., and Garcia-de-Palazzo, I. (1993). Cancer Res. 53, 94–100. Weiner, L. M., Clark, J. I., Davey, M., Li, W. S., Garcia de Palazzo, I., Ring, D. B., and Alpaugh, R. K. (1995). Cancer Res. 55, 4586–4593. Weiss, S. E., Tartter, P. I., Ahmed, S., Brower, S. T., Brusco, C., Bossolt, K. Amberson, J. B., and Bratton, J. (1995). Cancer (Philadelphia) 76, 268–274. Wells, A., Welsh, J. B., Lazar, C. S., Wiley, H. S., Gill, G. N., and Rosenfeld, M. G. (1990). Science 247, 962–964. Wells, J. A. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 1–6. Wen, D. Peles, E. ,Cupples, R. Suggs, S. V., Bacus, S. S., Luo, Y., Trail, G., Hu, S., Silbiger, S. M., Ben-Levy, R., Luo, Y., and Yarden, Y. (1992). Cell (Cambridge, Mass.) 69, 559–572. Willsher, P. C., Beaver, J., Pinder, S., Bell, J. A., Ellis, I. O., Blamey, R. W., and Robertson, J. F. (1996). Breast Cancer Res. Treat. 40, 251–255. Willsher, P. C., Leach, I. H., Ellis, I. O., Bell, J. A., Elston, C. W., Bourke, J. B., Blamey, R. W., and Robertson, J. F. (1997). Anticancer Res. 17, 2335–2338. Wiltschke, C., Kindas Muegge, I., Steininger, A., Reiner, A., Reiner, G., and Preis, P. N. (1994). J. Cancer Res. Clin. Oncol. 120, 737–742. Witters, L. M., Kumar, R., Chinchilli, V. M., and Lipton, A. (1997). Breast Cancer Res. Treat. 42, 1–5. Wong, A. J., Ruppert, J. M., Bigner, S. H., Grzeschik, C. H., Humphrey, P. A., Bigner, D. S., and Vogelstein, B. (1992). Proc. Natl. Acad. Sci. U.S.A. 89, 2965–2969. Wong, Y. F., Cheung, T. H., Lam, S. K., Lu, H. J., Zhuang, Y. L., Chan, M. Y., and Chung, T. K. (1995). Gynecol. Obstet. Invest. 40, 209–212. Wong, Y. F., Chung, T. K., Cehung, T. H., Lam, S. K., Tam, O. S., Lu, H. J., Xu, F. D., and Chang, A. M. (1996). J. Obstet. Gynaecol. Res. 22, 171–175.

ErbB/HER Signaling Network of Growth Factor Receptors

79

Wu, K., Salas, P. J., Yee, L., Fregien, N., and Carraway, K. L. (1994). Oncogene 9, 3139–3147. Wu, M. S., Shun, C. T., Sheu, J. C., Wang, H. P., Wang, J. T., Lee, W. J., Chen, C. J., Wang, T. H., and LIn, J. T. (1998). J. Gastroenterol. Hepatol. 13, 305–310. Yamamoto, T., Ikawa, S., Akiyama, T., Semba, K., Nomura, N., Miyajima, N., Saito, T., and Toyoshima, K. (1986). Nature (London) 319, 230–234. Yamanaka, Y., Friess, H., Kobrin, M., Buchler, M., Beger, H., and Korc, M. (1993a). Anticancer Res. 13, 565–570. Yamanaka, Y., Friess, H., Kobrin, M. S., Buchler, M., Kunz, J., Beger, H. G., and Korc, M. (1993b). Hum. Pathol. 24, 1127–1134. Yang, J. L., Yu, Y., Markovic, B., Russell, P. J., and Crowe, P. J. (1997). Anticancer Res. 17, 1023–1026. Yang, Y., Spitzer, E., Meyer, D., Sachs, M., Niemann, C., Hartmann, G., Weidner, K. M., Birchmeier, C., and Birchmeier, W. (1995). J. Cell Biol. 131, 215–226. Yarden, Y., and Peles, E. (1991). Biochemistry 30, 3543–3550. Yarden, Y., and Schlessinger, J. (1987). Biochemistry 26, 1443–1445. Yarden, Y., and Weinberg, R. A. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 3179–3183. Yonemura, Y., Ninomiya, I. Tsugawa, K., Fushida, S., Fujimura, T., Miyazaki, I., Uchibayashi, T., Endou, Y., and Sasaki, T. (1998). Cancer Detect. Prev. 22, 139–146. Yu, D., Liu, B., Jing, T., Sun, D., Price, J. E., Singletary, S. E., Ibrahim, N., Hortobagyi, G. N., and Hung, M. C. (1998). Oncogene 16, 2087–2094. Yu, D. H., and Hung, M. C. (1991). Oncogene 6, 1991–1996. Zafrani, B., Leroyer, A., Fourquet, A., Laurent, M.,Trophilme, D., Validire, P., and Sastre Garau, X. (1994). Semin. Diagn. Pathol. 11, 208–214. Zhang, D., Skowkowski, M. X., Mark, M., Frantz, G., Akita, R., Sun, Y., Hillan, K., Crowley, C., Brush, J., and Godowski, P. J. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 9562–9567. Zhang, K., Sun, J., Liu, N., Wen, D., Chang, D., Thomason, A., and Yoshinaga, S. K. (1996). J. Biol. Chem. 271, 3884–3890. Zhang, X. H., Takenaka, I., Sato, C., and Sakamoto, H. (1997). Urology 50, 636–642. Zhau, H. E., Zhang, X., von Eschenbach, A. C., Scorsone, K. Babaian, R. J., Ro, J. Y., and Hung, M. C. (1990). Mol. Carcinog. 3, 254–257.

This Page Intentionally Left Blank

p53 and Human Cancer: The First Ten Thousand Mutations Pierre Hainaut1 and Monica Hollstein2 1 IARC F69372 Lyon, France 2 DKFZ D69120 Heidelberg, Germany

I. Introduction II. Biology of the p53 Protein A. p53 Is a Sensor of Multiple Forms of Stress B. p53 Is a Member of a Multigene Family C. Structure of the p53 Protein D. Upstream of p53: Signaling Pathways Leading to p53 Activation E. Downstream of p53: Molecular Effectors of p53 Functions III. Mutations and Variations in the p53 Gene A. p53 Polymorphisms B. Somatic Mutations C. Germline Mutations in p53: Li–Fraumeni Cancer Syndrome IV. p53 Mutations in Sporadic Cancers: Host–Environment Interactions A. p53 Mutation Prevalence B. Type and Localization of Most Frequent Mutations C. Role of Environment in Shaping Tumor Mutation Patterns V. Clinical Relevance of p53 Mutations A. p53 Mutations as Markers of Clonality B. Detection and Follow-Up of Cancer Lesions C. Mutations and Tumor Prognosis D. p53 as a Target in New Therapeutic Strategies References

I. INTRODUCTION The p53 protein was codiscovered 20 years ago by several groups as a cellular protein of 53 kDa complexing with the large T antigen of the simian virus SV40 (Harris, 1996b). Initially cloned from human or rodent tumors, the gene was found to behave as an oncogene cooperating with RAS in the transformation of rat primary ﬁbroblasts. In the late 1980s, it became clear that this property belonged to some mutant forms, whereas the wild-type Advances in CANCER RESEARCH 0065-230X/00 $30.00

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

82

Pierre Hainaut and Monica Hollstein

gene was capable of suppressing transformation in vitro as well as tumor growth in vivo. In 1990, constitutive alterations of p53 were detected in patients of the Li–Fraumeni syndrome, a rare autosomal disease characterized by the early onset of several types of cancers. In 1992, Donehower and colleagues showed that p53 homozygous and hemizygous knockout mice had an essentially normal development but were prone to early, multiple cancers. These observations led to speculations that p53 was the “ultimate tumor suppressor gene” (Oren, 1992) or “all that lies between us and early death by cancer” (Lane, 1992). In human tumors, frequent loss of heterozygosity at the p53 locus (chromosome 17p13) was observed in 1988, and the ﬁrst point mutation in the coding sequence was described 10 years ago by Vogelstein and colleagues in a colorectal cancer (Baker et al., 1989; Nigro et al., 1989). This discovery has triggered a tidal wave of more than 13,000 publications on p53 structure, functions, and mutations. This unprecedented research effort is a cornerstone in our current understanding of the molecular biology and molecular epidemiology of human cancer. p53 is frequently mutated in most common human malignancies and behaves as a multifunctional transcription factor involved in the control of cell cycle, programmed cell death, senescence, differentiation and development, transcription, DNA replication, DNA repair, and maintenance of genomic stability. Moreover, researchers soon realized that point mutations were exceptionally diverse in their localization and nature and this has sparked widespread interest in the possibility that mutations reveal clues about the etiology and molecular pathogenesis of human cancer (M. Hollstein et al., 1991; M. C. Hollstein et al., 1990). A database of point mutations was initiated in 1990 and developed since as a collective effort by several scientists in Europe and in the United States (Hollstein et al., 1994). This database is the largest single-locus mutation database worldwide and now contains more than 10,000 somatic mutations identiﬁed by sequencing (Beroud and Soussi, 1998; Hainaut et al., 1998).* In this review, we summarize the major lessons from this vast sequencing effort. The biology of p53 has recently been the topic of several general reviews (Agarwal et al., 1998; Gottlieb and Oren, 1996; Hainaut, 1995; Harris, 1996d; Ko and Prives, 1996; Levine, 1997; Moll and Schramm, 1998; Morgan and Kastan, 1997; Selivanova and Wiman, 1995; Sidransky and Hollstein, 1996).

* Several p53 mutation databases are available on the Worldwide Web. Data in this review are from the IARC p53 mutation database (Hainaut et al., 1998; Hollstein et al., 1994), release R2 (July 1998, 9334 mutations), http:\\www.iarc.fr\p53\homepage.htm.

p53 and Human Cancer

83

II. BIOLOGY OF THE p53 PROTEIN A. p53 Is a Sensor of Multiple Forms of Stress The p53 protein is constitutively expressed in almost all cell types but has a very rapid turnover and appears to be latent under normal conditions. However, p53 is rapidly converted to an active form in response to a number of physical or chemical DNA-damaging agents such as X or gamma irradiation, UV rays, oxidizing agents, cytotoxic drugs, and cancer-causing chemicals. Induction of p53 implies nuclear retention, accumulation of the protein as a result of post-translational stabilization, and allosteric conversion to a form with high sequence-speciﬁc DNA-binding capacity. This has led to the concept that p53 is speciﬁcally activated in response to DNA damage, thus acting as a “guardian of the genome” against genotoxic stress (Kastan et al., 1991; Lane, 1992). With ionizing or UV irradiation, the critical molecular signal for p53 induction is the formation of DNA single- or double-strand breaks (Kastan, 1996; Nelson and Kastan, 1994; Huang et al., 1996a). Alkylating agents and chemicals that form bulky DNA adducts also induce p53 (Fritsche et al., 1993; Ramet et al., 1995; Siegel et al., 1995). Other inducers of p53 include factors that are essentially not DNA damaging such as hypoxia, hyperthermia, serum starvation, and suboptimal growth conditions, drugs that deplete ribonucleotide pools or microtubules, and cytokines of the TNF-␣ family (Di Leonardo et al., 1997; Donato and Perez, 1998; Graeber et al., 1994; Lanni and Jacks, 1998; Linke et al., 1996). Moreover, cellular adhesion has been shown to regulate p53 protein levels in primary keratinocytes (Nigro et al., 1997). Thus, p53 appears to be a sensor of multiple forms of stress rather than a speciﬁc DNA damage response factor. The pathway, extent, kinetics, and consequences of p53 induction depend on the cell type and the nature of stress. For example, induction by gamma irradiation is rapid and transient in most cultured cells (maximum after 2–4 hr) (Kuerbitz et al., 1992), whereas the response to benzo(a)pyrene is slow and persistent (maximum at 12–48 hr) (Ramet et al., 1995). These differences reﬂect the intracellular metabolism of genotoxic agents as well as the existence of distinct signaling pathways leading to p53 induction.

B. p53 Is a Member of a Multigene Family The p53 gene, located on chromosome 17p13.1 in humans, spans 20 kilobases and contains 11 exons, the ﬁrst one noncoding. The overall structure of the gene and the sequence of several domains corresponding to key struc-

84

Pierre Hainaut and Monica Hollstein

tural features of the protein are well conserved in all vertebrates. To date, however, no p53 homolog has been found in invertebrates (Soussi et al., 1990; Soussi and May, 1996). Recently, two p53-related genes have been isolated. The DNA-binding domains have the greatest homology with p53 (about 60% identity) and the residues required for sequence-speciﬁc DNA recognition are conserved and occupy identical positions as in p53 (Kaelin, 1998). p73 is located on human chromosome 1p36, within a region frequently deleted in neuroblastoma (Kaghad et al., 1997). p40 and p51 are two human cDNAs homologous to the rat Ket and correspond to a gene located on human chromosome 3p28 (Osada et al., 1998; Schmale and Bamberger, 1997; Trink et al., 1998). These genes are expressed in a limited number of tissues, have a complex expression pattern with several major splicing variants, and encode proteins that transactivate genes under the control of p53-response elements. Overexpression of p73 inhibits cell growth in a p53like manner by inducing apoptosis (Jost et al., 1997). However, neither p73 nor p51/p40/ket are responsive to DNA damage and, to date, they do not appear to be mutated at any signiﬁcant frequency in human cancer. The similarities between these genes and p53 indicate that they all derive from the same ancestral gene. p53 may represent a member of the family specializing in response to stress and resulting from gene duplication that occurred in the vertebrate lineage.

C. Structure of the p53 Protein The p53 protein has the anatomy of an oligomeric transcription factor of 393 residues organized in five structural and functional regions. These regions include an N-terminal, transcriptional activation domain (residues 1–44), a proline-rich regulatory domain (residues 62–94), a sequence-specific DNA-binding domain (residues 110–292), an oligomerization domain (residues 325–363), and a multifunctional, C-terminal domain involved in the regulation of DNA-binding (residues 363–393). The tertiary structure of several portions of the wild-type protein has been elucidated by crystallography or by NMR. Structural data are available for a stretch of the N terminus [residues 12–27, in complex with mdm-2 (Kussie et al., 1996)], the central globular domain [residues 102–296, in complex with specific DNA (Cho et al., 1994) or with the p53 binding protein BP2 (Gorina and Pavletich, 1996)], and the dimerization/tetramerization domain (residues 325–363, as a quaternary complex) (Clore et al., 1994, 1995; Jeffrey et al., 1995; W. Lee et al., 1994) (Fig. 1; see color plate). However, the structure of the full-length wild-type protein has not been elucidated and there is no published structural data on mutant p53. Overall, the p53 protein may be pictured as a tight, hydrophobic central globule containing the DNA-

p53 and Human Cancer

85

binding domain, flanked by more accessible N- and C-terminal regions (Pavletich et al., 1993). These regions are very sensitive to proteolysis and contain the dominant epitopes in mice immunized with purified p53 as well as in cancer patients carrying auto-antibodies to p53 (Legros et al., 1994).

1. N TERMINUS This acidic domain (1–44) provides interaction sites for TAF components of the basal transcription complex TFIID and for the p53-regulatory protein mdm-2 (X. Chen et al., 1993; Kussie et al., 1996; Oliner et al., 1993; Truant et al., 1993). Binding to these factors may be regulated by phosphorylation of p53 at several sites by two members of the PI3 kinase family, the dsDNA-dependent protein kinase (DNAPK, Ser 6, 15, and 37) (Lees-Miller et al., 1992; Shieh et al., 1997) and ATM (Ser 15) (Siliciano et al., 1997). In mouse p53, more extensive mapping has revealed additional sites for casein kinase I (Ser 4, 6, and 9) (Knippschild et al., 1997; Milne et al., 1992a), Raf-1 (within the 18 N-terminal residues) (Jamal and Ziff, 1995), and Jun N-terminal kinases ( JNK, Ser 34) (Meek, 1998; Milne et al., 1995). The 15residue N-terminal portion that has been crystallized is coiled into an amphipathic alpha helix that binds tightly into a deep hydrophobic cleft at the surface of the mdm-2 protein (Kussie et al., 1996). Binding to mdm-2 relies essentially on three hydrophobic residues that are also involved in transactivation (Phe 19, Trp 23, and Leu 26), supporting the hypothesis that mdm2 inactivates p53 by concealing its transactivation domain (Lin et al., 1994; Oliner et al., 1993).

2. PROLINE-RICH DOMAIN The proline-rich region (64–92) contains ﬁve repeats of the PXXP motif that resembles src-homology 3 (SH3) domains (Walker and Levine, 1996). SH3 domains are involved in controlling protein–protein interactions in signal transduction. However, it is unclear whether the proline-rich domain of p53 folds as a genuine SH3 recognition site. This domain is required to activate p53-dependent apoptosis but not growth arrest (Sakamuro et al., 1997) and may represent a docking site in the transduction of signals dependent on the growth arrest-speciﬁc gene Gas-1 (Ruaro et al., 1997). Deletion of the proline-rich domain does not affect transactivation of several promoters, such as WAF1, MDM2, and BAX, but it alters transcriptional repression, reactive oxygen species production, and sequence-speciﬁc transactivation of the PIG3 gene (Venot et al., 1998). This domain may contain phosphorylation sites for MAP kinases (mapped on Thr 73 and 83 in mouse p53) (Milne et al., 1994).

86

Pierre Hainaut and Monica Hollstein

3. DNA-BINDING DOMAIN p53 binds as a tetramer to DNA sequences containing four copies of the pentamer PuPuPuC(A/T) oriented in alternating directions. The human genome is expected to contain several hundred sequences that match this consensus (Bourdon et al., 1997; Tokino et al., 1994). The sequence-speciﬁc DNA-binding domain contains two ␤ sheets that provide a scaffold for a DNA-binding surface made of two noncontiguous regions, a large loop that binds in a minor groove of DNA and a loop-sheet-helix motif that binds in the major groove. The main contact in the minor groove is provided by Arg 248, whereas several residues make contacts in the major groove (Lys 120, Ser 241, Arg 273, Ala 276, Cys 277, and Arg 283) (Cho et al., 1994). This complex architecture is stabilized by the binding of a zinc atom on three cysteines (176, 238, 242) and on one histidine (179). Metal binding is dependent on the oxidation–reduction status of cysteines. Metal chelation and cysteine oxidation abrogate DNA binding by promoting the unfolding of the DNA-binding domain (Hainaut and Milner, 1993a, b; Verhaegh et al., 1998). 3-D structural modeling indicates that the steric constraints of the bound tretramers favor the bending of DNA toward the major groove at the pentamer junction, which corresponds to the intrinsically ﬂexible CATG sequence in most identiﬁed p53 response elements. DNA bending optimizes binding to the minor groove and stabilizes the tetramer through additional intersubunit contacts involving residues of the L2 loop (Pro 177, His 178, Glu 180, Arg 181) (Durell et al., 1998).

4. OLIGOMERIZATION DOMAIN The tetramerization domain encompasses one ␤ strand and one ␣ helix (residues 319–363). The domains of four p53 molecules associate to form a 20-kDa symmetric tetramer with a topology made up from a dimer of dimers (Arrowsmith and Morin, 1996; Clore et al., 1994, 1995; Jeffrey et al., 1995; W. Lee et al., 1994). The region upstream of the oligomerization domain contains a phosphorylation site for cyclin-dependent kinases (cyclin A-cdk2 and cyclin B-cdc2) (Wang and Prives, 1995) and the main nuclear localization signal (residues 316–322) (Shaulsky et al., 1991a,b).

Fig. 1 3-D structure of p53. This composite picture shows the structures of three p53 protein domains that have been elucidated so far. These structures include the mdm-2 binding site in the N terminus (residues 17–29; Kussie et al., 1996), the DNA-binding domain (residues 96–289; Cho et al., 1994), and residues 319–360 (Clore et al., 1994, 1995; W. Lee et al., 1994). 3-D models were produced with the RASMOL modeling software using the following coordinates: 1YCS (N terminus; Kussie et al., 1996), 1TUP (DNA-binding domain in complex with DNA; Cho et al., 1994), and 1PET (tretramerization domain, Lee et al., 1998).

p53 and Human Cancer

87

5. C TERMINUS This domain is rich in basic residues. It contains docking sites for several cellular proteins including the helicases ERCC2/XPD and ERCC3/XPB (Wang et al., 1994, 1996), the calcium binding protein S100b (Baudier et al., 1992) the heat-shock protein (hsp) 70 (Hainaut and Milner, 1992), the signal transduction protein 14-3-3␴ (Waterman et al., 1998) and the product of the breast cancer susceptibility gene BRCA1 (H. Zhang et al., 1998). The C terminus is responsible for maintenance of p53 in the latent form. This negative regulation may be due to a physical interaction between the C terminus and portions of the N terminus or of the DNA-binding domain. Neutralization of this domain by specific antibodies, by competing peptides, or by truncation brings about a structural change that converts p53 to the active form capable of binding to DNA (Arrowsmith and Morin, 1996; Halazonetis et al., 1993; Halazonetis and Kandil, 1993; Hupp and Lane, 1994; Selivanova et al., 1997). However, neutralization of the C terminus may not be required for binding to all genomic p53-response elements. For example, the p21waf-1 promoter contains two distinct p53 binding sites. Incubation with PAb421 enhances binding to the 5⬘ site, but inhibits binding to the 3⬘ site, suggesting that there are different classes of genomic DNA-binding sites for p53 (Resnick-Silverman et al., 1998). The C terminus also binds nonspecifically to single- and double-strand nucleic acid as well as to damaged DNA in vitro. It has been suggested that binding to damaged DNA may represent a signal for allosteric activation of p53 by neutralization of the C terminus ( Jayaraman and Prives, 1995; Shaw et al., 1996). The C terminus contains several sites of covalent modification by acetylation (Gu and Roeder, 1997), glycosylation (Shaw et al., 1996) and phosphorylation (by PKC; Ser 378; Takenaka et al., 1995), CDK7-Cyclin H/p36 (Ko et al., 1997; Lu et al., 1997), and casein kinase II (Ser 392; Fiscella et al., 1994; Herrmann et al., 1991; Milne et al., 1992b). Casein kinase II also forms stable, noncovalent complexes with the C terminus of p53 (Filhol et al., 1992).

D. Upstream of p53: Signaling Pathways Leading to p53 Activation Activation of p53 is a complex process involving the cooperation of multiple molecular mechanisms. Table I lists factors that have been implicated in the signalling pathways upstream of p53, and Table II lists downstream factors. Most of our current knowledge is derived from studies using ionizing radiation to induce DNA strand break damage. It is, however, likely that several distinct pathways coexist besides the ones activated by irradiation.

88 Table I Upstream of p53 Factor

Biochemical function/activated by

Interaction with p53

14-3-3s ATM

Signal transduction/ionizing radiations Kinase/ionizing radiations

Binding, C terminus (Ser 376) Phosphorylation, Ser 15

c-abl

Tyrosine kinase/irradiation, DNA-strand breaks Component of TFIIH

Binding, proline-rich region

cdk7-cyclin H Cdc2/Cdk2cyclin A/B CKI

Cell-cycle-dependent kinases/ Kinase/?

CKII

Kinase/UV

DNA-PK E6AP Hif-1 HMG-1

Kinase/UV E6 accessory protein/ubiquitin-mediated degradation Hypoxia-inducible factor/hypoxia High mobility group 1/?

MAPK

Mitogen-activated protein kinase/UV?

Phosphorylation of Ser 33 Binding/phosphorylation of Ser 315 Phosphorylation, several N-terminal serines Binding to C terminus, phosphorylation of Ser 389 Phosphorylation, Ser 15 and Ser 37 Binding Binding Binding to N terminus or to DNA-binding domain Phosphorylation, Thr 73 and 83 (mouse p53)

Reference Waterman et al. (1998) Siliciano et al. (1997) Waterman et al. (1998) Gogo et al. (1995) Yuan et al. (1996b) Ko et al. (1997) Wang and Prives (1995) Milne et al. (1992a) Hall et al. (1996) Kapoor and Lozano (1998) Shieh et al. (1997) Huibregtse et al. (1993b) An et al. (1998) Jayaraman et al. (1998) Milne et al. (1994)

mdm-2

Oncogene/negative control of p53

Binding, residues 13–29

NO

Nitric oxide/oxidative stress, inﬂamation, irradiation Cell-cycle inhibitor, product of CDKN2a/

Oxidation of cysteines in DNAbinding domain Prevents p53–mdm2 interactions

Histone acetyl-transferases/coactivators of transcription ADP-ribose polymerase/DNA strand breaks, nucleotide depletion Stress-activated kinases/UV Protein kinase C

binding, N terminus Acetylation, C terminus ADP-ribose polymers bind to p53

p19arf p300/CBP PARP JNK/p38 PKC Ref-1

Redox-repair enzyme/oxidative stress, hypoxia

Phosphorylation, Ser 34, mouse p53 Phosphorylation, Ser 378 Reduction of cysteine in DNA-binding region Binding to C terminus

Oliner et al. (1993) Kussie et al. (1996) Haupt et al. (1997) Kubbutat et al. (1997) Calmels et al. (1997) Kamijo et al. (1998) Pomerantz et al. (1998) Gu and Roeder (1997) Gu et al. (1997) Malanga et al. (1998) Whitacre et al. (1995) Milne et al. (1995) Takenaka et al. (1995) Baudier et al. (1992) Jayaraman et al. (1997) Meira et al. (1997)

For explanation, see text. Identical factors may be listed in Tables I and II, as for example 14-3-3s and mdm-2. Both genes are transactivated by p53 but their products also bind p53 and regulate its activity.

89

90 Table II Downstream of p53 Factor

Activity

Mode of regulation

14-3-3s

Signal transduction

Transcriptional activation

Bax-1

Dominant-negative inhibitor of bcl2 Repressor of apoptosis Inhibitor of proliferation Cell-cycle regulation, S phase Cell-cycle regulation Helicases, TFIIH complex Binding to PCNA Inhibitor of IGF-I Growth factor Survival factor Death signaling receptor Oncogene Multidrug resistance

Transcriptional activation

Bcl-2 BTG2 Cyclin A Cyclin G ERCC2/ERCC3 Gadd45 IGF-BP3 IGF-I IL-6 Killer/DR5 Mdm-2 MDR-1 NOS2/iNOS p21waf-1

Inducible nitric oxide synthase Inhibitor of CDK2-4 and 6

Transcriptional repression Transcriptional activation Transcriptional repression Transcriptional activation Activation by protein binding Transcriptional activation Transcriptional activation Transcriptional repression Transcriptional repression Transcriptional activation Transcriptional activation Transcriptional repression Transcriptional repression Transcriptional activation

Function G2 arrest via cdc2 block? Apoptosis Apoptosis ? Cell cycle arrest? Cell-cycle arrest? Transcription/DNA repair Cell cycle arrest? Apoptosis Apoptosis? Apoptosis? Apoptosis Repression of p53 Resistance to chemotherapy Control of oxidative stress Cell cycle arrest, G1 and G2/M

Reference Hermeking et al. (1997) Miyashita et al. (1994) Selvakumaran et al. (1994) Miyashita et al. (1994) Rouault et al. (1996) Desdouets et al. (1996) Okamoto and Beach (1994) Wang et al. (1996) Kastan et al. (1992) Buckbinder et al. (1995) Werner et al. (1996) Margulies and Sehgal (1993) Sheikh et al. (1998) Juven et al. (1993) Chin et al. (1992) Ambs et al. (1998); Forrester et al. (1996) El-Deiry et al. (1993)

p85

Pig-3 Pig-6 RPA TBP

Regulatory subunit of PI3 kinase Auxiliary subunit of polymerase s Galectin-7 Glutathione transferase homolog Quinone oxidase homolog Proline oxidase homolog Replication protein A TATA box-binding protein

TGF-alpha Thrombospondin-1 TopoIsomerase-I

Growth factor Inhibitor of angiogenesis Control of DNA topology

PCNA Pig-1 Pig-12

Transcriptional activation

Apoptosis

Yin et al. (1998)

Transcriptional activation

DNA repair?

Shivakumar et al. (1995)

Transcriptional activation Transcriptional activation

Differentiation? Apoptosis

Polyak et al. (1997) Polyak et al. (1997)

Transcriptional activation Transcriptional activation Inhibition by protein binding Inhibition by protein binding

Apoptosis Apoptosis Replication/DNA Repair Inhibition of transcription

Transcriptional activation Transcriptional activation Activation by protein binding

Growth stimulation? Apoptosis? Transcription/repair?

Polyak et al. (1997) Polyak et al. (1997) Dutta et al. (1993) Martin et al. (1993) Seto et al. (1993) Truant et al. (1993) Shin et al. (1995) Dameron et al. (1994) Gobert et al. (1996)

For explanation, see text. Identical factors may be listed in Tables I and II, as for example 14-3-3s and mdm-2. Both genes are transactivated by p53 but their products also bind p53 and regulate its activity.

91

92

Pierre Hainaut and Monica Hollstein

1. SIGNALING TO p53 Several molecules that are rapidly activated after DNA damage have been implicated in the signal transduction upstream of p53. Poly(ADP-ribose) polymerase (PARP) is an enzyme that uses NAD⫹ as a substrate to catalyze the formation of polymers or poly(ADP-ribose) in response to DNA strand interruptions (Satoh and Lindahl, 1992). In Chinese hamster cells, activation of p53 is impaired by PARP inhibitors and by culture in medium containing low levels of NAD⫹ (Whitacre et al., 1995). Poly(ADP-ribose) polymers bind strongly to several domains in the p53 protein (Malanga et al., 1998). Fibroblasts from PARP-null mice show increased sensitivity to DNA-damaging agents but retained at least partially the capacity to activate p53 (Agarwal et al., 1997; de Murcia et al., 1997). DNA strand interruptions also activate several kinases of the PI3 kinase family, including Atm and DNA-PK. These large-size molecules (⬎2500 residues) have related kinase domains in their C terminus but signiﬁcantly differ in the rest of their sequence. DNA-PK is directly involved in the recognition of DNA strand breaks through the binding of sensor subunits Ku 70 and Ku 80. Atm (AT-mutated) is the gene altered in ataxia telangiectasia (AT), a recessive childhood disorder characterized by neurologic deﬁciencies, infertility, immune deﬁciencies, and radiation sensitivity (Westphal, 1997). Atm initiates a cascade of signaling events in response to gamma irradiation, which involves Atr (a close homolog of Atm) and Chk1 (a homolog of a yeast mitotic checkpoint kinase coupling DNA repair with the cell-cycle machinery) (Hoekstra, 1997; Westphal, 1997). Atm and DNA-PK both phosporylate p53 in the N terminus (on Ser 15 and perhaps 37) (Shieh et al., 1997; Siliciano et al., 1997). Both of them also activate the c-Abl tyrosine kinase by direct binding through SH3 domains ( Jin et al., 1997; Kharbanda et al., 1997; Shafman et al., 1997). In addition, c-Abl has been reported to bind to p53 and to contribute to growth suppression in a p53-dependent manner (Yuan et al., 1996a,b; Goga et al., 1995). Mutation of Ser 15 reduces the ability of p53 to inhibit cell-cycle progression (Fiscella et al., 1993). Cells lacking Atm show impaired p53 induction and cell-cycle arrest after gamma irradiation, but not after other forms of DNA damage (Jongmans et al., 1996; Kastan et al., 1992; Xu et al., 1998). In cells lacking DNA-PK activity, recent data indicate that various forms of DNA damage fail to stimulate p53 DNA binding and transcriptional activity (Woo et al., 1998). However, primary ﬁbroblasts from severe-combined immunodeﬁcient (SCID) mice, which show low but detectable levels of DNA-PK activity, have no obvious defect in p53-dependent responses to stress (Huang et al., 1996b; Fried et al., 1996). Similar to Atm, Nibrin, the product of the gene involved in Nijmegen breakage syndrome (NBS), also appears to play a role in the pathway that links p53 to ionizing radiation but

p53 and Human Cancer

93

not to other DNA-damaging agents (Matsuura et al., 1998; Varon et al., 1998; Jongmans et al., 1997). The picture emerging from these complex interactions is that all these molecules are components of a multimolecular signaling machinery coupling the sensing of DNA strand interruptions to p53. Other kinases that may be involved in p53 activation include the Jun Nterminal kinases (JNK), which are activated within signaling cascades initiated by UV and ionizing radiation, cytokines such as TNF-␣, and heat shock, but not growth factors. Kinases that phosphorylate the extreme N terminus, such as casein kinase I and Raf-1, may be involved in the coordination of different stimuli linked to DNA damage and repair. MAP kinases are activated by growth factors, differentiating agents, and also ionizing radiation, but so far there is no report that they phosphorylate p53 from species other than mouse. The role of phosphorylation by casein kinase II (CKII), cyclindependent kinases, and PKC is still unclear. CKII has been shown to phosphorylate p53 in response to UV but not ionizing radiation (Kapoor and Lozano, 1998; see Meek, 1998, for review). Induction of p53 in response to hypoxia follows a pattern that differs from the one elicited by response to DNA damage. Hypoxia causes an increase in the nuclear levels of p53, but the G1 arrest induced by hypoxia is not strictly p53 dependent (Graeber et al., 1994). Cells expressing the HPV E6 protein fail to accumulate p53 in response to DNA damage but increase their p53 levels following hypoxia (Graeber et al., 1994). In tumors, hypoxia acts as a physiologic selection pressure for the expansion of cells with p53 mutations that have lost their apoptotic potential (Graeber et al., 1996). One of the speciﬁc mediators of the cellular responses to hypoxia is the transcription factor Hif-1␣ (hypoxia-inducible factor 1 alpha), which binds to p53, stabilizes the protein, and stimulates the transcription of p53-dependent reporter genes (An et al., 1998; Carmeliet et al., 1998). Binding of cellular proteins such as BRCA1 may also modulate p53 activity independently of DNA damage. BRCA1 increases p53-dependent transcription from the WAF1 and BAX promoters and cooperates with p53 in the induction of apoptosis of cancer cells (W. Zhang et al., 1998).

2. MODIFICATIONS OF p53 DURING INDUCTION Given the multiplicity of signals that affect p53, it is not surprising that the mechanism of p53 induction is in itself a multistep process involving coordinated covalent and noncovalent modiﬁcations in different domains of the molecule. Some of the modiﬁcations occurring in response to ionizing radiation have recently been described (Siliciano et al., 1997; Waterman et al., 1998). These ﬁndings provide a basis for the following three-step model of p53 induction (Fig. 2). The ﬁrst step involves the N terminus. Phosphorylation by Atm or DNA-

94

Pierre Hainaut and Monica Hollstein

Fig. 2 A three-step model of p53 activation by stress signals. Activation of p53 requires covalent and noncovalent modiﬁcations in distinct domains of the protein. The ﬁrst step involves phosphorylation in the N terminus, dissociation of complexes with mdm-2, and binding of histone-acetyl-transferases of the CBP/p300 family. The second step consists of coordinated modiﬁcations of the C terminus, including acetylation, changes in phosphorylation, and binding of speciﬁc proteins such as 14-3-3␴. The third step concerns the central portion of the protein and involves reduction of cysteines that play an important role in the conformation of the DNAbinding domain. For explanations and references, see text.

PK may contribute to the disruption of the complexes between p53 and mdm-2. Also, mdm-2 is the universal regulator of p53 protein stability, which targets p53 for rapid, proteasome-mediated degradation (Bottger et al., 1997; Haupt et al., 1997; Kubbutat et al., 1997; Lane and Hall, 1997). It is also a transcriptional target of p53 and therefore participates in a negative regulatory feedback loop that controls p53 levels (Momand and Zambetti, 1997; Piette et al., 1997). In addition, p19arf, the alternative product of the tumor suppressor locus CDKN2/INK4a, interacts with mdm-2 and

p53 and Human Cancer

95

neutralizes mdm-2’s inhibition of p53 (Kamijo et al., 1998; Pomerantz et al., 1998; Y. Zhang et al., 1998). Removal of mdm-2 not only allows p53 to escape degradation but also unmasks the transactivation domain, allowing interactions with other proteins such as members of the CBP/p300 family. CBP/p300 and their partners such as p/CAF bind to several transcription factors and play a broad role as coactivators of transcription through their histone-acetyl-transferase activity (Avantaggiati et al., 1997; Gu et al., 1997; Scolnick et al., 1997). The second step occurs at the C terminus. CBP acetylates ﬁve conserved lysine residues in the extreme C terminus (Lys 370, 372, 373, 381 and 382). Acetylation neutralizes the negative regulation exerted by the C terminus and activates the sequence-speciﬁc DNA-binding capacity of p53 (Gu and Roeder, 1997). Other covalent modiﬁcations of the C terminus include phosphorylation of Ser 378 and dephosphorylation of Ser 376 to generate a binding site for the signal transduction protein 14-3-3␴. In turn, association with 14-3-3␴ increases the afﬁnity of p53 for speciﬁc DNA. Neither Ser 376 dephosphorylation nor the association with 14-3-3␴ occurr in irradiated cells from AT patients, suggesting an involvement of Atm in the control of the modiﬁcations occuring at the C terminus of p53 (Waterman et al., 1998). The third step concerns the DNA-binding domain. This domain is intrinsically ﬂexible, and contains several cysteine residues that bind metals and are sensitive to oxidation–reduction. Metal chelation, high levels of nitric oxide, and agents that perturb the intracellular redox status have all been shown to modulate p53 DNA-binding activity in intact cells (Calmels et al., 1997; Rainwater et al., 1995; Verhaegh et al., 1997, 1998). Oxidized forms of full-length and C-terminally truncated p53, which are inactive for speciﬁc DNA-binding, are both stimulated by Ref-1, a dual-function protein that regulates the redox state of a number of proteins, and functions in DNA repair through its A/P endonuclease activity ( Jayaraman et al., 1997). These data indicate that redox regulation may play a role in the control of p53 and provide a plausible link between p53 activity and the mechanisms that control the intracellular redox balance in response to various forms of stress.

E. Downstream of p53: Molecular Effectors of p53 Functions Downstream p53 signaling is mediated by transcriptional activation of speciﬁc genes and by complex formation between p53 and heterologous proteins. The most relevant downstream effectors of p53 identiﬁed to date are listed in Table II.

96

Pierre Hainaut and Monica Hollstein

1. CELL-CYCLE CHECKPOINTS p53 is involved as part of several cell-cycle checkpoints, in G1, G2, and at mitosis following disruption of the mitotic spindle. p53 also plays a role in the control of centrosome duplication.

a. Arrest in G1

Arrest in G1 is a common response of many cultured cells to p53 induction. The central effector in this process is the cyclin-kinase inhibitor p21waf-1, which blocks several cyclin/cdk complexes at the G1 /S transition and prevents activation of the transcription factors of the E2F family (Del Sal et al., 1996; El-Deiry et al., 1993, 1994). GADD45 is another p53 target that may cause cell-cycle arrest when overexpressed (Chin et al., 1997; Kastan et al., 1992; Smith et al., 1997; Zhan et al., 1994, 1998). G1 arrest has often been interpreteted as a mechanism for allowing time for DNA repair before replication. Indeed, in some cancer cell lines, G1 arrest is transient. However, in normal human diploid ﬁbroblasts, the p53-dependent G1 arrest in response to gamma irradiation is irreversible and cells that escape initial arrest undergo long-term arrest in subsequent phases of cell cycle (Di Leonardo et al., 1994; Linke et al., 1997). Prolonged arrest may depend on the inability to repair damaged DNA, as ribonucleotide depletion in the absence of detectable DNA damage induces only transient arrest. Thus, p53dependent G1 arrest after DNA damage may function as a mechanism to eliminate damaged cells from the reproductively viable population (Di Leonardo et al., 1994; Linke et al., 1997).

b. Arrest in G2

Overexpression of p53 can inhibit entry into mitosis (Agarwal et al., 1995; Guillouf et al., 1995; Stewart et al., 1995). Activation of a temperature-sensitive p53 mutant induces cell-cycle arrest in G2 in primary rat embryo ﬁbroblasts enriched in the late G1 and early S phases before the temperature shift (Stewart et al., 1995). The molecular basis of G2 arrest is poorly understood. Effectors may include p21waf-1 (Niculescu et al., 1998) and 143-3␴, a transcriptional target of p53 that blocks cells in G2 by preventing the activation of cdc2 (Hermeking et al., 1997).

c. Mitotic Checkpoint Nocodazole, a drug that inhibits the assembly of microtubules, arrests normal mouse embryo ﬁbroblasts with a 4N content, whereas ﬁbroblasts from p53-deﬁcient mouse embryos undergo several rounds of DNA synthesis without chromosome segregation (Cross et al., 1995). This spindle checkpoint differs from the G2 /M checkpoint. Both p53+/+ and p53⫺/⫺ ﬁbroblasts treated with nocodazole transiently arrest at mitosis for the same length of

p53 and Human Cancer

97

time, but p53 is speciﬁcally required to prevent nocodazole-treated cells from reentering the cell cycle and initiating another round of DNA synthesis (Di Leonardo et al., 1997; Lanni and Jacks, 1998).

d. Control of Centrosome Duplication p53 localizes within the centrosome in Hela, COS, and 3T3 cells (Brown et al., 1994). In p53⫺/⫺ mouse embryo ﬁbroblasts, multiple centrosomes are generated during a single cell cycle, resulting in unequal segregation of chromosomes. In turn, abnormal centrosome duplication may lead to p53 activation and to cell-cycle arrest. Disruption of this control is one of the possible mechanisms by which loss of p53 causes genetic instability (Fukasawa et al., 1996).

2. APOPTOSIS p53 is involved in multiple distinct pathways leading to apoptosis. p53 induces apoptosis in a cell type-dependent manner (e.g., in lymphoid cells) in response to speciﬁc forms of DNA-damage, to hypoxia, to serum starvation or to overexpression of protooncogenes such as Myc or E1A (Canman and Kastan, 1997). On the other hand, expression of p53 at low levels may also exert an anti-apoptotic effect by promoting cell-cycle arrest after serum withdrawal (Lassus et al., 1996). First, p53 upregulates the expression of death receptors such as KiIller/DR5 (Sheikh et al., 1998) and APO1/fas/CD95 (Owen-Schaub et al., 1995). These receptors bind adaptor proteins to engage the caspase cascade and induce apoptosis. Second, p53 increases the expression of a number of genes predicted to encode proteins that generate or respond to oxidative stress (collectively called PIGs for p53-inducible genes) (Polyak et al., 1997). These proteins may be involved in the production of reactive oxygen species resulting in alterations of the mitochondrial membrane and cytoplasmic release of apoptogenic signals. The regulatory subunit of the phosphatidyl-3-OH kinase (PI3K), p85, is also upregulated by p53 and may participate as a signal transducer in the p53-dependent apoptosis induced by oxidative stress (Yin et al., 1998). Third, p53 further promotes mitochondrial leakage by upregulating Bax and downregulating Bcl2, two antagonistic proteins that insert into mitochondrial membranes to inhibit (Bcl2) or facilitate (Bax) the opening of mitochondrial permeability transition pores (Miyashita et al., 1994; Selvakumaran et al., 1994; Yin et al., 1997; Zamzami et al., 1996). Fourth, p53 decreases the availability of survival factor by upregulating the IGF-I binding protein IGF-BP3 (Buckbinder et al., 1995) and the anti-angiogenic factor thrombospondin-1 (Dameron et al., 1994). Fifth, p53 (and also p21waf-1 independently of p53) downregulates PS1 (presenilin-1), a member of a family of genes with antiapoptotic activity that are mutated in familial forms of Alzheimer’s disease

98

Pierre Hainaut and Monica Hollstein

(Roperch et al., 1998). Sixth, p53 induces apoptosis in a transcription-independent manner (Caelles et al., 1994). Proposed mechanisms include repression of helicases such as hERCC2/XPD and hERCC3/XPB (which bind to the C terminus of p53) (Bissonnette et al., 1997; Wang et al., 1996).

3. REPLICATION, TRANSCRIPTION, AND REPAIR p53 interacts with several components of the transcription, replication, and repair machineries to either repress or promote their activity. First, p53 binds to TBP (TATA box-binding protein) and TBP-associated factors TAFII40 and TAFII60 (X. Chen et al., 1993; Farmer et al., 1996a,b; Liu et al., 1993; Seto et al., 1992; Truant et al., 1993; Thut et al., 1995), and thus modulates the activity of general transcription factors controlling initiation at class II promoters. Initially, experiments performed with high amounts of ectopically expressed p53 have reported a capacity of p53 to act as a broad transcriptional repressor. However, recent evidence using better deﬁned experimental systems suggests that p53 may play a more speciﬁc role, either as an activator or a repressor of transcription. p53 also cooperates with the speciﬁc transcription factor WT1 in the transcriptional control of GADD45 (Bissonnette et al., 1997; Maheswaran et al., 1998; Wang et al., 1996; Zhan et al., 1998) and modulates the binding of Sp1 to DNA targets (Bargonetti et al., 1997; Ohlsson et al., 1998). By binding to TFIIH components such as hERCC2/XPD, hERCC3/XPB, and CSB, p53 may play a direct role in modulating nucleotide excision repair pathways (Wang et al., 1995). The p53 protein also stimulates the activity of topoisomerase I and this property is conserved by several mutant p53 proteins associated with cancer (Gobert et al., 1996). In addition, the p53 protein has been reported to exhibit 3⬘-to-5⬘ exonuclease activity and recent evidence suggests that this activity might provide a proofreading function for DNA polymerase alpha (Huang, 1998; Mummenbrauer et al., 1996). Moreover, p53 prevents replication by binding to RP-A (Dutta et al., 1993) and transactivates the proliferating cell nuclear antigen (PCNA) promoter in a concentration-dependent manner, suggesting that a moderate increase in p53 activity induces PCNA production to help in DNA repair (Shivakumar et al., 1995). Finally, p53 is speciﬁcally expressed in the meiotic pachytene phase of normal spermatogenesis and may participate in the control of the DNA reshufﬂing that occurs at this phase of cell cycle (Rotter et al., 1993).

4. DIFFERENTIATION AND DEVELOPMENT There is strong experimental evidence that overexpression of p53 may favor cell differentiation and modulate development. p53 is essential for the development of Xenopus embryo (Amariglio et al., 1997). Inhibition of p53 by ectopic expression of competing proteins blocked the ability of Xenopus

p53 and Human Cancer

99

early blastomeres to undergo differentiation and resulted in the formation of large cellular masses reminiscent of tumors (Wallingford et al., 1997). A fraction of female mice lacking p53 do not develop normally and display a lethal defect in neural tube closure resulting in exencephaly (Armstrong et al., 1995; Sah et al., 1995). Indirect evidence for a role of p53 in development comes from the study of mdm-2-deﬁcient mice, which die early in development, but are rescued and develop normally in a p53-null genetic background (Jones et al., 1995; Leveillard et al., 1998; Montes et al., 1995). The recent identiﬁcation of a rapidly growing family of p53-related genes calls for caution in the interpretation of the consequences of p53 overexpression on differentiation and development. Because these genes appear to be expressed in a restricted, tissue-speciﬁc pattern, it is likely that some of the identiﬁed p53 targets may turn out to be physiologic downstream effectors of p73 or p51/p40/ket.

5. SENESCENCE Ectopic expression of wild-type p53 triggers a rapid senescence program in human tumor cells lacking functional p53 (Sugrue et al., 1997). Normal human ﬁbroblasts generally undergo senescence after several passages in vitro and spontaneous immortalization in vitro is an extremely rare event. Mouse ﬁbroblasts heterozygous for a p53 mutation undergo senescence normally but eventually reenter proliferation, associated with either loss or mutation of the wild-type p53 allele. These data point to a role for p53 in the control of senescence (Wynford-Thomas, 1996, 1997).

6. CONTROL OF GENOMIC STABILITY The various mechanisms described earlier may all contribute to the maintenance of DNA integrity and chromosomal stability, either by enhancing DNA repair or by preventing proliferation of cells with damaged genomes. In vitro, p53 has been shown to protect against drug-selected gene ampliﬁcation, probably by preventing entry in S phase under conditions where DNA strand breaks can occur (Livingstone et al., 1992; Yin et al., 1992). In vivo, there is evidence that tumors from p53-deﬁcient mice exhibit increased aneuploidy (Donehower, 1996a, 1997). Moreover, normal cells from various organs of 4- to 6-week-old p53-null mice display a number of alterations such as aneuploidy, gene ampliﬁcation, multiple centrosomes, abnormally formed mitotic spindles, and increased apoptosis (Fukasawa et al., 1997). Fibroblasts from Li–Fraumeni patients show increased longevity, resistance to low-dose-rate ionizing radiation and accumulated structural chromosome aberrations in up to 100% of cells immediately before senescence (Boyle et al., 1998). These results demonstrate that the multiple downstream effectors of p53 cooperate to maintain the integrity and stability of the genome.

100

Pierre Hainaut and Monica Hollstein

III. MUTATIONS AND VARIATIONS IN THE p53 GENE A. p53 Polymorphisms Several DNA sequence variants (polymorphisms) present in human populations have been discovered, only two of which alter the amino acid sequence of the p53 protein (Buchman et al., 1988; Felley-Bosco et al., 1993; Gerwin et al., 1992; Matlashewski et al., 1987) (Table III). Most polymorphism are localized in introns, at sites that are not splice or branch sites. So far, there is no clear evidence of increased cancer linked to a sequence polymorphism that leaves the p53 protein unaltered (Table IIIA). TheoTable III Polymorphisms of the Human p53 Gene A. Polymorphisms in the p53 coding sequence that do not alter the amino acid sequence of the p53 protein Codon 21 36 213

Allele A

Allele a

GAC CGG CGA

GAT CCA CGG

Prevalence (a) ? 4% to 11%

Reference Ahuja et al. (1990) Felix et al. (1994) Carbone et al. (1991)

B. Polymorphisms in the p53 coding sequence that do alter the amino acid sequence of the p53 protein Codon

Allele A

Allele a

Prevalence (a)

47

CCG (pro)

TGC (ser)

To 1.5%

72

CGC (arg)

CCC (pro)

(Wide range)a

Reference Gerwin et al. (1992) Felley-Bosco et al. (1993) Harris et al. (1986) Beckman et al. (1994)

Examples of polymorphisms in noncoding regions of the p53 geneb Intron 1 1 1 1 2 3 6 7

Polymorphism nt 8545 (BgIII RFLP; T to A) nt 8703 [Alu seq.; (AAAAT)n] HaeIII RFLP VNTR G to C, 38 bp 3´ of exon 2 nt 11951 (16 bp duplication) A to G 61 bp 3´ of exon 6 C to T, and at 20bp 3´, T to G

a Depending on ethnic groups. b

Reference Willems et al. (1992) Futreal et al. (1991) Ito et al. (1994) Hahn et al. (1993) Pleasants and Hansen (1994) Lazar et al. (1993) Chumakov and Jenkins (1991) Prosser and Condie (1991)

For a full listing refer to the IARC p53 database, version R3, January 1999.

p53 and Human Cancer

101

retically, mechanisms by which these “silent” polymorphisms could affect p53 protein function include a synonymous “hungry” codon pairing with an aminoacyl-tRNA in limited supply, enhanced mutability due to altered DNA sequence context, or increased cryptic splicing events and altered transcript stability. However, experimental data on the relevance of such mechanisms for the human p53 gene are lacking. The two genetic variants that lead to protein polymorphisms are Ser to Pro at residue 47 and Arg to Pro at residue 72. The 47 Pro variant is a rare polymorphism affecting a codon conserved in evolution. Residue 72, although not conserved, is located within the proline-rich region and may affect the structure of the putative SH3 domain. Initially, neither functional nor genotyping studies on these polymorphisms offered data on biological differences that would be expected to affect cancer susceptibility (Matlashewski et al., 1987; Felley-Bosco et al., 1993). Sharp ethnic differences in the allele frequencies of codon 72 variants have been observed but there has been little consistent evidence of an association with cancer risk (Beckman et al., 1994; Weston et al., 1992, 1994). Recently, Storey et al. (1998) reported that the 72 Arg variant is more efﬁciently targeted for degradation by E6 proteins of cancer-associated HPV strains than the 72 Pro variant. However, epidemiologic studies to assess the link between this polymorphism and cervical cancer associated with infection by HPV show essentially negative results (Rosenthal et al., 1998). So far, studies investigating the various polymorphisms in relation to cancer risk have found no consistent associations (Olschwang et al., 1991; Zhang et al., 1992; Weston et al., 1994; Birgander et al., 1995; Mazars et al., 1992; Lazar et al., 1993). In several cases, putative associations were challenged by subsequent studies (Kawajiri et al., 1993; Jin et al., 1995; Murata et al., 1996; Runnebaum et al., 1995; Lancaster et al., 1995; Campbell et al., 1996; Peller et al., 1995b; Mavridou et al., 1998).

B. Somatic Mutations Among the human tumor mutations identiﬁed by sequencing, 87.2% are single base substitutions and 12.8% are complex mutations and short deletions or insertions. Missense mutations have been observed at 231 of the 393 codons, including all the codons of the DNA-binding domain except codon 123. This codon (ATC, threonine) is well conserved in evolution, but experimental mutation (to Alanine) at this codon has been shown to activate, rather than suppress DNA-binding activity (Freeman et al., 1994). The vast majority of the mutated codons are recurrent mutation sites that are likely to result in dysfunctional p53. Silent mutations represent up to 3.9% of the mutations in the database and it is possible that mutations occurring at rare

102

Pierre Hainaut and Monica Hollstein

positions may also represent unselected events that are comparable to silent mutations (Strauss, 1997). The DNA-binding domain contains 93% of all mutations identiﬁed to date. This high frequency may be overestimated because, after initial reports that mutations tended to cluster in the central portion of the coding sequence, most investigators have limited their analyses to exons 5 to 8. Insertions, deletions, and stop codons represent 48% of the mutations in the N terminus and 60% in the C terminus, but only 17% in the DNA-binding domain. Mutations are infrequent in two well-deﬁned domains, the N-terminal transactivation domain (residues 1–44, 0.2% of all mutations) and the C-terminal negative regulatory domain (residues 363– 393, 0.02% of all mutations). Both of these domains are, however, relatively conserved in evolution and play an important role in the control of p53 functions (see above). It is important to note that missense mutations have not been detected so far at important regulatory residues such as Ser 15 and 37 (phosphorylation by ATM and/or DNA-PK), Phe 19, Leu 22, Trp 23, and Leu 26 (interactions with mdm-2 and TBP), Ser 315 (phosphorylation by cyclin-dependent kinases), Ser 376 and 378 (interactions with 14-3-3␴), Lys 370, 372, 373, 381, and 382 (sites of acetylation) and Ser 392 (phosphorylation by CKII). These observations suggest that mutation of any single one of these residues is not sufﬁcient to inactivate p53. In the case of N-terminal residues, experimental studies have shown that mutation of at least two of the residues was required to inactivate p53 (Lin et al., 1994). That mutations cluster in the DNA-binding domain indicates that transcriptional activation is an essential biological mechanism of tumor suppression by p53 (Pietenpol et al., 1994). It also illustrates the exceptional structural ﬂexibility of the DNA-binding domain, since mutation of almost any of the residues can disrupt interaction with target DNA sequences. Five of the six most frequently mutated codons encode arginine residues (175, 248, 249, 273 282, representing 25% of all mutations). All play well-deﬁned roles in protein–DNA interactions, either by directly contacting DNA (248, 273) or by contributing to the stability of the DNA-binding surface (175, 249, 282) (Cho et al., 1994). Mutation of these residues inactivates p53 by abrogating essential DNA contacts or by destabilizing the binding surface. However, most of the other mutation sites are scattered in the ␤ sheet scaffold of the domain and probably inactivate p53 by disrupting the global architecture of the DNA-binding domain. Several human mutants are temperature sensitive for DNA-binding activity (including Val 143 to Ala, and Val 272 to Met), suggesting that they induce limited and reversible perturbation of the thermodynamic equilibrium of the DNA-binding domain (Hainaut et al., 1995; Rolley and Milner, 1994; Zhang et al., 1994). These observations have led to the classiﬁcation of mutants into several structural categories, as suggested by several authors (Hainaut, 1995; Rolley and Milner, 1994; Soussi and May, 1996).

p53 and Human Cancer

103

Functional studies show that all mutations are not equivalent. Mutants of p53 differ from each other by the degree of loss of activity toward speciﬁc targets (Ory et al., 1994) and by the capacity to exert dominant-negative effect over wild-type p53 (Zambetti and Levine, 1993; Forrester et al., 1995). These effects may be explained by the fact that the p53 protein does not bind with the same afﬁnity to all p53-responsive elements. The p21waf-1 promoter, for example, appears to contain high-afﬁnity binding sites, whereas the p53-binding sites of the bax promoter are of lower afﬁnity. Several mutants have been identiﬁed that activate p21waf-1 but are defective for activation of target genes involved in the pathway of apoptosis (Ludwig et al., 1996; Rowan et al., 1996; Ryan and Vousden, 1998; Flaman et al., 1998). These mutations are rarely found in tumors (together, they account for less than 0.5% of all mutations).

C. Germline Mutations in p53: Li–Fraumeni Cancer Syndrome Approximately half of the families with the Li–Fraumeni cancer syndrome carry one mutant p53 allele in somatic cells (Malkin et al., 1990; Srivastava et al., 1990; see review in Kleihues et al., 1997). This disease is characterized by familial clustering of various cancers, particularly early-onset breast cancer, sarcomas, leukemias, and brain and adrenocortical tumors. Mice with targeted nonfunctional p53 alleles provide an experimental model for this disorder. Early spontaneous tumors, primarily lymphomas and sarcomas, arise frequently in p53⫺/⫺ and p53+/⫺ mice, reﬂecting the Li–Fraumeni syndrome in these respects. Curiously, mammary and brain tumors are uncommon in p53 knockout mice (Donehower, 1996b; Eng et al., 1997). More than one-third of the 134 Li–Fraumeni families with germline p53 mutations registered in the IARC database carry mutations at one of the six most common mutated residues in sporadic cancers. The similarity in type and distribution of mutations is particularly striking between inherited p53 mutations in Li–Fraumeni pedigrees and somatic mutations found in sporadic brain tumors (astrocytomas and glioblastomas) in the general population. In both instances the most frequently observed mutations are G:C to A:T missense mutations at CpG dinucleotides within sequences encoding conserved residues at the p53–DNA interface (Kleihues et al., 1995, 1997). There is currently no evidence suggesting that certain germline mutations elicit a specific subset of cancer types within the tumor spectrum characterizing the Li–Fraumeni syndrome. Moreover, there is no explanation for the puzzling observation that members of Li–Fraumeni families are not at increased risk for cancers that show a high frequency of p53 mutations in sporadic disease, such as tumors of the upper aerodigestive tract (Kleihues et al., 1997).

104

Pierre Hainaut and Monica Hollstein

In patients from Li–Fraumeni families with an inherited p53 mutation, normal cells are heterozygous (p53 wt/mt), whereas in cancer cells, the wildtype allele is lost or otherwise inactivated by a somatic mechanism. Experimental studies have shown that the wild-type allele retained the capacity to suppress cell proliferation even in the presence of the mutant allele (Frebourg et al., 1992). The penetrance of the mutated gene is close to 100%, suggesting a causal link between the constitutive mutation and the somatic inactivation of the wild-type allele. In keeping with these observations, normal cells from individuals with germline p53 mutations show altered genomic stability (Boyle et al., 1998).

IV. p53 MUTATIONS IN SPORADIC CANCERS: HOST–ENVIRONMENT INTERACTIONS A. p53 Mutation Prevalence 1. FREQUENCY OF POINT MUTATIONS IN VARIOUS TUMORS Mutations in p53 coding sequences are found in almost every kind of human neoplasia by the time the capacity for invasive growth has been acquired (Greenblatt et al., 1994). Malignancies with mutation frequencies higher than 50% include skin cancers (except melanoma), late-stage cancers of the bladder, and most types of carcinomas of the aerodigestive tract. Lesions of the mucosa lining the aerodigestive tract are increasingly likely to carry a mutation as the degree of abnormality increases from mild to severe dysplasia and carcinoma in situ. Lymphomas and tumors of the brain, breast, prostate, and liver show an intermediate mutation frequency (15–35%). Malignancies with low mutation frequency include leukemia (10%), testicular cancer, and malignant melanoma (both less than 5%). It is important to note that the mutation frequency reported in any particular study is inﬂuenced by the techniques used for identiﬁcation of the mutation, as well as the stage of development of the cancer lesions. In most cancers, one allele carries a missense mutation and the other allele is lost (or, more rarely, also mutated) (Baker et al., 1989; Nigro et al., 1989). Loss or inactivation of the intact allele is not always required for disruption of p53 function. A number of mutants show the capacity to inactivate wild-type p53 in a dominant manner. Mutant p53 can oligomerize with wild-type p53 to form an inactive complex (Milner and Medcalf, 1991; Harvey et al., 1995; Inga et al., 1997). Some mutants also have acquired a gain-of-function phenotype that increases oncogenic transformation. The molecular basis of this effect is still a matter of debate (Dittmer et al., 1993;

p53 and Human Cancer

105

Gualberto et al., 1998; Forrester et al., 1995; Harvey et al., 1995; Inga et al., 1997). Cancer types in which mutations are rare include malignancies that have lost p53 function by a molecular mechanism other than point mutation. In cervical cancer caused by oncogenic types of human papilloma viruses, the p53 gene is often wild type. However, the protein is inactivated by binding of the viral E6 protein, in association with the cellular protein E6AP, which targets p53 for rapid, proteasome-mediated degradation (Huibregtse et al., 1993a; Maki et al., 1996; Molinari and Milner, 1995; Scheffner et al., 1994; Talis et al., 1998). In soft tissue tumors, the MDM-2 gene is ampliﬁed and overexpressed without evidence of p53 mutation in about 30% of the cases, leading to destabilization and inactivation of the transcriptional capacity of p53 (Oliner et al., 1992). Ampliﬁcation of MDM-2 has also been observed in other tumors, but is not always conﬁned to tumors with wild-type p53 (Momand et al., 1998). p53 mutations are also uncommon in mesothelioma, and interaction with the large T antigen of the simian virus SV40 has been proposed as an alternative mechanism of inactivation in mesothelial tumors (Carbone et al., 1997; Wiman and Klein, 1997). Alternative mechanisms of p53 inactivation are likely to exist. For example, p19arf, the alternative product of the CDKN2/INK4 tumor suppressor locus, interacts with mdm-2 and stabilizes p53 by neutralizing its mdm-2mediated degradation. The CDKN2/INK4 locus also encodes p16, a cyclindependent kinase inhibitor that controls the phosphorylation of Rb, and thus plays a central role in the control of both p53 and Rb pathways. This gene is relatively frequently mutated in some cancers and most of the mutations affect exon 2, which is common to both p16 and p19arf (Pollock et al., 1996). It remains to be determined whether inactivation of p19arf is an alternative to p53 mutation in cancer. Other modes of inactivation of structurally normal p53 have been proposed in testicular cancers, but the mechanism is unknown (Lutzker and Levine, 1996). Loss of function due to cytoplasmic retention has been observed in neuroblastoma (Moll et al., 1995, 1996) and in inﬂammatory breast cancer (Moll et al., 1992). A similar mechanism may be at work in some hepatocellular carcinomas associated with infection by hepatitis B virus (HBV), since HBx protein expressed in the liver of transgenic mice binds p53 and blocks its entry into the nucleus (Ueda et al., 1995). Many studies in recent years have been directed toward the identiﬁcation of mutations in downstream effectors of p53 in tumors with wild-type p53 gene, but few correlations have been discovered. Intensive sequencing efforts to identify inactivating mutations in WAF1 have produced largely negative results (Bhatia et al., 1995; Shiohara et al., 1994). Frameshift mutations have been observed in the BAX gene in a signiﬁcant proportion of colon and ovarian cancers with microsatellite mutator phenotype (Colella et al., 1998;

106

Pierre Hainaut and Monica Hollstein

Rampino et al., 1997). However, these alterations are rare in other tumors (Chou et al., 1996; Bilim et al., 1998). So far, the search for point mutations in the p53 relatives p73 and p51/p40/ket has also produced mostly negative results (Osada et al., 1998). Given the broad spectrum of p53 activities as a tumor suppressor, it is somewhat surprising that many cancers retain an apparently normal and functional p53 gene. As additional mechanisms of p53 inactivation are discovered, it may turn out that p53 function is disrupted in most, if not all, cancers by the time the capacity for invasive growth has been achieved (Kaelin, 1998). Alternatively, selection of a p53 mutation in tumorigenesis may depend on the cell context and on the extent to which wild-type p53 forms a rate-limiting step in the control of proliferative life span in that cell (Wynford-Thomas and Blaydes, 1998).

2. ROLE OF EXOGENOUS AND ENDOGENOUS FACTORS Besides inherent differences from one cell type to another, environmental factors are likely to contribute to the variable frequency of p53 mutations found in human cancers. The vast majority of head and neck cancers and lung cancers of smokers harbor a p53 mutation, whereas in cancers of nonsmokers the mutation frequency is around 20% (Brennan et al., 1995; Hernandez-Boussard and Hainaut, 1998; Kondo et al., 1996; Montesano et al., 1996). A high frequency of p53 point mutations in tumor cells of smokers is not surprising in view of the potent mutagenic activity of tobacco smoke. However, there are types of cancer in which p53 mutation is frequent without any evidence to date of the involvement of external mutagenic risk factors. For example, in astrocytomas and glioblastomas, p53 mutations are frequent and play an important role in disease progression, but it is not known whether they have an endogenous or exogenous origin (Kleihues et al., 1995). There is no doubt that mutations arise spontaneously in all organisms, and there are well-charted mechanisms by which they occur in mammalian cells (Lindahl, 1993). However, it is difﬁcult to estimate which mutations arise spontaneously in humans in vivo, and their frequencies in various genes, cell types, and differentiation states. Current knowledge is based largely on analysis of point mutations in human lymphocytes at the HPRT locus or in other genes that allow for phenotypic selection (Cariello et al., 1998). Recently, the design of high-throughput, sensitive techniques for mutation detection has allowed more direct approaches for revealing the presence of occasional mutants within a large population of wild-type p53 sequences (Aguilar et al., 1993; Cha et al., 1992; Flaman et al., 1995; Waridel et al., 1997). In particular, cell patches containing mutant p53 have been found in histologically normal skin and in the epithelium of the aerodigestive tract

p53 and Human Cancer

107

(Jonason et al., 1996; Ren et al., 1996a,b, 1997; Waridel et al., 1997). These results show that mutant cells are far more prevalent than would be anticipated from in vitro studies of mutation frequencies at the HPRT locus or other test loci (Loeb, 1991, 1998). Furthermore, mutation rates may vary considerably according to growth circumstances (Richards et al., 1997), and tumor cells may acquire a mutator phenotype during rounds of selection (Mao et al., 1997) and become hypermutable (Loeb, 1991; Strauss, 1992; Loeb and Christians, 1996). Conceivably, the p53 gene is inherently more vulnerable to point mutation than other genes, a conjecture suggested by the surprisingly high number of silent p53 tumor mutations (3.9% of all mutations) (Strauss, 1992, 1997, 1998), the plethora of different mutant p53 clones in the skin of basal carcinoma patients (Ponten et al., 1997a,b), and the high density of p53 “patches” (i.e., p53-immunopositive cell clusters, a considerable proportion of which have a mutant p53 gene) in normal skin of sun-exposed individuals (Jonason et al., 1996).

B. Type and Localization of Most Frequent Mutations Human cancers show differences in the type and localization of p53 mutations. In some well-characterized instances, the differences are conspicuous and have led to important insights on the nature of the mutagenic events involved in the etiopathogenesis of these cancers (see later discussion). These differences notwithstanding, generally the same large panoply of p53 mutants has been found in each major type of cancer. Of the 146 codons that are mutated in colon cancers, 123 (84%) are also codons where lung tumor mutations have been found. This extensive overlap suggests that the carcinogenic properties of a given mutant may not vary importantly from one tissue to another. There are isolated examples, however, suggesting that a given mutant may express tissue-speciﬁc tumorigenic properties. For example, the most common mutations at codon 248 are transitions resulting in substitution of Arg by either Gln or Trp, and the two mutants arise at equal frequencies in most cancer types. However, in hematologic malignancies, the Gln variant is much more frequent than the Trp variant. In addition, experimental studies have shown that the capacity of mutant p53 to inactive wildtype p53 in a dominant-negative manner varies signiﬁcantly from one cell type to another. An analysis of mutable sites in the DNA-binding domain of human p53 has been performed using a yeast functional assay. This assay is based on the expression of human p53 introduced into Saccharomyces cerevisiae containing a p53-dependent reporter gene. Random mutagenesis has revealed that mutation at 542 sites in a 1079-bp fragment could inactivate p53 DNA binding. This number is remarkably close to the number of observed sites at

108

Pierre Hainaut and Monica Hollstein

which missense or nonsense mutations occur in human cancer (457) (Flaman et al., 1994). This observation suggests that tumor mutations are found at almost every possible site that is important for p53 binding to DNA. In human tumors, the pattern of p53 mutations is dominated by base transitions (C to T or G to A) at CpG dinucleotides (23% of all mutations). Although this type of mutation has been observed at 35 different codons, 90% of human CpG transitions are at one of six “hotspot” codons 175, 213, 245, 248, 273 and 282 (Fig. 3A) known to be important in maintaining p53 biological activity. CpG dinucleotides are sites of cytosine methylation, and deamination of 5-methylcytosine producing thymine is the most characteristic event generating spontaneous mutations in mammalian cells (Barker et

Fig. 3A Role of exogenous and endogenous factors in shaping tumor mutation patterns. “Mini-hotspots” in the p53 coding sequence. Distribution of four mutation types in the coding sequence, GC to AT transitions at CpG (representing 23% of all mutations), GC to TA transversions (15%), GC to CG (8%), and AT to GC (12%). CpG transitions primarily appear as the consequence of an endogenous mutation mechanism and form well-deﬁned mutation “hotspots.” Because of these major hotspots, the mini-hotspots corresponding to other mutation types are difﬁcult to detect. Only sites that contain more than 1% of a speciﬁc mutation type are shown. Hotspot codons for each mutation type are indicated. Some of the nonCpG hotspots are well-characterized sites of alteration induced by exogenous carcinogens (see Fig. 5).

p53 and Human Cancer

109

al., 1984; Ehrlich and Wang, 1981; Rideout et al., 1990). In nine different human tissues and cell lines, the p53 sequences along exons 5–8 were found to be completely methylated at every CpG site, suggesting that tissue-specific methylation does not contribute to the differential mutation distribution at CpG sites seen in tumors. Biological selection of mutation at critical sites for p53 function and the intrinsic mutability of a sequence are thus both important hotspot determinants. Cancers with a high frequency of CpG transitions include colorectal cancer (45%), adenocarcinoma of the esophagus (48%) and stomach (34%), tumors of the brain (35%) and uterus (35%), and some hematopoietic malignancies (e.g., Burkitt’s lymphoma (32%) (Fig. 3B). In colorectal carcinogenesis, p53 mutation is typically a relatively late event that correlates with the acquisition of an invasive potential (Fearon and Vogelstein, 1990). Acquisition of a CpG mutation in this context may be a consequence of deregulated cell proliferation, DNA instability, or altered nitric oxide metabolism. In the development of adenocarcinoma of the esophagus, however, mutations are often detectable in the preneoplastic lesion, Barrett’s mucosa (Montesano et al., 1996). In the latter case, it has been speculated that

Fig. 3B Inverse relationship between CpG transitions and GC to TA transversions. Tumors have been classiﬁed according to the frequency of GC to TA transversions (increasing from top to bottom). Tumors with low frequency of GC to TA transversions show a high frequency of CpG transitions, and vice versa. Cancer types with low CpG transitions and high GC to TA transversions are associated with exogenous risk factors.

110

Pierre Hainaut and Monica Hollstein

the acquisition of a CpG mutation is favored by tissue-speciﬁc endogenous or exogenous factors in addition to the inherent mutability of this dinucleotide. Several factors may increase the formation of CpG transitions. They include oxidative stress, in particular nitric oxide, which enhances the rate of spontaneous deamination of 5-methycytosine. Interestingly, the expression of nitric oxide synthase 2 (iNOS) is negatively regulated by p53 and defective p53 function leads to increased intracellular levels of nitric oxide (Ambs et al., 1998; Forrester et al., 1996). Evidence that NO levels have an inﬂuence on the frequency of tumor CpG mutations in p53 is, however, circumstantial. In colon, iNOS activity increases with tumor progression (Geller et al., 1993; Singer et al., 1996; Thomsen et al., 1995; Wink et al., 1991, 1998). However, there is no major difference in the formation of CpG mutations in tumor tissue compared to normal mucosa (Schmutte et al., 1995, 1997). Another mechanism that may increase the frequency of CpG transitions is defective repair of G:T/U mismatches. The recent cloning of the human G/T mismatch-speciﬁc thymine-DNA glycosylase should help to clarify whether tissue-speciﬁc repair capacities have major effects on CpG transition prevalence (Neddermann et al., 1996). The inﬂuence of diet in the generation of CpG base transitions has also been considered. However, entries in the p53 mutation database from studies of colon cancer patients do not display distinct mutation patterns in different geographic regions despite divergence in dietary habits and dissimilar cancer risk. Other sources of naturally occurring mutations besides deamination of 5methylcytosine include spontaneous base hydrolysis (103 to 104 events/ genome/day), base alkylation, and unrepaired errors introduced during DNA replication (Kunkel, 1990; Lindahl, 1993; Xiao and Samson, 1993). Insertion and deletion mutations occur during replication from misalignment of complementary bases on template and nascent strands. DNA sequence context is the most important factor in determining the speciﬁcity of insertions and deletions. In addition, bulky adducts on speciﬁc bases can stabilize misalignments and increase frameshift mutagenesis (Bintz and Fuchs, 1990; Kunkel, 1990; Ripley, 1990). A survey of the p53 database shows that deletions and insertions account for approximately one tumor mutation in 10 (from 7% to 20% of all mutations, depending on the type of cancer). Analysis of the most common sites of insertion or deletion corroborates predictions from the DNA polymerase slippage/misalignment model. Most base gains or losses occur at repetitive sequences (such as monotonic base runs, adjacent or nonadjacent repeats of short tandem sequences, palindromes, and runs of purines or pyrimidines). These lesions usually result in the deletion or insertion of a single base pair or repeat element, and thus a frameshift mutation (Greenblatt et al., 1996; Jego et al., 1993).

p53 and Human Cancer

111

The ubiquitous features of p53 mutation spectra provide insights into spontaneous p53 mutations patterns in human tissues in vivo. At ﬁrst look, only heavy exposure to mutagenic carcinogens appears to produce an overlay of induced mutations of sufﬁcient magnitude to cause noticeable, statistically signiﬁcant shifts in what may represent the spontaneous pattern (Cariello et al., 1994a,b). At milder exposures, or in the presence of mixed exposures as would be expected in most human situations, the spontaneous pattern may hide clues about induced mutation proﬁles. For example, A:T to G:C mutations peak at codons 163, 179, and 220, sites that do not attract attention when all kinds of tumor mutations are mapped together, because of the looming CpG site transitions (Fig. 3A). The G:C to C:G base substitutions cluster in four conﬁned hotspots around codons 132, 158, 249, and especially 280 (codons 275–283), where about 25% of these mutations are found. The G:C to T:A transversions show a more punctual distribution of frequently mutated sites, peaking at codons 157, 176, and 249. The high prevalence of G to T substitutions at codons 157 and 249 has been reported in numerous, independent studies, and these codons also contain sequences targeted by human carcinogens that induce G to T transversions (Fig. 4; see also later discussion). Most of the mutations described at codon 176, however, are contributed by one single study on esophageal cancer from China, and this ﬁnding was not conﬁrmed in a subsequent study on a similar cohort. Therefore, whether codon 176 is a genuine hostpot for G:C to T:A transversions is still open to question. For a given tumor type, there is typically an inverse correlation between the proportions of G:C to T:A transversions and transitions at CpG sites (Fig. 3B). In particular, cancers associated with signiﬁcant exposure to mutagens in high-risk populations show a high proportion of G:C to T:A transversions and a low proportion of CpG transitions. As more speciﬁc and detailed information on mutation patterns in relation to speciﬁc exposures becomes available, a closer examination of the p53 mutation spectra may reveal the existence of “mini hotspots” that have been overlooked so far.

C. Role of Environment in Shaping Tumor Mutation Patterns Despite the common features of mutation proﬁles discussed earlier, the assembly of p53 mutation data provides several contrasting patterns that are statistically signiﬁcant in a number of well-deﬁned tumor types. Factors that inﬂuence the formation of distinctive mutation patterns can be seen as a succession of “ﬁlters” that allow for acquisition and persistence of particular mutations. First, the metabolic capacity of exposed cells and tissues largely determines the initial extent of DNA damage induced by a given mutagen.

112

Pierre Hainaut and Monica Hollstein

Second, the type of the damage caused by a mutagen can be speciﬁc in its nature and the DNA sequences at which it occurs. Third, the inherent capacity of cells to repair DNA corrects most alterations. Fourth, biological selection favors cells that have acquired a proliferative advantage as a consequence of the mutation. A tenable interpretation of a mutation pattern would need to take in account the contribution of these successive ﬁlters. The most distinctive mutation patterns have been observed in studies on speciﬁc high-risk populations exposed to prodigious levels of mutagens, and in general the spectrum of p53 mutations is in keeping with mutation patterns generated experimentally by the suspected agents. An “induced” mutation proﬁle is suspected when the following features appear: (1) tumortype or exposure-group-speciﬁc “hotspot” mutations other than CpG site transition mutations, (2) a strong strand bias in the orientation of the purine:pyrimidine base pairs at mutation sites, and (3) an unusual predominance of a particular type of base substitution. Strand bias in mutation patterns arises from lesions that are efﬁciently removed by repair processes brought forth by stalling of the transcription complex at a modiﬁed DNA base (Hanawalt and Mellon, 1993). For example, the damage formed by benzo(a)pyrene (an important mutagen from tobacco smoke), and lesions induced by UV radiation are less efﬁciently repaired on the nontranscribed strand of the p53 than on the transcribed strand of the p53 gene (Denissenko et al., 1998; Greenblatt et al., 1996). There are striking variations in strand bias among different base substitutions and among different tumor types (Fig. 4). In particular, mutations at G:C base pairs generate an interesting pattern: G to T transversions show a strong bias in most of the major cancer types, whereas there is only subtle, if any, strand bias for G to A transitions, particularly at CpG sites. Ninetytwo percent of G to T mutations in lung cancer occur at guanines located on the nontranscribed strand. This is what a large body of experimental mutagenesis data would predict if these mutations arose from the known G:C to T:A-inducing mutagens in tobacco smoke. A similar prominent strand bias in G to T transversions (90%) is seen in pancreatic cancers. A:T to G:C mutations have a more variable pattern, with essentially no bias in pancreatic cancers, in sharp contrast to lung and bladder cancers. One tentative explanation for this would be that a signiﬁcant fraction of A to G mutations in lung or bladder cancer arises from premutagenic, transcription complexstalling adenine adducts. In contrast, in pancreatic cancer, A to G mutations may arise essentially from spontaneous processes such as base hydrolysis or polymerase errors. Alternative explanations can also be considered, for example, that risk factors for pancreatic cancer may include agents that damage adenine and thymine equally, whereas in lung and bladder, the premutagenic damage from tobacco carcinogens occurs more frequently on adenine than on thymine. Such interpretations are speculative and more needs

p53 and Human Cancer

113

Fig. 4 Percentage of tumor mutations at p53 base pairs with purine on the nontranscribed strand. The strand bias of several types of mutations affecting purines is shown in several common human cancers. A strong strand bias is indicative of a possible perturbation of the transcription–repair complex at an adducted DNA base. CpG transitions show almost equal distribution on both strands. GC to TA transversions show a strong strand bias in most cancers. The strong bias of AT to GC transitions in lung and bladder cancer is a clue to the involvement of carcinogens in the genesis of these mutations.

to be learned about the tissue speciﬁcity of carcinogen metabolism, types of DNA damage that efﬁciently block transcription, DNA repair pathways, and about how cancer risk factors may affect these processes. The few mutation patterns that are unquestionably exposure-speciﬁc demonstrate that carcinogens can induce mutations in human cancer genes, which in turn elicit physiologic effects associated with malignant growth (Fig. 5). Four well-documented examples are summarized later (smoking and lung cancers; sunlight and nonmelanoma skin cancers; aﬂatoxins and hepatocellular carcinoma; tobacco, aromatic dyes, infection, and bladder cancer). Finally, the mutation spectrum of breast cancer is discussed as an illustration of present difﬁculties in gaining a better understanding of the origins of p53 mutations.

1. SMOKING AND LUNG CANCER More than 1000 p53 mutations found in lung tumors and biopsies are compiled in the p53 mutation database (Hernandez-Boussard and Hainaut,

114

Pierre Hainaut and Monica Hollstein

Fig. 5 Colocalization of DNA-base modiﬁcations induced by carcinogens, 5-methylcystosine residues, and tumor mutation hotspots in the p53 gene. p53 domains conserved in evolution are shaded (I–V). Portions of the coding sequences containing the major tumor mutation hotspots are given below. CpG sites are in bold type. Codons corresponding to DNA-binding residues of the p53 protein are underlined. Tumor hotspots are indicated above the DNA sequences. Tumor types: HCC (hepatocellular carcinoma); lung (primarily squamous cell, adeno-, large-cell, and small-cell carcinomas); colon; skin (primarily squamous and basal cell carcinomas; including tumors of Xeroderma pigmentosum patients). Numbers in brackets indicate the percent of total p53 mutations in that tumor type occurring at the designated codon. Sites where DNA is heavily modiﬁed by carcinogenic exposure are indicated below the DNA sequences shown in this diagram (Denissenko et al., 1996, 1997; Puisieux et al., 1991; Tornaletti et al., 1993). B, benzo(a)pyrene-diol-epoxide; Af, aﬂatoxin B1; UV, ultraviolet light.

1998). Mutations are detected in 50–80% of cases comprising all major histologic subtypes of lung cancers from smokers, with the lower prevalence being observed in adenocarcinoma. One-third of the base changes are G to T substitutions, which constitute only 13% of mutations in all other cancers combined. These transversions frequently occur at speciﬁc bases at codons 157, 158, 248, 249, and 273, show a strong strand bias (92% on the nontranscribed strand), and are uncommon in most other malignancies. Codons 157 and 158 are rarely mutated in cancers other than lung. Codons 273 and 248 are frequent mutation sites in all cancers. In all cancers combined except lung, 92% of these mutations are CpG transitions, whereas in lung tumors 50% of the mutations are G to T transversions, resulting in a substitution of Arg to Leu, a rare mutant in other cancers. Experimental studies

p53 and Human Cancer

115

with cultured primary bronchial cells have shown that hotspots for G to T transversions in lung cancers coincide with bases where metabolites of the tobacco carcinogen benzo(a)pyrene (BaP) form adducts in vitro (Denissenko et al., 1996). Metabolites of BaP bind preferentially to guanines next to a methylated cytosine, explaining the coincidence between target codons for adduct formation (codons 248 and 273) and frequent sites of CpG transitions (Denissenko et al., 1997). Although BaP is only one of many tobacco mutagens, it is thought to be the main inducer of G to T transversions in tobacco smoke, and the p53 mutation proﬁle in lung cancers from smokers is consistent with the one induced by tobacco smoke in an experimental microorganism (Salmonella) (DeMarini, 1998). The p53 mutation database contains only 36 mutations reported in individuals explicitly identiﬁed as nonsmokers. G to T transversions are infrequent (10%) whereas G:C to C:G transversions are unusually high (33%). These observations conform to the notion that the high frequency of G to T mutations in smokers is speciﬁcally associated with tobacco exposure. However, there are not enough data to warrant a detailed analysis of the mutation spectrum in tumors from nonsmokers. It is interesting to note that the high frequency of G to T mutations is also observed in tumors of the larynx but not in other forms of nonlung cancer associated with exposure to tobacco, such as cancer of the esophagus, oral cavity, and bladder (see later section). This observation suggests that the mutagenesis of p53 by metabolites of BaP occurs in a tissue-speciﬁc manner.

2. AFLATOXINS AND HEPATOCELLULAR CARCINOMA Two highly contrasting patterns of p53 mutations in HCC are seen in highrisk and low-risk populations. In areas of the world where HCC incidence rates are high (⬎40/100,000/year) and where diagnosis in the third decade of life is not uncommon, the p53 mutation proﬁle is dominated by a single codon 249 substitution (AGG to AGT, Arg to Ser), which represents up to 90% of all HCC mutations. Approximately half of the HCCs from such areas carry a mutant p53 (Bressac et al., 1991; Hsu et al., 1991; Ozturk, 1991). In these populations (e.g., in Qidong and Guangxi, China, in Senegal, or in Mozambique), the two most important and synergistically interacting risk factors are chronic HBV infection and ingestion of mold toxins, particularly aﬂatoxins. At the other extreme, in low-risk areas (less than 10 cases/ 100,000), roughly 10–15% of tumors contain a mutation and the base changes are diverse in kind and location, with GC to AT transitions in the majority. This pattern may resemble a proﬁle of spontaneous mutations in hepatocytes. Intermediate risk groups show a spectrum of mutations that falls between the two extremes (recently reviewed by Montesano et al., 1997).

116

Pierre Hainaut and Monica Hollstein

There is a direct association between the extent of exposure to aﬂatoxins in the diet and the prevalence of codon 249 Arg to Ser mutation. The most plausible interpretation of this association is that aﬂatoxins directly induce these mutations. This interpretation is supported by experimental studies showing that aﬂatoxin B1 binds at this codon and, preferentially induces this hotspot mutation in human cells in vitro. Moreover, the 249 Arg to Ser mutation is detected in the normal liver of aﬂatoxin-exposed individuals (Aguilar et al., 1993, 1994). HBV infection is a prominent risk factor wherever HCC surfaces, but the mechanism or extent to which the virus inﬂuences the prevalence or pattern of p53 mutation is still under discussion. There is evidence that the viral HBx protein affects transcription-coupled repair through interaction with p53 and also promotes the sequestration of the p53 protein in the cytoplasm (Harris, 1996c; Ueda et al., 1995; see earlier section). Inactivation of p53 function by HBV proteins could in theory eliminate a growth or survival advantage from gene mutation, yet patients in high-risk groups have both HBV DNA and p53 mutation. One possibility to explain this paradox is that interaction with HBV proteins decreases only some p53-mediated suppressive functions. Alternatively, the codon 249 mutation may possess biological properties that are essential in the context of hepatocarcinogenesis in HBVassociated cases (Ponchel et al., 1994).

3. SUNLIGHT AND SKIN CANCERS There is strong experimental and epidemiologic evidence linking UV radiation to the development of skin cancers (recently reviewed by Holmquist and Gao, 1997). p53 mutations are common in all skin cancers except melanoma, in which p53 appears to be inactivated by alternative, unknown mechanisms. The p53 mutation spectrum in nonmelanoma skin cancers (NMSC) shows a high frequency of C to T transitions (56% of all mutations), including tandem CC to TT transitions (6% of all mutations), a type of mutation that is extremely uncommon in any other tumor type. Both types of mutation are consistent with the mutagenic effects of UV in vitro. In particular, tandem CC to TT mutations are formed as a result of inefﬁcient repair of a common photoproduct, cyclobutane pyrimidine dimers. Exposure to UV radiation from sunlight is the major cause of human skin cancer. Skin tumors from patients with Xeroderma pigmentosum, a DNArepair deﬁciency associated with increased sensitivity to UV, show a particularly high frequency of these CC to TT transitions (Dumaz et al., 1993). Mutations at dipyrimidines have been observed in the normal skin of sunexposed skin cancer patients ( Jonason et al., 1996; Nakazawa et al., 1994; Ren et al., 1996a). The localization of mutations in skin shows striking differences with other types of cancers, with hotspots at codons 177–179 and

p53 and Human Cancer

117

at codon 278. To date, there is no evidence that these mutations are selected on the basis of particular biological properties in the skin. On the contrary, accumulation of these mutations is probably due to exceptionally slow repair of UV-induced lesions at codons 177, 196, and 278 (Tornaletti and Pfeifer, 1994). Half of the CC to TT transitions observed are within CpG sequences and recent evidence indicates that absorption of near-UV by 5methylcytosine is 5- to 10-fold higher than by cytosine (Denissenko et al., 1997). Because UV induces mutations in p53 in the normal skin cells of the skin, these alterations may represent a very early event in skin carcinogenesis (Ziegler et al., 1994). Surprisingly, however, a meticulous analysis of normal skin reveals that mutations can arise in normal skin cells without affecting their morphology and, apparently, their behavior (Jonason et al., 1996; Ren et al., 1996a, 1997). Mutation of p53 alone may thus not be sufﬁcient for initiation of skin carcinogenesis, and other genetic alterations are likely to be required before phenotypic changes associated with tumor development are observed.

4. TOBACCO, AROMATIC DYES, INFECTION, AND BLADDER CANCER Tobacco smoking, occupational exposure to certain chemical dyes, and inﬂammatory reactions to parasitic and other infections account for the majority of the bladder cancer cases in the world. The burden of DNA sequence alterations is expected to increase in cells exposed to mutagenic agents from these exposures, such as 4-aminobiphenyl and other aromatic amines derived from tobacco smoke and dye mixtures, or nitric oxide released during inﬂammatory responses. Forty to ﬁfty percent of bladder tumors contain mutant p53 alleles, and in a number of cases two distinct, nonsilent point mutations have been found in the same tumor. Mutagenic arylamine-derived products have been found in the urine and in the urothelium of smokers (Vineis et al., 1996). Because tobacco products and occupational exposures in the dye industry have mutagenic components in common, it is not surprising that the pattern of mutations in smokers shows considerable overlap with the one in occupationally exposed workers. In addition, chemical dye workers whose tumors have been examined by sequencing were mostly tobacco smokers (Esteve et al., 1995; Sorlie et al., 1998; Taylor et al., 1996; Yasunaga et al., 1997). Arylamines can induce a number of DNA lesions, in particular transversions at G bases (Essigmann and Wood, 1993). Recently, 4-aminobiphenyl, one of the major arylamines to which both smokers and chemical dye workers are exposed, has been shown to induce preferentially G to C transversions (Verghis et al., 1997). Bladder cancer shows the highest frequency of G:C to C:G mutations of all

118

Pierre Hainaut and Monica Hollstein

tumors listed in the p53 mutation database. The distribution of the 423 registered mutations in bladder cancer (primarily transitional cell carcinoma) shows hotspots at codon 280 (G AGA GA) (4.3% of the mutations, mostly G to C), and codon 285 (A GAG GA) (4.5% of the mutations, mostly G to C). G to A and G to T mutations are also found in bladder tumors, and the latter show a strong strand bias rivaling that described for lung cancer (Fig. 4), suggesting that they may have originated from bulky adduct premutagenic lesions. Other sites where mutations cluster include codon 249 (3%), codon 271 (2.5%), and two sites within exon 5 (codons 132–135, 3%; codons 151–158, 6.6%). Bladder cancer mutations thus show characteristics typical of mutations induced by exogenous DNA-damaging agents. Mapping of 4-aminobiphenyl (4AB) adduct sites along the p53 gene and further analysis of 4AB-induced mutations in experimental systems would provide additional clues regarding the role of 4AB in shaping this tumor mutation spectrum. Overall, the prevalence of CpG mutations is low (18%), except in tumors from a region of endemic schistosomal infection (predominantly squamous cell carcinomas) (Warren et al., 1995). This high prevalence of CpG transitions may be related to increased levels of nitric oxide in infected tissues. Comparison of mutations in lung and bladder cancers of smokers reveals that the same complex mixture, tobacco smoke, can induce different types of mutations in different tissues. Differences in exposure routes, tissue-speciﬁc metabolic pathways, and distribution of exposure components and their metabolites are expected to account for dissimilarities. DNA adduct measurements in various tissues following exposure to tobacco smoke or smoke condensate indicate there are shifts in the predominant tobacco smoke-derived DNA damaging agents in various organs.

5. BREAST CANCER: SEARCHING FOR CLUES About 850 mutations in breast cancers are listed in the p53 mutation database but despite this large collection, surprisingly little regarding the sources of these DNA changes can be read from these data. The heterogeneity of breast cancers, which may be seen as several distinct pathologies, may contribute to the difﬁculties in interpretation. Also, the prevalence of mutations is relatively low (20–30%), so the proportion of spontaneous versus (putatively) induced mutations may be high. Mutations are associated with the most aggressive tumor types, but there are no clear data to indicate that mutations are restricted to a subtype of breast cancer. Despite the somewhat low prevalence of mutations, p53 protein levels are elevated in a majority of the cancer cases examined, leading to the hypothesis that p53 protein function could be inactivated or otherwise deregulated by mechanisms other than mutation. However, functional analysis of primary breast tumor cells containing

p53 and Human Cancer

119

wild-type p53 alleles showed that the protein retained the capacity to accumulate and to transactivate p21waf-1 in response to DNA damage, suggesting that p53 is still at least partially functional (Guillot et al., 1997). Overall, the pattern of p53 mutation in breast cancer does not show any striking characteristics, apart from a relatively high prevalence of insertions, deletions, and nonsense mutations (together, 25%). The most frequent type of mutation is GC to AT transitions (40%), and they affect CpG and nonCpG sites equally. Comparisons among cohorts reveal differences in the nature, localization, and frequency of the mutations, but these studies concern small groups of patients and need to be substantiated on larger groups (Blaszyk et al., 1996). Breast cancer is one of the types of cancer that frequently arises in Li–Fraumeni syndrome families (Kleihues et al., 1997). The mutations found in Li–Fraumeni patients with breast cancer could be representative of background mutations that arise spontaneously in breast cancer. Comparison with sporadic breast cancer shows that two types of mutations, G to T and G to C transversions, have not been found in breast cancer patients in Li–Fraumeni families and thus could be speciﬁc for somatic breast cancer. These transversions represent only 18% of all breast cancer mutations. However, they show a strong strand bias and occur at sites also mutated in lung cancers from smokers (G to T transversions, codons 157, 248, 249 and 273) or in bladder cancers from smokers and/or dye-exposed workers (G to C transversions, codons 158 and 280). Overall, these data raise the possibility that most of the mutations in breast cancer might have a spontaneous origin, but that a small proportion of them bear signatures that suggest the involvement of exogenous carcinogens such as those present in tobacco smoke. These proposals await scrutiny by the analysis of large cohorts with deﬁned exposure histories.

V. CLINICAL RELEVANCE OF p53 MUTATIONS Potential clinical applications of the detection of p53 mutations in human tissues are being intensively studied (Lane, 1998; Sidransky and Hollstein, 1996). Despite a plethora of publications, conclusions that can be drawn are still limited, particularly with regard to prognosis, or are valid for only speciﬁc tumor types. A recurrent problem is that many studies are small scale and that little information is published on the exact typing of pathologies and therapeutic protocols. In the future, the development of new, reliable techniques for high-throughput mutation detection is likely to lead to larger studies where p53 mutations will be studied in parallel with other genetic alterations.

120

Pierre Hainaut and Monica Hollstein

A. p53 Mutations as Markers of Clonality Mutations in p53 are useful markers of tumor clonality. They have been used as such to compare, in individual patients, separate clusters of tumor cells from the same lesion, or multiple lesions arising in the same tissue (Franklin et al., 1997; Ponten et al., 1997a; Waridel et al., 1997; Jonason et al., 1996). Mutations are also useful for the follow-up of patients for detection of minimal residual disease, in comparison between primary and recurrent tumors and in tracing the origin of distant metastases (Franklin et al., 1997). Keep in mind, however, that in many cancers, p53 mutations appear to be a relatively late event in the sequence of genetic alterations that lead to tumor progression. Thus, the ﬁnding of different p53 mutations in separate clusters of a tumor does not exclude the possibility of a clonal origin and merely indicates that the two clusters might have diverged at some point in tumor progression.

B. Detection and Follow-Up of Cancer Lesions Identiﬁcation of p53 mutations may be of interest to identify early lesions that are at a high risk of malignant evolution, particularly in tumors where p53 mutation is considered as an early event. For example, in esophageal cancer detection of a mutation in a dysplastic lesion may be considered as an indicator of high risk of malignant transformation (Montesano and Hainaut, 1998). Circulating antibodies to p53 have been found in a subset of cancer patients (Angelopoulou et al., 1994; Crawford et al., 1982). In rare instances, these antibodies have been found even in blood samples collected months to years before cancer diagnosis (Trivers et al., 1995; Cawley et al., 1998; Lubin et al., 1995). The proportion of cancer patients with circulating p53 auto-antibodies is usually low (3–15%), but for some types of cancer this proportion is clearly higher. For example, in esophageal cancers (both squamous cell carcinoma and adenocarcinoma), it can be estimated that at least 25% of the patients are seropositive at time of surgery (Cawley et al., 1998; von Brevern et al., 1996). For a particular malignancy, the percent of patients reported to have circulating p53 auto-antibody can vary considerably from one study to another, probably attributable in part to the detection methods and protocol employed. Although there is a strong association between presence of a mutation in the tumor and humoral immune response, the auto-antibodies are directed against the (unmutated) N- and Cterminal domains of the protein rather than against the (mutated) DNA-binding domain (Lubin et al., 1993; Schlichtholz et al., 1994). The biological mechanisms leading to antibody production, the basis of interindividual differences, and the prognostic signiﬁcance, if any, are still far from understood.

p53 and Human Cancer

121

Recently, several studies have reported success in retrieving DNA from serum or plasma as a surrogate material to detect mutations (Blomeke et al., 1997; Chen et al., 1996; Nawroz et al., 1996; Sanchez-Cespedes et al., 1998). In 10 colon cancer cases with a p53 mutation in the tumor, the identical mutation was detected in seven corresponding serum samples. The frequent detection of p53 mutation in the serum of patients with early-stage tumors suggests a possible use of this approach for clinical prognosis and cancer monitoring of colorectal cancer patients (Hibi et al., 1998). Further work is needed to evaluate the reliability and prognostic signiﬁcance of this approach.

C. Mutations and Tumor Prognosis Many studies have addressed the possibility that the clinical behavior of tumors with p53 mutations may be signiﬁcantly different from that of tumors without p53 mutations. Inherent tumor aggressiveness and response to treatment by radio and/or chemotherapy are two potentially distinct aspects of this question. Several studies have shown that induction of apoptosis in cancer cells may depend on the presence of a wild-type p53 gene and that many antineoplastic drugs used clinically produce their therapeutic effects by inducing apoptosis in malignant cells (Lowe et al., 1993; Lowe, 1995; Nielsen and Maneval, 1998). Although substantiated in the clinics for hematologic malignancies (Diccianni et al., 1994; Marks et al., 1996), correlations between presence of a p53 mutation and clinical outcome have shown conﬂicting results for most types of cancer. Statistically signiﬁcant correlations between the presence of p53 mutations and poor tumor prognosis have been reported repeatedly in colon cancer (Ahnen et al., 1998; Paradiso et al., 1996; Pricolo et al., 1997), but not in breast cancers (Peller et al., 1995a; Rozan et al., 1998). In these later cancers, p53 mutations appear to be a late event in tumor progression, so whether a p53 mutation is an independent marker of prognosis is difﬁcult to establish. With the increasing knowledge of p53 protein structure and functions, several investigators have become interested in exploring whether tumor progression and tumor response to therapy may depend on the nature and localization of the p53 mutation. In colon cancer, Goh et al. (1995) have shown that tumors with mutations at codon 175 and conserved regions were more aggressive than tumors with point mutations located outside of these domains. In colon cancer, Børresen-Dale et al. (1998) found that patients with mutations located in the part of the DNA-binding domain stabilized by zinc had a signiﬁcantly shorter cancer-related survival after surgery than patients with mutations located elsewhere in the protein structure. In breast cancer, mutations in these same domains have been associated with a poor response

122

Pierre Hainaut and Monica Hollstein

to treatment by doxorubicyn (Aas et al., 1996). In head-and-neck cancer, analysis of 86 patients indicated that those with p53 mutations at residues in direct contact with target DNA fared less well than those with mutations elsewhere in the p53 structure (Erber et al., 1998). In lung cancer patients, Vega and collaborators (1997) showed a signiﬁcantly poorer survival in patients with mutations in exon 5 compared to patients with mutations in exons 6 to 9. Further studies are needed to determine whether the speciﬁc identiﬁcation of a p53 mutation may help clinicians in selecting the appropriate therapeutic approach, and whether these parameters vary with cancer type.

D. p53 as a Target in New Therapeutic Strategies The high proportion of tumors with defective p53 and the increasingly detailed information on the molecular biology of p53 make this molecule a particularly attractive focus in the development of new treatments. Reintroduction of wild-type p53 activity may help to eliminate p53-deﬁcient tumor cells by eliciting apoptosis, offering a diversity of p53-targeted strategies as illustrated by these examples: 1. peptides have been designed that regenerate p53 function in tumor cells by rescuing DNA-binding capacity of p53 mutants, an approach that is particularly attractive because it does not cause global DNA damage, in contrast to the antineoplastic drugs in current use (Halazonetis and Kandil, 1993; Hupp and Lane, 1994; Selivanova et al., 1997). 2. p53 function has been restored to tumors by viral delivery of wild-type p53 (Roth, 1998; Roth et al., 1996). 3. Selective destruction of p53-deﬁcient cells can be accomplished by genetically engineered adenovirus strains unable to replicate in cells with normal p53 (Bischoff et al., 1996). 4. Overexpression of the newly discovered homologs p73 and p51/p40/ ket can induce apoptosis in tumor cells lacking p53 ( Jost et al., 1997; Osada et al., 1998). Other potential strategies include induction of tumor rejection by idiotypic immunization with p53 antibody or derived peptides (Ruiz et al., 1998). In tumors that retain normal p53, stimulation of function by raising p53 levels, for example, by blocking its degradation with peptides that disrupt p53–mdm-2 interaction, is being examined (Bottger et al., 1997). As our knowledge of the biochemical mechanisms that regulate p53 functions improves, it may also become feasible to identify new targets for pharmaceutical intervention aimed at restoring p53 functions. It will be some time before enough experience and experimental data have accumulated to identify which strategies are most effective and workable in the clinic.

p53 and Human Cancer

123

ACKNOWLEDGMENTS The authors thank Dr. R. Montesano, Dr. J. Hall, Dr. D. Meek, and Prof. B. Strauss for critically reading the manuscript; Mrs. C. Meplan for discussions, suggestions, and help with literature searches; Mr. G. Mollon for illustration work; and Mrs. E. El Akroud for secretarial assistance. The authors are aware that although extensive this review is by no means exhaustive and they apologize to the authors of many important publications that are not cited due to space limitations.

REFERENCES Aas, T., Borresen, A. L., Geisler, S., Smith-Sorensen, B., Johnsen, H., Varhaug, J. E., Akslen, L. A., and Lonning, P. E. (1996). Nat. Med. 2(7), 811–814. Agarwal, M. L., Agarwal, A., Taylor, W. R., and Stark, G. R. (1995). Proc. Natl. Acad. Sci. U.S.A. 92(18), 8493–8497. Agarwal, M. L., Agarwal, A., Taylor, W. R., Wang, Z. Q., Wagner, E. F., and Stark, G. R. (1997). Oncogene 15(9), 1035–1041. Agarwal, M. L., Taylor, W. R., Chernov, M. V., Chernova, O. B., and Stark, G. R. (1998). J. Biol. Chem. 273(1), 1–4. Aguilar, F., Hussain, S. P., and Cerutti, P. (1993). Proc. Natl. Acad. Sci. U.S.A. 90(18), 8586– 8590. Aguilar, F., Harris, C. C., Sun, T., Hollstein, M., and Cerutti, P. (1994). Science 264, 1317– 1319. Ahnen, D. J., Feigl, P., Quan, G., Fenoglio-Preiser, C., Lovato, L. C., Bunn, P. A., Jr., Stemmerman, G., Wells, J. D., Macdonald, J. S., and Meyskens, F. L., Jr. (1998). Cancer Res. 58(6), 1149–1158. Ahuja, H. G., Testa, M. P., and Cline, M. J. (1990). Oncogene 5(9), 1409–1410. Amariglio, F., Tchang, F., Prioleau, M. N., Soussi, T., Cibert, C., and Mechali, M. (1997). Oncogene 15(18), 2191–2199. Ambs, S., Ogunfuskia, M. O., Merriam, W. G., Bennett, W. P., Billiar, T. R., and Harris, C. C. (1998). Proc. Natl. Acad. Sci. U.S.A. 95, 8823–8828. An, W. G., Kanekal, M., Simon, M. C., Maltepe, E., Blagosklonny, M. V., and Neckers, L. M. (1998). Nature (London) 392, 405–408. Angelopoulou, K., Diamandis, E. P., Sutherland, D. J., Kellen, J. A., and Bunting, P. S. (1994). Int. J. Cancer 58(4), 480–487. Armstrong, J. F., Kaufman, M. H., Harrison, D. J., and Clarke, A. R. (1995). Curr. Biol. 5(8), 931–936. Arrowsmith, C. H., and Morin, P. (1996). Oncogene 12(7), 1379–1385. Avantaggiati, M. L., Ogryzko, V., Gardner, K., Giordano, A., Levine, A. S., and Kelly, K. (1997). Cell (Cambridge, Mass.) 89(7), 1175–1184. Baker, S. J., Fearon, E. R., Nigro, J. M., Hamilton, S. R., Preisinger, A. C., Jessup, J. M., vanTuinen, P., Ledbetter, D. H., Barker, D. F., Nakamura, Y., White, R., and Vogelstein, B. (1989). Science 244, 217–221. Bargonetti, J., Chicas, A., White, D., and Prives, C. (1997). Cell. Mol. Biol. 43(7), 935–949. Barker, D., Schafer, M., and White, R. (1984). Cell (Cambridge, Mass.) 36(1), 131–138. Baudier, J., Delphin, C., Grunwald, D., Khochbin, S., and Lawrence, J. J. (1992). Proc. Natl. Acad. Sci. U.S.A. 89(23), 11627–11631.

124

Pierre Hainaut and Monica Hollstein

Beckman, G., Birgander, R., Sjalander, A., Saha, N., Holmberg, P. A., Kivela, A., and Beckman, L. (1994). Hum. Hered. 44(5), 266–270. Beroud, C., and Soussi, T. (1998). Nucleic Acids Res. 26(1), 200–204. Bhatia, K., Fan, S., Spangler, G., Weintraub, M., O’Connor, P. M., Judde, J. G., and Magrath, I. (1995). Cancer Res. 55(7), 1431–1435. Bilim, V. N., Kawasaki, T., Takahashi, K., and Tomita, Y. (1998). Anticancer Res. 18(3A), 1655 –1659. Bintz, R., and Fuchs, R. P. P. (1990). Mol. Gen. Genet. 221, 331–338. Birgander, R., Sjalander, A., Rannug, A., Alexandrie, A. K., Sundberg, M. I., Seidegard, J., Tornling, G., Beckman, G., and Beckman, L. (1995). Carcinogenesis (London) 16(9), 2233–2236. Bischoff, J. R., Kirn, D. H., Williams, A., Heise, C., Horn, S., Muna, M., Ng, L., Nye, J. A., Sampson-Johannes, A., Fattaey, A., and McCormick, F. (1996). Science 274, 373–376. Bissonnette, N., Wasylyk, B., and Hunting, D. J. (1997). Biochem. Cell Biol. 75(4), 351–358. Blaszyk, H., Hartmann, A., Sommer, S. S., and Kovach, J. S. (1996). Hum. Genet. 97(4), 543– 547. Blomeke, B., Bennett, W. P., Harris, C. C., and Shields, P. G. (1997). Carcinogenesis (London) 18(6), 1271–1275. Borresen-Dale, A. L., Lothe, R. A., Meling, G. I., Hainaut, P., Rognum, T. O., and Skovlund, E. (1998). Clin. Cancer Res. 4(1), 203–210. Bottger, A., Bottger, V., Sparks, A., Liu, W. L., Howard, S. F., and Lane, D. P. (1997). Curr. Biol. 7(11), 860–869. Bourdon, J. C., Deguin-Chambon, V., Lelong, J. C., Dessen, P., May, P., Debuire, B., and May, E. (1997). Oncogene 14(1), 85–94. Boyle, J. M., Mitchell, E. L., Greaves, M. J., Roberts, S. A., Tricker, K., Burt, E., Varley, J. M., Birch, J. M., and Scott, D. (1998). Br. J. Cancer 77(12), 2181–2192. Brennan, J. A., Mao, L., Hruban, R. H., Boyle, J. O., Eby, Y. J., Koch, W. M., Goodman, S. N., and Sidransky, D. (1995). N. Engl. J. Med. 332(7), 429–435. Bressac, B., Kew, M., Wands, J., and Ozturk, M. (1991). Nature (London) 350, 429–431. Brown, C. R., Doxsey, S. J., White, E., and Welch, W. J. (1994). J. Cell. Physiol. 160(1), 47– 60. Buchman, V. L., Chumakov, P. M., Ninkina, N. N., Samarina, O. P., and Georgiev, G. P. (1988). Gene 70(2), 245–252. Buckbinder, L., Talbott, R., Velasco-Miguel, S., Takenaka, I., Faha, B., Seizinger, B. R., and Kley, N. (1995). Nature (London) 377, 646–649. Caelles, C., Helmberg, A., and Karin, M. (1994). Nature (London) 370, 220–223. Calmels, S., Hainaut, P., and Ohshima, H. (1997). Cancer Res. 57(16), 3365–3369. Campbell, I. G., Eccles, D. M., Dunn, B., Davis, M., and Leake, V. (1996). Lancet 347, 393– 394. Canman, C. E., and Kastan, M. B. (1997). Adv. Pharmacol. 41, 429–460. Carbone, D., Chiba, I., and Mitsudomi, T. (1991). Oncogene 6(9), 1691–1692. Carbone, M., Rizzo, P., Grimley, P. M., Procopio, A., Mew, D. J., Shridhar, V., de Bartolomeis, A., Esposito, V., Giuliano, M. T., Steinberg, S. M., Levine, A. S., Giordano, A., and Pass, H. I. (1997). Nat. Med. 3(8), 908–912. Cariello, N. F., Cui, L., Beroud, C., and Soussi, T. (1994a). Cancer Res. 54(16), 4454–4460. Cariello, N. F., Piegorsch, W. W., Adams, W. T., and Skopek, T. R. (1994b). Carcinogenesis (London) 15(10), 2281–2285. Cariello, N. F., Douglas, G. R., Gorelick, N. J., Hart, D. W., Wilson, J. D., and Soussi, T. (1998). Nucleic Acids Res. 26(1), 198–199. Carmeliet, P., Dor, Y., Herbert, J. M., Fukumura, D., Brusselmans, K., Dewerchin, M., Neeman, M., Bono, F., Abramovitch, R., Maxwell, P., Koch, C. J., Ratcliffe, P., Moons, L., Jain, R. K., Collen, D., and Keshet, E. (1998). Nature (London) 394, 485–490.

p53 and Human Cancer

125

Cawley, H. M., Meltzer, S. J., De Benedetti, V. M., Hollstein, M. C., Muehlbauer, K. R., Liang, L., Bennett, W. P., Souza, R. F., Greenwald, B. D., Cottrell, J., Salabes, A., Bartsch, H., and Trivers, G. E. (1998). Gastroenterology 115(1), 19–27. Cha, R. S., Zarbl, H., Keohavong, P., and Thilly, W. G. (1992). PCR Methods Appl. 2(1), 14– 20. Chen, J., Marechal, V., and Levine, A. J. (1993). Mol. Cell. Biol 13(7), 4107–4114. Chen, X., Farmer, G., Zhu, H., Prywes, R., and Prives, C. (1993). Genes Dev. 7(10), 1837– 1849. Chen, X. Q., Stroun, M., Magnenat, J. L., Nicod, L. P., Kurt, A. M., Lyautey, J., Lederrey, C., and Anker, P. (1996). Nat. Med. 2(9), 1033–1035. Chin, K. V., Ueda, K., Pastan, I., and Gottesman, M. M. (1992). Science 255, 459–462. Chin, P. L., Momand, J., and Pfeifer, G. P. (1997). Oncogene 15(1), 87–99. Cho, Y., Gorina, S., Jeffrey, P. D., and Pavletich, N. P. (1994). Science 265, 346–355. Chou, D., Miyashita, T., Mohrenweiser, H. W., Ueki, K., Kastury, K., Druck, T., von Deimling, A., Huebner, K., Reed, J. C., and Louis, D. N. (1996). Cancer Genet. Cytogenet. 88(2), 136 –140. Chumakov, P. M., and Jenkins, J. R. (1991). Nucleic Acids Res. 19(24), 6969. Clore, G. M., Omichinski, J. G., Sakaguchi, K., Zambrano, N., Sakamoto, H., Appella, E., and Gronenborn, A. M. (1994). Science 265, 386–391. Clore, G. M., Ernst, J., Clubb, R., Omichinski, J. G., Kennedy, W. M., Sakaguchi, K., Appella, E., and Gronenborn, A. M. (1995). Nat. Struct. Biol. 2(4), 321–333. Colella, G., Vikhanskaya, F., Codegoni, A. M., Bonazzi, C., D’Incalci, M., and Broggini, M. (1998). Carcinogenesis (London) 19(4), 691–694. Crawford, L. V., Pim, D. C., and Bulbrook, R. D. (1982). Int. J. Cancer 30(4), 403–408. Cross, S. M., Sanchez, C. A., Morgan, C. A., Schimke, M. K., Ramel, S., Idzerda, R. L., Raskind, W. H., and Reid, B. J. (1995). Science 267, 1353–1356. Dameron, K. M., Volpert, O. V., Tainsky, M. A., and Bouck, N. (1994). Science 265, 1582– 1584. Del Sal, G., Murphy, M., Ruaro, E., Lazarevic, D., Levine, A. J., and Schneider, C. (1996). Oncogene 12(1), 177–185. DeMarini, D. M. (1998). Mutat. Res. 411(1), 11–18. de Murcia, J. M., Niedergang, C., Trucco, C., Ricoul, M., Dutrillaux, B., Mark, M., Oliver, F. J., Masson, M., Dierich, A., LeMeur, M., Waltzinger, C., Chambon, P., and de Murcia, G. (1997). Proc. Natl. Acad. Sci. U.S.A. 94(14), 7303–7307. Denissenko, M. F., Pao, A., Tang, M., and Pfeifer, G. P. (1996). Science 274, 430–432. Denissenko, M. F., Chen, J. X., Tang, M. S., and Pfeifer, G. P. (1997). Proc. Natl. Acad. Sci. U.S.A. 94(8), 3893–3898. Denissenko, M. F., Pao, A., Pfeifer, G. P., and Tang, M. (1998). Oncogene 16(10), 1241–1247. Desdouets, C., Ory, C., Matesic, G., Soussi, T., Brechot, C., and Sobczak-Thepot, J. (1996). FEBS Lett. 385(1–2), 34–38. Diccianni, M. B., Yu, J., Hsiao, M., Mukherjee, S., Shao, L. E., and Yu, A. L. (1994). Blood 84(9), 3105–3112. Di Leonardo, A., Linke, S. P., Clarkin, K., and Wahl, G. M. (1994). Genes Dev. 8(21), 2540– 2551. Di Leonardo, A., Khan, S. H., Linke, S. P., Greco, V., Seidita, G., and Wahl, G. M. (1997). Cancer Res. 57(6), 1013–1019. Dittmer, D., Pati, S., Zambetti, G., Chu, S., Teresky, A. K., Moore, M., Finlay, C., and Levine, A. J. (1993). Nat. Genet. 4(1), 42–46. Donato, N. J., and Perez, M. (1998). J. Biol. Chem. 273(9), 5067–5072. Donehower, L. A., Harvey, M., Slagle, B. L., McArthur, M. J., Montgomery, C. A. J., Butel, J. S., and Bradley, A. (1992). Nature 356, 215–221.

126

Pierre Hainaut and Monica Hollstein

Donehower, L. A. (1996a). Biochim. Biophys. Acta 1242(3), 171–176. Donehower, L. A. (1996b). Semin. Cancer Biol. 7(5), 269–278. Donehower, L. A. (1997). Cancer Surv. 29, 329–352. Dumaz, N., Drougard, C., Sarasin, A., and Daya-Grosjean, L. (1993). Proc. Natl. Acad. Sci. U.S.A. 90, 10529–10533. Durell, S. R., Jernigan, R. L., Appella, E., Nagaich, A. K., Harrington, R. E., and Zurkhin, V. B. (1998). Struct., Motion, Interaction Expression Biol. Macromol. 2, 277–295. Dutta, A., Ruppert, J. M., Aster, J. C., and Winchester, E. (1993). Nature (London) 365, 79– 82. Ehrlich, M., and Wang, R. Y. (1981). Science 212, 1350–1357. El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, W. E., Kinzler, K. W., and Vogelstein, B. (1993). Cell (Cambridge, Mass.) 75(4), 817– 825. El-Deiry, W. S., Harper, J. W., O’Connor, P. M., Velculescu, V. E., Canman, C. E., Jackman, J., Pietenpol, J. A., Burrell, M., Hill, D. E., and Wang, Y. (1994). Cancer Res. 54(5), 1169–1174. Eng, C., Schneider, K., Fraumeni, J. F., Jr., and Li, F. P. (1997). Cancer Epidemiol., Biomarkers Prev. 6(5), 379–383. Erber, R., Conradt, C., Homann, N., Enders, C., Finckh, M., Dietz, A., Weidauer, H., and Bosch, F. X. (1998). Oncogene 16(13), 1671–1679. Essigmann, J. M., and Wood, M. L. (1993). Toxicol. Lett. 67(1–3), 29–39. Esteve, A., Sorlie, T., Martel-Planche, G., Hollstein, M., Kusters, I., Lewalter, J., Vineis, P., Stephan-Odenthal, M., and Montesano, R. (1995). J. Occup. Environ. Med. 37(1), 59–68. Farmer, G., Colgan, J., Nakatani, Y., Manley, J. L., and Prives, C. (1996a). Mol. Cell. Biol. 16(8), 4295–4304. Farmer, G., Friedlander, P., Colgan, J., Manley, J. L., and Prives, C. (1996b). Nucleic Acids Res. 24(21), 4281–4288. Fearon, E. R., and Vogelstein, B. (1990). Cell (Cambridge, Mass.) 61, 759–868. Felix, C. A., Brown, D. L., Mitsudomi, T., Ikagaki, N., Wong, A., Wasserman, R., Womer, R. B., and Biegel, J. A. (1994). Oncogene 9(1), 327–328. Felley-Bosco, E., Weston, A., Cawley, H. M., Bennett, W. P., and Harris, C. C. (1993). Am. J. Hum. Genet. 53(3), 752–759. Filhol, O., Baudier, J., Delphin, C., Loue-Mackenbach, P., Chambaz, E. M., and Cochet, C. (1992). J. Biol. Chem. 267(29), 20577–20583. Fiscella, M., Ullrich, S. J., Zambrano, N., Shields, M. T., Lin, D., Lees-Miller, S. P., Anderson, C. W., Mercer, W. E., and Appella, E. (1993). Oncogene 8(6), 1519–1528. Fiscella, M., Zambrano, N., Ullrich, S. J., Unger, T., Lin, D., Cho, B., Mercer, W. E., Anderson, C. W., and Appella, E. (1994). Oncogene 9(11), 3249–3257. Flaman, J. M., Frebourg, T., Moreau, V., Charbonnier, F., Martin, C., Ishioka, C., Friend, S. H., and Iggo, R. (1994). Nucleic Acids Res. 22(15), 3259–3260. Flaman, J. M., Frebourg, T., Moreau, V., Charbonnier, F., Martin, C., Chappuis, P., Sappino, A. P., Limacher, I. M., Bron, L., Benhattar, J., Tada, M., Van Meir, E. G., Estreicher, A., and Iggo, R. D. (1995). Proc. Natl. Acad. Sci. U.S.A. 92(9), 3963–3967. Flaman, J. M., Robert, V., Lenglet, S., Moreau, V., Iggo, R., and Frebourg, T. (1998). Oncogene 16(10), 1369–1372. Forrester, K., Lupold, S. E., Ott, V. L., Chay, C. H., Band, V., Wang, X. W., and Harris, C. C. (1995). Oncogene 10(11), 2103–2111. Forrester, K., Ambs, S., Lupold, S. E., Kapust, R. B., Spillare, E. A., Weinberg, W. C., FelleyBosco, E., Wang, X. W., Geller, D. A., Tzeng, E., Billiar, T. R., and Harris, C. C. (1996). Proc. Natl. Acad. Sci. U.S.A. 93(6), 2442–2447. Franklin, W. A., Gazdar, A. F., Haney, J., Wistuba, I. I., La Rosa, F. G., Kennedy, T., Ritchey, D. M., and Miller, Y. E. (1997). J. Clin. Invest. 100(8), 2133–2137.

p53 and Human Cancer

127

Frebourg, T., Kassel, J., Lam, K. T., Gryka, M. A., Barbier, N., Andersen, T. I., Borresen, A. L., and Friend, S. H. (1992). Proc. Natl. Acad. Sci. U.S.A. 89(14), 6413–6417. Freeman, J., Schmidt, S., Scharer, E., and Iggo, R. (1994). EMBO J. 13(22), 5393–5400. Fried, L. M., Koumenis, C., Peterson, S. R., Green, S. L., van Zijl, P., Allalunis-Turner, J., Chen, D. J., Fishel, R., Giaccia, A. J., Brown, J. M., and Kirchgessner, C. U. (1996). Proc. Natl. Acad. Sci. U.S.A. 93(24), 13825–13830. Fritsche, M., Haessler, C., and Brandner, G. (1993). Oncogene 8(2), 307–318. Fukasawa, K., Choi, T., Kuriyama, R., Rulong, S., and Vande Woude, G. F. (1996). Science 271, 1744–1747. Fukasawa, K., Wiener, F., VandeWoude, G. F., and Mai, S. (1997). Oncogene 15(11), 1295– 1302. Futreal, P. A., Barrett, J. C., and Wiseman, R. W. (1991). Nucleic Acids Res. 19(24), 6977. Geller, D. A., Lowenstein, C. J., Shapiro, R. A., Nussler, A. K., Di Silvio, M., Wang, S. C., Nakayama, D. K., Simmons, R. L., Snyder, S. H., and Billiar, T. R. (1993). Proc. Natl. Acad. Sci. U.S.A. 90(8), 3491–3495. Gerwin, B. I., Spillare, E., Forrester, K., Lehman, T. A., Kispert, J., Welsh, J. A., Pfeifer, A. M., Lechner, J. F., Baker, S. J., Vogelstein, B., and Harris, C. C. (1992). Proc. Natl. Acad. Sci. U.S.A. 89(7), 2759–2763. Gobert, C., Bracco, L., Rossi, F., Olivier, M., Tazi, J., Lavelle, F., Kragh Larsen, A., and Riou, J. F. (1996). Biochemistry 35, 5778–5786. Goga, A., Liu, X., Hambuch, T. M., Senechal, K., Major, E., Berk, A. J., Witte, O. N., and Sawyers, C. L. (1995). Oncogene 11(4), 791–799. Goh, H. S., Yao, J., and Smith, D. R. (1995). Cancer Res. 55(22), 5217–5221. Gorina, S., and Pavletich, N. P. (1996). Science 274, 1001–1005. Gottlieb, T. M., and Oren, M. (1996). Biochim. Biophys. Acta 1287(2–3), 77–102. Graeber, T. G., Peterson, J. F., Tsai, M., Monica, K., Fornace, A. J., Jr., and Giaccia, A. J. (1994). Mol. Cell. Biol. 14(9), 6264–6277. Graeber, T. G., Osmanian, C., Jacks, T., Housman, D. E., Koch, C. J., Lowe, S. W., and Giaccia, A. J. (1996). Nature (London) 379, 88–91. Greenblatt, M. S., Bennett, W. P., Hollstein, M., and Harris, C. C. (1994). Cancer Res. 54(18), 4855–4878. Greenblatt, M. S., Grollman, A. P., and Harris, C. C. (1996). Cancer Res. 56(9), 2130–2136. Gu, W., and Roeder, R. G. (1997). Cell (Cambridge, Mass.) 90(4), 595–606. Gu, W., Shi, X., and Roeder, R. G. (1997). Nature (London) 387, 819–823. Gualberto, A., Aldape, K., Kozakiewicz, K., and Tlsty, T. D. (1998). Proc. Natl. Acad. Sci. U.S.A. 95(9), 5166–5171. Guillot, C., Falette, N., Paperin, M. P., Courtois, S., Gentil-Perret, A., Treilleux, I., Ozturk, M., and Puisieux, A. (1997). Oncogene 14(1), 45–52. Guillouf, C., Rosselli, F., Krishnaraju, K., Moustacchi, E., Hoffman, B., and Liebermann, D. A. (1995). Oncogene 10(11), 2263–2270. Hahn, M., Serth, J., Fislage, R., Wolfes, H., Allhoff, E., Jonas, U., and Pingoud, A. (1993). Clin. Chem. (Winston-Salem, N.C.) 39(3), 549–550. Hainaut, P. (1995). Curr. Opin. Oncol. 7(1), 76–82. Hainaut, P., and Milner, J. (1992). EMBO J 11(10), 3513–3520. Hainaut, P., and Milner, J. (1993a). Cancer Res. 53(8), 1739–1742. Hainaut, P., and Milner, J. (1993b). Cancer Res. 53(19), 4469–4473. Hainaut, P., Butcher, S., and Milner, J. (1995). Br. J. Cancer 71(2), 227–231. Hainaut, P., Hernandez, T., Robinson, A., Rodriguez-Tome, P., Flores, T., Hollstein, M., Harris, C. C., and Montesano, R. (1998). Nucleic Acids Res. 26(1), 205–213. Halazonetis, T. D., and Kandil, A. N. (1993). EMBO J. 12(13), 5057–5064. Halazonetis, T. D., Davis, L. J., and Kandil, A. N. (1993). EMBO J. 12(3), 1021–1028.

128

Pierre Hainaut and Monica Hollstein

Hall, S. R., Campbell, L. E., and Meek, D. W. (1996). Nucleic Acids Res. 24(6), 1119–1126. Hanawalt, P., and Mellon, I. (1993). Curr. Biol. 3, 67–69. Harris, C. C. (1996b). Carcinogenesis (London) 17(6), 1187–1198. Harris, C. C. (1996c). J. Natl. Cancer Inst. 88(20), 1442–1455. Harris, C. C. (1996d). Br. J. Cancer 73(3), 261–269. Harris, N., Brill, E., Shohat, O., Prokocimer, M., Wolf, D., Arai, N., and Rotter, V. (1986). Mol. Cell. Biol. 6(12), 4650–4656. Harvey, M., Vogel, H., Morris, D., Bradley, A., Bernstein, A., and Donehower, L. A. (1995). Nat. Genet. 9(3), 305–311. Haupt, Y., Maya, R., Kazaz, A., and Oren, M. (1997). Nature (London) 387, 296–299. Hermeking, H., Lengauer, C., Polyak, K., He, T. C., Zhang, L., Thiagalingam, S., Kinzler, K. W., and Vogelstein, B. (1997). Mol. Cell. 1(1), 3–11. Hernandez-Boussard, T. M., and Hainaut, P. (1998). Environ. Health Perspect. 106(7), 385– 391. Herrmann, C. P., Kraiss, S., and Montenarh, M. (1991). Oncogene 6(5), 877–884. Hibi, K., Robinson, C. R., Booker, S., Wu, L., Hamilton, S. R., Sidransky, D., and Jen, J. (1998). Cancer Res. 58(7), 1405–1407. Hoekstra, M. F. (1997). Curr. Opin. Genet. Dev. 7(2), 170–175. Hollstein, M., Sidransky, D., Vogelstein, B., and Harris, C. C. (1991). Science 253, 49–53. Hollstein, M., Rice, K., Greenblatt, M. S., Soussi, T., Fuchs, R., Sorlie, T., Hovig, E., SmithSorensen, B., Montesano, R., and Harris, C. C. (1994). Nucleic Acids Res. 22(17), 3551– 3555. Hollstein, M. C., Metcalf, R. A., Welsh, J. A., Montesano, R., and Harris, C. C. (1990). Proc. Natl. Acad. Sci. U.S.A. 87(24), 9958–9961. Holmquist, G. P., and Gao, S. (1997). Mutat. Res. 386, 69–101. Hsu, I. C., Metcalf, R. A., Sun, T., Welsh, J. A., Wang, N. J., and Harris, C. C. (1991). Nature (London) 350, 427–428. Huang, L. C., Clarkin, K. C., and Wahl, G. M. (1996a). Proc. Natl. Acad. Sci. U.S.A. 93(10), 4827–4832. Huang, L. C., Clarkin, K. C., and Wahl, G. M. (1996b). Cancer Res. 56(13), 2940–2944. Huang, P. (1998). Oncogene 17(3), 261–270. Huibregtse, J. M., Scheffner, M., and Howley, P. M. (1993a). Mol. Cell. Biol. 13(2), 775–784. Huibregtse, J. M., Scheffner, M., and Howley, P. M. (1993b). Mol. Cell. Biol. 13(8), 4918– 4927. Hupp, T. R., and Lane, D. P. (1994). Curr. Biol. 4(10), 865–875. Inga, A., Cresta, S., Monti, P., Aprile, A., Scott, G., Abbondandolo, A., Iggo, R., and Fronza, G. (1997). Carcinogenesis (London) 18(10), 2019–2021. Ito, T., Seyama, T., Hayashi, T., Mizuno, T., Iwamoto, K. S., Tsuyama, N., Dohi, K., Nakamura, N., and Akiyama, M. (1994). Hum. Genet. 93(2), 222. Jamal, S., and Ziff, E. B. (1995). Oncogene 10(11), 2095–2101. Jayaraman, J., and Prives, C. (1995). Cell (Cambridge, Mass.) 81(7), 1021–1029. Jayaraman, L., Murthy, K. G., Zhu, C., Curran, T., Xanthoudakis, S., and Prives, C. (1997). Genes Dev. 11(5), 558–570. Jayaraman, L., Moorthy, N. C., Murthy, K. G., Manley, J. L., Bustin, M., and Prives, C. (1998). Genes Dev. 12(4), 462–472. Jeffrey, P. D., Gorina, S., and Pavletich, N. P. (1995). Science 267, 1498–1502. Jego, N., Thomas, G., and Hamelin, R. (1993). Oncogene 8(1), 209–213. Jin, S., Kharbanda, S., Mayer, B., Kufe, D., and Weaver, D. T. (1997). J. Biol. Chem. 272(40), 24763–24766. Jin, X., Wu, X., Roth, J. A., Amos, C. I., King, T. M., Branch, C., Honn, S. E., and Spitz, M. R. (1995). Carcinogenesis (London) 16(9), 2205–2208.

p53 and Human Cancer

129

Jonason, A. S., Kunala, S., Price, G. J., Restifo, R. J., Spinelli, H. M., Persing, J. A., Leffell, D. J., Tarone, R. E., and Brash, D. E. (1996). Proc. Natl. Acad. Sci. U.S.A. 93(24), 14025– 14029. Jones, S. N., Roe, A. E., Donehower, L. A., and Bradley, A. (1995). Nature (London) 378, 206– 208. Jongmans, W., Artuso, M., Vuillaume, M., Bresil, H., Jackson, S. P., and Hall, J. (1996). Oncogene 13(6), 1133–1138. Jongmans, W., Vuillaume, M., Chrzanowska, K., Smeets, D., Sperling, K., and Hall, J. (1997). Mol. Cell. Biol. 17(9), 5016–5022. Jost, C. A., Marin, M. C., and Kaelin, W. G., Jr. (1997). Nature (London) 389, 191–194. Juven, T., Barak, Y., Zauberman, A., George, D. L., and Oren, M. (1993). Oncogene 8, 3411– 3416. Kaelin, W. G., Jr. (1998). Science 281, 57–58. Kaghad, M., Bonnet, H., Yang, A., Creancier, L., Biscan, J. C., Valent, A., Minty, A., Chalon, P., Lelias, J. M., Dumont, X., Ferrara, P., McKeon, F., and Caput, D. (1997). Cell (Cambridge, Mass.) 90(4), 809–819. Kamijo, T., Weber, J. D., Zambetti, G., Zindy, F., Roussel, M. F., and Sherr, C. J. (1998). Proc. Natl. Acad. Sci. U.S.A. 95(14), 8292–8297. Kapoor, M., and Lozano, G. (1998). Proc. Natl. Acad. Sci. U.S.A. 95(6), 2834–2837. Kastan, M. B. (1996). BioEssays 18(8), 617–619. Kastan, M. B., Onyekwere, O., Sidransky, D., Vogelstein, B., and Craig, R. W. (1991). Cancer Res. 51(23 Pt. 1), 6304–6311. Kastan, M. B., Zhan, Q., El-Deiry, W. S., Carrier, F., Jacks, T., Walsh, W. V., Plunkett, B. S., Vogelstein, B., and Fornace, A. J., Jr. (1992). Cell (Cambridge, Mass.) 71(4), 587–597. Kawajiri, K., Nakachi, K., Imai, K., Watanabe, J., and Hayashi, S. (1993). Carcinogenesis (London) 14(6), 1085–1089. Kharbanda, S., Pandey, P., Jin, S., Inoue, S., Bharti, A., Yuan, Z. M., Weichselbaum, R., Weaver, D., and Kufe, D. (1997). Nature (London) 386, 732–735. Kleihues, P., Aguzzi, A., and Ohgaki, H. (1995). Toxicol. Lett. 82–83, 601–605. Kleihues, P., Schauble, B., zur Hausen, A., Esteve, J., and Ohgaki, H. (1997). Am. J. Pathol. 150(1), 1–13. Knippschild, U., Milne, D. M., Campbell, L. E., DeMaggio, A. J., Christenson, E., Hoekstra, M. F., and Meek, D. W. (1997). Oncogene 15(14), 1727–1736. Ko, L. J., and Prives, C. (1996). Genes Dev. 10(9), 1054–1072. Ko, L. J., Shieh, S. Y., Chen, X., Jayaraman, L., Tamai, K., Taya, Y., Prives, C., and Pan, Z. Q. (1997). Mol. Cell. Biol. 17(12), 7220–7229. Kondo, K., Tsuzuki, H., Sasa, M., Sumitomo, M., Uyama, T., and Monden, Y. (1996). J. Surg. Oncol. 61(1), 20–26. Kubbutat, M. H., Jones, S. N., and Vousden, K. H. (1997). Nature (London) 387, 299– 303. Kuerbitz, S. J., Plunkett, B. S., Walsh, W. V., and Kastan, M. B. (1992). Proc. Natl. Acad. Sci. U.S.A. 89(16), 7491–7495. Kunkel, T. A. (1990). Biochemistry 29(35), 8003–8011. Kussie, P. H., Gorina, S., Marechal, V., Elenbaas, B., Moreau, J., Levine, A. J., and Pavletich, N. P. (1996). Science 274, 948–953. Lancaster, J. M., Brownlee, H. A., Wiseman, R. W., and Taylor, J. (1995). Lancet 346, 182. Lane, D. P. (1992). Nature (London) 358, 15–16. Lane, D. P. (1998). Curr. Biol. 2, 581–583. Lane, D. P., and Hall, P. A. (1997). Trends Biochem. Sci. 22(10), 372–374. Lanni, J. S., and Jacks, T. (1998). Mol. Cell. Biol. 18(2), 1055–1064. Lassus, P., Ferlin, M., Piette, J., and Hibner, U. (1996). EMBO J. 15(17), 4566–4573.

130

Pierre Hainaut and Monica Hollstein

Lazar, V., Hazard, F., Bertin, F., Janin, N., Bellet, D., and Bressac, B. (1993). Oncogene 8(6), 1703–1705. Lee, C. W., Sorensen, T. S., Shikama, N., and La Thangue, N. B. (1998). Oncogene 16(21), 2695–2710. Lee, J. M., Abrahamson, J. L., Kandel, R., Donehower, L. A., and Bernstein, A. (1994). Oncogene 9(12), 3731–3736. Lee, W., Harvey, T. S., Yin, Y., Yau, P., Litchﬁeld, D., and Arrowsmith, C. H. (1994). Nat. Struct. Biol. 1(12), 877–890. Lees-Miller, S. P., Sakaguchi, K., Ullrich, S. J., Appella, E., and Anderson, C. W. (1992). Mol. Cell. Biol. 12(11), 5041–5049. Legros, Y., Lafon, C., and Soussi, T. (1994). Oncogene 9(7), 2071–2076. Leveillard, T., Gorry, P., Niederreither, K., and Wasylyk, B. (1998). Mech. Dev. 74(1–2), 189–193. Levine, A. J. (1997). Cell (Cambridge, Mass.) 88(3), 323–331. Lin, J., Chen, J., Elenbaas, B., and Levine, A. J. (1994). Genes Dev. 8(10), 1235–1246. Lindahl, T. (1993). Nature (London) 362, 709–715. Linke, S. P., Clarkin, K. C., Di Leonardo, A., Tsou, A., and Wahl, G. M. (1996). Genes Dev. 10(8), 934–947. Linke, S. P., Clarkin, K. C., and Wahl, G. M. (1997). Cancer Res. 57(6), 1171–1179. Liu, X., Miller, C. W., Koefﬂer, P. H., and Berk, A. J. (1993). Mol. Cell. Biol. 13(6), 3291–3300. Livingstone, L. R., White, A., Sprouse, J., Livanos, E., Jacks, T., and Tlsty, T. D. (1992). Cell (Cambridge, Mass.) 70(6), 923–935. Loeb, L. A. (1991). Cancer Res. 51(12), 3075–3079. Loeb, L. A. (1998). Adv. Cancer Res. 72, 25–56. Loeb, L. A., and Christians, F. C. (1996). Mutat. Res. 350(1), 279–286. Lowe, S. W. (1995). Curr. Opin. Oncol. 7(6), 547–553. Lowe, S. W., Schmitt, E. M., Smith, S. W., Osborne, B. A., and Jacks, T. (1993). Nature (London) 362, 847–849. Lu, H., Fisher, R. P., Bailey, P., and Levine, A. J. (1997). Mol. Cell. Biol. 17(10), 5923–5934. Lubin, R., Schlichtholz, B., Bengoufa, D., Zalcman, G., Trédaniel, J., Hirsch, A., de Fromentel, C. C., Preudhomme, C., Fenaux, P., Fournier, G., Mangin, P., Laurent-Puig, P., Pelletier, G., Schlumberger, M., Desgrandchamps, F., Le Duc, A., Peyrat, J. P., Janin, N., Bressac, B., and Soussi, T. (1993). Cancer Res. 53(24), 5872–5876. Lubin, R., Zalcman, G., Bouchet, L., Tredanel, J., Legros, Y., Cazals, D., Hirsch, A., and Soussi, T. (1995). Nat. Med. 1(7), 701–702. Ludwig, R. L., Bates, S., and Vousden, K. H. (1996). Mol. Cell. Biol. 16(9), 4952–4960. Lutzker, S. G., and Levine, A. J. (1996). Nat. Med. 2(7), 804–810. Maheswaran, S., Englert, C., Lee, S. B., Ezzel, R. M., Settleman, J., and Haber, D. A. (1998). Oncogene 16(16), 2041–2050. Maki, C. G., Huibregtse, J. M., and Howley, P. M. (1996). Cancer Res. 56(11), 2649–2654. Malanga, M., Pleschke, J. M., Kleczkowska, H. E., and Althaus, F. R. (1998). J. Biol. Chem. 273(19), 11839–11843. 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., and Tainsky, M. A. (1990). Science 250, 1233–1238. Mao, E. F., Lane, L., Lee, J., and Miller, J. H. (1997). J. Bacteriol. 179(2), 417–422. Margulies, L., and Sehgal, P. B. (1993). J. Biol. Chem. 268(20), 15096–15100. Marks, D. I., Kurz, B. W., Link, M. P., Ng, E., Shuster, J. J., Lauer, S. J., Brodsky, I., and Haines, D. S. (1996). Blood 87(3), 1155–1161. Martin, D. W., Munoz, R. M., Subler, M. A., and Deb, S. (1993). J. Biol. Chem. 268(18), 13062–13067. Matlashewski, G. J., Tuck, S., Pim, D., Lamb, P., Schneider, J., and Crawford, L. V. (1987). Mol. Cell. Biol. 7(2), 961–963.

p53 and Human Cancer

131

Matsuura, S., Tauchi, H., Nakamura, A., Kondo, N., Sakamoto, S., Endo, S., Smeets, D., Solder, B., Belohradsky, B. H., Der Kaloustian, V. M., Oshimura, M., Isomura, M., Nakamura, Y., and Komatsu, K. (1998). Nat. Genet. 19(2), 179–181. Mavridou, D., Gornall, R., Campbell, I. G., and Eccles, D. M. (1998). Br. J. Cancer 77(4), 676–677. Mazars, G. R., Jeanteur, P., Lynch, H. T., Lenoir, G., and Theillet, C. (1992). Oncogene 7(4), 781–782. Meek, D. W. (1998). Cell. Signal. 10(3), 159–166. Meira, L. B., Cheo, D. L., Hammer, R. E., Burns, D. K., Reis, A., and Friedberg, E. C. (1997). Nat. Genet. 17(2), 145. Milne, D. M., Palmer, R. H., Campbell, D. G., and Meek, D. W. (1992a). Oncogene 7(7), 1361–1369. Milne, D. M., Palmer, R. H., and Meek, D. W. (1992b). Nucleic Acids Res. 20(21), 5565–5570. Milne, D. M., Campbell, D. G., Caudwell, F. B., and Meek, D. W. (1994). J. Biol. Chem. 269(12), 9253–9260. Milne, D. M., Campbell, L. E., Campbell, D. G., and Meek, D. W. (1995). J. Biol. Chem. 270(10), 5511–5518. Milner, J., and Medcalf, E. A. (1991). Cell (Cambridge, Mass.) 65(5), 765–774. Miyashita, T., Krajewski, S., Krajewska, M., Wang, H. G., Lin, H. K., Liebermann, D. A., Hoffman, B., and Reed, J. C. (1994). Oncogene 9(6), 1799–1805. Molinari, M., and Milner, J. (1995). Oncogene 10(9), 1849–1854. Moll, U. M., and Schramm, L. M. (1998). Crit. Rev. Oral Biol. Med. 9(1), 23–37. Moll, U. M., Riou, G., and Levine, A. J. (1992). Proc. Natl. Acad. Sci. U.S.A. 89(15), 7262–7266. Moll, U. M., LaQuaglia, M., Benard, J., and Riou, G. (1995). Proc. Natl. Acad. Sci. U.S.A. 92(10), 4407–4411. Moll, U. M., Ostermeyer, A. G., Haladay, R., Winkﬁeld, B., Frazier, M., and Zambetti, G. (1996). Mol. Cell. Biol. 16(3), 1126–1137. Momand, J., and Zambetti, G. P. (1997). J. Cell. Biochem. 64(3), 343–352. Momand, J., Jung, D., Wilczynski, S., and Niland, J. (1998). Nucleic Acids Res. 26(15), 3453–3459. Montes de Oca Luna, R. Wagner, D. S., and Lozano, G. (1995). Nature (London) 378, 203–206. Montesano, R., and Hainaut, P. (1998). Cancer Surv. 32, 53 –68. Montesano, R., Hollstein, M., and Hainaut, P. (1996). Int. J. Cancer 69(3), 225–235. Montesano, R., Hainaut, P., and Wild, C. P. (1997). J. Natl. Cancer Inst. 89(24), 1844–1851. Morgan, S. E., and Kastan, M. B. (1997). Adv. Cancer Res. 71, 1–25. Mummenbrauer, T., Janus, F., Muller, B., Wiesmuller, L., Deppert, W., and Grosse, F. (1996). Cell (Cambridge, Mass.) 85(7), 1089–1099. Murata, M., Tagawa, M., Kimura, M., Kimura, H., Watanabe, S., and Saisho, H. (1996). Carcinogenesis (London) 17(2), 261–264. Nakazawa, H., English, D., Randell, P. L., Nakazawa, K., Martel, N., Armstrong, B. K., and Yamasaki, H. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 360–364. Nawroz, H., Koch, W., Anker, P., Stroun, M., and Sidransky, D. (1996). Nat. Med. 2(9), 1035–1037. Neddermann, P., Gallinari, P., Lettieri, T., Schmid, D., Truong, O., Hsuan, J. J., Wiebauer, K., and Jiricny, J. (1996). J. Biol. Chem. 271(22), 12767–12774. Nelson, W. G., and Kastan, M. B. (1994). Mol. Cell. Biol. 14(3), 1815–1823. Niculescu, A. B., Chen, X., Smeets, M., Hengst, L., Prives, C., and Reed, S. I. (1998). Mol. Cell. Biol. 18(1), 629–643. Nielsen, L. L., and Maneval, D. C. (1998). Cancer Gene Ther. 5(1), 52–63. Nigro, J. M., Baker, S. J., Preisinger, A. C., Jessup, J. M., Hostetter, R., Cleary, K., Bigner, S. H.,

132

Pierre Hainaut and Monica Hollstein

Davidson, N., Baylin, S., Devilee, P., Glover, T., Collins, F. S., Weston, A., Modali, R., Harris, C. C., and Vogelstein, B. (1989). Nature (London) 342, 705–708. Nigro, J. M., Aldape, K. D., Hess, S. M., and Tlsty, T. D. (1997). Cancer Res. 57(17), 3635–3639. Ohlsson, C., Kley, N., Werner, H., and LeRoith, D. (1998). Endocrinology (Baltimore) 139(3), 1101–1107. Okamoto, K., and Beach, D. (1994). EMBO J. 13(20), 4816–4822. Oliner, J. D., Kinzler, K. W., Meltzer, P. S., George, D. L., and Vogelstein, B. (1992). Nature (London) 358, 80–83. Oliner, J. D., Pietenpol, J. A., Thiagalingam, S., Gyuris, J., Kinzler, K. W., and Vogelstein, B. (1993). Nature 362, 857–860. Olschwang, S., Laurent-Puig, P., Vassal, A., Salmon, R. J., and Thomas, G. (1991). Hum. Genet. 86(4), 369–370. Oren, M. (1992). FASEB J. 6(13), 3169–3176. Ory, K., Legros, Y., Auguin, C., and Soussi, T. (1994). EMBO J. 13(15), 3496–3504. Osada, M., Ohba, M., Kawahara, C., Ishioka, C., Kanamaru, R., Katoh, I., Ikawa, Y., Nimura, Y., Nakagawara, A., Obinata, M., and Ikawa, S. (1998). Nat. Med. 4(7), 839–843. Owen-Schaub, L. B., Zhang, W., Cusack, J. C., Angelo, L. S., Santee, S. M., Fujiwara, T., Roth, J. A., Deisseroth, A. B., Zhang, W. W., and Kruzel, E. (1995). Mol. Cel. Biol. 15(6), 3032–3040. Ozturk, M. (1991). Lancet 338, 1356–1359. Paradiso, A., Rabinovich, M., Vallejo, C., Machiavelli, M., Romero, A., Perez, J., Lacava, J., Cuevas, M. A., Rodriquez, R., Leone, B., Sapia, M. G., Simone, G., and De Lena, M. (1996). Int. J. Cancer. 69(6), 437–441. Pavletich, N. P., Chambers, K. A., and Pabo, C. O. (1993). Genes Dev. 7(12B), 2556–2564. Peller, S., Halevy, A., Slutzki, S., Kopilova, Y., and Rotter, V. (1995a). Mol. Carcinog. 13(3), 166–172. Peller, S., Kopilova, Y., Slutzki, S., Halevy, A., Kvitko, K., and Rotter, V. (1995b). DNA Cell Biol. 14(12), 983–990. Pietenpol, J. A., Tokino, T., Thiagalingam, S., El-Deiry, W. S., Kinzler, K. W., and Vogelstein, B. (1994). Proc. Natl. Acad. Sci. U.S.A. 91(6), 1998–2002. Piette, J., Neel, H., and Marechal, V. (1997). Oncogene 15(9), 1001–1010. Pleasants, L. M., and Hansen, M. F. (1994). Hum. Genet. 93(5), 607–608. Pollock, P. M., Pearson, J. V., and Hayward, N. K. (1996). Genes Chromosomes Cancer 15(2), 77–88. Polyak, K., Xia, Y., Zweier, J. L., Kinzler, K. W., and Vogelstein, B. (1997). Nature (London) 389, 300–305. Pomerantz, J., Schreiber-Agus, N., Liegeois, N. J., Silverman, A., Alland, L., Chin, L., Potes, J., Chen, K., Orlow, I., Lee, H. W., Cordon-Cardo, C., and DePinho, R. A. (1998). Cell (Cambridge, Mass.) 92(6), 713–723. Ponchel, F., Puisieux, A., Tabone, E., Michot, J. P., Froschl, G., Morel, A. P., Frebourg, T., Fontaniere, B., Oberhammer, F., and Ozturk, M. (1994). Cancer Res. 54(8), 2064–2068. Ponten, F., Berg, C., Ahmadian, A., Ren, Z. P., Nister, M., Lundeberg, J., Uhlen, M., and Ponten, J. (1997a). Oncogene 15(9), 1059–1067. Ponten, F., Williams, C., Ling, G., Ahmadian, A., Nister, M., Lundeberg, J., Ponten, J., and Uhlen, M. (1997b). Mutat. Res. 382(1–2), 45–55. Pricolo, V. E., Finkelstein, S. D., Hansen, K., Cole, B. F., and Bland, K. I. (1997). Arch. Surg. (Chicago) 132(4), 371–374. Prosser, J., and Condie, A. (1991). Nucleic Acids Res. 19, 4799. Puisieux, A., Lim, S., Groopman, J., and Ozturk, M. (1991). Cancer Res. 51(22), 6185– 6189.

p53 and Human Cancer

133

Rainwater, R., Parks, D., Anderson, M. E., Tegtmeyer, P., and Mann, K. (1995). Mol. Cell. Biol. 15(7), 3892–3903. Ramet, M., Castren, K., Jarvinen, K., Pekkala, K., Turpeenniemi-Hujanen, T., Soini, Y., Paakko, P., and Vahakangas, K. (1995). Carcinogenesis (London) 16(9), 2117–2124. Rampino, N., Yamamoto, H., Ionov, Y., Li, Y., Sawai, H., Reed, J. C., and Perucho, M. (1997). Science 275, 967–969. Ren, Z. P., Hedrum, A., Ponten, F., Nister, M., Ahmadian, A., Lundeberg, J., Uhlen, M., and Ponten, J. (1996a). Oncogene 12(4), 765–773. Ren, Z. P., Ponten, F., Nister, M., and Ponten, J. (1996b). Int. J. Cancer 69(3), 174–179. Ren, Z. P., Ahmadian, A., Ponten, F., Nister, M., Berg, C., Lundeberg, J., Uhlen, M., and Ponten, J. (1997). Am. J. Pathol. 150(5), 1791–1803. Resnick-Silverman, L., St.Clair, S., Maurer, M., Zhao, K., and Manfredi, J. J. (1998). Genes Dev. 12, 2102–2107. Richards, B., Zhang, H., Phear, G., and Meuth, M. (1997). Science 277, 1523–1526. Rideout, W. M., Coetzee, G. A., Olumi, A. F., and Jones, P. A. (1990). Science 249, 1288–1290. Ripley, L. S. (1990). Annu. Rev. Genet. 24, 189–213. Rolley, N., and Milner, J. (1994). Oncogene 9(10), 3067–3070. Roperch, J. P., Alvaro, V., Prieur, S., Tuynder, M., Nemani, M., Lethrosne, F., Piouffre, L., Gendron, M. C., Israeli, D., Dausset, J., Oren, M., Amson, R., and Telerman, A. (1998). Nat. Med. 4(7), 835–838. Rosenthal, A. N., Ryan, A., Al-Jehani, R. M., Storey, A., Harwood, C. A., and Jacobs, I. J. (1998). Lancet 352, 871–872. Roth, J. A. (1998). Curr. Opin. Oncol. 10(2), 127–132. Roth, J. A., Nguyen, D., Lawrence, D. D., Kemp, B. L., Carrasco, C. H., Ferson, D. Z., Hong, W. K., Komaki, R., Lee, J. J., Nesbitt, J. C., Pisters, K. M., Putnam, J. B., Schea, R., Shin, D. M., Walsh, G. L., Dolormente, M. M., Han, C. I., Martin, F. D., Yen, N., Xu, K., Stephens, L. C., McDonnell, T. J., Mukhopadhyay, T., and Cai, D. (1996). Nat. Med. 2(9), 985–991. Rotter, V., Schwartz, D., Almon, E., Goldﬁnger, N., Kapon, A., Meshorer, A., Donehower, L. A., and Levine, A. J. (1993). Proc. Natl. Acad. Sci. U.S.A. 90(19), 9075–9079. Rouault, J. P., Falette, N., Guehenneux, F., Guillot, C., Rimokh, R., Wang, Q., Berthet, C., Moyret-Lalle, C., Savatier, P., Pain, B., Shaw, P., Berger, R., Samarut, J., Magaud, J. P., Ozturk, M., Samarut, C., and Puisieux, A. (1996). Nat. Genet. 14(4), 482–486. Rowan, S., Ludwig, R. L., Haupt, Y., Bates, S., Lu, X., Oren, M., and Vousden, K. H. (1996). EMBO J. 15(4), 827–838. Rozan, S., Vincent-Salomon, A., Zafrani, B., Validire, P., De Cremoux, P., Bernoux, A., Nieruchalski, M., Fourquet, A., Clough, K., Dieras, V., Pouillart, P., and Sastre-Garau, X. (1998). Int. J. Cancer 79(1), 27–33. Ruaro, E. M., Collavin, L., Del Sal, G., Haffner, R., Oren, M., Levine, A. J., and Schneider, C. (1997). Proc. Natl. Acad. Sci. U.S.A. 94(9), 4675–4680. Ruiz, P. J., Wolkowicz, R., Waisman, A., Hirschberg, D. L., Carmi, P., Erez, N., Garren, H., Herkel, J., Karpuj, M., Steinman, L., Rotter, V., and Cohen, I. R. (1998). Nat. Med. 4(6), 710–712. Runnebaum, I. B., Tong, X. W., Konig, R., Hong, Z., Korner, K., Atkinson, E. N., Kreienberg, R., and Kieback, D. G. (1995). Lancet 345, 994. Ryan, K. M., and Vousden, K. H. (1998). Mol. Cell. Biol. 18(7), 3692–3698. Sah, V. P., Attardi, L. D., Mulligan, G. J., Williams, B. O., Bronson, R. T., and Jacks, T. (1995). Nat. Genet. 10(2), 175–180. Sakamuro, D., Sabbatini, P., White, E., and Prendergast, G. C. (1997). Oncogene 15(8), 887–898. Sanchez-Cespedes, M., Monzo, M., Rosell, R., Pifarre, A., Calvo, R., Lopez-Cabrerizo, M. P., and Astudillo, J. (1998). Ann. Oncol. 9(1), 113–116.

134

Pierre Hainaut and Monica Hollstein

Satoh, M. S., and Lindahl, T. (1992). Nature (London) 356, 356–358. Scheffner, M., Huibregtse, J. M., and Howley, P. M. (1994). Proc. Natl. Acad. Sci. U.S.A. 91(19), 8797–8801. Schlichtholz, B., Tredaniel, J., Lubin, R., Zalcman, G., Hirsch, A., and Soussi, T. (1994). Br. J. Cancer 69(5), 809–816. Schmale, H., and Bamberger, C. (1997). Oncogene 15(11), 1363–1367. Schmutte, C., Yang, A. S., Beart, R. W., and Jones, P. A. (1995). Cancer Res. 55(17), 3742–3746. Schmutte, C., Baffa, R., Veronese, L. M., Murakumo, Y., and Fishel, R. (1997). Cancer Res. 57(14), 3010–3015. Scolnick, D. M., Chehab, N. H., Stavridi, E. S., Lien, M. C., Caruso, L., Moran, E., Berger, S. L., and Halazonetis, T. D. (1997). Cancer Res. 57(17), 3693–3696. Selivanova, G., and Wiman, K. G. (1995). Adv. Cancer Res. 66, 143–180. Selivanova, G., Iotsova, V., Okan, I., Fritsche, M., Strom, M., Groner, B., Grafström, R. C., and Wiman, K. G. (1997). Nat. Med. 3(6), 632–638. Selvakumaran, M., Lin, H. K., Miyashita, T., Wang, H. G., Krajewski, S., Reed, J. C., Hoffman, B., and Liebermann, D. (1994). Oncogene 9(6), 1791–1798. Seto, E., Usheva, A., Zambetti, G. P., Momand, J., Horikoshi, N., Weinmann, R., Levine, A. J., and Shenk, T. (1992). Proc. Natl. Acad. Sci. U.S.A. 89(24), 12028–12032. Shafman, T., Khanna, K. K., Kedar, P., Spring, K., Kozlov, S., Yen, T., Hobson, K., Gatei, M., Zhang, N., Watters, D., Egerton, M., Shiloh, Y., Kharbanda, S., Kufe, D., and Lavin, M. F. (1997). Nature (London) 387, 520–523. Shaulsky, G., Goldﬁnger, N., Peled, A., and Rotter, V. (1991a). Cell Growth Differ. 2(12), 661–667. Shaulsky, G., Goldﬁnger, N., Tosky, M. S., Levine, A. J., and Rotter, V. (1991b). Oncogene 6(11), 2055–2065. Shaw, P., Freeman, J., Bovey, R., and Iggo, R. (1996). Oncogene 12(4), 921–930. Sheikh, M. S., Burns, T. F., Huang, Y., Wu, G. S., Amundson, S., Brooks, K. S., Fornace, A. J., Jr., and El-Deiry, W. S. (1998). Cancer Res. 58(8), 1593–1598. Shieh, S. Y., Ikeda, M., Taya, Y., and Prives, C. (1997). Cell (Cambridge, Mass.) 91(3), 325– 334. Shin, T. H., Paterson, A. J., and Kudlow, J. E., (1995). Mol. Cell Biol. 15, 4694–4701. Shiohara, M., El-Deiry, W. S., Wada, M., Nakamaki, T., Takeuchi, S., Yang, R., Chen, D. L., Vogelstein, B., and Koefﬂer, H. P. (1994). Blood 84(11), 3781–3784. Shivakumar, C. V., Brown, D. R., Deb, S., and Deb, S. P. (1995). Mol. Cell. Biol. 15(12), 6785–6793. Sidransky, D., and Hollstein, M. (1996). Annu. Rev. Med. 47, 285–301. Siegel, J., Fritsche, M., Mai, S., Brandner, G., and Hess, R. D. (1995). Oncogene 11(7), 1363–1370. Siliciano, J. D., Canman, C. E., Taya, Y., Sakaguchi, K., Appella, E., and Kastan, M. B. (1997). Genes Dev. 11(24), 3471–3481. Singer, I. I., Kawka, D. W., Scott, S., Weidner, J. R., Mumford, R. A., Riehl, T. E., and Stenson, W. F. (1996). Gastroenterology 111(4), 871–885. Smith, M. L., Kontny, H. U., Bortnick, R., and Fornace, A. J., Jr. (1997). Exp. Cell Res. 230(1), 61–68. Sorlie, T., Martel-Planche, G., Hainaut, P., Lewalter, J., Holm, R., Borresen-Dale, A. L., and Montesano, R. (1998). Br. J. Cancer 77(10), 1573–1579. Soussi, T., and May, P. (1996). J. Mol. Biol. 260(5), 623–637. Soussi, T., Caron de Fromentel, C., and May, P. (1990). Oncogene 5(7), 945–952. Srivastava, S., Zou, Z. Q., Pirollo, K., Blattner, W., and Chang, E. H. (1990). Nature (London) 348, 747–749. Stewart, N., Hicks, G. G., Paraskevas, F., and Mowat, M. (1995). Oncogene 10(1), 109–115.

p53 and Human Cancer

135

Storey, A., Thomas, M., Kalita, A., Harwood, C., Gardiol, D., Mantovani, F., Breuer, J., Leigh, I. M., Matlashewski, G., and Banks, L. (1998). Nature (London) 393, 229–234. Strauss, B. S. (1992). Cancer Res. 52(2), 249–253. Strauss, B. S. (1997). Carcinogenesis (London) 18(8), 1445–1452. Strauss, B. S. (1998). Genetics 148(4), 1619–1626. Sugrue, M. M., Shin, D. Y., Lee, S. W., and Aaronson, S. A. (1997). Proc. Natl. Acad. Sci. U.S.A. 94(18), 9648–9653. Takenaka, I., Morin, F., Seizinger, B. R., and Kley, N. (1995). J. Biol. Chem. 270(10), 5405–5411. Talis, A. L., Huibregtse, J. M., and Howley, P. M. (1998). J. Biol. Chem. 273(11), 6439–6445. Taylor, J. A., Li, Y., He, M., Mason, T., Mettlin, C., Vogler, W. J., Maygarde, S., and Liu, E. (1996). Cancer Res. 56, 294–298. Thomsen, L. L., Miles, D. W., Happerﬁeld, L., Bobrow, L. D., Knowles, R. G., and Moncada, S. (1995). Br. J. Cancer 72, 41–44. Thut, C. J., Chen, J. L., Klemm, R., and Tjian, R. (1995). Science 267, 100–104. Tokino, T., Thiagalingam, S., El-Deiry, W. S., Waldman, T., Kinzler, K. W., and Vogelstein, B. (1994). Hum. Mol. Genet. 3(9), 1537–1542. Tornaletti, S., and Pfeifer, G. P. (1994). Science 263, 1436–1438. Tornaletti, S., Rozek, D., and Pfeifer, G. P. (1993). Oncogene 8(8), 2051–2057. Trink, B., Okami, K., Wu, L., Sriuranpong, V., Jen, J., and Sidransky, D. (1998). Nat. Med. 4(7), 747–748. Trivers, G. E., Cawley, H. L., DeBenedetti, V. M., Hollstein, M., Marion, M. J., Bennett, W. P., Hoover, M. L., Prives, C. C., Tamburro, C. C., and Harris, C. C. (1995). J. Natl. Cancer Inst. 87(18), 1400–1407. Truant, R., Xiao, H., Ingles, C. J., and Greenblatt, J. (1993). J. Biol. Chem. 268(4), 2284–2287. Ueda, H., Ullrich, S. J., Gangemi, J. D., Kappel, C. A., Ngo, L., Feitelson, M. A., and Jay, G. (1995). Nat. Genet. 9(1), 41–47. Varon, R., Vissinga, C., Platzer, M., Cerosaletti, K. M., Chrzanowska, K. H., Saar, K., Beckmann, G., Seemanova, E., Cooper, P. R., Nowak, N. J., Stumm, M., Weemaes, C. M., Gatti, R. A., Wilson, R. K., Digweed, M., Rosenthal, A., Sperling, K., Concannon, P., and Reis, A. (1998). Cell (Cambridge, Mass.) 93(3), 467–476. Vega, F. J., Iniesta, P., Caldes, T., Sanchez, A., Lopez, J. A., de Juan, C., Diaz-Rubio, E., Torres, A., Balibrea, J. L., and Benito, M. (1997). Br. J. Cancer 76(1), 44–51. Venot, C., Maratrat, M., Dureuil, C., Conseiller, E., Bracco, L., and Debussche, L. (1998). EMBO J. 17(16), 4668–4679. Verghis, S. B., Essigmann, J. M., Kadlubar, F. F., Morningstar, M. L., and Lasko, D. D. (1997). Carcinogenesis (London) 18(12), 2403–2414. Verhaegh, G. W., Richard, M. J., and Hainaut, P. (1997). Mol. Cell. Biol. 17(10), 5699– 5706. Verhaegh, G. W., Parat, M. O., Richard, M. J., and Hainaut, P. (1998). Mol. Carcinog. 21(3), 205–214. Vineis, P., Talaska, G., Malaveille, C., Bartsch, H., Martone, T., Sithisarankul, P., and Strickland, P. (1996). Int. J. Cancer 65(3), 314–316. von Brevern, M. C., Hollstein, M. C., Cawley, H. M., De Benedetti, V. M., Bennett, W. P., Liang, L., He, A. G., Zhu, S. M., Tursz, T., Janin, N., and Trivers, G. E. (1996). Cancer Res. 56(21), 4917–4921. Walker, K. K., and Levine, A. J. (1996). Proc. Natl. Acad. Sci. U.S.A. 93(26), 15335–15340. Wallingford, J. B., Seufert, D. W., Virta, V. C., and Vize, P. D. (1997). Curr. Biol. 7(10), 747– 757. Wang, X. W., Forrester, K., Yeh, H., Feitelson, M. A., Gu, J. R., and Harris, C. C. (1994). Proc. Natl. Acad. Sci. U.S.A. 91(6), 2230–2234.

136

Pierre Hainaut and Monica Hollstein

Wang, X. W., Yeh, H., Schaeffer, L., Roy, R., Moncollin, V., Egly, J. M., Wang, Z., Freidberg, E. C., Evans, M. K., Taffe, B. G., Bohr, V. A., Weeda, G., Hoeijmakers, J. H. J., Forrester, K., and Harris, C. C. (1995). Nat. Genet. 10(2), 188–195. Wang, X. W., Vermeulen, W., Coursen, J. D., Gibson, M., Lupold, S. E., Forrester, K., Xu, G., Elmore, L., Yeh, H., Hoeijmakers, J. H., and Harris, C. C. (1996). Genes Dev. 10(10), 1219–1232. Wang, Y., and Prives, C. (1995). Nature (London) 376, 88–91. Waridel, F., Estreicher, A., Bron, L., Flaman, J. M., Fontolliet, C., Monnier, P., Frebourg, T., and Iggo, R. (1997). Oncogene 14(2), 163–169. Warren, W., Biggs, P. J., el-Baz, M., Ghoneim, M. A., Stratton, M. R., and Venitt, S. (1995). Carcinogenesis (London) 16(5), 1181–1189. Waterman, M. J., Stavridi, E. S., Waterman, J. L., and Halazonetis, T. D. (1998). Nat. Genet. 19(2), 175–178. Werner, H., Karnieli, E., Rauscher, F. J., and LeRoith, D. (1996). Proc. Natl. Acad. Sci. U.S.A. 93(16), 8318–8323. Weston, A., Caporaso, N. E., Perrin, L. S., Sugimura, H., Tamai, S., Krontiris, T. G., Trump, B. F., Hoover, R. N., and Harris, C. C. (1992). Environ. Health Perspect. 98, 61–67. Weston, A., Ling-Cawley, H. M., Caporaso, N. E., Bowman, E. D., Hoover, R. N., Trump, B. F., and Harris, C. C. (1994). Carcinogenesis (London) 15(4), 583–587. Westphal, C. H. (1997). Curr. Biol. 7(12), R789–R792. Whitacre, C. M., Hashimoto, H., Tsai, M. L., Chatterjee, S., Berger, S. J., and Berger, N. A. (1995). Cancer Res. 55(17), 3697–3701. Willems, P. M., Meyerink, J. P., van de Locht, L. T., Smetsers, T. F., de Vries, N., and Mensink, E. J. (1992). Nucleic Acids Res. 20(5), 1172. Wiman, K. G., and Klein, G. (1997). Nat. Med. 3(8), 839–840. Wink, D. A., Kasprzak, K. S., Maragos, C. M., Elespuru, R. K., Misra, M., Dunams, T. M., Cebula, T. A., Koch, W. H., Andrews, A. W., Allen, J. S., and Kiefer, L. K. (1991). Science 254, 1001–1003. Wink, D. A., Vodovotz, Y., Laval, J., Laval, F., Dewhirst, M. W., and Mitchell, J. B. (1998). Carcinogenesis (London) 19, 711–721. Woo, R. A., McLure, K. G., Lees-Miller, S. P., Rancourt, D. E., and Lee, P. W. (1998). Nature (London) 394, 700–704. Wynford-Thomas, D. (1996). J. Pathol. 180(2), 118–121. Wynford-Thomas, D. (1997). Eur. J. Cancer 33(5), 716–726. Wynford-Thomas, D., and Blaydes, J. (1998). Carcinogenesis (London) 19(1), 29–36. Xiao, W., and Samson, L. (1993). Proc. Natl. Acad. Sci. U.S.A. 90(6), 2117–2121. Xu, Y., Yang, E. M., Brugarolas, J., Jacks, T., and Baltimore, D. (1998). Mol. Cell. Biol. 18(7), 4385–4390. Yasunaga, Y., Nakanishi, H., Naka, N., Miki, T., Tsujimura, T., Itatani, H., Okuyama, A., and Aozasa, K. (1997). Lab. Invest. 77(6), 677–684. Yin, Y., Tainsky, M. A., Bischoff, F. Z., Strong, L. C., and Wahl, G. M. (1992). Cell (Cambridge, Mass.) 70(6), 937–948. Yin, C., Knudson, C. M., Korsmeyer, S. J., and Van Dyke, T. (1997). Nature (London) 385, 637–640. Yin, Y., Terauchi, Y., Solomon, G. G., Aizawa, S., Rangarajan, P. N., Yazaki, Y., Kadowaki, T., and Barrett, J. C. (1998). Nature (London) 391, 707–710. Yuan, Z. M., Huang, Y., Fan, M. M., Sawyers, C., Kharbanda, S., and Kufe, D. (1996a). J. Biol. Chem. 271(43), 26457–26460. Yuan, Z. M., Huang, Y., Whang, Y., Sawyers, C., Weichselbaum, R., Kharbanda, S., and Kufe, D. (1996b). Nature (London) 382, 272–274. Zambetti, G. P., and Levine, A. J. (1993). FASEB J. 7(10), 855–865.

p53 and Human Cancer

137

Zamzami, N., Susin, S. A., Marchetti, P., Hirsch, T., Gomez-Monterrey, I., Castedo, M., and Kroemer, G. (1996). J. Exp. Med. 183(4), 1533–1544. Zhan, Q., Lord, K. A., Alamo, I., Jr., Hollander, M. C., Carrier, F., Ron, D., Kohn, K. W., Hoffman, B., Liebermann, D. A., and Fornace, A. J., Jr. (1994). Mol. Cell. Biol. 14(4), 2361–2371. Zhan, Q., Chen, I. T., Antinore, M. J., and Fornace, A. J., Jr. (1998). Mol. Cell. Biol. 18(5), 2768–2778. Zhang, H., Somasundaram, K., Peng, Y., Tian, H., Bi, D., Weber, B. L., and El-Deiry, W. S. (1998). Oncogene 16(13), 1713–1721. Zhang, W., Hu, G., and Deisseroth, A. (1992). Gene 117(2), 271–275. Zhang, W., Guo, X. Y., Hu, G. Y., Liu, W. B., Shay, J. W., and Deisseroth, A. B. (1994). EMBO J. 13(11), 2535–2544. Zhang, Y., Xiong, Y., and Yarbrough, W. G. (1998). Cell (Cambridge, Mass.) 92(6), 725–734. Ziegler, A., Jonason, A. S., Leffell, D. J., Simon, J. A., Sharma, H. W., Kimmelman, J., Remington, L., Jacks, T., and Brash, D. E. (1994). Nature (London) 372, 773–776.

This Page Intentionally Left Blank

Macrophage Stimulating Protein Edward J. Leonard and Alla Danilkovitch Laboratory of Immunobiology NCI–Frederick Cancer Research and Development Center Frederick, Maryland 21702

I. II. III. IV. V.

VI.

VII.

VIII.

IX.

Introduction Structural Aspects of Native and Recombinant Pro-MSP Binding of MSP to its Receptor A Model of MSP-Induced Receptor Dimerization Regulation of MSP Activity: Pathways for Pro-MSP Cleavage A. Pro-MSP Convertase Activity of Bovine or Human Serum B. Pro-MSP Cleavage Activity on the Surface of Murine Peritoneal Macrophages C. Pro-MSP Convertase Activity in Human Wound Exudates MSP–Ron Signaling A. Activation of Ras-Dependent Pathways B. Activation of PI3K Pathways C. Activation of Focal Adhesion Kinase D. Activation of Src Kinase E. Activation of JNK F. Possible Role of MSP–Ron Signaling Transduction Components in Tumor Progression Modes of MSP Receptor Regulation or Activation A. Upregulation of Normal Receptor Expression as Part of a Host Defense Response B. Increased Ron Receptor Expression in Abnormal Cells C. Ligand-Independent Activation of Mutant Ron Receptors Target Cells for Macrophage Stimulating Protein A. Macrophages B. Osteoclasts C. Cells of Ectodermal Origin D. Vascular Marrow Cells E. Bone Marrow Cells F. Insights from Message Expression and from Knockout Mice Perspective References

I. INTRODUCTION MSP belongs to a family of proteins, typical members of which are serine proteases such as plasminogen or prothrombin. They are secreted as singlechain molecules that have no biological or enzymatic activity until they are proteolytically cleaved, generally at a single site to form a disulﬁde-linked Advances in CANCER RESEARCH 0065-230X/00 $30.00

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

140

Edward J. Leonard and Alla Danilkovitch

chain heterodimer (Fig. 1) (Francis and Marder, 1990). The family is characterized by a conserved triple disulﬁde loop structure (kringle) that occurs in multiple copies in the chain of the protein. MSP has a 45% sequence similarity to hepatocyte growth factor/scatter factor (HGF/SF) (Yoshimura et al., 1993). Hence, MSP is called hepatocyte growth factor-like protein by some investigators (Han et al., 1991). Whereas most of the proteins in this family are serine proteases, MSP and HGF have no enzymatic activity because of amino acid substitutions in the catalytic triad. Instead they are thought to have evolved from an ancient coagulation protein (Donate et al., 1994) to become growth and motility factors, while retaining the proteasedependent activation mechanism of the kringle protein family. The receptor for MSP (called Ron in the human and Stk in the mouse) is

Fig. 1 Schematic representation of the structure and processing of pro-MSP to mature MSP. Pro-MSP is a single-chain, 711-residue protein, which is biologically inactive until it is cleaved at Arg483-Val484 (solid arrow) by trypsin-like serine proteases to form mature disulﬁde-linked -chain MSP. The interchain disulﬁde (dashed line) is formed by -chain Cys468 and -chain Cys588. The chain is a 53-kDa protein containing an N-terminal hairpin loop and four kringle domains (K1–K4). The chain is a 25-kDa protein containing a serine protease-like domain, which is enzymatically inactive because the catalytic triad His, Asp, and Ser was replaced by Gln, Gln, and Tyr (open circles). Five Arg residues (639, 641, 683, 687, 689) forming positively charged cluster and/or S1 substrate pocket Glu648 and Asn682 are hypothesized to play a role in binding of MSP to its receptor. In some recombinant pro-MSP molecules, the interchain disulﬁde is absent, because Cys672 coopts Cys588 to form an intrachain disulﬁde.

Macrophage Stimulating Protein

141

one of a small family of receptor tyrosine kinases (RTKs) that comprises human Met (Bottaro et al., 1991) and chicken Sea (Huff et al., 1993). Met and Sea can undergo neoplastic transformation. Met was ﬁrst isolated from a human osteosarcoma cell line as an oncogene fused with the tpr locus (Cooper et al., 1984; Park et al., 1986), and oncogenic Sea causes avian erythroblastosis (Huff et al., 1993). Hence, MSP/Ron is an appropriate subject for review in this series, because of the possibility of oncogenic receptor mutants.

II. STRUCTURAL ASPECTS OF NATIVE AND RECOMBINANT PRO-MSP The MSP gene was cloned by screening a human genomic library with a probe coding for a kringle region of prothrombin (Han et al., 1991); the cDNA was cloned independently from a hepatocarcinoma cell line (Yoshimura et al., 1993). Development of a high production expression system in CHO cells provided recombinant pro-MSP for structural and biological studies (Yoshikawa et al., 1999). It was shown that MSP is a glycoprotein, with N-linked sugars at the 3 Asn loci that add a total of about 5000 Da to the mass of the molecule. The oligosaccharides at these loci are heterogeneous; four different sugars are bound to Asn72, six to Asn296, and eight to Asn615. Cleavage of pro-MSP to MSP was shown to occur at Arg483Val484. When recombinant pro-MSP was cleaved by kallikrein at the activation site to produce the expected -chain disulﬁde-linked heterodimer, SDS–PAGE under nonreducing conditions showed not only the heterodimer, but also free and chains (Yoshikawa et al., 1999). Thus, the CHO cell recombinant pro-MSP was heterogeneous: Some molecules had the -chain disulﬁde linkage, others did not. In the aberrant molecules, -chain Cys588 formed an intrachain disulﬁde with Cys672 instead of with -chain Cys468. When Cys672 was mutated to Ala, almost all recombinant pro-MSP had the interchain disulﬁde (Wahl et al., 1997). Interestingly, circular dichroism analysis of recombinant pro-MSP did not reveal detectable secondary structure (D. Xie and E. J. Leonard, unpublished data). This suggests that in the absence of a stable folded structure, formation in the CHO cell of either the interchain or intrachain disulﬁde is a random event, which results in the secretion of both types of molecules. The liver constitutively produces single-chain pro-MSP, which is released into the circulating blood. In contrast to the recombinant protein, isoelectric focusing studies show that the native material has secondary structure. The isoelectric point (pI) of the native protein in different samples of human serum was 5.5 to 6.2. In 6 M urea, the pI shifted to 7.6 (Leonard et al., 1982),

142

Edward J. Leonard and Alla Danilkovitch

a value close to the pI of 7.0 calculated from the amino acid composition. The pI shift suggests that the native protein is folded in a way that buries some of the alkaline amino acid residues. The difference between native and recombinant pro-MSP with respect to secondary structure injects a cautionary note for studies that used the recombinant protein for pharmacokinetics (Leonard and Skeel, 1996) or for identifying speciﬁc pro-MSP convertases. It would be prudent to repeat these studies with natural pro-MSP.

III. BINDING OF MSP TO ITS RECEPTOR The receptor for MSP is the human Ron gene product (Wang et al., 1994a; Gaudino et al., 1994), a transmembrane protein tyrosine kinase cloned from a keratinocyte cDNA library (Ronsin et al., 1993). The STK gene, cloned from hematopoietic stem cells of mouse bone marrow (Iwama et al., 1994), is the murine homolog of Ron (Wang et al., 1995). The Ron gene encodes a 190-kDa protein, the mature form of which is a disulﬁde-linked heterodimer comprising a 40-kDa extracellular chain and a 150-kDa chain (Ronsin et al., 1993). The chain has an extracellular domain, a transmembrane segment, and a large cytoplasmic tail with an intrinsic tyrosine kinase, the activity of which is increased by ligand–receptor binding. In a study to identify the domains of MSP that interact with its receptor (Wang et al., 1997), ﬁve recombinant proteins were used, including pro-MSP, MSP, MSP and chains, and MSP-NK2 (an IgG Fc fusion protein comprising the MSP N terminus including the ﬁrst two kringles). The free and chains were puriﬁed from preparations of recombinant MSP molecules with the mismatched disulﬁde described earlier. Saturable binding of 125IMSP was shown for various cells expressing the MSP receptor, including mouse and human epithelial cell lines as well as MDCK cells transfected with human Ron (MDCK-RE7 cells). The binding of radiolabeled MSP was completely inhibited by a 10- to 15-fold molar excess of unlabeled MSP, partially inhibited by a comparable excess of chain, and not inhibited by chain. Interaction of free chain with the MSP receptor was conﬁrmed by showing saturable binding of 125I- chain to MDCK-RE7 cells, in contrast to no speciﬁc binding of radiolabeled chain or MSP-NK2. The estimated Kd of 1.7 nM for chain binding was higher than the Kd of 0.6–0.8 nM for MSP, reﬂecting the differences in potency as competitive inhibitors noted earlier. HGF and MSP are thought to have evolved from a common ancestor gene and, in addition to a 45% sequence similarity, they have comparable domain structures and mechanisms of activation. Furthermore, the receptors for HGF and MSP are in the same subfamily, and have many unique structural features in common (Ronsin et al., 1993; Huff et al., 1993; Park et al., 1987).

Macrophage Stimulating Protein

143

The ﬁnding that MSP chain, not chain, bound to Ron was therefore completely unanticipated, because the high-afﬁnity binding site of HGF for its receptor is in the chain. A critical region for HGF binding is a hairpin loop in the N-terminal domain, deletion of which eliminates binding (Matsumoto et al., 1991; Okigaki et al., 1992). An energy minimization model of the hairpin loops of HGF and MSP revealed a three-arginine positively charged face of the HGF loop, in contrast to only one arginine for the corresponding region of MSP (Donate et al., 1994). A role for the arginines in HGF receptor binding was suggested by diminished activity of an HGF mutant in which alanines were substituted for the 3 arginines (though the authors did not rule out the possibility of agonist instability to account for their results) (Sakata et al., 1997). The absence of a highly charged MSP hairpin loop region and undetectable MSP -chain binding to Ron are consistent with this suggested role. We then considered the possibility that binding of the MSP chain to Ron might also be via a positively charged region. Accordingly, an energy-minimized model of the MSP chain was constructed (M. Miller, A. Danilkovitch, and E. J. Leonard, unpublished data) using coordinates from HGF modeling (Donate et al., 1994) based on the structures of three mammalian serine proteases. This revealed a cluster of ﬁve arginines on surface loops of the MSP chain. A critical role of this region for binding was established when substitution of one of the arginine residues for alanine caused complete loss of -chain binding to Ron.

IV. A MODEL OF MSP-INDUCED RECEPTOR DIMERIZATION It is generally accepted that ligand binding to transmembrane receptor protein tyrosine kinases causes receptor dimerization and autophosphorylation (Ullrich and Schlessinger, 1990). Two different examples of receptor dimerization by growth factors may be relevant to activation of Ron by MSP. One is the binding of stem cell factor (SCF) to its receptor, kit: SCF dimers bind to pairs of Kit receptors, the stoichiometry being 2:2 (Philo et al., 1996; Lemmon et al., 1997). In the case of human growth hormone (HGH), the stoichiometry is 1:2: One region of an HGH monomer binds with high afﬁnity to its receptor (R1), after which another region of HGH binds to a second receptor (R2). Thus, the HGH monomer is bivalent with respect to receptor binding. Although the second site binding afﬁnity is lower, the ligand–receptor complex is stabilized by an R1/R2 interaction as the receptors are brought into proximity by HGH (Cunningham et al., 1991; de Vos et al., 1992).

144

Edward J. Leonard and Alla Danilkovitch

The following considerations lead us to favor the HGH model for MSPinduced dimerization of Ron (Miller and Leonard, 1998). Although the MSP chain binds to cells expressing Ron, it does not induce a cellular response. Only the MSP -chain disulﬁde-linked heterodimer induces biological activity and presumably receptor dimerization. Additional evidence for the importance of both and chains for biological activity of MSP has recently been published (Bezerra et al., 1998). Although the requirement for intact MSP might reﬂect the formation of ligand dimers, there is no evidence by gel ﬁltration of MSP dimers at physiologic concentrations. An alternative is that MSP, like HGH, is bivalent, with a high-afﬁnity -chain binding site and a second low-afﬁnity binding site for Ron on the chain. The converse might apply to HGF, for which a high-afﬁnity binding site has been demonstrated on the chain, and for which a low-afﬁnity site on the chain can be postulated, since critical amino acid substitutions in the chain cause decreases in biological activity to 2% of wild-type HGF. Detection of MSP -chain binding to Ron would provide support for the HGH model. We did not observe any speciﬁc binding of 125I- chain to Ronexpressing cells (Wang et al., 1997). However, after equilibration of unlabeled chain with MDCK-RE7 cells at 4C, followed by washing of ﬂuid phase chain, we detected by Western blot bound chain in immunoprecipitated cell lysates. No chain was adsorbed to parental MDCK cells that did not express Ron. This ﬁnding supports the idea that MSP may be bivalent with respect to receptor binding, so that one ligand molecule is capable of inducing receptor dimerization. When soluble receptor becomes available, it should be possible to determine if ligand:receptor stoichiometry is indeed 1:2.

V. REGULATION OF MSP ACTIVITY: PATHWAYS FOR PRO-MSP CLEAVAGE Like kringle proteins of the coagulation system, MSP is constitutively synthesized by hepatic parenchymal cells. In a study of the biochemical basis for liver-speciﬁc transcription of the MSP gene, both positive and negative regulatory elements have been found. The latter are thought to account for inhibition of transcription in nonhepatic cells (Ueda et al., 1998). The gene product is secreted by the liver into the circulating blood as biologically inactive pro-MSP. The mean concentration of pro-MSP in the plasma of a series of normal human subjects is 4 nM, which is in the range for optimal biological activity (Nanney et al., 1998). The level is not changed in the course of an acute phase reaction (Wang et al., 1993). To act on target cells in extravascular sites, this protein of hepatic origin must diffuse into tissues and

Macrophage Stimulating Protein

145

be cleaved by a pro-MSP convertase to biologically active MSP. Although R483-V484 is a cleavage site, a pro-MSP mutant with that site blocked (R483E) has biological activity, which indicates that cleavage at an alternative locus can yield a functional protein (Waltz et al., 1997).

A. Pro-MSP Convertase Activity of Bovine or Human Serum The R483-V484 scissile bond is a typical cleavage site for trypsin-like serine proteases. Indeed, several such puriﬁed proteases of the coagulation system, including factors XIa and XIIa and serum kallikrein cleave pro-MSP to active MSP (Wang et al., 1994b). Cleavage does not occur when blood clots, indicating that pro-MSP is not a preferred substrate for these enzymes (Wang et al., 1996c). However, as noted later, serum has pro-MSP convertase activity. CHO cells transfected with the pro-MSP cDNA have been used to express recombinant protein. When the cells were grown in serum-free medium, no cleavage of secreted pro-MSP was detected in culture ﬂuid sampled over a 36-hr period. In contrast, in cultures containing 10% heat-activated fetal bovine serum, partial cleavage of pro-MSP to MSP was detected after 16 hr in culture, which was almost complete at 36 hr. Cleavage was prevented by inhibitors of trypsin-like serine proteases (aprotinin, leupeptin, and soybean trypsin inhibitor) (Wang et al., 1994a). The cleavage product was biologically active. In a cell-free system, fetal bovine serum, normal calf serum, and normal human serum partially cleaved 125I–pro-MSP to MSP within 3 hr at 37C (A. Skeel and E. J. Leonard, unpublished data). These results indicate that fetal bovine serum accounts for pro-MSP cleavage in transfected CHO cell cultures. Although cleavage of pro-MSP in vitro by serum is important in the laboratory, it probably does not relate to the requirement for convertase activity in local tissue sites. Two pro-MSP convertase activities of potential in vivo relevance have been described.

B. Pro-MSP Cleavage Activity on the Surface of Murine Peritoneal Macrophages When labeled pro-MSP was incubated with murine peritoneal macrophages and the culture ﬂuids sampled within 3 hr and analyzed by immunoprecipitation and radioautography, several cleavage products were detected (Wang et al., 1996c). The and chains of mature MSP (about 53 and 30 kDa) comprised only a small fraction of the total; most of the cleav-

146

Edward J. Leonard and Alla Danilkovitch

age product comprised 47- and 35-kDa fragments. When the experiment was repeated with a panel of serine protease inhibitors, AEBSF and soybean trypsin inhibitor prevented the degradative cleavage, so that the predominant product was mature MSP. It thus appeared that this macrophage population had two pro-MSP cleavage activities that could be differentially inhibited by selected protease inhibitors. Low concentrations of murine or human serum had effects similar to that of STI. The protein in human serum that inhibited the degradative cleavage of pro-MSP by macrophages is 1antichymotrypsin (A. Skeel and E. J. Leonard, unpublished data). These results suggest an interesting mode of localizing the site of MSP action, in which the peritoneal macrophage, one of the target cells for MSP, also has the capacity to cleave pro-MSP to active MSP. An additional level of control is apparently provided by a pro-MSP degrading enzyme and its inhibitor, 1-antichymotrypsin. Blood monocytes have neither the MSP receptor nor pro-MSP cleavage activities, indicating that both properties are late events in mononuclear phagocyte maturation. We recently showed that all macrophages in normal human dermis express the MSP receptor (Nanney et al., 1998). It would be important to know if these cells, and tissue macrophages in general, have pro-MSP convertase activity, providing a mechanism for physiologic regulation of this growth and motility factor.

C. Pro-MSP Convertase Activity in Human Wound Exudates In contrast to the undetectable or minimal amount of mature MSP in normal human blood plasma, MSP comprises about half the total pro-MSP MSP found in surgical drainage ﬂuid or human burn wound ﬂuid (Nanney et al., 1998). We conclude that the circulating pro-MSP that diffuses into the wound site is cleaved by cell-associated or ﬂuid phase enzymes or both. In a large series of tested wound exudates, all cleaved labeled pro-MSP to MSP. The enzymatic activity was inhibited by leupeptin and aprotinin, but not by STI, 2-macroglobulin or C1-inhibitor. This inhibition proﬁle is the same as for human tissue kallikrein, which brought to mind the fact that in a series of serine proteases tested for pro-MSP convertase activity, two murine tissue kallikreins were the most potent (Wang et al., 1994c). However, in studies of four wound ﬂuids, Chao showed that the kallikrein inhibitor kallistatin was present in great molar excess over tissue kallikrein; and highly puriﬁed tissue kallikrein (KLK1) did not cleave pro-MSP ( J. Chao, unpublished data). If we succeed in our current efforts to purify and characterize the wound ﬂuid pro-MSP convertase activity, it may be possible to determine if this is a product of one of the cell types at the site of tissue injury.

Macrophage Stimulating Protein

147

VI. MSP–RON SIGNALING Interaction of MSP with Ron leads to reactions typical for receptor tyrosine kinases, transphosphorylation of receptor cytoplasmic domains and Ron kinase activation (Lemmon and Schlessinger, 1994; Heldin, 1995) (Fig. 2). Upregulation of kinase activity is caused by phosphorylation of Y1238 and Y1239 of the kinase domain (Gaudino et al., 1994). The active kinase phosphorylates two C-terminal tyrosine residues, Y1353 and Y1360, which creates a high-afﬁnity binding site for intracellular transducing molecules containing Src homology 2 (SH2) and/or phosphotyrosine binding (PTB) domains (Iwama et al., 1996). Y1353 and Y1360 represent a binding site for a number of SH2 proteins that selectively recognize the motif pTyr-hydrophobic-X-hydrophobic (Songyang et al., 1993, 1994). This doublet of tyrosine residues is also found in the other members of the Ron receptor family, Met and c-Sea (Ponzetto et al., 1994). The doublet is a multifunctional docking site that can directly interact with PLC, PI3K, p85, Shc, and Grb2 (Iwama et al., 1996). Mutations of these two tyrosines abolished MSPmediated biological effects despite an unchanged kinase activity (Iwama et

Fig. 2 MSP–Ron interaction induces receptor dimerization and subsequent Ron kinase activation and tyrosine phosphorylation. Phosphorylated tyrosine residues are docking sites for numerous intracellular molecules. PLC, PI3K, Grb2, and Shc directly interact with Ron and activate multiple signaling pathways. Ras, MAPK, and AKT are downstream components that mediate the indicated MSP activities. FAK, Src, and JNK are also involved in MSP/Ron signaling, but their upstream signals and/or their direct interaction with Ron have not yet been identiﬁed.

148

Edward J. Leonard and Alla Danilkovitch

al., 1996). Although the cytoplasmic part of Ron contains 14 tyrosine residues that might serve as phosphorylation sites, there are no data about the functional role of most of them, and it is proposed that the pleotropic response induced by MSP results from the activation of multiple signaling pathways by recruitment of intracellular components to the Y1353/Y1360 multifunctional docking site.

A. Activation of Ras-Dependent Pathways Ras is a 21-kDa protein belonging to the small molecular weight GTPbinding protein family. Ras proteins regulate cell proliferation and differentiation (Heimbrook et al., 1997). Ras is active in GTP-bound form and acts through protein–protein interactions involving the Ras effector loop (amino acid residues 32–40) (Marshall, 1995; Hwang et al., 1996). The Ras-dependent pathway in cells involves activation of the MAPK cascade by ligands of protein tyrosine kinases. GTP-bound RAS binds cytoplasmic Raf1, the ﬁrst member of the MAPK cascade that terminates with activation of transcriptional factors required for the S phase of the cell cycle (Marshall, 1995). Ras can also regulate Raf-independent signals via Rac/Rho and PI3K (Rodriguez-Viciana et al., 1994; Qiu et al., 1995a,b). In MDCK cells with overexpressed Ron receptor, the SH2-domain containing adaptor protein Grb2 bound to activated Ron chain (Li et al., 1995). Grb2, which recognizes phosphotyrosine-X-N sequences (Songyang et al., 1994), may therefore bind to Ron at its C-terminal Y1360/N1362 motif (Iwama et al., 1996). Binding of Grb2 to Ron was associated with increased GTP-Ras due to activation of SOS (Li et al., 1995), a protein that catalyzes the exchange of GDP for GTP on Ras and activates Ras-dependent signaling pathways (Aronheim et al., 1994). MSP caused translocation of Grb2–SOS complexes from cytosol to plasma membrane, where the target Ras is located. In contrast to the absence of binding of the Shc adaptor protein to Ron (Li et al., 1995), association of Shc with activated Stk (the murine MSP receptor) was detected in murine hematopoietic cell lines stably transfected with Stk (Iwama et al., 1996). Thus, it is possible for Sch to recruit Grb2 to Stk, and it has been shown that Grb2 recognizes phosphorylated Shc (Clark et al., 1992). Experiments with overexpression of dominant negative N17Ras demonstrated that Ras does not mediate MSPinduced cell motility (Wang et al., 1996). Activated Ras is implicated in MSP mitogenic signaling (A. Danilkovitch, unpublished observations), although it is possible that in cells where activation of PI3K is Ras dependent, Ras might also mediate MSP-induced motility. The role of PI3K in MSP signaling is discussed next.

Macrophage Stimulating Protein

149

B. Activation of PI3K Pathways Phosphatidylinositol-3-kinase (PI3K) is a heterodimeric protein consisting of an 85-kDa regulatory subunit with two SH2 domains and one SH3 domain, and a 110-kDa catalytic subunit. PI3K phosphorylates the D-3 position of the inositol ring of phosphatidylinositol, phosphatidylinositol-4phosphate, and phosphatidylinositol-4,5-bisphosphate (Cantley et al., 1991). PI3K is associated with numerous activated growth factor receptors via SH2 domains and is involved in growth factor-mediated signaling (Kapeller and Cantley, 1994; Hu et al., 1992). MSP induces PI3K tyrosine phosphorylation, and PI3K catalytic activity is necessary to transduce MSP signaling: Wortmannin, a speciﬁc inhibitor of PI3K activity, inhibits MSP-dependent adhesion, motility, and cell shape change. Addition of MSP to epithelial cells expressing Ron causes PI3K to associate directly with Ron (Wang et al., 1996b). Studies of the binding speciﬁcity of the p85 SH2 domains reveal a core phosphotyrosine containing motif p Tyr-X-X-Met on target proteins (Varticovski et al., 1994), and Ron has such a motif around Y1317 (Ronsin et al., 1993). On the other hand, it has been shown that PI3K may interact with Y1353 and Y1360 of the multifunctional docking site (Iwama et al., 1996). In contrast to signaling by many other growth factors, where PI3K activation is Ras dependent (RodriguezViciana et al., 1994), it appears that activated Ron binds PI3K directly and stimulates catalytic activity. The lack of inhibition of MSP-mediated cell motility by dominant negative N17Ras supports this idea (Wang et al., 1996b). Downstream components of PI3K signaling include ribosomal protein kinase p70S6K (Chung et al., 1994), Rho family GTP-binding protein Rac (Parker, 1995), PKC (Akimoto et al., 1996; Moriya et al., 1996), and AKT (Franke et al., 1997). Rac is important for PI3K-dependent actin rearrangement (Reif et al., 1996), and thus it may mediate MSP-induced cell shape change, adhesion, and motility. AKT is a serine/threonine-speciﬁc kinase that is important for protection of cells from apoptosis (Kauffmann-Zeh et al., 1997; Songyang et al., 1997; Dudek et al., 1997). We found that MSP may protect cells from apoptosis via PI3K and AKT. MSP induces PI3K-dependent AKT activation, and dominant-negative AKT abolishes the MSP antiapoptotic effect (A. Danilkovitch and E. J. Leonard, unpublished data). Thus, PI3K is important for MSP-induced cell survival as well as adhesion and motility. PI3K is required for HGF-induced mitogenic signals in epithelial cells (Rahimi et al., 1996). We found that PI3K is also involved in MSP mitogenic signaling, but this effect is related to the ability of PI3K to protect cells from apoptosis via its downstream component AKT kinase (A. Danilkovitch and E. J. Leonard, unpublished data).

150

Edward J. Leonard and Alla Danilkovitch

C. Activation of Focal Adhesion Kinase Focal adhesion kinase (FAK) is a nonreceptor cytoplasmic tyrosine kinase implicated in cell growth, survival, adhesion, and motility (Hanks and Polte, 1997). FAK is spatially and functionally associated with the chain of integrins, cell surface receptors that are responsible for the interaction of cells with extracellular matrix (ECM) (Schaller et al., 1992; Hanks et al., 1992). ECM-induced integrin receptor aggregation is accompanied by FAK tyrosine phosphorylation and kinase activation, followed by activation of downstream signals (Hanks and Polte, 1997). FAK tyrosine phosphorylation and activation may also be regulated by growth factors (Abedi and Zachary, 1997; Rankin and Rozengurt, 1994; Baron et al., 1998; Zhu et al., 1998). MSP, like HGF (Matsumoto et al., 1994), induces FAK tyrosine phosphorylation and activation (Danilkovitch and Leonard, 1997). FAK is a component of the signaling path that mediates MSP-induced cell proliferation: Inhibition of FAK activity via overexpression of dominant-negative FAK leads to signiﬁcant reduction of MSP-induced mitogenic signaling (Danilkovitch and Leonard, 1998, also unpublished data). These data demonstrate that there is crosstalk between signal transduction pathways coming from two different receptors, Ron and integrins.

D. Activation of Src Kinase Src and Src related kinases are included in the large family of cytoplasmic tyrosine kinases on the basis of their common domain structure: acetylated amino terminus, SH2 and SH3 domains, and highly conserved catalytic domain (Williams et al., 1998). Src kinase has little activity in cells in the absence of an activating signal, but many stimuli increase Src enzymatic activity via phosphorylation mechanisms (Cooper and Howell, 1993). Src is involved in growth factor receptor signaling: Src interacts with PDGF (Kypta et al., 1990), EGF (Maa et al., 1995), bFGF (Zhan et al., 1994), and CSF-1 (Courtneidge et al., 1993) receptors. Ligand-dependent association of Src with growth factor receptors is mediated through the Src SH2 domain (Williams et al., 1998; Erpel and Courtneidge, 1995). Src is involved in HGF/Met signaling; it directly interacts with tyrosine residues of the multifunctional docking site (Ponzetto et al., 1994). Although association of Src with Ron has not yet been shown, Src is a component of MSP–Ron signaling. In bone marrow-derived osteoclast-like cells, MSP caused ruffled border formation, which was associated with a rapid redistribution of Src from a diffuse cytoplasmic distribution to a peripheral location near the plasma membrane (Kurihara et al., 1996). Src also mediates MSP-induced cell proliferation, and MAPK activity is regulated by Src

Macrophage Stimulating Protein

151

in MSP stimulated cells: Dominant negative Src significantly reduced MSPdependent proliferation and MAPK activation (Danilkovitch and Leonard, 1998).

E. Activation of JNK The c-Jun amino-terminal kinases (JNKs) are members of the MAPK family, which are activated by cytokines and/or environmental stress (Whitmarsh and Davis, 1996). The JNK pathway regulates cell proliferation, apoptosis, and tissue morphogenesis (Ip and Davis, 1998). JNK activation is mediated by dual phosphorylation on Thr and Tyr by two MAP kinases, MKK4 and MKK7 (Whitmarsh and Davis, 1996). Rho family GTPases and PI3K also may serve as intermediates in pathways leading to JNK activation (Joneson et al., 1996; Lamarche et al., 1996; Logan et al., 1997; LopezIlasaca et al., 1997). MSP and HGF both activate the JNK pathway (Rodrigues et al., 1997). MSP-induced JNK activation in a murine erythroleukemia cell line transfected with Stk was associated with induction of apoptosis (Iwama et al., 1996). MSP may simultaneously activate two different pathways leading to either cell growth or apoptosis; cells undergo apoptosis if the apoptotic pathway overcomes the mitogenic signals. It has been proposed that MSP-induced JNK activation is related to expression of two unidentiﬁed proteins, p61 and p65. These proteins interact directly with the multifunctional docking site at the C terminus of Ron in a ligand-dependent manner (Iwama et al., 1996). In contrast to the above hematopoietic cell line, MSP-induced JNK activation in epithelial cells does not induce apoptosis (A. Danilkovitch, unpublished data). However MSP has been reported to induce apoptosis in selected human pulmonary carcinoma cell lines (Willett et al., 1997).

F. Possible Role of MSP–Ron Signal Transduction Components in Tumor Progression Growth factor receptor tyrosine kinases can activate multiple pathways that lead to mitogenesis. There are many examples of growth factor receptor overexpression in tumor cells, and their constitutive activation by various mechanisms is related to tumor formation or progression (Karunagaran et al., 1996; Tsujimura, 1996; Empereur et al., 1997; Borset et al., 1996; Moscatello et al., 1996; Schmidt et al., 1997; Santoro et al., 1995). Many of the above-described MSP–Ron signaling molecules are known to cause cell transformation. For example, Ras genes are mutated in 20 –30% of human cancers and the encoded proteins are believed to contribute to the patho-

152

Edward J. Leonard and Alla Danilkovitch

genesis of these malignancies (Lowy and Willumsen, 1993; Bos, 1989). Constitutively active Ras leads to cell transformation, which is related to activation of the Ras downstream component MAPK (Oldham et al., 1998). Active PI3K is also an essential component for efﬁcient Ras transformation (Rodriguez-Viciana et al., 1997). Activation of c-Src kinase occurs with high frequency in cells with malignant potential (Cartwright et al., 1990), in primary tumors (Cartwright et al., 1994), and in metastatic tumors (Talamonti et al., 1993). Analysis of human tumors with an invasive phenotype revealed FAK overexpression (Owens et al., 1995; Weiner et al., 1993), and this is highly correlated with invasive potential (Partin et al., 1989). Thus, FAK may be involved in tumor invasion and metastasis, and it may also be useful as a marker of tumor cell invasion. Activation of the JNK pathway is essential for cell transformation by the tpr-Met oncoprotein and, thus, JNK may also be implicated in tumor development and progression (Rodrigues et al., 1997). The relationship of Ron gene mutations to tumor formation or invasiveness is reviewed in Section VII.

VII. MODES OF MSP RECEPTOR REGULATION OR ACTIVATION Control of MSP actions on cells can be at the level of ligand or receptor (Table I). Generation of active MSP from pro-MSP has been reviewed in Section V. Receptor-based controls or activation are summarized in this section. Table I Modes of Ron Receptor Activation I.

Activation of pro-MSP (Section V) Cleavage of plasma pro-MSP in extravascular sites Cell-bound pro-MSP convertase Fluid phase pro-MSP convertase Cleavage of locally produced pro-MSP: testis; embryonic development? II. Upregulation of receptor expression as part of host defense response Human wounds (Section VII,A) Response to cytokines or growth factors (Sections VII,A and VII,B) III. Increased receptor expression in abnormal cells (Section VII,B) Primary cancers Cancer cell lines IV. Ligand-independent activation of mutant receptors (Section VII,C) Tpr-Ron -Ron Ron point mutations akin to those in Kit and Ret

Macrophage Stimulating Protein

153

A. Upregulation of Normal Receptor Expression as Part of a Host Defense Response An example is from a recent study that evaluated Ron expression in histologic sections of normal skin and of human burn wounds (Nanney et al., 1998). Ron was detected immunohistochemically with a monoclonal antibody against the extracellular domain of the receptor. Ron expression by keratinocytes was undetectable in four of six specimens of normal skin. In contrast, Ron was strongly expressed in epithelial cells of human burn wounds. The cellular distribution of increased Ron expression was variable in different wound specimens, occurring in proliferative and/or differentiated populations, the latter including cells in accessory structures such as sweat ducts and hair follicles. Frequency of Ron detection on endothelial cells of dermal capillaries was also higher in burn wounds than in normal skin, suggesting the possibility of an angiogenic effect of MSP. All dermal macrophages in both normal and burn wound skin expressed Ron. Inasmuch as biologically active MSP was found in wound exudates, the ﬁndings are consistent with a host defense response in which MSP interacts with the Ron receptor on macrophages, keratinocytes and vascular endothelium—all participants in wound healing. It is reasonable to assume that Ron upregulation observed in the preceding study is caused by one or more of the many cytokines that are generated in response to tissue injury (Martin, 1997). Chen et al. (1997) tested the effects of a series of cytokines and growth factors on MSP receptor protein expression in murine Hep 1-6 cells (a cell line derived from a murine hepatocellular carcinoma). The receptor was upregulated by IL-1, IL-6, TNF, and HGF, but not by TGF- or EGF (Chen et al., 1997).

B. Increased Ron Receptor Expression in Abnormal Cells Maggiora et al. (1998) recently studied Ron expression in 74 human primary breast carcinomas, 12 benign tumors, and 8 normal mammary glands. Solubilized protein from surgical specimens was immunoprecipitated with an anti-Ron antibody and analyzed by Western blot after SDS–PAGE. Whereas Ron protein was barely detectable in blots from normal or benign tumor tissue, Ron was overexpressed in 35 of the 74 carcinomas; in 12 of the 35, the increase was more than 20-fold. In a number of cases, Ron was phosphorylated, indicating that the receptor was activated. High expression was conﬁrmed by Ron immunoreactivity in parafﬁn sections of selected patients. Because Southern blots showed no evidence of gene ampliﬁcation, it was presumed that overexpression was mediated at the transcriptional lev-

154

Edward J. Leonard and Alla Danilkovitch

el. In a similar evaluation of three normal liver samples and seven primary hepatocellular carcinomas, Ron was overexpressed in two of the carcinomas (Chen et al., 1997). The cause of Ron overexpression in these tumors could be paracrine or autocrine. The paracrine possibility would be another example of a host response to tissue pathology in which cytokine release by normal cells that participate in the host response causes upregulation of the receptor on the tumor cells (item II in Table I). On the other hand, because Ron has been readily detected in a number of cancer cell lines that are removed from the host tissue environment (Wang et al., 1996b; Gaudino et al., 1994; Maggiora et al., 1998), Ron expression may be enhanced in autocrine fashion by factors secreted by the tumor cells. What are the functional effects of Ron overexpression in these cancer cells? In the case of Met, cells of the murine C127 cell line that are engineered to overexpress normal Met and HGF/SF become morphologically transformed and tumorigenic in vivo (for review, see Jeffers et al., 1996). In contrast, there are no examples to date of tumorigenicity for cells that overexpress normal Ron (see later section). The in vitro effects of MSP on tumor cell lines include proliferation, stimulation of motility, and capacity to migrate through an artiﬁcial ECM (Wang et al., 1996a,b; Maggiora et al., 1998), properties that might contribute to a metastatic potential in vivo provided that active MSP is present at the site.

C. Ligand-Independent Activation of Mutant Ron Receptors Three mutant Ron receptors have been described with constitutively activated tyrosine kinase. Of these three mutants, only -Ron occurs naturally; the other two mutants were engineered.

1. -RON In a screening of tumor cell lines for Ron, KATO-III, a gastric carcinoma cell line, was unique in that the predominant Ron band under reducing conditions was 165-kDa uncleaved, truncated Ron. This was in contrast to the products of other cells, which showed 170-kDa uncleaved Ron and the 150kDa chain of the normal disulﬁde-linked heterodimer. The unique -Ron band was the product of an alternatively spliced mRNA that resulted in deletion of 147 bp of message coding for a segment of Ron (Collesi et al., 1996). When the cDNA of -Ron was transfected into COS-1 cells, the product was uncleaved -Ron, which was not expressed on the cell surface and which

Macrophage Stimulating Protein

155

was tyrosine phosphorylated. In both transfected cells and KATO cells, Ron oligomers were found by SDS–PAGE under nonreducing conditions, which became monomers under reducing conditions. The authors suggest that an uneven number of cysteines in -Ron allows a free cysteine to make a disulﬁde linkage with another receptor molecule, since oligomerization was prevented by treating cells with mercaptoethanol. Furthermore, there are three highly conserved cysteines in the 49-residue deletion fragment that normally contribute to intramolecular disulﬁde bond formation. When one of these cysteines in normal Ron was mutated to alanine, the receptor was retained in the intracellular compartment and was tyrosine phosphorylated, presumably due to oligomerization. Several standard assays have been used to test for transformation of cells transfected with mutated receptors: (1) proliferation rates in low-serum medium, with estimation of a rate decrease at conﬂuence; (2) focus formation; (3) anchorage-independent growth in soft agar; and (4) formation of tumors in nude mice. In addition, the Met/Ron/Sea receptor family can induce complex patterns of motility and morphogenesis: (5) increased motility or scatter of MDCK epithelial cells, one of the original assays for the Met ligand, HGF/SF; (6) chemotaxis to ligand or increased motility of cells with constitutively activated receptor, measured as migration through pores of polycarbonate membranes; (7) “invasive” behavior, measured as migration through matrigel, an extracellular matrix mixture of collagen IV, laminin, and glycosaminoglycans; (8) in vivo invasive behavior, measured by the capacity of intravenously injected cells to colonize lungs of nude mice; (9) morphogenetic responses in three-dimensional collagen gels, typically the formation of multicellular branched tubular structures. By criteria 1 and 2, cells transfected with activated -Ron were not transformed. However, they did have an invasive phenotype by tests 6 and 7.

2. TPR-RON The oncogenic potential of Met was revealed by the discovery of Tpr-Met, in which the extracellular, transmembrane, and intracellular juxtamembrane portions of Met are replaced by the N-terminal sequence of the Tpr gene product (Cooper et al., 1984; Park et al., 1986). Tpr-Met is constitutively dimerized (Rodrigues and Park, 1993), which causes ligand-independent transphosphorylation and activation of the Met kinase. Stable transfectants of Tpr-Met in murine NIH 3T3 cells are transformed (focus formation, growth in soft agar). To test for the oncogenic potential of the other members of the Met family, transfectants of Tpr-Ron and Tpr-Sea were generated. Whereas NIH 3T3 cells transfected with Tpr-Sea were transformed, the Tpr-Ron transfectants were not. Cells with all three chimeras had higher pro-

156

Edward J. Leonard and Alla Danilkovitch

liferation rates than the medium–2% FCS control, but the rate for Tpr-Ron cells reached a plateau between 5 and 7 days, indicative of contact-inhibited growth (Santoro et al., 1996). The absence of Tpr-Ron transforming ability was possibly related to a ﬁvefold lower catalytic efﬁciency (Vmax /Km) of the Ron kinase relative to the Met kinase. From focus forming and anchorage-independent growth experiments on cells with domain swapping, it was apparently not related to the C-terminal multifunctional docking site, but to the kinase. For example, a Tpr-Ron with a Met kinase domain became transforming, and a Tpr-Met with a Ron C terminus retained its transforming ability. Although Tpr-Ron did not induce transformation, transfected 3T3 cells migrated spontaneously through 8-m pores in a transwell assay as well as through matrigel. And expression in MDCK cells induced scattering, as well as tube formation, in three-dimensional collagen gels. Thus, Tpr-Ron is similar to -Ron in its capacity to induce cellular motility and morphogenesis.

3. RON POINT MUTATIONS Santoro et al. (1998) showed that Ron does indeed have oncogenic potential, by engineering speciﬁc point mutations in the tyrosine kinase domain. The approach was based on the fact that a number of human neoplastic syndromes are associated with activating point mutations in highly conserved loci in the TK domains of Kit and Ret receptors (Santoro et al., 1995). Mutations in these receptors are found in mast cell leukemia and multiple endocrine neoplasia respectively. Likewise, similar point mutations in Met have been found in hereditary and sporadic human renal papillary carcinomas (Schmidt et al., 1997). Two different point mutations were constructed in the TK domains of both Ron and Tpr-Ron, a Kit-type mutant (D1232V) and a Ret-type mutant (M1254T). Both mutations converted Ron and Tpr-Ron to transforming genes, as shown by NIH 3T3 transfectants that caused focus formation and tumors when injected into nude mice. All transfectants also colonized murine lungs after intravenous injection. Comparison of the TK catalytic efﬁciency of nontransforming Tpr-Ron with the Tpr-Ron mutants showed that the latter were three to four times higher than Tpr-Ron. This observation lends support to the suggested importance of TK activity noted in the comparison of nontransforming TprRon and oncogenic Tpr-Met. Also of interest was the detection of two phosphorylated proteins in lysates from cells transfected with mutant, but not wild-type, receptors. This is the ﬁrst demonstration that the Ron gene shares with other members of the Met family the potential to become an oncogene. It should provide impetus to use an efﬁcient screening program to determine if there are any human cancers caused by mutations in this receptor tyrosine kinase.

Macrophage Stimulating Protein

157

VIII. TARGET CELLS FOR MACROPHAGE STIMULATING PROTEIN A. Macrophages Expression of Ron or Stk is restricted to speciﬁc subpopulations of the mononuclear phagocyte lineage and is a late maturational event. Whereas murine resident peritoneal macrophages responded to MSP, acute exudate macrophages, which are recent arrivals from the circulation, did not (Leonard and Skeel, 1980; Iwama et al., 1995). Functional responses correlated with expression of Stk, as well as with expression of F4/80, a marker of macrophage maturation. By FACS analysis, resident peritoneal macrophages were Stkhigh-F4/80high; acute exudate macrophages were Stknegative-F4/ 80low. Within 3 days after induction, the exudate macrophages acquired the mature expression pattern of resident macrophages. Other Stk-negative murine mononuclear phagocyte subpopulations by FACS analysis included gated cells from bone marrow, blood, spleen, and bronchoalveolar lavage (Iwama et al., 1995). In humans, peripheral blood monocytes do not express Ron and do not respond to MSP. In contrast, Ron is detectable by immunostaining in all resident macrophages of normal human dermis (Nanney et al., 1998). It will be of interest to test for Ron expression by macrophages at other interfaces between host and external environment, notably lung and intestinal tract. The cellular responses to MSP are remarkably diverse. In the case of murine resident peritoneal macrophages they can be grouped into two broad categories: motility and mediator production. Effects on motility have been documented in various assays. Before MSP was puriﬁed, activity in serum was detected as induction of macrophage responsiveness to the complementderived chemoattractant, C5a (Leonard and Skeel, 1976). When added to the bottom wells of a multiwell chemotaxis chamber, puriﬁed MSP is an attractant for murine macrophages (Skeel and Leonard, 1994). Unlike chemoattractants that can recruit circulating leukocytes to extravascular sites, MSP is incapable of this action because of the absence of Ron or Stk receptors on circulating monocytes. Addition of MSP to macrophages in tissue culture wells induces a shape change characterized by cytoplasmic projections that is fully developed within an hour (Leonard and Skeel, 1976). This reﬂects induction of a complex motility pattern that can be observed by time-lapse photography (K. Fuller and E. J. Leonard, unpublished data). Signs of membrane stimulation are evident within 15 min, as spiky projections (ﬁlipodia) extend and withdraw from the cell surface. This is followed shortly by broader projections (lamellipodia), which appear to envelop ﬂuid droplets (macropinocytosis) that enter the cell and migrate to the interior.

158

Edward J. Leonard and Alla Danilkovitch

Membrane activity is accompanied by cellular motility, characterized by extension and withdrawal of cell projections, unaccompanied by signiﬁcant translational movement. Although this cellular motility program is initiated by MSP, it persists for at least 2 hr without a requirement for continuing MSP stimulation. This conclusion is based on the fact that MSP action is dependent on PI3K, and can be prevented by prior administration of the PI3K inhibitor wortmannin. However, once MSP induces the cellular motility pattern, addition of wortmannin has no effect. The functional importance of this MSP-induced macrophage membrane motility is unclear. Even unstimulated resident macrophages have a remarkably high plasma membrane turnover; they internalize the equivalent of all of their cell surface every 33 min (Steinman et al., 1976). This is associated with micropinocytosis, which can be quantiﬁed by uptake of a nonbinding solute such as Lucifer Yellow (Knight et al., 1992). Our preliminary data show that Lucifer Yellow uptake by macrophages is not increased by MSP (Casas-Finet and E. J. Leonard, unpublished data). Is it possible that MSP enhances phagocytosis of certain types of nonopsonized particles, which might enhance wound debridement? In the case of opsonized particles, the effect of MSP on uptake via the CR3 receptor of erythrocytes coated with C3bi is unequivocal: In the absence of MSP, these cells adhere to the macrophages, but are not ingested. Addition of MSP causes ingestion within minutes (Skeel et al., 1991). The receptor speciﬁcity of these phagocytic events is striking. In contrast to the C3bi opsonin, erythrocytes coated with IgG are ingested via the macrophage Fc receptor without a requirement for MSP. The other broad category of MSP actions on macrophages relates to mediator production. Endotoxin, or combinations of proinﬂammatory cytokines, causes expression of murine macrophage-inducible nitric oxide synthase, an effect that can be detected by Northern blots for the mRNA or by measurement of nitrate in the culture ﬂuid. MSP prevents induction of NOsynthase by any of the above stimuli (Wang et al., 1994d). The inhibitory action of MSP is conﬁned to this speciﬁc mediator. MSP did not inhibit endotoxin-induced expression of mRNA for monocyte chemoattractant protein-1. Furthermore, MSP caused secretion of IL-6 (but not IL-1 or TNF) within 6 hr, and did not inhibit endotoxin-induced secretion of IL-1, IL-6, or TNF (A. Skeel and E. J. Leonard, unpublished data). The in vitro modulation by MSP of endotoxin-induced NO production now has an in vivo counterpart. Concentrations of nitrate in serum of Stk / mice that received endotoxin intravenously were higher than in serum of comparably treated normal mice; and at a critical endotoxin dose, only 20% of the Stk / mice survived, compared to 80% survival for normal mice (Correll et al., 1997). If MSP plays a role in the host response to endotoxemia, pro-MSP must be cleaved to biologically active MSP. Within 4 hr after i.v. administration of

Macrophage Stimulating Protein

159

endotoxin to normal human volunteers, mature MSP was detected in EDTA plasma (E. J. Leonard and A. F. Suffredini, unpublished data). In view of a proposed role of NO in inﬂammation (Grigoriadis et al., 1994), it was also of interest that DTH responses to oxazalone, measured as ear swelling, were more intense in Stk / mice than in normal mice (Correll et al., 1997).

B. Osteoclasts It has been suggested that macrophages and osteoclasts are derived from a common hematopoietic precursor (Grigoriadis et al., 1994). Of great interest, therefore, is the transient expression of Stk mRNA in foci of bone formation in mouse embryos (Gaudino et al., 1995; Quantin et al., 1996). Localization of Stk in cells that also contained tartrate-resistant acid phosphatase (TRAP) suggested that expression was in differentiated osteoclasts (Quantin et al., 1996). Stk protein was detected by immunohistochemical staining in multinucleated osteoclasts isolated from femurs of 14-day-old mice (Kurihara et al., 1996). Together, these ﬁndings suggest that MSP could play a role in remodeling of developing bone. However reported functional effects of MSP on osteoclasts—stimulation of membrane rufﬂing and pit formation—have been limited to osteoclast-like cells derived from cultured bone marrow (Kurihara et al., 1996). In contrast to effects of HGF on osteoclasts isolated from rat (Fuller et al., 1995) or mouse long bones, MSP did not stimulate formation of pits in bone slices, and there was no effect on motility as determined by time-lapse video (K. Fuller and E. J. Leonard, unpublished data). Likewise, MSP did not increase release of 45Ca from fetal rat long bones in tissue culture (L. Raisz and E. J. Leonard, unpublished data).

C. Cells of Ectodermal Origin Ron, the MSP receptor, was cloned from the cDNA of a human keratinocyte library (Ronsin et al., 1993). This led to studies showing that MSP binds with high afﬁnity to human (Wang et al., 1996b) and murine epithelial cell lines (Wang et al., 1996a). MSP stimulated thymidine incorporation into A549 lung carcinoma cells, T47D mammary carcinoma cells, PC12 pheochromocytoma cells (Gaudino et al., 1994, 1995), and CMT-93 rectal carcinoma cells (Waltz et al., 1997). When tested on BK-1, a murine normal keratinocyte cell line, MSP caused an increase in cell number over a 10-day culture period, with an efﬁcacy comparable to that of EGF (Wang et al., 1996a). Induction by MSP of migration in chemotaxis chambers was observed for A549 cells (Gaudino et al., 1994), and for human and murine neoplastic cell lines as well as HK-NOC, a cell line established from normal hu-

160

Edward J. Leonard and Alla Danilkovitch

man foreskin (Wang et al., 1996a,b). Stimulation of motility is dependent on PI3K (Wang et al., 1996b), as elaborated in Section VI. In response to human wounds, Ron is upregulated in keratinocytes and in accessory epithelial structures including sweat ducts and hair follicles. This ﬁnding, along with the presence of active MSP in wound ﬂuid, suggests a possible role for MSP in wound healing (Nanney et al., 1998). In immunolocalization studies of ciliated epithelium, Ron was detected in nasal and bronchial epithelium and in normal oviduct (Sakamoto et al., 1997). Primary cultures of normal human bronchial epithelium expressed Ron as assessed by ﬂow cytometry and bound MSP with a Kd of 0.5 nM. MSP caused tyrosine phosphorylation of the receptor. Ron and Met were both expressed in ciliated bronchial epithelium, the latter in a basolateral location, the former at the apical surface just below the base of the cilia. This led to the ﬁnding that MSP caused a transient increase of about 20% in the ciliary beat frequency of nasal mucosal cells. In view of Ron mRNA expression in sperm and MSP mRNA in epithelium of epididymis (Ohshiro et al., 1996), the authors suggest a possible role for MSP in sperm motility as well as in mucociliary transport.

D. Vascular Endothelial Cells Ron protein was detected by immunolocalization on vascular endothelial cells in human dermis. The frequency of detection was higher in burn wounds than in normal skin (Nanney et al., 1998). HGF has been reported to be angiogenic. MSP is also angiogenic by the mouse cornea assay (Y. Cao and E. J. Leonard, unpublished data).

E. Bone Marrow Cells MSP was evaluated for effects on erythroid and myeloid progenitor cells of human bone marrow, and no colony stimulating activity was detected. However when marrow progenitors were maximally stimulated by colony stimulating factor plus either steel factor or Flt3 ligand, MSP inhibited by about 50% the formation of granulocyte-macrophage colonies (Broxmeyer et al., 1996). MSP stimulated maturation of human megakaryocyte cell lines as well as primary bone marrow megakaryocytes. The criterion for maturation was an increase in ploidy, which was quantiﬁed as DNA content per cell, determined by ﬂow cytometry. MSP stimulated increased secretion of IL-6 by these cells. Since the effect on ploidy was abolished by antibodies to IL-6, it appears that MSP acts via IL-6 in this system (Banu et al., 1996). MSP also stimulates se-

Macrophage Stimulating Protein

161

cretion of IL-6 by murine resident peritoneal macrophages, in amounts comparable to that induced by endotoxin (A. Skeel and E. J. Leonard, unpublished data).

F. Insights from Message Expression and from Knockout Mice Two papers on Stk mRNA in mouse embryos show that it is expressed transiently in speciﬁc locations in the developing central nervous system (Quantin et al., 1996; Gaudino et al., 1995). Receptor message is expressed in mucosal cells of stomach, small intestine, and colon in both embryonic and adult mice (Quantin et al., 1996). The possibility of paracrine effects of MSP should be considered in light of low MSP message expression in organs other than liver, the locus of constitutive secretion of pro-MSP into the circulation. This includes kidney and pancreas (Yoshimura et al., 1993), epithelium of rat epididymis (Ohshiro et al., 1996), rat lung, adrenal and placenta (Degen et al., 1991) and central nervous system of the chick embryo (Thery et al., 1995). MSP mRNA was also detected by RT–PCR in some, but not all, samples of nonneoplastic human lung tissue adjacent to surgically removed lung tumor as well as selected lung carcinoma cell lines (Willett et al., 1998). In contrast to developmental defects in mice with targeted mutations in Met (Bladt et al., 1995) or its ligand (Uehara et al., 1995; Schmidt et al., 1995), Stk / mice developed normally (Correll et al., 1997). Abnormalities reported to date relate to endotoxin challenge and DTH reactions as noted in section VIIIA. MSP / mice also grow to adulthood without obvious abnormalities except for lipid-containing vacuoles in hepatocytes. Notable is the fact that these animals healed an incisional wound as rapidly as normal mice (Bezerra et al., 1998). This should be considered in relation to the generation of active MSP and upregulation of Ron in human burn wounds (Nanney et al., 1998). The possibilities are that the human data are epiphenomena unrelated to wound healing, that there is redundancy in the systems, that an unidentiﬁed additional ligand can activate Ron, or that a more challenging wound model is required to reveal a defect in the MSP / mice.

IX. PERSPECTIVE From the structural and in vitro biological data summarized in this review, we can conclude that MSP is a growth and motility factor that activates a typical cell membrane protein tyrosine kinase receptor. However, despite an

162

Edward J. Leonard and Alla Danilkovitch

increasing body of knowledge about MSP and Ron in vitro, we do not know the role of MSP/Ron in vivo. “In vivo veritas” was posted on the ofﬁce door of the ﬁrst chief of our laboratory, an adaptation for medical research of the ancient insight (“in vino veritas”) into the effects of the Roman equivalent of the two-martini lunch. We hope to elicit this in vivo truth incrementally, by using the in vitro data to design animal experiments and to make clinical correlations in pathological conditions. For example, the inhibition by MSP of endotoxin-induced macrophage synthesis of NO synthase (Wang et al., 1994d) provided a rational basis for studying the response of Stk / mice to endotoxin challenge (Correll et al., 1997). And the study of MSP and Ron in human wounds (Nanney et al., 1998) derived from a postulated role in tissue injury, which was based on the evolution of MSP from coagulation proteins (Donate et al., 1994) and on the presence of Ron on keratinocyte cell lines (Wang et al., 1996a,b). We hope eventually to know if MSP has a physiologic role in normal turnover of cells in skin and intestinal mucosa, and whether any of the actions described in Section VIII on target cells have in vivo relevance. Finally, the recent demonstration that cells transfected with an appropriately mutated Ron are tumorigenic (Santoro et al., 1998) will doubtless stimulate a search for similar mutations in human cancer. Note added in proof: A vital role for the Ron/Stk receptor in development was recently established by Muraoka et al. (1999) who reported that Stk / murine embryos remained viable through the blastocyst stage, but failed to survive thereafter. Hemizygous mice had abnormally high nitric oxide production in response to endotoxin, conﬁrming the results of Correll et al., (1997).

REFERENCES Abedi, H., and Zachary, I. (1997). J. Biol. Chem. 272, 15442–15451. Akimoto, K., Takahashi, R., Moriya, S., Nishioka, N., Takayanagi, J., Kimura, K., Fukui, Y., Osada, S., Mizuno, K., Hirai, S., Kazlauskas, A., and Ohno, S. (1996). EMBO J. 15, 788– 798. Aronheim, A., Engelberg, D., Li, N., al-Alawi, N., Schlessinger, J., and Karin, M. (1994). Cell (Cambridge, Mass.) 78, 949 –961. Banu, N., Price, D. J., London, R., Deng, B., Mark, M., Godowski, P. J., and Avraham, H. (1996). J. Immunol. 156, 2933–2940. Baron, V., Calleja, V., Ferrari, P., Alengrin, F., and Van Obberghen, E. (1998). J. Biol. Chem. 273, 7162–7168. Bezerra, J. A., Carrick, T. L., Degen, J. L., Witte, D., and Degen, S. F. (1998). J. Clin. Invest. 101, 1175–1183.

Macrophage Stimulating Protein

163

Bladt, F., Riethmacher, D., Isenmann, S., Aguzzi, A., and Birchmeier, C. (1995). Nature (London) 376, 768 –771. Borset, M., Lien, E., Espevik, T., Helseth, E., Waage, A., and Sundan, A. (1996). J. Biol. Chem. 271, 24655–24661. Bos, J. L. (1989). Cancer Res. 49, 4682–4689. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M., Kmiecik, T. E., Vande Woude, G. F., and Aaronson, S. A. (1991). Science 251, 802–804. Broxmeyer, H. E., Cooper, S., Li, Z.-H., Lu, L., Sarris, A., Wang, M.-H., Chang, M.-S., Donner, D. B., and Leonard, E. J. (1996). Ann. Hematol. 73, 1– 9. Cantley, L. C., Auger, K. R., Carpenter, C., Duckworth, B., Graziani, A., Kapeller, R., and Soltoff, S. (1991). Cell (Cambridge, Mass.) 64, 281–302. Cartwright, C. A., Meisler, A. I., and Eckhart, W. (1990). Proc. Natl. Acad. Sci. U.S.A. 87, 558– 562. Cartwright, C. A., Coad, C. A., and Egbert, B. M. (1994). J. Clin. Invest. 93, 509–515. Chen, Q., Seol, D. W., Carr, B., and Zarnegar, R. (1997). Hepatology 26, 59 –66. Chung, J., Grammer, T. C., Lemon, K. P., Kazlauskas, A., and Blenis, J. (1994). Nature (London) 370, 71–75. Clark, S. G., Stern, M. J., and Horvitz, H. R. (1992). Nature (London) 356, 340–344. Collesi, C., Santoro, M. M., Gaudino, G., and Comoglio, P. M. (1996). Mol. Cell.. Biol. 16, 5518–5526. Cooper, C. S., Park, M., Blair, D. G., Tainsky, M. A., Huebner, K., Croce, C. M., and Vande, W. G. (1984). Nature (London) 311, 29 –33. Cooper, J. A., and Howell, B. (1993). Cell (Cambridge, mass.) 73, 1051–1054. Correll, P. H., Iwama, A., Tondat, S., Mayrhofer, G., Suda, T., and Bernstein, A. (1997). Genes Funct. 1, 1–15. Courtneidge, S. A., Dhand, R., Pilat, D., Twamley, G. M., Waterﬁeld, M. D., and Roussel, M. F. (1993). EMBO J. 12, 943 –950. Cunningham, B. C., Ultsch, M., de Vos, A. M., Mulkerrin, M. G., Clauser, K. R., and Wells, J. A. (1991). Science 254, 821–825. Danilkovitch, A., and Leonard, E. J. (1997). J. Leukocyte Biol., Suppl., p. 19 (abstr.). Danilkovitch, A., and Leonard, E. J. (1998). J. Leukocyte Biol., Suppl., p. 32 (abstr.). Degen, S. J., Stuart, L. A., Han, S., and Jamison, C. S. (1991). Biochemistry 30, 9781– 9791. de Vos, A. M., Ultsch, M., and Kossiakoff, A. A. (1992). Science 255, 306–312. Donate, L. E., Gherardi, E., Srinivasan, N., Sowdhamini, R., Aparicio, S., and Blundell, T. L. (1994). Protein Sci. 3, 2378–2394. Dudek, H., Datta, S. R., Franke, T. F., Birnbaum, M. J., Yao, R., Cooper, G. M., Segal, R. A., Kaplan, D. R., and Greenberg, M. E. (1997). Science 275, 661–665. Empereur, S., Djelloul, S., Di Gioia, Y., Bruyneel, E., Mareel, M., Van Hengel, J., Van Roy, F., Comoglio, P., Courtneidge, S., Paraskeva, C., Chastre, E., and Gespach, C. (1997). Br. J. Cancer 75, 241–250. Erpel, T., and Courtneidge, S. A. (1995). Curr. Opin. Cell biol. 7, 176–182. Francis, C. W., and Marder, V. J. (1990). In “Hematology” (W. J. Williams, E. Beutler, A. J. Erslev, and M. A. Lichtman, eds.), p. 1313. McGraw-Hill, New York. Franke, T. F., Kaplan, D. R., and Cantley, L. C. (1997). Cell (Cambridge, Mass.) 88, 435 –437. Fuller, K., Owens, J., and Chambers, T. J. (1995). Biochem. Biophys. Res. Commun. 212, 334– 340. Gaudino, G., Follenzi, A., Naldini, L., Collesi, C., Santoro, M., Gallo, K. A., Godowski, P. J., and Comoglio, P. M. (1994). EMBO J. 13, 3524–3532. Gaudino, G., Avantaggiato, V., Follenzi, A., Acampora, D., Simeone, A., and Comoglio, P. M. (1995). Oncogene 11, 2627–2637.

164

Edward J. Leonard and Alla Danilkovitch

Grigoriadis, A. E., Wang, Z. Q., Cecchini, M. G., Hofstetter, W., Felix, R., Fleisch, H. A., and Wagner, E. F. (1994). Science 266, 443–448. Han, S., Stuart, L. A., and Degen, S. J. F. (1991). Biochemistry 30, 9768–9780. Hanks, S. K., and Polte, T. R. (1997). BioEssays 19, 137. Hanks, S. K., Calalb, M. B., Harper, M. C., and Patel, S. K. (1992). Proc. Natl. Acad. Sci. U.S.A. 89, 8487–8491. Heimbrook, D. C., Oliff, A., and Gibbs, J. B. (1997). In “Cancer Principles and Practice of Oncology” (V. DeVita, S. Hellman, and S. Rosenberg, eds.), pp. 35 –43. Lippincott-Raven press, Philadelphia. Heldin, C. H. (1995). Cell (Cambridge, Mass.) 80, 213 –223. Hu, P., Margolis, B., Skolnik, E. Y., Lammers, R., Ullrich, A., and Schlessinger, J. (1992). Mol. Cell. Biol. 12, 981–990. Huff, J. L., Jelinek, M. A., Borgman, C. A., Lansing, T. J., and Parsons, J. T. (1993). Proc. Natl. Acad. Sci. U.S.A. 90, 6140–6144. Hwang, M. C., Sung, Y. J., and Hwang, Y. W. (1996). J. Biol. Chem. 271, 8196–8202. Ip, Y. T., and Davis, R. J. (1998). Curr. Opin. Cell Biol. 10, 205–219. Iwama, A., Okano, K., Sudo, T., Matsuda, Y., and Suda, T. (1994). Blood 83, 3160–3169. Iwama, A., Wang, M.-H., Yamaguchi, N., Okano, K., Sudo, T., Gervais, F., Morissette, C., Leonard, E. J., and Suda, T. (1995). Blood 86, 3394–3403. Iwama, A., Yamaguchi, N., and Suda, T. (1996). EMBO J. 15, 5866–5875. Jeffers, M., Rong, S., and Woude, G. F. (1996). J. Mol. Med. 74, 505 –513. Joneson, T., McDonough, M., Bar-Sagi, D., and Van Aelst, L. (1996). Science 274, 1374–1376. Kapeller, R., and Cantley, L. C. (1994). BioEssays 16, 565. Karunagaran, D., Tzahar, E., Beerli, R. R., Chen, X., Graus-Porta, D., Ratzkin, B. J., Seger, R., Hynes, N. E., and Yarden, Y. (1996). EMBO J. 15, 254 –264. Kauffmann-Zeh, A., Rodriguez-Viciana, P., Ulrich, E., Gilbert, C., Coffer, P., Downward, J., and Evan, G. (1997). Nature (London) 385, 544–548. Knight, K. R., Vairo, G., and Hamilton, J. A. (1992). J. Leukocyte Biol. 51, 350–359. Kurihara, N., Iwama, A., Tatsumi, J., Ikeda, K., and Suda, T. (1996). Blood 87, 3704–3710. Kypta, R. M., Goldberg, Y., Ulug, E. T., and Courtneidge, S. A. (1990). Cell 62, 481–492. Lamarche, N., Tapon, N., Stowers, L., Burbelo, P. D., Aspenstrom, P., Bridges, T., Chant, J., and Hall, A. (1996). Cell (Cambridge, Mass.) 87, 519–529. Lemmon, M. A., and Schlessinger, J. (1994). Trends Biochem. Sci. 19, 459–463. Lemmon, M. A., Pinchasi, D., Zhou, M., Lax, I., and Schlessinger, J. (1997). J. Biol. Chem. 272, 6311–6317. Leonard, E. J., and Skeel, A. (1976). Exp. Cell Res. 102, 434–438. Leonard, E. J., and Skeel, A. (1980). J. Reticuloendothel. Soc. 28, 437–447. Leonard, E. J., and Skeel, A. (1996). J. Leukocyte Biol. 60, 453–458. Leonard, E. J., Skeel, A., and Allenmark, S. (1982). Arch. Biochem. Biophys. 214, 12–16. Li, B. Q., Wang, M. H., Kung, H. F., Ronsin, C., Breathnach, R., Leonard, E. J., and Kamata, T. (1995). Biochem. Biophys. Res. Commun. 216, 110–118. Logan, S. K., Falasca, M., Hu, P., and Schlessinger, J. (1997). Mol. Cell. Biol. 17, 5784–5790. Lopez-Ilasaca, M., Li, W., Uren, A., Yu, J. C., Kazlauskas, A., Gutkind, J. S., and Heidaran, M. A. (1997). Biochem. Biophys. Res. Commun. 232, 273–277. Lowy, D. R., and Willumsen, B. M. (1993). Annu. Rev. Biochem. 62, 851–891. Maa, M. C., Leu, T. H., McCarley, D. J., Schatzman, R. C., and Parsons, S. J. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 6981–6985. Maggiora, P., Marchio, S., Stella, M., Giai, M., Belﬁore, A., De Bortoli, M., Di Renzo, M., Costantino, A., Sismond, P., and Comoglio, P. M. (1998). Oncogene 16, 2927–2933. Marshall, M. S. (1995). FASEB J. 9, 1311–1318. Martin, P. (1997). Science 276, 75–81.

Macrophage Stimulating Protein

165

Matsumoto, K., Takehara, T., Inoue, H., Hagiya, M., Shimizu, S., and Nakamura, T. (1991). Biochem. Biophys. Res. Commun. 181, 691–699. Matsumoto, K., Nakamura, T., and Kramer, R. H. (1994). J. Biol. Chem. 269, 31807–31813. Miller, M., and Leonard, E. J. (1998). FEBS Lett. 429, 1– 3. Moriya, S., Kazlauskas, A., Akimoto, K., Hirai, S., Mizuno, K., Takenawa, T., Fukui, Y., Watanabe, Y., Ozaki, S., and Ohno, S. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 151–155. Moscatello, D. K., Montgomery, R. B., Sundareshan, P., McDanel, H., Wong, M. Y., and Wong, A. J. (1996). Oncogene 13, 85–96. Muraoka, R. S., Sun, W. Y., Colbert, M. C., Waltz, S. E., Witte, D. P., Degen, J. L., and Degen, S. J. (1999). J. Clin. Invest. 103, 1277–1285. Nanney, L. B., Skeel, A., Luan, J., Polis, S., Richmond, A., Wang, M.-H., and Leonard, E. J. (1998). J. Invest. Dermatol. 111, 573–581. Ohshiro, K., Iwama, A., Matsuno, K., Ezaki, T., Sakamoto, O., Hamaguchi, I., Takasu, N., and Suda, T. (1996). Biochem. Biophys. Res. Commun. 227, 273–280. Okigaki, M., Komada, M., Uehara, Y., Miyazawa, K., and Kitamura, N. (1992). Biochemistry 31, 9555–9561. Oldham, S. M., Cox, A. D., Reynolds, E. R., Sizemore, N. S., Coffey, R. J. J., and Der, C. J. (1998). Oncogene 16, 2565–2573. Owens, L. V., Xu, L., Craven, R. J., Dent, G. A., Weiner, T. M., Kornberg, L., Liu, E. T., and Cance, W. G. (1995). Cancer Res. 55, 2752–2755. Park, M., Dean, M., Cooper, C. S., Schmidt, M., O’Brien, S. J., Blair, D. G., and Vande, W. G. (1986). Cell (Cambridge, Mass.) 45, 895–904. Park, M., Dean, M., Kaul, K., Braun, M. J., Gonda, M. A., and Vande Woude, G. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 6379–6383. Parker, P. J. (1995). Curr. Biol. 5, 577–579. Partin, A. W., Schoeniger, J. S., Mohler, J. L., and Coffey, D. S. (1989). Proc. Natl. Acad. Sci. U.S.A. 86, 1254–1258. Philo, J. S., Wen, J., Wypych, J., Schwartz, M. G., Mendiaz, E. A., and Langley, K. E. (1996). J. Biol. Chem. 271, 6895–6902. Ponzetto, C., Bardelli, A., Zhen, Z., Maina, F., dalla Zonca, P., Giordano, S., Graziani, A., Panayotou, G., and Comoglio, P. M. (1994). Cell (Cambridge, Mass.) 77, 261–271. Qiu, R. G., Chen, J., Kirn, D., McCormick, F., and Symons, M. (1995a). Nature (London) 374, 457–459. Qiu, R. G., Chen, J., McCormick, F., and Symons, M. (1995b). Proc. Natl. Acad. Sci. U.S.A. 92, 11781–11785. Quantin, B., Schuhbaur, B., Gesnel, M.-C., Dollé, P., and Breathnach, R. (1996). Dev. Dyn. 204, 383–390. Rahimi, N., Tremblay, E., and Elliott, B. (1996). J. Biol. Chem. 271, 24850–24855. Rankin, S., and Rosengurt, E. (1994). J. Biol. Chem. 269, 704–710. Reif, K., Nobes, C. D., Thomas, G., Hall, A., and Cantrell, D. A. (1996). Curr. Biol. 6, 1445– 1455. Rodrigues, G. A., and Park, M. (1993). Mol. Cell. Biol. 13, 6711–6722. Rodrigues, G. A., Park, M., and Schlessinger, J. (1997). EMBO J. 16, 2634–2645. Rodriguez-Viciana, P., Warne, P. H., Dhand, R., Vanhaesebroeck, B., Gout, I., Fry, M. J., Waterﬁeld, M. D., and Downward, J. (1994). Nature (London) 370, 527–532. Rodriguez-Viciana, P., Warne, P. H., Khwaja, A., Marte, B. M., Pappin, D., Das, P., Waterﬁeld, M. D., Ridley, A. and Downward, J. (1997). Cell (Cambridge, Mass.) 89, 457–467. Ronsin, C., Muscatelli, F., Mattei, M. G., and Breathnach, R. (1993). Oncogene 8, 1195– 1202. Sakamoto, O., Iwama, A., Amitani, R., Takehara, T., Yamaguchi, N., Yamamoto, T., Masuyama, K., Yamanaka, T., Ando, M., and Suda, T. (1997). J. Clin. Invest. 99, 701–709.

166

Edward J. Leonard and Alla Danilkovitch

Sakata, H., Stahl, S. J., Taylor, W. G., Rosenberg, J. M., Sakaguchi, K., Wingﬁeld, P. T., and Rubin, J. S. (1997). J. Biol. Chem. 272, 9457–9463. Santoro, M., Carlomagno, F., Romano, A., Bottaro, D. P., Dathan, N. A., Grieco, M., Fusco, A., Vecchio, G., Matoskova, B., and Kraus, M. H. (1995). Science 267, 381–383. Santoro, M., Penengo, L., Minetto, M., Orecchia, S., Cilli, M., and Gaudino, G. (1998). Oncogene 17, 741–749. Santoro, M. M., Collesi, C., Grisendi, S., Gaudino, G., and Comoglio, P. M. (1996). Mol. Cell. Biol. 16, 7072–7083. Schaller, M. D., Borgman, C. A., Cobb, B. S., Vines, R. R., Reynolds, A. B., and Parsons, J. T. (1992). Proc. Natl. Acad. Sci. U.S.A. 89, 5192–5196. Schmidt, C., Bladt, F., Goedecke, S., Brinkmann, V., Zschiesche, W., Sharpe, M., Gherardi, E., and Birchmeier, C. (1995). Nature (London) 373, 699–702. Schmidt, L., Duh, F. M., Chen, F., Kishida, T., Glenn, G., Choyke, P., Scherer, S. W., Zhuang, Z., Lubensky, I., Dean, M., Allikmets, R., Chidambaram, A., Bergerheim, U. R., Feltis, J. T., Casadevall, C., Zamarron, A., Bernues, M., Richard, S., Lips, C. J., Walther, M. M., Tsui, L. C., Geil, L., Orcutt, M. L., Stackhouse, T., and Zbar, B. (1997). Nat. Genet. 16, 68–73. Skeel, A., and Leonard, E. J. (1994). J. Immunol. 152, 4618–4623. Skeel, A., Yoshimura, T., Showalter, S., Tanaka, S., Appella, E., and Leonard, E. (1991). J. Exp. Med. 173, 1227–1234. Songyang, Z., Shoelson, S. E., Chaudhuri, M., Gish, G., Pawson, T., Haser, W. G., King, F., Roberts, T., Ratnofsky, S., and Lechleider, R. J. (1993). Cell (Cambridge, Mass.) 72, 767– 778. Songyang, Z., Shoelson, S. E., McGlade, J., Olivier, P., Pawson, T., Bustelo, X. R., Barbacid, M., Sabe, H., Hanafusa, H., and Yi, T. (1994). Mol. Cell. Biol. 14, 2777–2785. Songyang, Z., Baltimore, D., Cantley, L. C., Kaplan, D. R., and Franke, T. F. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 11345–11350. Steinman, R. M., Brodie, S. E., and Cohn, Z. A. (1976). J. Cell. Biol. 68, 665–687. Talamonti, M. S., Roh, M. S., Curley, S. A., and Gallick, G. E. (1993). J. Clin. Invest. 91, 53 – 60. Thery, C., Sharpe, M. J., Batley, S. J., Stern, C. D., and Gherardi, E. (1995). Dev. Genet. 17, 90–101. Tsujimura, T. (1996). Pathol. Int. 46, 933–938. Ueda, A., Takeshita, F., Yamashiro, S., and Yoshimura, T. (1998). J Biol. Chem. 273, 19339–19347. Uehara, Y., Minowa, O., Mori, C., Shiota, K., Kuno, J., Noda, T., and Kitamura, N. (1995). Nature (London) 373, 702–705. Ullrich, A., and Schlessinger, J. (1990). Cell (Cambridge, Mass.) 61, 203–212. Varticovski, L., Harrison-Findik, D., Keeler, M. L., and Susa, M. (1994). Biochim. Biophys. Acta 1226, 1–11. Wahl, R. C., Costigan, V. J., Batac, J. P., Chen, K., Cam, L., Courchesne, P. L., Patterson, S. D., Zhang, K., and Paciﬁci, R. E. (1997). J. Biol. Chem. 272, 15053–15056. Waltz, S. E., McDowell, S. A., Muraoka, R. S., Air, E. L., Flick, L. M., Chen, Y. Q., Wang, M. H., and Degen, S. J. (1997). J. Biol. Chem. 272, 30526–30537. Wang, M.-H., Skeel, A., Yoshimura, T., Copeland, T. D., Sakaguchi, K., and Leonard, E. J. (1993). J. Leukocyte Biol. 54, 289 –295. Wang, M.-H., Ronsin, C., Gesnel, M.-C., Coupey, L., Skeel, A., Leonard, E. J., and Breathnach, R. (1994a). Science 266, 117–119. Wang, M.-H., Yoshimura, T., Skeel, A., and Leonard, E. J. (1994b). J. Biol. Chem. 269, 3436– 3440. Wang, M.-H., Gonias, S. L., Skeel, A., Wolf, B. B., Yoshimura, T., and Leonard, E. J. (1994c). J. Biol. Chem. 269, 13806–13810.

Macrophage Stimulating Protein

167

Wang, M.-H., Cox, G. W., Yoshimura, T., Shefﬂer, L. A., Skeel, A., and Leonard, E. J. (1994). J. Biol. Chem. 269, 14027–14031. Wang, M.-H., Iwama, A., Skeel, A., Suda, T., and Leonard, E. J. (1995). Proc. Natl. Acad. Sci. U.S.A. 92, 3933–3937. Wang, M.-H., Dlugosz, A. A., Sun, Y., Skeel, A., Yuspa, S. H., Suda, T., and Leonard, E. J. (1996a). Exp. Cell Res. 226, 39–46. Wang, M.-H., Montero-Julian, F. A., Dauny, I., and Leonard, E. J. (1996b). Oncogene 13, 2167–2175. Wang, M.-H., Skeel, A., and Leonard, E. J. (1996c). J. Clin. Invest. 97, 720–727. Wang, M.-H., Julian, F. M., Breathnach, R., Godowski, P. J., Takehara, T., Yoshikawa, W., Hagiya, M., and Leonard, E. J. (1997). J. Biol. Chem. 272, 16999–17004. Weiner, T. M., Liu, E. T., Craven, R. J., and Cance, W. G. (1993). Lancet 342, 1024–1025. Whitmarsh, A. J., and Davis, R. J. (1996). J. Mol. Med. 74, 589–607. Willett, C. G., Smith, D. I., Shridhar, V., Wang, M. H., Emanuel, R. L., Patidar, K., Graham, S. A., Zhang, F., Hatch, V., Sugarbaker, D. J., and Sunday, M. E. (1997). J. Clin. Invest. 99, 2979–2991. Willett, C. G., Wang, M. H., Emanuel, R. L., Graham, S. A., Smith, D. I., Shridhar, V., Sugarbaker, D. J., and Sunday, M. E. (1998). Am. J. Respir. Cell Mol. Biol. 18, 489–496. Williams, J. C., Wierenga, R. K., and Saraste, M. (1998). Trends Biochem. Sci. 23, 179 –184. Yoshikawa, W., Hara, H., Takehara, T., Shimonishi, M., Sakai, H., Shimizu, N., Shimizu, S., Wang, M.-H., Hagiya, M., Skeel, A., and Leonard, E. J. (1999). Arch. Biochem. Biophys. 363, 356 – 360. Yoshimura, T., Yuhki, N., Wang, M.-H., Skeel, A., and Leonard, E. J. (1993). J. Biol. Chem. 268, 15461–15468. Zhan, X., Plourde, C., Hu, X., Friesel, R., and Maciag, T. (1994). J. Biol. Chem. 269, 20221– 20224. Zhu, T., Goh, E. L., and Lobie, P. E. (1998). J. Biol. Chem. 273, 10682–10689.

This Page Intentionally Left Blank

CD44 Glycoproteins in Colorectal Cancer: Expression, Function, and Prognostic Value Vera J. M. Wielenga, Ronald van der Neut, G. Johan A. Offerhaus, and Steven T. Pals Department of Pathology, Academic Medical Center University of Amsterdam, 1105 AZ Amsterdam, The Netherlands

I. Introduction II. Structure and Function of CD44 A. The Structure of CD44 B. CD44 Glycoproteins: Receptors for ECM Components C. CD44 Isoforms Decorated with Heparan Sulfate Bind and Present Growth Factors III. CD44 in Tumor Progression and Metastasis A. Expression and Regulation of CD44 in Colorectal Cancer B. CD44 as a Prognosticator in Colorectal Cancer IV. Conclusions References

I. INTRODUCTION Colorectal cancer is a common disease in the Western world and represents the second leading cause of cancer-related death (Coleman et al., 1993; American Cancer Society, 1994). It evolves through a series of morphologically recognizable stages known as the adenoma carcinoma sequence (Muto et al., 1975). Recent advances in molecular genetics have greatly increased our understanding of the development of colorectal cancer (Kinzler and Vogelstein, 1996; Korinek et al., 1997; Morin et al., 1997; Liu et al., 1996; Yingling et al., 1996). Tumor progression in the colorectum is characterized by a stepwise accumulation of speciﬁc molecular genetic alterations, ultimately resulting in invasive and metastatic cancer. Most of the molecules that have thus far been implicated in this neoplastic process either cause genetic instability or act on regulation of the cell cycle, thereby resulting in a disturbed homeostasis between cell proliferation and apoptosis. The main cause of cancer-related death, however, is not growth of the primary tumor, but the formation of metastases in distant organs. Although relatively little is known about the molecular mechanisms underlying this complicated process, a Advances in CANCER RESEARCH 0065-230X/00 $30.00

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

170

Vera J. M. Wielenga et al.

large body of studies indicate an important role for CD44 (reviewed in Lesley et al., 1993; Naor et al., 1997). CD44 was originally described as a homing receptor on lymphocytes, mediating lymphocyte interactions with high endothelial venules (HEV) (Jalkanen et al., 1986, 1987). Metastasizing tumor cells and recirculating (activated) lymphocytes share several properties including motility and invasive behavior involving reversible adhesive contacts, accumulation and expansion in draining lymph nodes, release into the ciruculation, and adhesion to vascular endothelium and extravasation. This analogy between lymphocyte recirculation and tumor dissemination prompted the hypothesis that malignant cells might use molecules like CD44 for metastasis formation (Herrlich et al., 1993). Support for this hypothesis comes from experimental studies in laboratory animals showing a causal role of speciﬁc CD44 isoforms in metastasis formation, as well as from clinical studies documenting deregulated CD44 expression in human cancer (Lesley et al., 1993; Naor et al., 1997). On the other hand, contradictory reports concerning both the biological role of CD44 in tumorigenesis and its clinical usefulness as a prognosticator have been published. In this paper, we address these contradictions by summarizing and discussing the current literature on the expression, regulation, and prognostic value of CD44 in colorectal cancer. Furthermore, we present a model for CD44 function in colorectal tumorigenesis.

II. STRUCTURE AND FUNCTION OF CD44 A. The Structure of CD44 CD44 is a family of type I transmembrane glycoproteins that are widely expressed on a variety of cells including cells of epithelial, mesenchymal, and hematopoietic origin. All CD44 family members are encoded by a single gene on chromosome 11p13 that consists of 19 exons (Fig. 1) (Stamenkovic et al., 1989; Screaton et al., 1992). They share the N-terminal cartilage link protein homology domain encoded by exons 1–5, which bind hyaluronic acid (HA) (Stamenkovic et al., 1989; Aruffo et al., 1990; Peach et al., 1993). However, as a result of extensive alternative splicing of exons 6–14 (also referred to as exons v2–v10), the extracellular membrane proximal domain of CD44 is highly variable (Günthert et al., 1991; Dougherty et al., 1991; Screaton et al., 1992; Tölg et al., 1993). Additional diversity of CD44 results from post-translational modiﬁcations with N- and O-linked sugars and glycosaminoglycan (GAG) side chains (Stamenkovic et al., 1989; Brown et al., 1991; Faassen et al., 1992; Jalkanen and Jalkanen, 1992; D. G. Jackson et al., 1995). The expression of CD44 isoforms is tissue speciﬁc. For example,

CD44 Glycoproteins in Colorectal Cancer

171

Fig. 1 Schematic representation of the CD44 gene and its encoded proteins. The extracellular domain and cytoplasmic tail of CD44 isoforms vary in size as the result of alternative splicing. The alternatively spliced exons are indicated by open boxes. The human v1 exon contains a stop codon. In the model of the protein, all putative glycosylation sites are indicated: O-glycosylation (open circles); N-glycosylation (closed circles); chondroitin sulfate (open squares); hep-aran sulfate chain (rod). As indicated, the heparan sulfate binding site in exon v3 has the ability to bind growth factors. In addition, the HA-binding sites (double lines); the disulﬁde bonds (S–S); the ankyrin binding site (...); the ezrin binding sites (---); the phosphorylation sites (P); and the putative interaction sites for SRC family kinases, are indicated.

172

Vera J. M. Wielenga et al.

the shortest isoform of CD44 (CD44s), which lacks all variable exons, is the most common form on hematopoietic cells, whereas larger CD44 splice variants dominate on several normal epithelia. These variant isoforms are also expressed on neoplastic epithelia and on activated lymphocytes and malignant lymphomas (Günthert et al., 1991; Dougherty et al., 1991; Screaton et al., 1992; Heider et al., 1993, 1995; Fox et al., 1994; Arch et al., 1992; Koopman et al., 1993; Stauder et al., 1995).

B. CD44 Glycoproteins: Receptors for ECM Components CD44 has been implicated in lymphocyte homing and activation, hematopoiesis, and tumor progression and metastasis (Lesley et al., 1993; Naor et al., 1997). It is believed to function in these processes as a cell adhesion receptor, linking the cell and the cytoskeleton to extracellular matrix (ECM) molecules. The CD44 cytoplasmic tail associates with the actin cytoskeleton via ankyrin and proteins of the ERM family. These cytoskeletal interactions may regulate HA binding and CD44-dependent cell motility as well as inside-out signaling events (Lesley et al., 1993; Kalomiris and Bourguignon, 1988; Tsukita et al., 1994; Legg and Isacke, 1998; Sheikh et al., 1998). Importantly, a number of studies have indicated that CD44 can function as a signal-transducing receptor. Engagement of CD44 costimulates antigen-speciﬁc lymphocyte activation and proliferation and leads to activation of integrins on the lymphocyte cell surface (Haynes et al., 1989; Shimizu et al., 1989; Koopman et al., 1990). CD44 can associate with the Src family member of p56lck and triggering of CD44 results in an increased p56lck activity and in phosphorylation of a number of intracellular proteins including ZAP70 (Taher et al., 1996, 1999). CD44 is a major receptor for hyaluronate, a GAG that is abundant in the ECM of mesenchymal tissues and that plays a regulating role in cell migration during inﬂammation, wound healing, and development (Aruffo et al., 1990; Lesley et al., 1993; Knudson and Knudson, 1993). The binding site for HA is located on the N-terminal part of CD44, which is present in all isoforms. However, HA-binding capacity is not a constant feature of CD44, but is subject to complex regulation by mechanisms involving both alternative splicing, modulation of cytoskeletal interaction, and post-translational modiﬁcation of CD44 (Van der Voort et al., 1995; Bennett et al., 1995b; Katoh et al., 1995; Sheikh et al., 1998). By altering cellular interactions with HA, CD44 might facilitate tumor metastasis at two distinct levels. First, it might promote cell migration through the ECM. Second, it might facilitate rolling of tumor cells on HA expressed on the surface of vascular endothelium, thereby promoting extravasation (DeGrendele et al., 1996, 1997). In-

CD44 Glycoproteins in Colorectal Cancer

173

deed, Bartolazzi et al. (1994) reported that HA binding is essential for the metastasis-promoting effect of CD44s in melanoma cell lines. In pancreatic carcinoma cell lines, however, Sleeman et al. (1996) found no relation between HA-binding capacity and metastatic potential, indicating that in this tumor type other properties of CD44 promote metastasis. Additional molecules that have been reported to interact with CD44 are collagen IV, ﬁbronectin, serglycin, and osteopontin; the biologic signiﬁcance of these interactions needs further study (Carter and Wayner, 1988; Jalkanen and Jalkanen, 1992; Toyama-Sorimachi et al., 1995; Weber et al., 1996). As will be discussed later, the recently discovered novel function of CD44, that is, binding and presentation of growth factors, might shed new light on the mechanism of tumor formation.

C. CD44 Isoforms Decorated with Heparan Sulfate Bind and Present Growth Factors CD44 can be modiﬁed by both chondroitin and heparan sulfate side chains and is thus a “facultative” cell-surface proteoglycan (Brown et al., 1991; Faassen et al., 1992; Jalkanen et al., 1988; D. G. Jackson et al., 1995). Heparan sulfate (HS) modiﬁcations are associated with v3 containing CD44 isoforms (D. G. Jackson et al., 1995), which possess a consensus motif SGSG for HS addition (Bourdon et al., 1987). HS proteoglycans are involved in regulation of cell growth and motility (Kjellén and Lindahl, 1991; Ruoshlati and Yamaguchi, 1991; Schlessinger et al., 1995). These molecules bind growth factors via their HS side chains and target these factors to their high afﬁnity signal transducing receptors. This process has been particularly well explored for ﬁbroblast growth factor 2 (FGF-2). Binding of FGF-2 to the low-afﬁnity proteoglycan receptor on the cell surface allows more frequent encounters with the high-afﬁnity receptor. Furthermore, formation of a multivalent FGF–proteoglycan complex promotes dimerization of the highafﬁnity receptors essential for signaling (Yayon et al., 1991; Ruoslahti and Yamaguchi, 1991; Schlessinger et al., 1995). Importantly, structural modiﬁcations of HS side chains determine their speciﬁcity for a given heparin-binding growth factor, creating a mechanism for cell or tissue selective growth factor binding (David, 1993; Lindahl et al., 1994; Lyon et al., 1994; Tanaka et al., 1998). Tanaka and colleagues (1993) have shown that heparan sulfate proteoglycan forms of CD44 (CD44-HS) can present the chemokine MIP-1 to T lymphocytes, resulting in integrin activation. Subsequently, studies by Bennett and colleagues (1995a) demonstrated binding of FGF-2. During embryogenesis, presentation of this growth factor by CD44-HS expressed on the apical epidermal ridge appears to be crucial for limb bud formation

174

Vera J. M. Wielenga et al.

(Sherman et al., 1998). Recent studies from our own laboratory have indicated that CD44-HS can also bind hepatocyte growth factor/scatter factor (HGF/SF). Presentation of HGF/SF by CD44-HS strongly promotes signaling through its high-afﬁnity receptor, the receptor tyrosine kinase c-Met, leading to enhanced activation of the ERK1/2 MAP kinases as well as to hyperphosphorylation of several other cytoplasmic proteins (Van der Voort et al., 1999; Taher et al., 1999). This observation provides an important novel functional link between CD44 and metastasis formation: HGF/SF is a member of the plasminogen-related growth factor family that induces growth, and motility of target epithelial cells, endothelial cells, myoblasts, and lymphocytes (Stoker et al., 1987; Bottaro et al., 1991; Bussolino et al., 1992; Tajima et al., 1992; Weidner et al., 1993; Donate et al., 1994; Bladt et al., 1995; Boros and Miller, 1995; Brinkmann et al., 1995; Van der Voort et al., 1997). Apart from these physiologic functions, the c-Met–HGF/SF pathway promotes invasion and metastasis. Stimulation of c-Met by HGF/ SF induces epithelial cells to invade collagen matrices in vitro, whereas cotransfection of c-met and HGF/SF induces an invasive and metastatic phenotype in NIH 3T3 cells (Weidner et al., 1990; Rong et al., 1992, 1994; Giordano et al., 1993). In HGF/SF transgenic mice, tumorigenesis was observed in many different tissues including mammary glands, skeletal muscles, and melanocytes (Takayama et al., 1997). Furthermore, in human renal papillary carcinomas activating c-met mutations have recently been identiﬁed (Schmidt et al., 1997). NIH 3T3 expressing these mutant c-Met molecules are transformed in vitro and tumorigenic in vivo (Jeffers et al., 1998). We hypothesize that tumor cells, by overexpressing CD44-HS, acquire an increased sensitivity to HGF/SF, leading to a growth advantage and to an invasive and metastatic phenotype similar to that observed in c-Met mutants (Fig. 2). This hypothesis is supported by the fact that CD44-HS is overexpressed in conjunction with c-Met in a variety of tumors including colorectal cancer (Liu et al., 1992; Yamashita et al., 1994; Di Renzo et al., 1995; Tuck et al., 1996; Wielenga et al., 1999). Hence, functional collaboration between CD44-HS and the c-Met-HGF/SF pathway presumably is an important factor in tumor growth and metastasis.

III. CD44 IN TUMOR PROGRESSION AND METASTASIS The initial observations relating CD44 expression to tumor dissemination were made in human non-Hodgkin’s lymphomas by us and the group of Jalkanen (Pals et al., 1989; Horst et al., 1990; Koopman et al., 1993; Jalkanen et al., 1990, 1991). In these studies, expression of CD44 on human lym-

CD44 Glycoproteins in Colorectal Cancer

175

Fig. 2 Regulation and function of CD44 in colorectal tumorigenesis. Due to an APC mutation, the Wnt pathway is constitutively activated in colorectal cancer and uncontrolled amounts of -catenin will complex with Tcf-4 and enter the nucleus. Transcription of target genes, including CD44 will be promoted, resulting in an enhanced level of CD44 protein expression on the cell membrane. The tumor-promoting effect of CD44 can be due to either hyaluronate binding resulting in enhanced cell motility or in modulation of signaling involved in regulation of motility and proliferation. Signaling can be mediated via the cytoplasmic tail of CD44 or by presentation of heparin-binding growth factors, such as HGF/SF.

phomas was found to be linked to tumor dissemination and to unfavorable prognosis. Sy and colleagues (1991) demonstrated that CD44s enhances growth and metastasis of human lymphoma cells in nude mice. The observation by Günthert et al. (1991) that a CD44 variant containing exon v6 could confer metastatic potential to a rat pancreas carcinoma cell line provoked numerous studies on CD44 variants in a wide variety of human tumors. In gastrointestinal cancer (Matsumura and Tarin, 1992; Heider et al., 1993; Tanabe et al., 1993; Wielenga et al., 1993; Finn et al., 1994; Kim et al., 1994; Mulder et al., 1994; Orzechowski et al., 1995; Rodriguez et al., 1995; Imazeki et al., 1996; Yamaguchi et al., 1996), breast cancer (Matsumura and Tarin, 1992; Friedrich et al., 1995; Kaufmann et al., 1995), nonHodgkin’s lymphomas (Koopman et al., 1993; Stauder et al., 1995), cervical (Dall et al., 1994) and bladder cancer (Cooper, 1995), and central nervous system malignancies (Kaaijk et al., 1995), increased levels of CD44

176 Table I CD44 Protein Expression in Colorectal Carcinoma—Correlation with Tumor Parameters Correlation withb Author

W/P/Fa

mAbs(exon)/pAbs(exon)

Expression inc Carcinoma (n)d

Metastases (n)d

(3) (17)

(19) (17)

(3) (8)

/

(22)

(38)

nt

nt

/

nt

(5)

nt

nt nt

nt yes

/ nt

(30) nt

(8) (68)

nt nt

Stage

Grade

Prognosis

Normal

no nt

no nt

nt nt

/ /

yes

yes

nt

nt

nt

nt nt

Adenoma (n)d

Abbasi et al. (1993) Heider et al. (1993)

P W/F

Wielenga et al. (1993)

F

Fox et al. (1994)

P/F

Kim et al. (1994) Mulder et al. (1994) P. A. Jackson et al. (1995) Orzechowski et al. (1995) Gotley et al. (1996)

W/P F

F10-44-2(1) NKI-PI(1–5)/CD44v(v3–10), DIII(v6/7), DI(v3) NKI-P1(1–5), VFF4(v6), VFF7(v6), VFF8(v5), VFF11(v3), VFF14(v10), VFF16(v10), VFF17(v7/8)/ CD44v(v3–10) 2C5(2/3), 3G5(v3), 3D2(v4/5), 2F10(v6), 4B3(v6), IE8(v8/9) U9M2 (1–5) VFF4 (v6), VFF7 (v6)

P

F10-44-2 (1)

no

no

nt

/

(32)

(68)

nt

F F

BMS116 (v6) F10-44-2 (1), 2C5 (2/3), 3G5 (v3), 3D2 (v4/5), 2F10 (v6), IE8 (v8/9)

nt no

nt nt

nt nt

/

(13) (13)

(11) (61)

(7) (35)

Gorham et al. (1996)

F

Kawahara et al. (1996) Woodman et al. (1996)

P F

Yamaguchi et al. (1996) Bhataydekar et al. (1998) Coppola et al. (1998) Givechian et al. (1998) Ropponen et al. (1998) Wielenga et al. (1998)

P P W/P F P F

F10-44-2 (1), Hermes-3 (5), 23.6.1(v2), 2F10 (v6) 11.24 (v9) F10-44-2 (1), Hermes-3 (5), 23.6.1(v2), 3G5 (v3), VFF8 (v5), 2F10 (v6) 44-IV (v8–10) F10-44-2 (1) VFF7(v6) VFF7 (v6), VFF8 (v5), SFF2(S) Hermes-3(5), 3G5(v3), VFF18(v6) VFF4(v6), VFF7(v6), VFF18(v6), VFF8(v5)

no

nt

nt

/

(2)

(10)

nt

yes nt

nt nt

nt nt

/ /

(30) nt

(35) (19)

nt nt

yes nt

no nt

yes yes

/ nt

nt (8)

(215) (98)

nt nt

no no yes yes

no nt yes no

no nt yes yes

/ / / /

(35) nt nt nt

(34) (30) (194) (68)

(26) (8) nt nt

aW, Western blotting; F, immunohistochemistry on frozen tissue; P, immunohistochemistry on formalin ﬁxed, parafﬁn embedded tissue. bCorrelation with stage of tumorigenesis, histologic grade, and prognosis. cExpression in normal ( normal crypts), adenomas, carcinomas, and metastases: nt, not tested; , not present; / , present at the base of the crypts; , , , present at

low, intermediate, high levels, respectively. dn, number of specimens examined.

177

178

Vera J. M. Wielenga et al.

or altered splicing patterns have been found. Recently it was shown that expression of antisense CD44v6 mRNA in the colorectal tumor cell line HT29 suppresses growth and metastatic behavior in nude mice (Reeder et al., 1998). In the next paragraphs, we discuss the currently available data on CD44 expression in colorectal cancer.

A. Expression and Regulation of CD44 in Colorectal Cancer Colorectal cancer serves as a paradigm of multistep tumorigenesis in epithelial cancer. It evolves through a series of morphologically well-deﬁned stages accompanied by progressive accumulation of genetic changes involving components of the Wnt-signaling cascade (APC or -catenin), the ras oncogene family, the p53 and the TGF-2 receptor suppressor pathways (Kinzler and Vogelstein, 1996). This prompted us and others to use colorectal cancer as a model to study CD44 expression in relation to tumor progression. These studies have provided compelling evidence for grossly enhanced and deregulated expression of CD44 during colorectal cancer development (summarized in Tables I and II). In the normal colorectal mucosa, CD44 protein is expressed at low levels and is conﬁned to the base of the crypts. A similar distribution pattern was found for CD44s and for variant exons (Abbasi et al., 1993; Heider et al., 1993, 1995; Wielenga et al., 1993, 1998; Fox et al., 1994; P. A. Jackson et al., 1995b; Gorham et al., 1996; Gotley et al., 1996; Kawahara et al., 1996; Woodman et al., 1996; Yamaguchi et al., 1996; Coppola et al., 1998; Givechian et al., 1998; Ropponen et al., 1998). In colorectal tumors, expression of CD44 protein, including domains encoded by variant exons, is generally strongly enhanced in comparison to normal mucosa (Fig. 3; see color plate), although there is marked inter- and intratumor heterogeneity. The enhanced CD44 protein expression reﬂects (de)regulation at the transcriptional level, as more and larger transcripts were consistently ampliﬁed from tumor tissue (Table II). The relative increase in mRNA containing variant exon sequences over CD44s mRNA was interpreted to reﬂect inﬁdelity of the splicing machinery within the tumor cells. A recent study, however, has challenged this interpretation: By selective RT–PCR of enriched colon crypt cells the same set of isoforms was found in the normal and in the neoplastic colorectal epithelium (Givechian et al., 1998). Further studies are needed to clarify this issue. The major upregulation of CD44 occurs at the transition from normal mucosa to adenoma, indicating that deregulation of CD44 represents an early event in the adenoma-carcinoma sequence (Abbasi et al., 1993; Heider et al., 1993; Wielenga et al., 1993; Kim et al., 1994; P. A. Jackson et al., 1995; Got-

Table II CD44 mRNA Expression in Colorectal Cancer—Correlation with Tumor Parameters Correlation withc Author Stamenkovic et al. (1991) Matsumura and Tarin (1992) Tanabe et al. (1993) Wielenga et al. (1993) Finn et al. (1994) Orzechowski et al. (1995) Rodriguez et al. (1995) Yoshida et al. (1995) Gorham et al. (1996) Gotley et al. (1996) Gotley et al. (1996) Higashikawa et al. (1996) Imazeki et al. (1996) Woodman et al. (1996) Wong et al. (1997) Givechian et al. (1998)

N/RT/ISHa N RT RT RT RT ISH RT RT ISH N RT RT RT RT RT RT

Exonb v8–10, CD44s v6/7, CD44s v8–10, CD44s v5, v6, v9, CD44s v8–10, v6–10, CD44s v6 v2–v7, CD44s v2–v7 v2–10, CD44s v3, v6, v7, v8–10 v2–v7, v8–10, CD44s v6–9 v6, v7 v2, v6, CD44s v2, v7, CD44s v2–v10, CD44s

Expression ind

Stage

Grade

Prognosis

nt nt nt nt nt yes yes nt no nt no no nt nt nt nt

nt nt nt nt nt yes nt nt nt nt nt nt nt nt nt nt

nt nt nt nt yes nt nt nt nt nt nt nt nt nt nt nt

Normal

Adenoma (n)e nt nt nt nt nt (13) nt nt (2) (6) nt (5) (16) nt (8) nt

Carcinoma (n)e

Metastases (n)e

(2) (9) (14) (6) (24) (11) (44) (24) (10) (31) (57) (23) (32) (19) (59) (30)

nt nt (16) nt nt (7) nt nt nt (17) (3) nt (9) nt (1) (8)

aN, Northern blotting; RT, reverse transcriptase PCR; ISH, in situ hybridization. bExon, the exons or exon compositions that were examined. cCorrelation with stage of tumorigenesis, histologic grade, and prognosis. dExpression in normal ( normal crypts), adenomas, carcinomas, and metastases: nt, not tested; , not present; , , , present at low, intermediate, high levels, respec-

tively. en, number of specimens examined.

179

180

Vera J. M. Wielenga et al.

ley et al., 1996; Kinzler and Vogelstein, 1996; Korinek et al., 1997; Morin et al., 1997; Coppola et al., 1998). This suggests a causal relation for mutations involving components of the Wnt-signaling cascade, because these play a key role in the initial neoplastic transformation of colon epithelium (Kinzler and Vogelstein, 1996; Korinek et al., 1997; Morin et al., 1997). A recent study from our laboratory, analyzing CD44 expression in the intestinal mucosa of mice and humans with genetic defects in Apc or Tcf-4 indeed strongly supports a regulatory role of Wnt-pathway in CD44 expression (Wielenga et al., 1999). We observed enhanced CD44 expression in tumors arising in the intestinal mucosa of humans with a germline APC mutation, that is, familial adenomatous polyposis patients, as well as in Apc mutant mice. Importantly, this CD44 overexpression was not only present in adenomas and invasive carcinomas, but already in aberrant crypt foci with dysplasia, that is, in the earliest detectable neoplastic lesions. These lesions show loss of the wild-type copy of the APC gene but neither ras nor p53 mutations. In sharp contrast, Tcf-4 mutant mice, which have a disrupted Wnt-signaling pathway, show a complete absence of CD44 in the intestinal mucosa. Taken together, these results indicate that CD44 is upregulated in an early phase of colorectal tumorigenesis, as a direct or indirect result of a constitutive activation of the Wnt pathway (Fig. 2) (Wielenga et al., 1999). Transfection of mutant APC cells with wild-type APC leads to apoptosis (Morin et al., 1996). Interestingly, COX2 inhibition by NSAIDs such as Sulindac has a similar effect, and is accompanied by redistribution of catenin from the nucleus to the cytoplasm, suggesting parallelism between APC function and COX2 inhibition (Morin et al., 1996; Beazer-Barclay et al., 1996; Chan et al., 1998). Since CD44 expression was also reported to have a negative effect on apoptosis (Koopman et al., 1994; Ayroldi et al., 1995; Günthert et al., 1996; Yu et al., 1997), it could be one of the molecules, together with COX2, via which a constitutive active Wnt pathway results in a decrease in apoptosis.

B. CD44 as a Prognosticator in Colorectal Cancer Because CD44 splice variants containing v6 were shown to confer metastatic potential to rat carcinoma cell lines (Günthert et al., 1991), overexpression of CD44 variants on colorectal carcinomas might also increase metastatic propensity, leading to disseminated disease and tumor-related death. To Fig. 3 Expression of CD44 during colorectal tumorigenesis. Whereas CD44 expression in normal colorectal epithelium is low and conﬁned to the base of the crypts, it is enhanced in adenomas and in carcinomas. Stainings for CD44v6 are shown.

CD44 Glycoproteins in Colorectal Cancer

181

address this hypothesis, a number of studies have assessed the relation between CD44 expression on the primary tumor at surgery and tumor dissemination (Dukes’ stage) and prognosis. Although a positive correlation between CD44 expression and Dukes’ stage was found in only about half of the studies (Tables I and II), a strong correlation between CD44 expression and tumor-related death was observed in all major studies that have thus far been published (Table I). In a study group of 68 patients with a complete long-term follow-up (6.5– 9.5 years), we observed a strong correlation between CD44v6 expression and tumor-related death. In patients who had undergone an apparently radical resection of their primary tumor, the level of CD44v6 expression had prognostic value independent of Dukes’ stage, tumor grade, and tumor localization (Mulder et al., 1994; Wielenga et al., 1998). By using a panel of three anti-CD44v6 antibodies for detection, recognizing the same epitope but with different afﬁnity, an increased prognostic signiﬁcance was obtained (Wielenga et al., 1998). The results of the preceding study were conﬁrmed by three large immunohistochemical studies by Yamaguchi et al. (1996), Ropponen et al. (1998), and Bhatavdekar et al. (1998). In the study by Yamaguchi et al., CD44v8-10 expression instead of CD44v6, was assessed. The study group consisted of 215 patients with a median follow-up of 10 years. In the study by Ropponen, CD44s, CD44v3, and CD44v6, were assessed in 194 patients with a median follow-up of 14 years. Bhatavdekar et al. demonstrate a signiﬁcant correlation of CD44s with survival in 98 patients with a median follow-up of 60 months. In all three studies, expression of CD44 had independent prognostic value in patients after radical surgery. This relation was similar for CD44s, CD44v8-10, CD44v3, and CD44v6. In contrast, Coppola et al. (1998) did not observe a correlation between CD44 expression and survival in colorectal cancer patients. This study, however, suffers from a number of major drawbacks. Most importantly, the authors used a single low-afﬁnity mAb (VFF7) on formalin-ﬁxed material to detect CD44v6. Strong CD44v6 expression was detected in only 5% of the carcinomas, and was often localized in the cytoplasm instead of on the cell membrane. Added to a relatively small patient group (n 34) and short follow-up (mean 17.6 months), this presumably explains the discrepant results. The use of the same low-afﬁnity antibody can explain the negative results of Koretz et al. (1995) and Pals (1995). In conclusion, CD44 expression, assessed with mAbs against either CD44v3, v6, or v8-10, is a strong independent prognosticator in patients with colorectal cancer. CD44 expression reﬂects propensity for metastasis after apparently curative surgery and may be used to select patients that might beneﬁt from adjuvant therapy.

182

Vera J. M. Wielenga et al.

IV. CONCLUSIONS Studies from several laboratories have explored the expression and prognostic value of CD44 expression in colorectal cancer. Despite discrepancies, which can largely be attributed to technical differences, the following important conclusions can be drawn. First, CD44 mRNA and protein are upregulated in colorectal cancer. In the adenoma-carcinoma sequence the major increase of CD44 occurs at the transition from normal mucosa to adenoma, that is, in dysplastic aberrant crypt foci. Hence, it is closely associated with loss of APC/-catenin tumor suppressor function, and presumably results from constitutive activation of the Wnt-signaling pathway. Growth factor binding, as well as apoptosis inhibition by CD44, likely are instrumental in the growth advantage of CD44-bearing tumor cells. Second, several independent studies indicate that CD44 expression is an important independent prognosticator in colorectal cancer patients. In patients who had an apparently radical resection, CD44v6 (as well as v3, v8-10, and CD44s) expression predicts tumor-related death. CD44 expression reﬂects propensity for metastasis after apparently curative surgery and may be used to select patients that might beneﬁt from adjuvant therapy.

ACKNOWLEDGMENTS We thank Dr. Frank van den Berg for helpful discussions. This work was supported by grants from the Praeventiefonds (project 28-2575), and the Dutch Cancer Society (projects UVA 981712 and RUL 94-817).

REFERENCES Abbasi, A. M., Chester, K. A., Talbot, I. C., Macpherson, A. S., Boxer, G., Forbes, A., Malcom, A. D. B., and Begent, R. H. J. (1993). Eur. J. Cancer 29A, 1995–2002. American Cancer Society (1994). “Cancer Facts and Figures—1994.” American Cancer Society, Atlanta, GA. Arch, R., Wirth, K., Hofmann, M., Ponta, H., Matzku, S., Herrlich, P., and Zöller, M. (1992). Science 257, 682–685. Aruffo, A., Stamenkovic, I., Melnick, M., Underhill, C. B., and Seed, B. (1990). Cell (Cambridge, Mass.) 61, 1303–1313. Ayroldi, E., Cannarile, L., Migliorati, G., Bartoli, A., Nicoletti, I., and Riccardi, C. (1995). Blood 86, 2672–2678. Bartolazzi, A., Peach, R., Aruffo, A., and Stamenkovic, I. (1994). J. Exp. Med. 180, 53–66. Beazer-Barclay, Y., Levy, D. B., Moser, A. R., Dove, W. F., Hamilton, S. R., Vogelstein, B., and Kinzler, K. W. (1996). Carcinogenesis (London) 17, 1757–1760.

CD44 Glycoproteins in Colorectal Cancer

183

Bennett, K. L., Jackson, D. G., Simon, J. C., Tanczos, E., Peach, R., Modrell, B., Stamenkovic, I., Plowman, G., and Aruffo, A. (1995a). J. Cell Biol. 128, 687–698. Bennett, K. L., Modrell, B., Greenﬁeld, B., Bartolazzi, A., Stamenkovic, I., Peach, R., Jackson, D., Spring, F., and Aruffo, A. (1995b). J. Cell Biol. 131, 1623–1633. Bhatavdekar, J. M., Patel, D. D., Chikhlikar, P. R., Trivedi, T. I., Gosalia, N. M., Ghosh, N., Shah, N. G., Vora, H. H., and Suthar, T. P. (1998). Ann. Surg. Oncol. 5, 495–501. Bladt, F., Riethmacher, D., Isenmann, S., Aguzzi, A., and Birchmeier, C. (1995). Nature (London) 376, 768 –771. Boros, P., and Miller, C. M. (1995). Lancet 345, 293 –295. Bottaro, D. P., Rubin, J. S., Faletto, D. L., Chan, A. M. L., Kmiecik, T. E., van de Woude, G. F., and Aaronson, S. A. (1991). Science 251, 802–804. Bourdon, M. A., Krusius, T., Campbell, S., Schwartz, N. B., and Ruoslahti, E. (1987). Proc. Natl. Acad. Sci. U.S.A. 84, 3194–3198. Brinkmann, V., Foroutan, H., Sachs, M., Weidner, K. M., and Birchmeier, W. (1995). J. Cell Biol. 131, 1573–1586. Brown, T., Bouchard, T., St. John, T., Wayner, E., and Carter, W. G. (1991). J. Cell Biol. 113, 207–221. Bussolino, F., Di Renzo, M. F., Ziche, M., Bocchietto, E., Olivero, M., Naldini, L., Gaudino, G., Tamagnone, L., Coffer, A., and Comoglio, P. M. (1992). J. Cell Biol. 119, 629– 641. Carter, W. G., and Wayner, E. A. (1988). J. Biol. Chem. 263, 4193–4201. Chan, T. A., Morin, P. J., Vogelstein, B., and Kinzler, K. W. (1998). Proc. Natl. Acad. Sci. U.S.A. 95, 681–686. Coleman, M. P., Esteve, J., Damieki, P., Arslan, A., and Renard, H. (1993). IARC Sci. Publ. 121, 1–806. Cooper, D. L. (1995). J. Pathol. 177, 1–3. Coppola, D., Hyacinthe, M., Fu, L., Cantor, A. B., Karl, R., Marcet, R., Cooper, D. L., Nicosia, S. V., and Cooper, H. S. (1998). Hum. Pathol. 29, 627–635. Dall, P., Heider, K. H., Hekele, A., von Minckwitz, G., Kaufmann, M., Ponta, H., and Herrlich, P. (1994). Cancer Res. 54, 3337–3341. David, G. (1993). FASEB J. 7, 1023–1030. DeGrendele, H. C., Estess, P., Picker, L. J., and Siegelman, M. H. (1996). J. Exp. Med. 183, 1119–1130. DeGrendele, H. C., Estress, P., and Siegelman, M. H. (1997). Science 278, 672–675. Di Renzo, M. F., Olivero, M., Giacomini, A., Porte, H., Chastre, E., Mirossay, L., Nordlinger, B., Bretti, S., Bottardi, S., Giordano, S., Plebani, M., Gespach, C., and Comoglio, P. M. (1995). Clin. Cancer Res. 1, 147–154. Donate, L. E., Gherardi, E., Srinivasan, N., Sowdhamini, R., Aparicio, S., and Blundell, T. L. (1994). Protein Sci. 3, 2378–2394. Dougherty, G. J., Lansdorp, P. M., Cooper, D. L., and Humphries, R. K. (1991). J. Exp. Med. 174, 1–5. Faasen, A. E., Schrager, J. A., Klein, D. J., Oegema, T. R., Couchman, J. R., and McCarthy, J. B. (1992). J. Cell Biol. 116, 521–531. Finn, L., Dougherty, G., Finley, G., Meisler, A., Becich, M., and Cooper, D. L. (1994). Biochem. Biophys. Res. Commun. 200, 1015–1022. Fox, S. B., Fawcett, J., Jackson, D. G., Collins, I., Gatter, K. C., Harris, A. L., Gearing, A., and Simmons, D. L. (1994). Cancer Res. 54, 4539–4546. Friedrichs, K., Franke, F., Lisboa, B. W., Kügler, G., Gille, I., Terpe, H. J., Hölzel, F., Maass, H., and Günthert, U. (1995). Cancer Res. 55, 5424–5433. Giordano, S., Zhen, Z., Medico, E., Gaudino, G., Galimi, F., and Comoglio, P. M. (1993). Proc. Natl. Acad. Sci. U.S.A. 90, 649–653.

184

Vera J. M. Wielenga et al.

Givechian, M., Wörner, S., Sträter, J., Zöller, M., Heuschen, U., Heuschen, G., Lehnert, T., Herfarth, C., and von Knebel Doeberitz, M. (1998). Eur. J. Cancer 34, 1099–1104. Gorham, H., Sugino, T., Woodman, A. C., and Tarin, D. (1996). J. Clin. Pathol. 49, 482– 488. Gotley, D. C., Fawcett, J., Walsh, M. D., Reeder, J. A., Simmons, D. L., and Antalis, T. M. (1996). Br. J. Cancer 74, 341–351. Günthert, A. R., Sträter, J., von Reyher, U., Henn, C., Joos, S., Koretz, K., Moldenhauer, G., Krammer, P. H., and Möller, P. (1996). J. Cell Biol. 134, 1089–1096. Günthert, U., Hofmann, M., Rudy, W., Reber, S., Zöller, M., Haussmann, I., Matzku, S., Wenzel, A., Ponta, H., and Herrlich, P. (1991). Cell (Cambridge, Mass.) 65, 13 –24. Haynes, B. F., Telen, M. J., Hale, L. P., and Denning, S. M. (1989). Immunol. Today 10, 423– 428. Heider, K. H., Horst, E., van den Berg, F., Hofmann, M., Ponta, H., Herrlich, P., and Pals, S. T. (1993). J. Cell Biol. 120, 227–233. Heider, K. H., Mulder, J. W. R., Ostermann, E., Susani, S., Patzelt, E., Pals, S. T., and Adolf, G. R. (1995). Eur. J. Cancer 31A, 2385–2391. Herrlich, P., Zöller, M., Pals, S. T., and Ponta, H. (1993). Immunol. Today 14, 395–399. Higashikawa, K., Yokozaki, H., Ue, T., Taniyama, K., Ishikawa, T., Tarin, D., and Tahara, E. (1996). Int. J. Cancer 66, 11–17. Horst, E., Meijer, C. J. L. M., Radaskiewicz, T., Ossekoppele, G. J., van Krieken, J. H. J. M., and Pals, S. T. (1990). Leukemia 4, 595–599. Imazeki, F., Yokosuka, O., Yamaguchi, T., Ohto, M., Isono, K., and Omata, M. (1996). Gastroenterology 110, 362–368. Jackson, D. G., Bell, J. I., Dickinson, R., Timans, J., Shields, J., and Whittle, N. (1995). J. Cell Biol. 128, 673–685. Jackson, P. A., Green, M. A., Pouli, A., Hubbard, R., Marks, C. G., and Cook, M. G. (1995). J. Clin. Pathol. 48, 1098–1101. Jalkanen, S., and Jalkanen, M. (1992). J. Cell Biol. 116, 817 –825. Jalkanen, S., Bargatze, R. F., Herron, L. R., and Butcher, E. C. (1986). Eur. J. Immunol. 16, 1195–1202. Jalkanen, S., Bargatze, R. F., de los Toyos, J., and Butcher, E. C. (1987). J. Cell Biol. 105, 983– 990. Jalkanen, S., Jalkanen, M., Bargatze, R., Tammi, M., and Butcher, E. C. (1988). J. Immunol. 141, 1615–1623. Jalkanen, S., Joensuu, H., and Klemi, P. (1990). Blood 76, 1559–1566. Jalkanen, S., Joensuu, H., Söderström, K. O., and Klemi, P. (1991). J. Clin. Invest. 87, 1835– 1840. Jeffers, M., Fiscella, M., Webb, C. P., Anver, M., Koochekpour, S., and Vandewoude, G. F. (1998). Proc. Natl. Acad. Sci. U.S.A. 95, 14417–14422. Kaaijk, P., Troost, D., Morsink, F., Keehnen, R. M. J., Leenstra, S., Bosch, D. A., and Pals, S. T. (1995). J. Neuro-Oncol. 26, 190–195. Kalomiris, E. L., and Bourguignon, L. Y. W. (1988). J. Cell Biol. 106, 319–327. Katoh, S., Zheng, Z., Oritani, K., Shimozato, T., and Kincade, P. W. (1995). J. Exp. Med. 182, 419–429. Kaufmann, M., Heider, K. H., Sinn, H. P., von Minckwitz, G., Ponta, H., and Herrlich, P. (1995). Lancet 345, 615–619. Kawahara, K., Yoshino, T., Kawasaki, N., Miyake, K., and Akagi, T. (1996). J. Clin. Pathol. 49, 478–481. Kim, H., Yang, X. L., Rosada, C., Hamilton, S. R., and August, J. T. (1994). Arch. Biochem. Biophys. 310, 504 –507. Kinzler, K. W., and Vogelstein, B. (1996). Cell (Cambridge, Mass.) 87, 159–170.

CD44 Glycoproteins in Colorectal Cancer

185

Kjellén, L., and Lindahl, U. (1991). Annu. Rev. Biochem. 60, 443–475. Knudson, C. H., and Knudson, W. (1993). FASEB J. 7, 1233–1241. Koopman, G., van Kooyk, Y., de Graaf, M., Meijer, C. J. L. M., Figdor, C. G., and Pals, S. T. (1990). J. Immunol. 145, 3589–3593. Koopman, G., Heider, K. H., Horst, E., Adolf, G. R., van den Berg, F., Ponta, H., Herrlich, P., and Pals, S. T. (1993). J. Exp. Med. 177, 897–904. Koopman, G., Keehnen, R. M., Lindthout, E., Newman, W., Shimigu, Y., van Seventer, G. A., de Groot, C., and Pals, S. T. (1994). J. Immunol. 152, 3760–3767. Koretz, K., Möller, P., Lehnert, T., Hinz, U., Otto, H. F., and Herfarth, C. (1995). Lancet 345, 327–328. Korinek, V., Barker, N., Morin, P. J., van Wichen, D., de Weger, R., Kinzler, K. W., Vogelstein, B., and Clevers, H. (1997). Science 275, 1784–1787. Legg, J. W., and Isacke, C. M. (1998). Curr. Biol. 8, 705–708. Lesley, J., Hyman, R., and Kincade, P. W. (1993). Adv. Immunol. 4, 271–335. Lindahl, U., Lidholt, K., Spillman, D., and Kjellén, L. (1994). Thromb. Res. 75, 1– 32. Liu, C., Park, M., and Tsao, M. S. (1992). Oncogene 7, 181–185. Liu, F., Hata, A., Baker, J., Doody, J., Carcamo, J., Harland, R., and Massague, J. A. (1996). Nature (London) 381, 620–623. Lyon, M., Deakin, J. A., Mizuno, K., Nakamura, T., and Gallagher, J. T. (1994). J. Biol. Chem. 269, 11216–11223. Matsumura, Y., and Tarin, D. (1992). Lancet 340, 1053–1058. Morin, P. J., Vogelstein, B., and Kinzler, K. W. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 7950– 7954. Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B., and Kinzler, K. W. (1997). Science 275, 1787–1790. Mulder, J. W. R., Kruyt, P. M., Sewnath, M., Oosting, J., Seldenrijk, C. A., Weidema, W., Offerhaus, G. J. A., and Pals, S. T. (1994). Lancet 344, 1470–1472. Muto, T., Bussey, H. J. R., and Morson, B. C. (1975). Cancer (Philadelphia) 36, 2251–2270. Naor, D., Sionov, R. V., and Ish-Shalom, D. (1997). Adv. Cancer Res. 71, 241–319. Orzechowski, H. D., Beckenbach, C., Herbst, H., Stölzel, U., Riecken, E. O., and Stallmach, A. (1995). Eur. J. Cancer 31A, 2073–2079. Pals, S. T. (1995). Lancet 345, 328. Pals, S. T., Horst, E., Ossekoppele, G. J., Figdor, C. G., Scheper, R. J., and Meijer, C. J. L. M. (1989). Blood 73, 885–888. Peach, R. J., Hollenbaugh, D., Stamenkovic, I., and Aruffo, A. (1993). J. Cell Biol. 122, 257– 264. Reeder, J. A., Gotley, D. C., Walsh, M. D., Fawcett, J., and Antalis, T. M. (1998). Cancer Res. 58, 3719–3726. Rodriguez, C., Monges, G., Rouanet, P., Dutrillaux, B., Lefrançois, D., and Theillet, C. (1995). Int. J. Cancer 64, 347–354. Rong, S., Bodescot, M., Blair, D., Dunn, J., Nakamura, T., Mizuno, K., Park, M., Chan, A., Aaronson, S., and van de Woude, G. F. (1992). Mol. Cell. Biol. 12, 5152–5158. Rong, S., Segal, S., Anver, M., Resau, J. H., and van de Woude, G. F. (1994). Proc. Natl. Acad. Sci. U.S.A. 91, 4731–4735. Ropponen, K. M., Eskelinen, M. J., Lipponen, P. K., Alhava, E., and Kosma, V. M. (1998). Scand. J. Gastroenterol. 33, 303–309. Ruoslahti, E., and Yamaguchi, Y. (1991). Cell (Cambridge, Mass.) 64, 867–969. Schlessinger, J., Lax, I., and Lemmon, M. (1995). Cell (Cambridge, Mass.) 83, 357–360. Schmidt, L., Duh, F. M., Chen, F., Kishida, T., Glenn, G., Choyke, P., Scherer, S. W., Zhuang, Z., Lubensky, I., Dean, M., Allikmets, R., Chidambaram, A., Bergerheim, U. R., Feltis, J. T., Casadevall, C., Zamarron, A., Bernues, M., Richard, S., Lips, C. J. M., Walther, M. M., Tsui,

186

Vera J. M. Wielenga et al.

L. C., Geil, L., Orcutt, M. L., Stackhouse, T., Lipan, J., Slife, L., Brauch, H., Decker, J., Niehans, G., Hughson, M. D., Moch, H., Storkel, S., Lerman, M. I., Linehan, W. M., and Zbar, B. (1997). Nat. Genet. 16, 68–73. Screaton, G. R., Bell, M. V., Jackson, D. G., Cornelis, F. B., Gerth, U., and Bell, J. I. (1992). Proc. Natl. Acad. Sci. U.S.A. 89, 12160–12164. Sheikh, H., Legg, J., Lewis, C., Peck, D., and Isacke, C. (1998). Cell Adhes. Commun. 6, 149– 156. Sherman, L., Wainwright, D., Ponta, P., and Herrlich, P. (1998). Genes Dev. 12, 1058–1071. Shimizu, Y., Seventer van, G. A., Siraganian, R., Wahl, L., and Shaw, S. (1989). J. Immunol. 143, 2457–2463. Sleeman, J. P., Arming, S., Moll, J. F., Hekele, A., Rudy, W., Sherman, L., Kreil, G., Ponta, H., and Herrlich, P. (1996). Cancer Res. 56, 3134–3141. Stamenkovic, I., Amiot, M., Pesando, J. M., and Seed, B. (1989). Cell (Cambridge, Mass.) 56, 1057–1062. Stamenkovic, I., Aruffo, A., Amiot, M., and Seed, B. (1991). EMBO J. 10, 343–348. Stauder, R., Eisterer, W., Thaler, J., and Günthert, U. (1995). Blood 85, 2885–2899. Stoker, M., Gherardi, E., Perryman, M., and Gray, J. (1987). Nature (London) 327, 239 –242. Sy, M. S., Guo, Y. J., and Stamenkovic, I. (1991). J. Exp. Med. 174, 859–866. Taher, T. E. I., Smit, L., Grifﬁoen, A. W., Schilder-Tol, E. J. M., Borst, J., and Pals, S. T. (1996). J. Biol. Chem. 271, 2863–2867. Taher, T. E. I., van der Voort, R., Smit, L., Keehnen, R. M. J., Schilder-Tol, E. J. M., Spaargaren, M., and Pals, S. T. (1999). Curr. Top. Microbiol. Immunol. 246, 31–43. Tajima, H., Matsumoto, K., and Nakamura, T. (1992). Exp. Cell Res. 202, 423–431. Takayama, H., LaRochelle, W. J., Sharp, R., Otsuka, T., Kriebel, P., Anver, M., Aaronson, S. A., and Merlino, G. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 701–706. Tanabe, K. K., Ellis, L. M., and Saya, H. (1993). Lancet 341, 725–726. Tanaka, Y., Adams, D. H., Hubscher, S., and Shaw, S. (1993). Nature (London) 361, 79–82. Tanaka, Y., Kimata, K., Adams, D. H., and Eto, S. (1998). Proc. Assoc. Am. Physicians 110, 118–125. Tölg, C., Hofmann, M., Herrlich, P., and Ponta, H. (1993). Nucleic Acids Res. 21, 1225– 1229. Toyama-Sorimachi, N., Sorimachi, H., Tobita, Y., Kitamura, F., Yagita, H., Suzuki, K., and Miyasaka, M. (1995). J. Biol. Chem. 270, 7437–7444. Tsukita, S., Oishi, K., Sagara, J., Kawai, A., and Tsukita, S. (1994). J. Cell Biol. 126, 391–399. Tuck, A. B., Park, M., Sterns, E. E., Boag, A., and Elliot, B. E. (1996). Am. J. Pathol. 148, 225– 232. Van der Voort, R., Manten-Horst, E., Smit, L., Ostermann, E., van den Berg, F., and Pals, S. T. (1995). Biochem. Biophys. Res. Commun. 214, 137–144. Van der Voort, R., Taher, T. I. E., Keehnen, R. M. J., Smit, C., Groenink, M., and Pals, S. T. (1997). J. Exp. Med. 185, 2121–2131. Van der Voort, R., Taher, T. E. I., Wielenga, V. J. M., Spaargaren, M., Prevo, R., Smit, C., David, G., Hartmann, G., Gherardi, E., and Pals, S. T. (1999). J. Biol. Chem. 274, 6499–6506. Weber, G. F., Ashkar, S., Glimcher, M. J., and Cantor, H. (1996). Science 271, 509–512. Weidner, K. M., Behrens, J., Vandekerckhove, J., and Birchmeier, W. (1990). J. Cell Biol. 111, 2097–2108. Weidner, K. M., Sachs, M., and Birchmeijer, W. (1993). J. Cell Biol. 121, 145 –154. Wielenga, V. J. M., Heider, K. H., Offerhaus, G. J. A., Adolf, G. R., van den Berg, F. M., Ponta, H., Herrlich, P., and Pals, S. T. (1993). Cancer Res. 53, 4754–4756. Wielenga, V. J. M., van der Voort, R., Mulder, J. W. R., Kruyt, P. M., Weidema, W. F., Oosting, J., Seldenrijk, C. A., van Krimpen, C., Offerhaus, G. J. A., and Pals, S. T. (1998). Scand. J. Gastroenterol. 33, 82–87.

CD44 Glycoproteins in Colorectal Cancer

187

Wielenga, V. J. M., Smits, R., Korinek, V., Smit, C., Kielman, M., Fodde, R., Clevers, H., and Pals, S. T. (1999). Am. J. Pathol. 154, 515–523. Wong, L. S., Cantrill, J. E., Morris, A. G., and Fraser, I. A. (1997). Br. J. Surg. 84, 363 –367. Woodman, A. C., Sugiyama, M., Yoshida, K., Sugino, T., Borgya, A., Goodison, S., Matsumura, Y., and Tarin, D. (1996). Am. J. Pathol. 149, 1519–1530. Yamaguchi, A., Urano, T., Goi, T., Saito, M., Hiroso, K., Nakagawa, G., Shiku, H., and Furukawa, K. (1996). J. Clin. Oncol. 14, 1122–1127. Yamashita, J., Ogawa, M., Yamahita, S., Nomura, K., Kuramoto, M., Saishoji, T., and Shin, S. (1994). Cancer Res. 54, 1630–1633. Yayon, A., Klagsbrun, M., Esko, J. D., Leder, P., and Ornitz, D. M. (1991). Cell (Cambridge, Mass.) 64, 841–848. Yingling, J., Das, P., Savage, S., Zhang, M., Padgett, R., and Wang, X. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 8940–8944. Yoshida, J., Bolodeoku, J., Sugino, T., Goodison, S., Matsumura, Y., Warren, B., Toge, T., Tahara, E., and Tarin, D. (1995). Cancer Res. 55, 4273–4277. Yu, Q., Toole, B. P., and Stamenkovic, I. (1997). J. Exp. Med. 12, 1985–1996.

This Page Intentionally Left Blank

A Simple Model for Carcinogenesis of Colorectal Cancers with Microsatellite Instability Nicolas Janin* Comité de Patholgie Mammaire Institut Gustave Roussy 94805 Villejuif, France

I. Introduction II. A Simple Model of RER Tumor Carcinogenesis A. The Model’s General Features B. Application to Digestive Tract Tumors in Lynch Syndrome III. Evidence to Support the Model A. Explanation of the Curious Features of RER CRCs B. Explanation of the Lynch Syndrome-Associated Tumor Spectrum C. Explanation of the Absence of Variant Tumors During Childhood D. Data About MMR Deﬁciency in Normal Somatic Cells IV. Discussion A. Necessary Changes of Paradigms: Testable Hypotheses B. Accelerated Aging Process in Nontransformed RER Cells: Possible Role of Very Early Telomerase Reactivation in Senescence Rescue and Possible Impact on Karyotype C. Human Cancers Developing in Pathologically Hyperproliferative Tissues D. Violations of Multistep Carcinogenesis in Experimental Cancers E. A Way Out of Confusion in the Epidemiology of Human CRCs? F. Confrontation of the Model with the Basic Concepts in Genetic Predisposition to Cancer G. An Interesting Historical Problem V. Conclusion References

Lynch syndrome is a hereditary predisposition to colorectal cancer (CRC) and other cancers caused by germline mutations in mismatch repair (MMR) genes. Almost all the cancers diagnosed in Lynch syndrome have an acquired MMR deﬁciency, a replication error (RER) mutator phenotype that is also found in a minority of sporadic cancers developed in the target organs of Lynch syndrome. Lynch syndrome displays many curious features that cannot be accounted for by the prevailing concepts of carcinogenesis and genetics: (1) CRCs occur preferentially in the right side of the colon, whereas the majority of sporadic cases develop in the left colon; (2) the increased risk of CRC is not associated with an increased incidence of adenomatous polyps, which are necessary precancerous lesions in the development of common CRCs; (3) the tumor spectrum in Lynch syndrome is restricted to the colon and some extracolonic sites, whereas the responsible MMR genes are ubiquitously expressed; (4) the tumor risk, which is negligible during

*Corresponding address: 4, place du 11 Novembre, 92 240 Malakoff, France

Advances in CANCER RESEARCH 0065-230X/00 $30.00

Copyright © 2000 by Academic Press. All rights of reproduction in any form reserved.

190

Nicolas Janin

childhood, becomes signiﬁcant during adulthood at the age of 25 and thereafter remains essentially constant throughout the ages. (5) Finally, the sporadic counterparts to the CRCs diagnosed in the setting of Lynch syndrome very curiously develop almost exclusively in the right colon, whereas this right-sidedness is much less pronounced in Lynch syndrome. To explain these anomalies, we propose a model of RER carcinogenesis based on the simple idea that the RER mutator phenotype has only short-term viability in normal cells. The proposed model states that the RER carcinogenesis is divided into two clearly distinct evolutive phases: (1) a preliminary phase starting with the counterselective loss of mismatch repair function, in which most clones with the RER mutator phenotype are eliminated through apoptosis or an accelerated aging process; (2) an explosive phase that is initiated only if mutations blocking apoptosis and senescence, rapidly acquired during the short life span of the nontransformed RER clones, eventually rescue one RER cell that gives rise to the malignant clone. It will be shown that this theoretical framework with its heterodox initiation process not only possesses the virtue of allowing an understanding of Lynch syndrome, but may also have broader applications to all research ﬁelds dealing with carcinogenesis. © 2000 Academic Press.

I. INTRODUCTION The carcinogenesis of common colorectal cancers (CRCs) evolves slowly and goes through multiple pathological stages. The obligatory precancerous lesion is the adenoma, which typically forms polyps bulging in the colonic lumen. Few polyps degenerate into cancers, a process that takes approximately 10 years (the polyp dwell time). Common CRCs develop preferentially in the distal part of the colon (see Winaver et al., 1997, for review). In molecular terms, colorectal carcinogenesis may be adequately described as a multistep evolutionary process, each step corresponding to an acquired genetic anomaly that confers a selective advantage to the transforming cell (opening a door to clonal expansion). The mutations underlying the development of CRCs have been deﬁned and analyzed, and correlations have been made between the genetic anomalies and the pathological features (Fearon and Vogelstein, 1990). Mutations inactivating the two alleles of the APC gene occur very early in the course of the disease, for they are present in the majority of common adenomas (Levy et al., 1994). Figure 1 summarizes these basic data about common colonic carcinogenesis. Deﬁciency in the DNA mismatch repair (MMR) pathway deﬁnes a variant subset of colorectal cancers (Lynch and Smyrk, 1996; Kinzler and Vogelstein, 1996). Replication errors (RER) are responsible for their main characteristic, the large-scale accumulation of frameshift mutations affecting microsatellite tracts (Lynch and Smyrk, 1996; Kinzler and Vogelstein, 1996), a phenomenon that is often referred to as microsatellite instability. The cancers without MMR deﬁciency will thereafter be called the common cancers, whereas the cancers with MMR deﬁciency will be alluded to as variant cancers. The RER phenotype can be acquired early in carcinogenesis, evidence

Carcinogenesis of Colorectal Cancers

191

Fig. 1 Cartoon presentation of the classical multistep model depicting the evolution of common colorectal cancers. Inactivation of the two alleles of APC appears to be required for the development of common adenomas. Familial adenomatous polyposis is due to germline mutations inactivating one allele of APC.

having been demonstrated at the adenoma stage (Ionov et al., 1993; Aaltonen et al., 1994; Shibata et al., 1994). RER variants account for approximately 15% of sporadic CRCs and have no 100%-speciﬁc pathological or clinical features. However, they are more often poorly differentiated and are more often associated with an abundant mucin production and a dense lymphocytic inﬁltrate than common type RER CRCs. Moreover, they have a better prognosis when compared to the common CRCs, after adjustment for the usual prognostic factors. The most intriguing clinical feature of these RER variants is their quasi-constant localization to the proximal colon (Table I). Lynch syndrome, also called hereditary nonpolyposis colorectal cancer, is the hereditary predisposition to the RER variant CRC (Lynch and Smyrk, 1996) and is also associated with an increased risk of carcinoma of the endometrium and other speciﬁc extracolonic tumors, which in sporadic cases sometimes display a RER phenotype (Peltomäki et al., 1993). Lynch syndrome is typically caused by germline mutations in MMR genes (Lynch and Smyrk, 1996; Kinzler and Vogelstein, 1996). Consistent with the fact that cancer-prone individuals in Lynch syndrome have no phenotypic anomaly that would allow for identiﬁcation of their risk, the MMR gene mutations typically behave recessively at the cellular level with regard to the RER mutator phenotype (Casares et al., 1995). The MMR genes are ubiquitously ex-

192

Nicolas Janin

pressed and the restricted tumor spectrum of Lynch syndrome remains unexplained. Variant CRCs diagnosed in the setting of Lynch syndrome are often diagnosed at an early age, and are associated with an increased risk of multiple CRCs (Lynch and Smyrk, 1996). As their sporadic counterparts, they also have a tendency to develop in the proximal colon, but this rightsidedness is less pronounced: The distribution of RER tumors is very curiously shifted to the distal colon in Lynch syndrome (Table I). The largest unbiased data reveal another curiosity of Lynch syndrome: Cancer cases are very evenly distributed throughout the age classes between 25 and 75 (Aarnio et al., 1995; Voskuil et al., 1997), giving a cumulative risk that increases linearly above the third decade of life (Fig. 2). This very atypical age distribution of variant CRCs in Lynch syndrome would be compatible with a completely nonorthodox carcinogenesis for variant cancers that would entail a single rate-limiting step in the case of a preexisting germline mutation in a MMR gene. Finally at-risk people genetically prone to RER cancers seem to have no more colorectal adenomas than the general population (Lynch and Smyrk, 1996). Clinical evidence suggests that their cancer risk could be due to an accelerated tumorigenesis from the adenoma stage, a rapid evolution corresponding to the aggressive polyp concept: (1) Several cases of advancedstage CRCs have been reported less than 3 years after a normal screening colonoscopy among at-risk individuals (Järvinen et al., 1995; Vasen et al., 1995), suggesting that polyp dwell time could here be considerably shorter than in common (RER ) colorectal carcinogenesis (Winaver et al., 1997). (2) Screened HNPCC individuals develop signiﬁcantly fewer CRCs than unTable I Features of Common (RER ) and Variant (RER) Colorectal Cancersa RER sporadic Percent of total CRCs Typical age Natural History Pathology

85% 60 Common polyp Well differentiated

Genetic proﬁle

Complex caryotype APC mut. 75% TGFb RII mut.: rare LOH frequent 30% proximal 70% distal

Distribution

RER sporadic 13% 60 ? Poorly differentiated Mucin production Lymphoid inﬁltrate Normal caryotype APC mutations: rare TGFb RII mut. 90% LOH rare 95% proximal 5% distal

RER Lynch 2% Wide range 25–75 Aggressive polyp Idem

Idem

60% proximal 40% distal

aComparison of the sporadic RER tumors with their hereditary counterparts. The ﬁgures are only indicative. LOH, loss of heterozygoty.

Carcinogenesis of Colorectal Cancers

193

Fig. 2 Cumulative risk of CRC among gene carriers in HNPCC. The gene carriers’ cumulative risk of CRC increases as an almost perfectly linear function of time above the third decade of life, with a CRC risk of 1.6% per year. This linear relationship does not ﬁt the classical multistep model of carcinogenesis, but is compatible with a “one-hit” model of carcinogenesis. (Redrawn with the data of Voskuil et al., 1997.)

screened patients, the diminished risk being interpreted as the consequence of the removal of adenomas during colonoscopies (Järvinen et al., 1995). The absence of polyp excess in Lynch syndrome still remains unexplained: It has been argued that acquisition of a mutator phenotype could favor mutations of the APC gene and hence lead to the development of many adenomas (Beck et al., 1997). We tentatively propose later a simple model of RER carcinogenesis that provides an explanation to the right-sidedness of the variant CRCs and to the curious features of Lynch syndrome.

II. A SIMPLE MODEL OF RERⴙ TUMOR CARCINOGENESIS A. The Model’s General Features This model (Fig. 3) derives from data unambiguously showing that the RER phenotype very often precedes and actually causes the APC mutations observed in variant CRCs: signiﬁcantly different from RER tumors, the spectrum of APC mutations in RER tumors bears the signature of an MMR deﬁcit, that is, frameshift mutations affecting microsatellite tracts (Huang et al., 1996). Because APC mutations are acquired very early in the course of

194

Nicolas Janin

Fig. 3 The two phases of variant colorectal carcinogenesis. The dots symbolize the numerous mutations accumulated because of the RER mutator phenotype. The arrows symbolize the three assumptions of the model: (1) The ﬁrst step is acquisition of the RER mutator phenotype. Cells with MMR deﬁciency have no growth advantage. (2) Cells with an RER mutator phenotype may undergo malignant transformation through a several-step process that is purely mutation driven initially. (3) The mutation load rapidly causes the elimination of nontransformed cells with an RER mutator phenotype. Apoptosis or senescence restrict their life span to a limited number of replications.

common CRCs evolution, these data very strongly suggest that acquisition of genetic instability precedes initiation of carcinogenesis in variant CRCs (Huang et al., 1996). The model’s ﬁrst assuumption is that loss of the wild-type allele of the MMR gene occurs many times during the lifetime of a cancer-prone individual, in all the tissues where continuous proliferation throughout life allows the apparition of mutations. The second assumption is that the ensuing genetic instability, adding an enormous mutation load to each cell generation, favors the initiation of carcinogenesis through a mutation-driven process (numerous mutations are accumulated before a change of ﬁtness is eventually achieved). The third assumption is that the mutator phenotype is tolerable in normal cells only during a short period of time, because DNA damage normally elicits apoptosis (Kerr et al., 1996). Evidence for DNAdamage-induced apoptosis has been demonstrated in the bowel epithelium (Potten, 1992). The life span of a clone of nonmalignant RER cells is thus limited to a ﬁnite number of mitoses. Acquired genetic instability takes the cells out of their normal living frame and seals their destiny: either death (in most of the cases) or survival through malignant transformation. Completion of RER tumorigenesis beyond the initiating steps imposes the accumulation of numerous mutations in a process that normally goes on over many years, but is here completed in a short time because MMR-deﬁcient cells accumulate more mutations after a single division than RER cells after years of mitoses. In this model of RER tumorigenesis, the only rate-limiting steps are those leading to the initiation of the process.

Carcinogenesis of Colorectal Cancers

195

Impairment of apoptosis appears clearly as a necessary condition of initiation in this model, but it should not be sufﬁcient: nonmalignant RER clones escaping apoptosis should nevertheless have a limited life span. Admitting that the number of accumulated mutations is not only the principal trace of the elapsed time in somatic cells, but also a major cause of aging (Cristofalo et al., 1994), it may be said that the RER cells experience an enormous contraction of time: What goes on in the clone of cells stemming from a RER cell may be compared to the situation in people with Werner’s syndrome (adult-onset progeria) who accumulate DNA damage because of a defective helicase (Gray et al., 1997) and experience premature senescence and death. Unless they are saved by neoplastic transformation, we propose that nonmalignant RER cells escaping apoptosis are doomed to undergo an accelerated aging process and a dramatic shortening of their replicative life span, a phenomenon that has been demonstrated in Werner’s syndrome cells cultured in vitro (Faragher et al., 1993). The RER carcinogenesis process is thus divided up into two clearly distinct evolutive phases: (1) a preliminary phase concerning normal or nearly normal cells in which the acquired mutations necessary to initiate a RER carcinogenesis are paradoxically counterselected mainly because of apoptosis and possibly because of senescence; and (2) an explosive phase that takes place when the RER hypermutator state is no longer counterselective.

B. Application to Digestive Tract Tumors in Lynch Syndrome Mature functional cells lining the gastrointestinal tract live only 3–5 days and are continuously being replaced by new cells produced in the proliferative compartment of the crypts (Fig. 4A). The only cells that are not short lived in this epithelium are the stem cells, which have the unique capacity of undergoing self-maintenance divisions (Potten and Loefﬂer, 1992). Considering the normally low spontaneous mutation rate, the stem cells are supposed to be the only cells that can transform into cancer cells (Potten and Loefﬂer, 1992) because their longevity endows them with the time necessary to accumulate the mutations, eventually leading to the development of cancers. However, these considerations may not be relevant to RER mutator cells: Because the probability of cancer initiating mutations occurring in a post-stem cell is no longer negligible, we propose that all of the cells of the proliferative compartment of the crypts can be involved in RER tumorigenesis up to the moment they enter the postmitotic compartment (Fig. 4B). Consequently, a tiny difference in the number of replicating cells in the crypt should have a huge impact on the probability of variant carcinogenesis, because of the exponential relationship linking the number of mitoses in the

196

Nicolas Janin

Fig. 4 Illustration of two key elements of the model applied to the digestive tract: (1) the RER stem cells have a ﬁnite replicative life span; and (2) the mutation-driven initiation process is tightly coupled to the mitoses in the post-stem cells, which are permissive to malignant transformation. (A) The histology of the colonic crypt with the location of the stem cells. (Adapted from Potten and Loefﬂer, 1992.) (B) All the cells theoretically permissive to malignant transformation. Each dot represents the RER neomutations load added to the daughter cells after one mitosis. Simplifying assumptions are made: (1) All the stem cells mitoses are asymmetric, always yielding one differentiating cell and one remaining stem cell; (2) all of the ephemeral cell mitoses are symmetrical and no cell loss occurs in the proliferation compartment; and (3) replication errors induce the same number of mutations each time the cell divides, neglecting the many mutations occurring at loci that have already been hit during precedent rows of replication. Under these conditions, the number of viable cells susceptible to transformation in the progeny of a single RER stem cell is a function of two parameters: X, number of self-maintenance mitoses before the stem cell is destroyed by the mutation load; N, number of mitoses an ephemeral colonocyte normally undergoes in the crypt on its way to maturation.

crypts to the number of cells permissive to transformation (see legend for Fig. 4B). Since APC mutations have been described in RER tumors (Huang et al., 1996) and since APC mutations are initiating mutations in the evolution of common CRCs, we describe the model for the case in which loss of APC is the initiating event of variant cancer. It is proposed that loss of APC function in an RER enterocyte can simultaneously have two consequences: (1) Block apoptosis, thus allowing the continuation of the mutation-driven process; (2) lead to adenoma formation, thus initiating the selection-driven

Carcinogenesis of Colorectal Cancers

197

process of carcinogenesis. We try to estimate the probability of the event “initiation of the variant carcinogenesis process” in the case for which there is a germline MMR gene mutation. Loss of APC in a RER cell can be achieved in two mutually exclusive ways: Loss of the wild-type MMR allele can occur either before or after inactivation of the ﬁrst allele of APC. The problem is thus to evaluate the probabilities associated with the following sequences: either MMR2rAPC1rAPC2, or APC1rMMR2rAPC2. These two pathways must be considered separately if we want to estimate the probability of development of a variant CRC. The probability of the former sequence is proportional to the product of two probabilities: p1 probability of losing the MMR gene wild-type allele in one enterocyte, and p2 and p3 the probability of thereafter inactivating the two alleles of APC within the few viable generations of RER enterocytes. Since it is not conceivable that the three independent mutations all occur in the short-lived cells of the colonic epithelium, the ﬁrst step must occur in a stem cell. If we try to evaluate the probability of one stem cell being mutated, we must take into account that the mutation rate varies according to the environmental mutagenic inﬂuences, which in turn depend both on the diet and on the location of the stem cell in the digestive tract: Because colon content varies greatly from one point to another, the mutagenicity of the colonic lumen also may be expected to vary from one point to another. If we consider a precise location in the colon of an individual and consider that this person never changes his diet, p1 should remain constant lifelong. The probability p2 and p3 is proportional to the number of viable progeny of a RER stem cell, which depends on two parameters: the number X of viable self-maintenance divisions of the stem cell (which in turn essentially depends on the threshold of mutations induced apoptosis), and the number N of mitoses that ephemeral colonocytes undergo in the crypt on their way to maturation (Fig. 4B). Since neither X (deﬁned by the tolerable mutation load) nor N should vary at a given point in the colon, p2 and p3 should also remain constant throughout life, and likewise the product p1 (p2 and p3). The accumulation of these rate-limiting steps of RER carcinogenesis can thus be assimilated to an accident that can happen any time with a constant risk. Considering the second pathway leading to cancer initiation in Lynch syndrome, the probability of the sequence APC1rMMR2rAPC2 is the product of three probabilities: p 1 probability of APC1, p 2 probability of MMR2 when APC1, p 3 probability of APC2 when (APC1 and MMR2). As we have seen above, p 1 is a function of the basal mutation rate and should therefore remain constant throughout life. Considering APC as a typical anti-oncogene (with a recessive expression of the mutations at the cell level), we may assume that loss of APC1 is not associated with the acquisition of a selective advantage. Therefore, p 2 should also remain constant throughout life. As discussed earlier, p 3 depends both on X and on N and

198

Nicolas Janin

should also remain constant throughout life assuming these parameters are not modiﬁed. When a germline MMR gene mutation is present, the model therefore predicts that the risk of a colonic stem cell degenerating into a variant cancer should remain constant throughout life: It is a combination of the probabilities of the two independent pathways of transformation of a stem cell into an RER adenoma. Passing from the level of the cells to the level of the organism, the model predicts that the cumulative risk that this accident happens in an individual should be a linear function of time: The cumulative risk in one individual can be defined as the sum of all linear functions of time corresponding to all the colonic stem cells. The RER disease strikingly appears to be a “one-hit” disease in this model. It is due to the fact that the proposed evolution of variant CRCs entails only a single rate-limiting step conferring a selective advantage: the last mutation of the initiation process. The mutations are either neutral or counterselective before that step, and they are no longer rate limiting after that. The model predicts that this key step is tightly dependent on the tissue turnover (defining N) and on the tolerance of the stem cells to a huge mutation load (defining X).

III. EVIDENCE TO SUPPORT THE MODEL A. Explanation of the Curious Features of RERⴙ CRCs The model predicts that the CRC risk in Lynch syndrome should remain constant lifelong. The largest unbiased data from the population-based Finnish and Dutch registry (Aarnio et al., 1995; Voskuil et al., 1997) seem compatible with the model, because the cumulative risk of CRC increases quite linearly over the third decade of life (Fig. 2). However, the prediction that CRCs should occur throughout life and, therefore, also during childhood does not match with these empirical data. Fortunately, CRCs are exceptional before the age of 20 in Lynch syndrome. But age-dependent variations in the physiology of the colonic epithelium turnover or in the tolerance of the stem cells to genetic instability could distort the linear relationship predicted from the model. This suggestion of a physiologic difference between children and adults is in agreement with the reported cumulative risk of metachronous CRC (CRMet) in the Finnish registry after a ﬁrst CRC diagnosed in adulthood: The drawing of CRMet makes an almost perfect straight line up until the 45 years of maximal follow-up (Aarnio et al., 1995). The relation linking time with risk can be written CRMet 1.6 N%, where N is the number of years after the ﬁrst CRC. This strikingly linear relationship

Carcinogenesis of Colorectal Cancers

199

cannot be explained by the classical multistep model of carcinogenesis but is a logical consequence in the frame of this model. The model also predicts that the RER adenoma should rapidly degenerate, as expected from the clinical data suggesting an aggressive behavior of polyp in Lynch syndrome (Järvinen et al., 1995; Vasen et al., 1995), because none of the events leading to carcinoma beyond that stage appears rate limiting. Of course, the completion of colorectal carcinogenesis beyond the adenoma stage imposes acquisition of numerous other mutations, but genetic instability accelerates the evolutionary process beyond the RER adenoma stage. Lack of polyposis in Lynch syndrome appears like another logical consequence: (1) There is no reason that at-risk people should develop more sporadic (RER ) adenomas than the general population; (2) They are protected from the development of RER adenomas by apoptosis; (3) Whenever a RER adenoma appears, it is difﬁcult to observe because it degenerates so rapidly into a carcinoma. The model also provides a logical explanation to the skewed distribution toward the proximal colon of CRCs in Lynch syndrome or in their sporadic RER counterparts. Available data from humans (Goodlad et al., 1991; Patchett et al., 1997; Green et al., 1998) and rats (Pozharisski et al., 1980; Cooke et al., 1984; Hall et al., 1992a) show that the number of mitoses in colonic crypts varies according to the anatomic situation of the crypt: There is a decreasing proximal-to-distal proliferative gradient in the colon. According to the model, one or two of the rate-limiting steps of RER tumorigenesis (whether APC ﬁrst allele is mutated ﬁrst or not) are tightly linked to cell replication because of the MMR deﬁciency. The model therefore suggests a causal link between physiologic differences in the kinetics of epithelium turnover along the different segments of the colon and the skewed distribution of CRCs toward the proximal colon: Initiation of RER tumors is expected to be more frequent in the places where colonocytes physiologically undergo more mitoses before they become terminally differentiated. The difference could be even greater because the crypt density (and therefore the number of stem cells) has been reported to follow a proximal-to-distal decreasing gradient in the rat colon (Park et al., 1997). Considered alone, the model cannot explain the difference of right-sidedness between HNPCC-associated CRCs and their sporadic RER counterparts. However, this can be very easily achieved by a prior understanding of the left-sidedness of sporadic RER tumors and the adjunction of the following very simple hypothesis: When we consider the mutations occurring under physiologic conditions (and not the mutations occurring in cells with a RER mutator phenotype, which very heavily depend on the presence of microsatellite hotspots for mutations), the basal mutation rate of two genes that are both actively transcribed in the same normal somatic cell should be gross-

200

Nicolas Janin

ly proportional to their respective length. Accordingly, the APC gene (cDNA 8500 pb) should mutate more frequently than the MMR gene (2500 pb for hMLH1 and 2800 pb for hMSH2, which are the two MMR gene most frequently involved in Lynch syndrome). Importantly, APC and MMR genes can be inactivated by mutations located throughout their coding sequence (Lynch and Smyrk, 1996; Kinzler and Vogelstein, 1996). Common RER tumors are the most frequent of sporadic CRCs and about 70% of sporadic CRCs and of sporadic polyps are located distal to the splenic ﬂexure (Winaver et al., 1997). Because the distributions of colorectal polyps and cancers are almost identical, this suggests that an initiating event of sporadic CRC tumorigenesis happens with a different probability throughout the colon. Because of the importance of APC mutations in the genesis of adenoma, it may be proposed that the skewed distribution of RER tumors reﬂects variations in the probability of inactivating APC, which would be more frequently inactivated in an enterocyte located in the sigmoid than in an enterocyte located in the ascending colon. If one accepts the idea that distribution of common tumors reﬂects regional differences in the mutagenicity of the fecal bolus, the absence of left-sided sporadic variant tumors appears as an obvious consequence of the competitive probabilities of inactivating APC or any one of the MMR genes. Two facts hinder the development of left-sided RER tumors in someone whose germline DNA is wild-type both for APC and MMR: (1) APC is much more likely to be mutated ﬁrst, which leads to the development of a left-sided RER tumor. (2) If somatic mutations inactivate the two alleles of a MMR gene, development of a RER tumor is less likely in the left colon than in the right colon because of differences in epithelium turnover. The distribution shift of CRCs to the left colon in Lynch syndrome now has a simple explanation: The development of left-sided RER tumors is more likely in people who have a germline MMR gene mutation, because the competing risk of RER tumor development becomes negligible. The inactivation of a single MMR gene allele is an event that is much more likely to happen than the inactivation of the two alleles of the APC gene. Inactivation of the wild-type MMR gene is of course more likely to occur in the left colon than in the right colon (because of differences in exposure to mutagens) and the relative preponderance of right-sided RER tumors remains only because of differences in epithelium proliferation.

B. Explanation of the Lynch Syndrome-Associated Tumor Spectrum The model predicts that RER tumor initiation depends both on X (number of mitoses of the stem cell with mismatch repair deﬁciency) and on N

Carcinogenesis of Colorectal Cancers

201

(the number of mitoses ephemeral cells undergo in the tissue before terminal differentiation). In order of importance, X ranks above N since a variant tumor can arise only if the stem cell tolerates a certain number of replication error-prone cell divisions. The lifetime risk of CRC in Lynch syndrome is close to 80%, whereas the lifetime risk of small bowel tumors is close to 1% (Aarnio et al., 1995). It is logical to assume that this interorgan difference in tumor incidence is causally related to some properties intrinsic to the mucosa, and not to differences in carcinogen exposures due to the differences in the water content of the feces: The switch in tumor incidence is extremely brutal at the ileocæcal transition and not gradual, as would be expected if it were related to the progressive lowering of the colonic water content. The reason for this discrepancy in tumor incidence does not appear from the analysis of the anatomy of the digestive tract below the stomach: The ﬁrst impression that one gets from looking at slides of the tissues is the strikingly similar morphology of the mucosa from the duodenum down to the rectum. Since kinetic studies also reveal a close similitude in turnover throughout the epithelium, the model suggests looking for variations of the parameter X along the bowel. Indeed, available data suggest a signiﬁcant difference of stem cell tolerance to mutations between the small and the large bowel. It has been shown that apoptotic cells appear in the digestive tract of mice a few hours after the administration of mutagenic chemicals (Potten et al., 1992). By comparing the distributions of the apoptotic bodies along the crypts, the authors observed that the apoptotic bodies colocalize with the stem cells in the small bowel but not in the large bowel, thus suggesting that colonic stem cells could be more resistant to an experimentally induced mutation load than the small bowel stem cells. The authors suggested that a difference in the balance between stem cell deletion or persistence must affect the susceptibility of the tissue to neoplasia (Potten et al., 1992). Their experimental data could at least in part explain the epidemiologic data regarding the relative difference in cancer incidence between the small bowel and the colon (small bowel cancers are exceptional). We suggest extending this interpretation to mutations resulting from replication errors, and propose that small bowel cancers also occur very rarely in Lynch syndrome because stem cell apoptosis is very efﬁciently triggered in this tissue, immediately eliminating the mutation loaded progeny of RER stem cells. Considering the digestive tract alone, available data are thus consistent with the idea that variations of X explain the restriction of the tumors to the colon, while differences in the spatial distribution of N are responsible for the characteristic distribution of the CRCs among the segments of the colon. More generally, it can be proposed that analyzing the variations of X and N would give the clue to the particular cancer spectrum associated with Lynch syndrome. Analysis of the distribution of endometrial cancer throughout the ages among NHPCC gene carriers reveals a very striking decrease of the in-

202

Nicolas Janin

cidence in the postmenopausal period; about 80% of the affected patients have their endometrial cancer diagnosed under the age of 55, and the incidence of endometrial cancer very sharply decreases between the ages of 50 and 60 (Aarnio et al., 1995). These data perfectly ﬁt the model’s prediction of a close link between hormone-dependent endometrium proliferation and cancer risk.

C. Explanation of the Absence of Variant Tumors During Childhood The distribution of RER tumors over the ages could similarly be explained by an aging process of the stem cells, associated with an increasing difﬁculty to trigger apoptosis in response to a mutation load. Experimental evidence in the mouse indirectly supports this idea: It has been demonstrated that the number of somatic mutations acquired by colonic stem cells 30 days after administration of a single dose of mutagen is 75% higher in 27month-old mice than in 3-month-old animals (Lowes et al., 1992), the life expectancy of a laboratory mouse being approximately 3 years. These data can be interpreted by assuming that old mice colonic stem cells are more resistant to induction of apoptosis by a mutation load than young mice colonic stem cells. The quasi absence of CRCs in Lynch syndrome during childhood could thus be explained by a very efﬁcient induction of apoptosis by a mutation load in the young colonic stem cells, a process that would become less efﬁcient with increasing age of the stem cells.

D. Data About MMR Deﬁciency in Normal Somatic Cells The model predicts the existence of short-lived normal cells with MMR deﬁciency. In individuals carrying a MMR gene germline mutation, a minority of enterocytes isolated from a phenotypically normal epithelium have been found to exhibit microsatellite instability, but their life span (or rather the life span of their whole parenthood) has not been assessed (Parsons et al., 1995a). It has been shown that mice with a germline inactivation of the two alleles of a MMR gene are viable and fertile (de Wind et al., 1995; Reitmair et al., 1995). Considering the MSH-2 gene, it should, however, be stressed that the successful creation of MSH2 / mice does not prove that microsatellite instability does not interfere with mouse development and is compatible with adult life: It was clearly shown that none of the normal tissues displays instability of microsatellite sequences, a phenomenon restricted to lymphomas and organ samples inﬁltrated by the tumors (Reitmair et

203

Carcinogenesis of Colorectal Cancers

al., 1995). It seems therefore that constitutional knockout of the MSH2 gene in these mice is not synonymous with generalized genomic instability. If these mice are viable, it is not although their normal cells display a mutator phenotype but on the contrary because their normal cells have no catastrophic deﬁciency in MMR function.

IV. DISCUSSION The concept that genomic instability contributes to carcinogenesis has been reviewed by Loeb (1991), who has speciﬁcally argued, on the basis of the probability of spontaneous mutations, that acquisition of a mutator phenotype should necessarily be an early step in carcinogenesis. This idea is very clearly incorporated in the proposed model of variant carcinogenesis, in which the RER mutator phenotype is the very ﬁrst anomaly acquired by cells that may eventually undergo malignant transformation. The idea that apoptosis could be a safeguard in preventing neoplasia in MMR-deﬁcient cells has already been very clearly expressed (Kinzler and Vogelstein, 1996), a point that is central in the proposed model, but the authors did not consider specifically that apoptosis would eliminate normal or near-normal enterocytes with an acquired RER phenotype. Reminiscent of the two faces of the roman god Janus, the MMR genes have thus already been attributed the antagonistic functions considered in the model. The very simple novelty that has enabled the construction of the model has consisted of setting separate reference positions for the analysis of the dual properties of MMR genes (Table II). This shift in perspective led us to consider Lynch syndrome to be an inherited predisposition to short-lived foci of RER cells, which appear when somatic cells lose the wild-type allele of the MMR gene involved, and dis-

Table II Two Points of View About the RER Mutator Phenotype and the Function of MMR Genes Healthy organism RER cell fate MMR gene classiﬁcation

Elimination Housekeeping genes

Consequence of RER phenotype

Activation of anti-oncogenes Activation of apoptosis

Variant carcinogenesis Clonal expansion Keepers of the gate to necessary genomic instability Inactivation of antioncogenes Activation of oncogenes

204

Nicolas Janin

appear when the mutation load becomes incompatible with life after a ﬁnite number of mitoses. In turn, this led us to hypothesize the existence of a heterodox initiation process for RER carcinogenesis that has the following features: (1) It entails several steps. (2) These steps are the only rate-limiting steps of the whole variant carcinogenesis. (3) All of these steps except for the last one are either neutral or counterselective. This theoretical framework provides the clue allowing demonstrations that Lynch syndrome can be understood in terms of variations of the physiologic parameters (tolerance of the stem cells to mutations and epithelial turnover), which modulate the probability of completing this initiation process, leading to the accidental survival and malignant transformation of somatic cells with an acquired RER mutator phenotype. One major assumption of the model is that apoptosis impairment is closely linked to initiation, an idea consistent with the data showing that elimination of mature cells through apoptosis has a central role in the control of cell number in the gastrointestinal epithelium (Hall et al., 1994b). It has been proposed that mutations of the APC gene could block apoptosis and elicit the switch from the ﬁrst to the second evolutionary phase of the RER carcinogenesis model. This proposal is compatible with the following data: (1) APC expression is restricted in the bowel epithelium to regions in which terminal differentiation is established (Midgley et al., 1997); and (2) transfection of APC in a colon cell line induces apoptosis (Morin et al., 1996). However, alternative initiation pathways are surely quite common in RER carcinogenesis, because APC mutations have been reported to be rare in RER tumors (Konishi et al., 1996; Olschwang et al., 1997). This fact does not change the relevance of the model, which relies on the evolutionary logic of RER carcinogenesis and not on the individual genes involved in the process. Analysis of the mutation proﬁle of RER tumors shows that acquired frameshift mutations inactivating the TGF- RII receptor gene (Parsons et al., 1995b) and the BAX gene (Rampino et al., 1997) are detected both with high speciﬁcity and high frequency. These data are compatible with the proposed evolution of RER carcinogenesis, because both genes are supposed to be normally involved in the apoptotic process: (1) The growth factor TGF- is normally expressed by fully differentiated enterocytes near the luminal surface of the crypt, where it is thought to inhibit growth and induce apoptosis through an autocrine loop (Avery et al., 1993), and it has been shown in vitro that TGF- can induce apoptosis in a nontransformed human colon adenoma cell line (Wang et al., 1995). (2) The BAX gene is deﬁnitely not a classical anti-oncogene and its only known function is to be directly involved in the induction of apoptosis. When the APC gene is not involved, impairment of the TGF and BAX proapoptotic signals could favor survival of the RER enterocytes and contribute to the initiation or later survival of RER tumors.

Carcinogenesis of Colorectal Cancers

205

In the description of the model, APC has been chosen as the target of the initiating mutations although it is involved in only a minority of variant CRCs. This debatable choice was made purposely, because there is no doubt about the fact that loss of APC in an enterocyte can very strongly contribute to the development of an adenoma. Loss of APC second allele enables to understand the switch from a purely mutation-driven process (counterselective) to a mutation- and selection-driven process. On the other hand, available data do not allow us to state that an adenoma would result as the consequence of inactivation of any of the genes known to be frequently involved in variant tumorigenesis. The choice of APC allows making a simple and biologically plausible description of the initiation process of variant CRCs. The main problem that remains untouched by the choice of APC is the speculated very early requirement to escape senescence.

A. Necessary Changes of Paradigms: Testable Hypotheses If the model is correct, the basic concepts of initiation and promotion, which are adapted to the classical multistep evolution of RER-tumors, become irrelevant to describe the evolution of RER tumors. The concept of promotion does not make sense for RER carcinogenesis: The genetic instability of RER premalignant cells endows them with such a strong “endogenous promotion” that carcinogenesis cannot be modiﬁed by exogenous exposures. A corollary of this is that a clear-cut benign/malignant separation is no longer possible in this model: Because the genetic instability drives a rapid progression to malignancy, we must accept the paradox that RER carcinogenesis is already an irreversible malignant process at the “benign” of RER adenoma stage. The traditional terminology of initiation, which merely corresponds to the ﬁrst mutation in the classical multistep model, cannot be applied to the “initiation” of RER tumors, which is a process including several steps. We propose to name the initiation process of RER tumors run-initiation. The preﬁx run can be regarded either as an allusion to the post-stem cell mitosis-coupled mechanism of escape to cellular death (a run to escape death), or as a reminder of the fact that RER carcinogenesis follows a hit-and-run evolutionary logic, in which the cells switch without transition from initiation to progression. We must accept another paradox when we consider the RER disease: The process is reversible only during the peculiar run-initiation phase (up until the last step before completion of the process), because of the limited life span of the nonneoplastic RER cells. We propose to name the proposed evolutionary model of RER tumors the few-hits-and-run model. Because of this change of concepts, we must also change the way we clas-

206

Nicolas Janin

sify carcinogens when we switch from the common to the variant mechanism of carcinogenesis: Promotors of RER tumors become run-initiation facilitators, or run-initiators of RER tumors. In fact, all the carcinogens of RER tumors are necessarily run initiators because this phase entails all the rate-limiting and environment sensitive steps of the carcinogenic process. Mitogens and/or apoptosis blockers are expected to be very powerful run initiators, whereas pure mutagens are expected to be only weakly carcinogenic in this model. Figure 5 and Table III tentatively summarize the necessary conceptual switch from RER to RER tumors. These new concepts give a way to test the model: Modiﬁcations of proliferation or apoptosis rate must have an extremely strong impact on RER carcinogenesis. A clinically very important prediction of the model is that slowing down proliferation and/or enhancing the apoptosis rate should be very effective in preventing the development of RER tumors.

B. Accelerated Aging Process in Nontransformed RERⴙ Cells: Possible Role of Very Early Telomerase Reactivation in Senescence Rescue and Possible Impact on Karyotype A reactivation of telomerase activity has been reported in 8 of 24 normal colonic mucosa samples from patients affected by Lynch syndrome, a phenomenon marginally observed in only one of the non-Lynch controls (Cheng et al., 1998). This observation is unique for Lynch syndrome patients, be-

Fig. 5 Environmental inﬂuences on RER and RER diseases. The precancerous neoplastic lesion of RER carcinogenesis degenerates very rapidly and has not been represented.

207

Carcinogenesis of Colorectal Cancers

Table III Inﬂuence of Carcinogens on RER and RER Tumorigenesisa Inﬂuence RER RER

Mutagens Mitogens Mutagens Mitogens

Latency

Reversibility

Long Long Short Short

No Yes Yes (cell death) Yes (cell death)

aMitogens and apoptosis blockers could in theory both favor the run-initiation, the magnitude of their inﬂuence depending on the basal physiologic parameters.

cause it is well documented that telomerase activity is rarely detected in normal somatic tissues (Shay and Bacchetti, 1997). Since reactivation of telomerase could be one of the mechanisms allowing the normal cells to escape the RER-accelerated senescence process, it would be interesting to test the hypothesis that this telomerase activity coincides with the foci of cells that have acquired a MMR deﬁciency. If reactivation of telomerase is necessary extremely early in the course of carcinogenesis of variant CRCs, these cancers should have chromosomes with a normal or an increased telomere length compared to normal mucosa. We are not aware of studies addressing this question speciﬁcally. However, in a study analyzing telomere length in CRCs, it has been observed that the majority of CRCs have a reduced telomere length compared to normal mucosa, only a small minority of CRCs having a normal or above normal telomere length (Hastie et al., 1990). Moreover, when the analysis could be made, it was demonstrated that this shortening of telomeres associated with malignancy had already occurred at the adenoma stage (Hastie et al., 1990), an observation consistent with the results of studies addressing the question of the temporal pattern of telomerase activation during colorectal carcinogenesis. These studies have shown that adenomatous polyps have no detectable telomerase activity, the detection of telomerase activity being correlated with the acquisition of malignancy (Chadeneau et al., 1995; Kim et al., 1994). It has been suggested that the early telomere shortening could play a causal role in the acquisition of the chromosome rearrangements usually observed in the CRCs (Hastie et al., 1990). If this is correct, it implies that CRCs with normal karyotypes should have a precancerous adenoma stage with no telomere shortening. It is therefore tempting to consider the hypothesis that telomerase reactivation in nontransformed cells with MMR deﬁciency, possibly associated with the activation of other enzymes allowing escape from replicative senescence, would explain the normal karyotype characteristic of variant CRCs. If this is correct, an RER phenotype should be found in the small subset of CRCs with normal or above-normal telomere length, and variant adenomas should have an abnormally detectable

208

Nicolas Janin

telomerase activity compared to common adenomatous polyps. These points are easily testable.

C. Human Cancers Developing in Pathologically Hyperproliferative Tissues The model predicts that pathological states associated with an increased cellular proliferation should favor the apparition of RER carcinomas. Only one example of a clear-cut and well-studied association between proliferation and solid tumors is given here. Microsatellite instability has been reported to be present with high frequency (50% of cases) in the colonic epithelium of patients suffering from ulcerative colitis (Brentnall et al., 1996). Importantly, the studied mucosa was histologically free from benign or malignant neoplasms. Ulcerative colitis is clearly associated both with an increased proliferation rate in the crypts (Deschner and Salmon, 1981) and with an increased risk of CRC (Devroede et al., 1971; Ekbom et al., 1990). Microsatellite instability has already been reported in approximately 20% of a series of CRCs occurring in the setting of ulcerative colitis (Suzuki et al., 1994). However, the proportion of RER tumors could be higher, because epidemiologic, clinicopathologic, and biologic data in ulcerative colitis are so strikingly reminiscent of the Lynch syndrome: (1) The cumulative incidence of CRC increases quite linearly with time in adulthood with a cancer rate of approximately 2% per year, CRCs being exceptional among affected children (Devroede et al., 1971; Ekbom et al., 1990). (2) There is no distinct precancerous adenomatous polyp stage ( Jain and Peppercorn, 1997). (3) Advanced stage CRCs have been reported 10 –23 months after a colonoscopy reportedly free of cancer or dysplasia (Connell et al., 1994). (4) Poorly differentiated and mucinous carcinomas are more common in colitis-associated CRCs than in sporadic CRCs ( Jain and Peppercorn, 1997). (5) The molecular genetic proﬁle of the CRCs diagnosed in the setting of ulcerative colitis (Benhattar and Saraga, 1995 and references therein) is closer to that of RER than RER tumors, with a reduced frequency of APC mutations and a reduced frequency of allelic losses (Table I). The follow-up colonoscopies proposed to patients suffering from ulcerative colitis may perhaps allow testing the prediction that microsatellite instability is essentially a focal and transient phenomenon of the mucosa, associated with short-lived RER stem cells. Search for microsatellite instability in leukemia and lymphoma has revealed that this phenomenon is rare among hematological cancers, with the exception of lymphoid tumors in immunosuppressed patients and in lymphomas derived from mucosa-associated lymphoid tissue (Fey, 1997), diseases which are both usually preceded by a benign B-cell hyperplasia state.

Carcinogenesis of Colorectal Cancers

209

In the frame of the few-hits-and-run model, the risk among immunosuppressed patients would be interpreted as follows: B-cell stem cells sometimes accidentally acquire an MMR deﬁciency and thereby generate B cells loaded with new antigens because of the RER phenotype. Because B lymphocytes are very efﬁcient antigen-presenting cells, these RER B cells should elicit a strong immune reaction and should be very rapidly eliminated in an immunocompetent host. In case of immunosuppression, these RER lymphocytes would have a prolonged life span that would enable the run-initiation to take place. Of course, the immunosuppression-associated cancer risk can also be interpreted in the frame of the standard multiple-hit model of carcinogenesis. In fact, analysis of the available data suggests that both mechanisms of carcinogenesis take place: The clinicopathologic and molecular particularities of the immunosuppression-associated malignant lymphoproliferative disorders (LPD) allow the distinction of two kinds of lymphomas among post-transplantation (PT) patients (Chadburn et al., 1998 and references therein), both diseases being monoclonal regarding their EBV status. The two subsets of malignant PT-LPD are (1) the polymorphic PTLPD, which develops in extranodal sites in the majority of patients, and responds satisfactorily to treatment in two-thirds of cases; and (2) the malignant lymphoma/multiple myeloma PT-LPD, a very aggressive disease that is very often diffuse at the diagnosis, does not respond to treatment, and which tends to be diagnosed after a longer time interval following transplantation. These two clinicopathologic entities have a very different genetic proﬁle. The former shows no mutations in the usual oncogenes and anti-oncogenes (cmyc, bcl-1, bcl-2, p53, etc.), whereas such alterations are present in the latter subset. We suggest that these two PT-LPD correspond, respectively, to a RER and a RER process by analogy with the known differences in genetic proﬁles and in evolution characterizing the two subsets of colonic cancers. This hypothesis is easily testable. Instead of trying to test the model by demonstrating the existence of facts predicted by the theory, one could also follow the reverse pathway and try to ﬁnd out whether the few-hits-and-run model could explain ﬁndings that cannot be satisfactorily interpreted in the frame of the multistep model (with the associated classical concepts of initiation and promotion). Flagrant violations of the classical model of carcinogenesis have been reported in the literature about experimental models of chemically induced CRCs and other cancers.

D. Violations of Multistep Carcinogenesis in Experimental Cancers The CRCs induced in rats by administration of alkylating carcinogens can be classiﬁed in two clearly distinct pathologic categories (Sunter et al., 1978):

210

Nicolas Janin

(1) well-differentiated carcinomas of the left colon, often associated with adenomatous tissue at their periphery; and (2) poorly differentiated and mucin-secreting carcinomas of the right colon, which are not associated with benign adenoma lesions, but are instead conspicuously associated with lymphoid aggregates (Ward, 1974). These two subsets of experimental CRCs do not correspond to the same disease observed at different time of evolution: It has clearly been demonstrated that carcinomas paradoxically appear ﬁrst after injection of the carcinogen, and are later followed by the benign adenomas (Maskens and Dujardin-Loits, 1981). Reminiscent of the HNPCC problematic lack of polyp excess, the undifferentiated experimental rat CRCs thus seem to arise de novo from a normal mucosa, without the prior formation of a benign precancerous adenomatous lesion. The anatomic distribution, the evidence for or against a distinct precancerous adenoma stage, and the pathologic ﬁndings of the two subsets of experimental tumors are strikingly reminiscent of the characteristics of the human common and variant CRCs (Table I). The match is not perfect though, and this interspecies comparison reveals a very striking discrepancy: The frequency ratio of the two CRCs subsets is inverted. Opposite to the human situation, rats most frequent tumors are de novo carcinomas (Pozharisski, 1975a). Two lines of evidence suggest that the chemically induced de novo rat carcinomas evolve according to the proposed few-hits-and-run model. Analysis of the chemical carcinogenesis in rat CRCs has been performed with the aim of establishing the relationship between dose, number of cells at risk, time, and tumor yield (Maskens, 1981). The analysis proved that this carcinogenesis cannot be accounted for by the classical multistep model. In complete agreement with the predictions of the few-hits-and-run model, the analysis led to the following conclusions: (1) Two and only two distinct and essential changes in the affected cells are necessary to induce the carcinomatous phenotype in normal enterocytes. (2) Only the ﬁrst stage depends on carcinogen exposure. (3) The cells having undergone the ﬁrst change do not possess a proliferative advantage allowing them to grow into clones. (4) The passage from the ﬁrst to the second stage of carcinogenesis does not require carcinogen exposure, but randomly affects cells that have undergone the ﬁrst change at a rate that is essentially constant. The inﬂuence of environmental modiﬁcations on the evolution of experimental rat CRCs has been tested, sometimes giving clear-cut evidence of violation of the multistep model. Only two examples are given here: 1. A chronic irritation in the caecum of rats caused by a nonabsorbable suture has been shown to enhance dramatically the yield of carcinogen-induced CRCs in the Caecum, most of the tumors being undifferentiated carcinomas (Pozharisski, 1975b). In rats not receiving the carcinogen, no tumors were observed and the most conspicuous changes related to the injury

Carcinogenesis of Colorectal Cancers

211

were lesions of chronic inﬂammation and a very signiﬁcant increase of the proliferation rate in the epithelial zone within 1 mm from the wound. More than 90% of the wounded rats receiving the carcinogen developed malignant tumors, in comparison with only one-fourth of the nonwounded rats receiving the carcinogen. The author gave the following interpretation of the impressive gain in tumor yield by a device normally expected to act at the promotion stage of carcinogenesis: The intensiﬁcation of carcinogenesis by an injury that causes an activation of tissue proliferation seems to differ in nature from cocarcinogenesis . . . cocarcinogens exert their effect at the stage of “activation” (promotion), whereas injuries inﬂicted upon tissues prior to carcinogen treatment are conducive to the intensiﬁcation of proliferation and contribute to “initiation” (Pozharisski, 1975b). 2. The anti-inﬂammatory pyroxicam has been shown to reduce very efﬁciently the incidence of proximal tumors and to have only a modest effect distally (T. Liu et al., 1995). The likely mechanism of action of pyroxicam inferred from the data (T. Liu et al., 1995) appears to be prevention of initiation: Only the number of tumors was diminished but not their individual volumes. This result is paradoxical since anti-inﬂammatory drugs induce enterocyte apoptosis (Lee, 1993), and no data suggest that pyroxicam could interfere with the metabolic activation of the carcinogen. These ﬁndings gathered from experimental rat CRCs are thus very difﬁcult to interpret within the frame of the classical concepts of initiation and promotion: Neither changes in proliferation nor changes in apoptosis should signiﬁcantly inﬂuence the classical initiation stage. However, these ﬁndings perfectly ﬁt the predictions of the few-hits-and-run model, in which the rate-limiting steps are highly sensitive both to proliferation and to apoptosis. An acquired MMR deﬁciency can thus be predicted in the right-sided undifferentiated mucin-producing experimental CRCs: It would not be logical to try to ﬁgure out another explanation to account for all the similitudes of these tumors with the human RER CRCs, including their striking morphologic parenthood. To the best of our knowledge, MMR deﬁciency has never been tested in chemically induced rat CRCs with the standard protocols using azoxymethane or dimethylhydralazine. Investigation of this point should both conﬁrm the model and unify the human data with those observed in the animal experiments. The few-hits-and-run model also explains the reason why this model of CRC reproduces the human pathology in a very misleading way, by inverting the ratio between variant (the most frequent in the rat) and common CRCs (the most frequent in human). It is well recognized that all carcinogens are mitogenic at the high doses used in the animal experiments (Ames and Gold, 1990). Indeed, an important increase of crypt cell proliferation has been reported as the ﬁrst detectable anomaly in the course of experimental colonic carcinogenesis in the rat (Deschner and Maskens, 1982).

212

Nicolas Janin

Many other experimental ﬁndings in the ﬁeld of experimental cancers cannot be satisfactorily interpreted in the frame of the classic initiation-promotion paradigm. Only two examples of such problems are given here: 1. It has long been recognized that mutagens classiﬁed as initiators in the paradigmatic initiation-promotion model of skin carcinogenesis can no longer be characterized as initiators but become full carcinogens when used at high doses. Whereas the benign skin papillomas are necessary precancerous lesions when the initiators are given at low doses, this intermediate stage disappears at high doses (along with its associated promotion phase): A great majority of de novo skin carcinomas are observed when the mutagen is given at high doses and behaves like a complete carcinogen (Fisher et al., 1988). The terminology complete carcinogen suggests that the drug has both initiating and promoting properties in the frame of the classical two-stage model. This interpretation is not satisfying because promotion is a long process that obviously does not exist in the case of de novo skin carcinomas. It is tempting to suggest that these de novo skin carcinomas are RER tumors. 2. It has been reported that de novo hepatocellular carcinomas (developing without any detectable adenomatous precancerous stage requiring promotion to transform eventually in carcinomatous foci) are induced in rats receiving high doses of a DNA-reactive liver carcinogen, whereas the same carcinogen produces only premalignant tumors at low doses (Yamamoto et al., 1994; Enzman et al., 1995; Williams et al., 1996). Moreover, the dose– response curve regarding tumor yield is curiously nonlinear with a slight positive slope at low doses and a markedly increased slope at higher doses, data suggesting the onset of another mechanism of carcinogenesis at high doses (Enzman et al., 1995; Williams et al., 1996). It appears reasonable to test the hypothesis that this other mechanism is a run-initiation.

E. A Way Out of Confusion in the Epidemiology of Human CRCs? The ﬁeld of epidemiology has been wrought by many conﬂicting results about the relationship between food and colorectal cancer. If the model is correct, it can be predicted that our understanding of the inﬂuence of diet on CRC risk will change dramatically if the genetic heterogeneity of the disease is taken into account, with the associated differences in sensibility to carcinogens. The RER phenotype of CRCs has never been investigated in the published epidemiologic studies. In the absence of any molecular data that would give an unambiguous estimate of RER tumor incidence, the measure of right-sided CRCs in epidemiologic studies can be used as a poor-quality surrogate for RER tumors: If we assume that the mostly right-sided

Carcinogenesis of Colorectal Cancers

213

RER tumors make up approximately 15% of sporadic tumors and that 30% of RER tumors develop in the proximal part of the colon (see Table I), we can estimate that roughly half of the proximal tumors are of the RER type. We therefore suggest that all the studies showing a tendency for a different risk in the proximal colon should be reanalyzed in the light of the proposed model: A nonsigniﬁcant trend toward right-sided “colorectal cancer” may reﬂect a strong effect related to the minority of RER tumors. Data that might turn out to be of paramount importance for the HNPCC at-risk people could be gathered from the prevention studies in the general population. One example is given here: A meta-analysis of analytical epidemiologic studies investing the role of calcium in chemoprevention of CRCs has concluded that the results do not support the hypothesis that calcium prevents colorectal cancer (Bergsma-Kadijk et al., 1996). Because of a dilution bias, this conclusion might be wrong regarding the possible chemopreventive effect of calcium on variant CRCs. The following facts seriously raise the suspicion that there might be a dilution bias: (1) The relative risk (RR) of CRC is signiﬁcantly reduced whereas the RR of adenomas (Ad) is not: RRCRC 0.86 (0.74 0.98) compared to RRAd 1.13 (0.91 1.39). (2) The deﬁned cancer subsites were colon, rectum, and colorectum; they did not allow estimation of the distribution of cancers in the proximal colon. (3) The proximal colon location was only taken into account after pooling the adenomas and the carcinomas, a confusion that should be avoided in order to render the analysis relevant to the RER tumors. If the genetic heterogeneity could be taken into account in those studies, the conclusion might well turn out to be that calcium supplementation is an extremely effective way of preventing the variant CRCs (15% of all CRCs), whereas it has no inﬂuence on the development of common CRCs (85% of all CRCs). The few-hits-and-run model predicts that calcium should speciﬁcally prevent variant cancers: Calcium is supposed to exert a protective effect by binding to mitogenic factors in the colonic lumen (Newmark et al., 1984), the most important of them probably being the bile acids (van der Meer et al., 1991; Lupton et al., 1995). By rendering mitogenic factors biologically inert, calcium slows down colonic proliferation (Lupton et al., 1995), which should block the mitosis-linked run-initiation. Since the RER phenotype is now relatively easy to assess, all the right-sided CRCs should be tested for microsatellite instability in the setting of the epidemiologic studies dealing with CRCs causes or prevention. Because the polymerase chain reaction can be applied to the analysis of parafﬁn tumor blocks, this investigation could perhaps be done in a retrospective way if the blocks of the right-sided tumors are accessible, thus allowing use of all data already gathered. Table IV gives some examples of possible strategies to test the few-hitsand-run model.

214

Nicolas Janin

Table IV Some Ways to Test the Few-Hits-and-Run Model Experimental cancers

Human cancers

1. Look for an RER phenotype in the proposed counterpart to the human variant CRC. 2. Look for an RER phenotype in the de novo carcinomas and/or when high doses of carcinogens are used (when initiators must be reclassiﬁed as complete carcinogens). 3. Try the few-hits-and-run model when stuck with the prevailing concepts. 1. Test predictions: Microsatellite instability in nontumoral tissues should be a focal and transient phenomenon (in at-risk people in HNPCC families, in ulcerative colitis, etc.). Cancers with microsatellite instability should occur in the pathological states associated with a chronically increased cellular proliferation. 2. Reappraise the CRC epidemiologic data after testing the rightsided cancers for microsatellite instability.

F. Confrontation of the Model with the Basic Concepts in Genetic Predisposition to Cancer The mechanisms underlying the autosomal dominant syndrome predisposing to cancer rely on two basic concepts that have been respectively put forward by Knudson (1971) and by Comings (1973) about the retinoblastoma, the former providing a logical link between the genetics of a somatic disease and the genetics of a hereditary predisposition to that disease, the latter providing a mechanism for the evolution of the somatic disease. The data of genetic epidemiology showed that the clinical presentation of sporadic retinoblastoma (always unifocal and unilateral) differed from the presentation of retinoblastoma diagnosed in the setting of hereditary autosomic dominant predisposition to the disease (often multifocal and/or bilateral tumors). Knudson’s two-hits model provides a logical link between the genetics of retinoblastoma and the genetics of the hereditary predisposition to this disease (Fig. 6A). Supposing the development of any retinoblastoma is caused by the accumulation of two mutations within a retinoblast explains the clinical presentation of the disease by assuming two additional hypotheses: (1) One of these mutations spontaneously occurs with a very high probability within at least one of the retinoblasts during any child’s ontogeny; and (2) genetic predisposition to retinoblastoma is the consequence of a constitutional ﬁrst mutation in one of the genes mutated in sporadic cases. According to Knudson’s hypothesis, a retinoblastoma-prone child often has multifocal and/or bilateral retinoblastoma because each of the spontaneously occurring mutations gives rise to a clone of malignant cells. Comings (1973) proposed that Knudson’s elusive genetic targets of the two hits could be the two alleles of a single gene, whose product would negatively

Carcinogenesis of Colorectal Cancers

215

regulate cellular growth. At the cellular level, the function of the gene imagined by Comings has to be lost in order to confer a selective advantage to the transforming cell, a concept that only later was given the name anti-oncogene (Fig. 6B). From the point of view of a geneticist dealing with hereditary predisposition to cancer, one can consider Knudson’s hypothesis as a central theme that applies to all the known autosomal dominant syndromes. On the other hand, Comings’s hypothesis might be considered as a variation on this theme that applies to the paradigmatic retinoblastoma case and to most, but not all, of the known syndromes of predisposition to cancer. Obviously, there are few exceptions where the gene involved is not an anti-oncogene, but rather an oncogene, multiple endocrine neoplasia type II (MENII), a familial form of medullary thyroid carcinoma and pheochromocytomas, being a good example (Ponder and Smith, 1996). The proposed model of RER colorectal cancers suggests that the Lynch syndrome represents a new variation on the central theme of autosomal dominant cancer predisposition. Knudson’s prediction of a shortened carcinogenesis among at-risk people applies to Lynch syndrome: Sporadic RER CRCs exist and a subset of them have acquired mutations inactivating both alleles of a known MMR gene (B. Liu et al., 1995). On the other

Fig. 6 Cartoon presentation of the basic concepts of genetic predisposition to cancer (see text for explanations).

216

Nicolas Janin

hand, Comings’s prediction applies to Lynch syndrome only in a very restricted manner: The MMR genes can be considered as anti-oncogenes only after completion of the run initiation process (Fig. 7). The basic assumption of the model is that MMR genes are essential housekeeping genes, that is, necessary to cell survival. Loss of MMR function appears as a step in carcinogenesis only from a misleading teleological perspective. Actually, the history of carcinogenesis starts only when the run-initiation process is completed. What happens before does not belong to the history of carcinogenesis, but to the history of the cancer-free organism. The organism must be considered as healthy as long as cells acquiring a MMR deﬁciency are

Fig. 7 Conciliation of antagonistic points of view through the few-hits-and-run model. Application to Lynch syndrome of the concepts of genetic predisposition to cancer. The MMR gene housekeeping and oncogenic functions are incompatible with each other in nontransformed cells. The last step of run-initiation operates the switch allowing the paradoxical selection of the intrinsically counterselective RER mutator phenotype. Loss of function of the MMR genes can be considered to favor carcinogenesis only after the switch-point. Knudson’s hypothesis applies from the moment of the conception of cancer-prone individuals, whereas Comings’s hypothesis in a broad sense does not apply but from a misleading teleological tumor endpoint. Note: APC mutations initiate the malignant RER process in this diagram although other genes are involved in the majority of variant CRCs.

Carcinogenesis of Colorectal Cancers

217

eliminated, giving rise to multiple and unrelated ephemeral foci of microsatellite instability. In the light of the few-hits-and-run model, it is nonsense to consider the acquisition of a RER mutator phenotype as the initiating step of “the abortive phase of variant carcinogenesis”; the two elements of this intellectual construction are clearly incompatible with each other, since carcinogenesis can only take place when the preliminary phase does not abort. The onset of MMR deﬁciency in a cell is therefore not an initiating event of carcinogenesis, but rather an initiating event of an improbable prehistory of variant carcinogenesis. The few-hits-and-run model allows the integration of two points of view: (1) the teleological point of view of cancer cells, in which MMR deﬁciency persists as the unique trace of the prehistorical phase; and (2) the point of view of the organism, in which microsatellite instability foci in healthy tissues witness the abortion of most prehistorical phases.

G. An Interesting Historical Problem It is interesting to consider the factors that may have impeded the integration of known facts and the formulation of the few-hits-and-run model. By applying Thomas Kuhn’s analysis about the evolution of scientiﬁc ideas (Kuhn, 1996), it is obvious that this model of variant carcinogenesis could not be built on the basic two-stage, initiation-promotion model of carcinogenesis initially put forward more than 50 years ago by Berenblum (1941). On the other hand, the new model does not challenge the basic concepts of autosomal dominant predisposition to cancer. However, these latter concepts have been presented in an illegitimate way that may have impeded the conception of the few-hits-and-run model. As a matter of fact, analysis of the literature reveals that Knudson’s (1971) and Comings’s (1973) hypotheses were presented in one block in innumerable occasions during the 1980s and the 1990s, with an implicit confusion of the two concepts. Although this presentation is neither necessary nor legitimate, it is so common that it has acquired the status of a very peculiar paradigm, which appears to be false from the beginning instead of becoming false through scientiﬁc progress. In the frame of this illegitimate concept of genetic predisposition to cancer, MMR genes have logically been classiﬁed into a special category of anti-oncogenes named “caretakers” because of their DNA-repair function (Kinzler and Vogelstein, 1996). Because the inactivation of anti-oncogenes confers a selective advantage, it is clear that the preliminary phase of the few-hits-and-run model was inconceivable from this reference position. The run-initiation gives a solution allowing us to consider the MMR genes both as housekeeping genes (their primary function) and as indirect anti-oncogenes (a function that they only acquire very rarely by accident). The run-initiation model

218

Nicolas Janin

states that the MMR genes do not express these antagonistic and mutually exclusive functions simultaneously, but sequentially in the course of carcinogenesis (Fig. 7). It may turn out to be interesting and instructive to make an historical analysis of the reasons that have led to the apparition and the maintenance of the illegitimate paradigm of predisposition to cancer.

V. CONCLUSION The few-hits-and-run model is a new concept of carcinogenesis built on well-established facts about Lynch syndrome and on few hypotheses, which have already been proposed individually. This model not only allows the understanding of the genotype–phenotype relationship of Lynch syndrome, but also makes possible the connection and the integration of a great number of other data and concepts stemming from seemingly unrelated ﬁelds. The simplest way to test the model will be to assess MMR deﬁciency in rodent models of carcinogenesis. An understanding of RER tumor biology may bring us closer to an effective chemoprevention of CRC in people who have a germline mutation in a MMR gene. The simple biologic plausibility of the model and its unifying and creative power justify its adoption until new data suggest a better interpretation.

ACKNOWLEDGMENTS The ideas presented in this paper developed as a consequence of Dr. Teresa Fiuza’s conﬁdence and friendly insistence. Most of them were laid down while I was on the leave from the Institut Gustav Roussy in Villejuif. Many thanks for critical reading of the manuscript to Drs. Jean Bénard, Marie-Christine Dokhelar, Jean-Marc Guinebretière, Thierry Heidmann, Monique Lê, Bert Vogelstein, and Robert Weinberg. The work would not have been possible without the dedication of Mrs. Francine Courtial and her colleagues in documentation searching. I am deeply indebted to Mrs. Jane Gledhill for her great help with the English language and to Dr. Geneviève Contesso for her constant support.

REFERENCES Aaltonen, L. A., Peltomäki, P., Mecklin, J.-P., Järvinen, H., Jass, J. R., Green, J. S., Lynch, H. T., Watson, P., Tallqvist, G., Juhola, M., Sistonen, P., Hamilton, S. R., Kinzler, K. W., Vogelstein, B., and de la Chapelle, A. (1994). Cancer Res. 54, 1645–1648. Aarnio, M., Mecklin, J.-P., Aaltonen, L. A., Nyström-Lahti, M., and Järvinen, H. J. (1995). Int. J. Cancer (Pred. Oncol.) 64, 430 –433.

Carcinogenesis of Colorectal Cancers

219

Ames, B., and Gold, L. S. (1990). Science 249, 970–971. Avery, A., Paraskeva, C., Hall, P., Flanders, K. C., Sporn, M., and Moorghen, M. (1993). Br. J. Cancer 68, 137–139. Beck, N. E., Tomlinson, I. P. M., Homfray, T. F. R., Frayling, I. M., Hodgson, S. V., and Bodmer, W. F. (1997). Gut 41, 235–238. Benhattar, J., and Saraga, E. (1995). Eur. J. Cancer 31A, 1171–1173. Berenblum, I. (1941). Cancer Res. 1, 807–814. Bergsma-Kadijk, J. A., van’t Veer, P., Kampman, E., and Burema, J. (1996). Epidemiology 7, 590–597. Brentnall, T. A., Crispin, D. A., Bronner, M. P., Cherian, S. P., Hueffed, M., Rabinovitch, P. S., Rubin, C. E., Haggitt, R. C., and Boland, R. (1996). Cancer Res. 56, 1237–1240. Casares, S., Ionov, Y., Ge, H.-Y., Stanbridge, E., and Perucho, M. (1995). Oncogene 11, 2303– 2310. Chadburn, A., Chen, J. M., Hsu, D. T., Frizzera, G., Cesarman, E., Garret, T. J., Mears, J. G., Zangwill, S. D., Addonizio, L. J., Michler, R. E., and Knowles, D. M. (1998). Cancer (Philadelphia) 82, 1978–1987. Chadeneau, C., Hay, K., Hirte, H. W., Gallinger, S., and Bacchetti, S. (1995). Cancer Res. 55, 2533–2536. Cheng, A.-J., Reiping, T., Wang, J.-Y., See, L.-C., and Wang, T.-C. (1998). J. Natl. Cancer Inst. 90, 316–321. Comings, D. E. (1973). Proc. Natl. Acad. Sci. U.S.A. 70, 3324–3328. Connell, W. R., Lennard-Jones, J. E., Williams, C. B., Talbot, I. C., Price, A. B., and Wilkinson, K. H. (1994). Gastroenterology 107, 934–944. Cooke, T., Kirkham, N., Sainthorp, D. H., Inman, C., Goeting, N., and Taylor, I. (1984). Gut 25, 748–755. Cristofalo, V. J., Gerhard, G. S., and Pignolo, R. J. (1994). Surg. Clin. North Am. 74, 1–21. Deschner, E. E., and Maskens, A. P. (1982). Cancer (Philadelphia) 50, 1136–1141. Deschner, E. E., and Salmon, R. J. (1981). In “Gastrointestinal Cancer 1” ( J. J. Decosse and P. Sherlock, eds.), pp. 1–26. Martinus Nijhoff, The Hague. Devroede, G. J., Taylor, W. F., Sauer, W. G., Jackman, R. J., and Stickler, G. B. (1971). N. Engl. J. Med. 285, 17–21. de Wind, N., Dekker, M., Berns, A., Radman, M., and te Riele, H. (1995). Cell (Cambridge, Mass.) 82, 321–330. Ekbom, A., Helmick, C., Zack, M., and Adami, H.-O. (1990). N. Engl. J. Med. 323, 1228– 1233. Enzman, H., Zerban, H., Kopp-Schneider, A., Löser, E., and Bannasch, P. (1995). Carcinogenesis (London) 16, 1503–1518. Faragher, R. G. A., Kill, I. R., Hunter, J. A. A., Pope, F. M., Tannock, C., and Shall, S. (1993). Proc. Natl. Acad. Sci. U.S.A. 90, 12030–12034. Fearon, E. R., and Vogelstein, B. (1990). Cell (Cambridge, Mass.) 61, 759 –767. Fey, M. (1997). Leuk. Lymphoma 28, 11–22. Fisher, S. M., Reiners, J. J. Jr., Pence, B. C., Aldaz, C. M., Conti, C. J., Morris, R. J., O’Connel, J. F., Rotstein, J. B., and Slaga, T. J. (1988). In “Tumor Promoters: Biological Approaches for Mechanistic Studies and Assay Systems” (R. Langenbach et al., eds.), pp. 11–30. Raven Press, New York. Goodlad, R. A., Levi, S., Lee, C. L., Mandir, N., Hodgson, H., and Wright, N. A. (1991). Gastroenterology 101, 1235–1241. Gray, M. D., Shen, J.-C., Kamath-Loeb, A. S., Blank, A., Sopher, B. L., Martin, G. M., Oshima, J., and Loeb, L. A. (1997). Nat. Genet. 17, 100 –103. Green, S. E., Chapman, P., Burn, J., Burt, A. D., Bennett, M., Appleton, D. R., Varma, J. S., and Mathers, J. C. (1998). Gut 43, 85–92.

220

Nicolas Janin

Hall, C., Youngs, D., and Keighley, M. R. (1992). Gut 33, 1528–1531. Hall, P. A., Coates, P. J., Ansari, B., and Hopwood, D. (1994). J. Cell Sci. 107, 3569–3577. Hastie, N. D., Dempster, M., Dunlop, M. G., Thompson, A. M., Green, D. K., and Allshire, R. C. (1990). Nature (London) 346, 866–868. Huang, J., Papadopoulos, N., McKinley, A., Farrington, S. M., Curtis, L. J., Wyllie, A. H., Zheng, S., Willson, J. K. W., Markowitz, S. D., Morin, P., Kinzler, K. W., Vogelstein, B., and Dunlop, M. D. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 9049–9054. Ionov, Y., Peinado, M. A., Malkhosyan, S., Shibata, D., and Perucho, M. (1993). Nature (London) 363, 558 –561. Jain, K. J., and Peppercorn, M. A. (1997). Dig. Dis. 15, 243–252. Järvinen, H. J., Mecklin, J.-P., and Sipponen, P. (1995). Gastroenterology 108, 1405– 1411. Kerr, J. F. R., Wyllie, A. H., and Currie, A. R. (1996). Br. J. Cancer 56, 2922–2926. Kim, N. W., Piatyszeck, M. A., Prowse, K. R., Harley, C. B., West, M. D., Ho, P. L. C., Coviello, G. M., Wright, W. E., Weinrich, S. L., and Shay, J. W. (1994). Science 266, 2011–2015. Kinzler, K. W., and Vogelstein, B. (1996). Cell (Cambridge, Mass.) 87, 159–170. Knudson, A. G. (1971). Proc. Natl. Acad. Sci. U.S.A. 68, 820–823. Konishi, M., Kikuchi-Yanoshita, R., Tanaka, K., Muraoka, M., Onda, A., Okumura, Y., Kishi, N., Iwama, T., Mori, T., Koike, M., Ushio, K., Chiba, M., Nomizu, S., Konishi, F., Utsunomiya, J., and Miyaki, M. (1996). Gastroenterology 111, 307–317. Kuhn, T. (1996). “The Structure of Scientiﬁc Revolution,” 3rd ed. University of Chicago Press, Chicago. Lee, F. D. (1993). J. Clin. Pathol. 46, 118–122. Levy, D. B., Smith, K. J., Beazer-Barclay, Y., Hamilton, S. R., Vogelstein, B., and Kinzler K. (1994). Cancer Res. 54, 5933–5938. Liu, B., Nicolaides, N. C., Markowitz, S., Willson, J. K. V., Parsons, R. E., Jen, J., Papadopoulos, P., Peltomäki, P., de la Chapelle, A., Hamilton, S. R., Kinzler, K. W., and Vogelstein, B. (1995). Nat. Genet. 9, 48–55. Liu, T., Mokuolu, A. O., Rao, C. V., Reddy, B. S., and Holt, P. R. (1995). Gastroenterology 109, 1167–1172. Loeb, L. (1991). Cancer Res. 51, 3075–3079. Lowes, A., Williams, D., Williams, G. T., and Williams, E. D. (1992). J. Pathol. 167, 135A. Lupton, J. R., Chen, X.-P., and Frølich, W. (1995). Nutr. Cancer 23, 221–231. Lynch, H. T., and Smyrk, T. C. (1996). Cancer (Philadelphia) 78, 1149–1167. Maskens, A. P. (1981). Cancer Res. 41, 1240–1245. Maskens, A. P., and Dujardin-Loits, R.-M. (1981). Cancer (Philadelphia) 47, 81–89. Midgley, C. A., White, S., Howitt, R., Save, V., Dunlop, M. G., Hall, P., Lane, D. P., Wyllie, A. H., and Bubb, V. (1997). J. Pathol. 181, 426–433. Morin, P. J., Vogelstein, B., and Kinzler, K. W. (1996). Proc. Natl. Acad. Sci. U.S.A. 93, 7950– 7954. Newmark, H. L., Wargovitch, M. J., and Bruce, W. R. (1984). JNCI, J. Natl. Cancer Inst. 72, 1323–1325. Olschwang, S., Hamelin, R., Laurent-Puig, P., Thuille, B., de Rycke, Y., Li, Y.-J., Muzeau, F., Girodet, J., Salmon, R.-J., and Thomas, G. (1997). Proc. Natl. Acad. Sci. U.S.A. 94, 12122– 12127. Park, H.-S., Goodlad, R. A., Ahnen, D. J., Winnett, A., Sasieni, P., Lee, C. Y., and Wright, N. A. (1997). Am. J. Pathol. 151, 843–852. Parsons, R., Li, G.-M., Longley, M., Modrich, P., Liu, B., Berk, T., Hamilton, S. R., Kinzler, K. W., and Vogelstein, B. (1995a). Science 268, 738–740. Parsons, R., Myeroff, L. L., Liu, B., Willson, J. K. V., Markowitz, S. D., Kinzler, K. W., and Vogelstein, B. (1995b). Cancer Res. 55, 5548–5550.

Carcinogenesis of Colorectal Cancers

221

Patchett, S. E., Alstead, E. M., Saunders, B. P., Hodgson, S. V., and Farthing, M. J. G. (1997). Dis. Colon Rectum 40, 168 –171. Peltomäki, P., Lothe, R. A., Aaltonen, L. A., Pylkkänen, L., Nyström-Lahti, M., Seruca, R., David, L., Holm, R., Ryberg, D., Haugen, A., Brøgger, A., Børresen, A.-L., and de la Chapelle, A. (1993). Cancer Res. 53, 5853–5855. Ponder, B. A. J., and Smith, D. (1996). Adv. Cancer Res. 70, 179–222. Potten, C. S. (1992). Cancer Metab. Rev. 11, 179–195. Potten, C. S., and Loefﬂer, M. (1992). Development (Cambridge, UK) 110, 1001–1020. Potten, C. S., Li, Y. Q., O’Connor, P. J., and Winton, D. J. (1992). Carcinogenesis (London) 13, 2305–2312. Pozharisski, K. M. (1975a). J. Natl. Cancer Inst. 54, 1115–1123. Pozharisski, K. M. (1975b). Cancer Res. 35, 3824–3830. Pozharisski, K. M., Klimashevski, V. F., and Gushchin, V. A. (1980). Exp. Pathol. 18, 387–406. Rampino, N., Yamamoto, H., Ionov, Y., Li, Y., Sawai, H., Reed, J. C., and Perucho, M. (1997). Science 275, 967–969. Reitmair, A. H., Schmits, R., Ewel, A., Bapat, B., Redston, M., Mitri, A., Waterhouse, P., Mittrücker, H.-W., Wakeham, A., Liu, B., Thomason, A., Griesser, H., Gallinger, S., Ballhausen, W. G., Fishel, R., and Mak, T. W. (1995). Nat. Genet. 11, 64–70. Shay, J. W., and Bacchetti, S. (1997). Eur. J. Cancer 33, 787–791. Shibata, D., Peinado, M. A., Ionov, Y., Malkhosyan, S., and Perucho, M. (1994). Nat. Genet. 6, 273–281. Sunter, J. P., Appleton, D. R., Wright, N. A., and Watson, A. J. (1978). Virchows Arch. B 29, 211–233. Suzuki, H., Harpaz, N., Tarmin, L., Yin, J., Bell, J. D., Hontanosas, M., Groisman, G. M., Abraham, J. M., and Meltzer, S. J. (1994). Cancer Res. 54, 4841–4844. van der Meer, R., Kleibeuker, J. K., and Lapré, J. A. (1991). Eur. J. Cancer Prev. 1(Suppl. 2), 55–62. Vasen, H. F. A., Nagengast, F. M., and Meera Khan, P. (1995). Lancet 345, 1183–1184. Voskuil, D. W., Vasen, H. F. A., Kampman, E., van’t Veer, P., and the National Collaborative Group on HNPCC (1997). Int. J. Cancer 72, 205–209. Wang, C. Y., Eshleman, J., Willson, J. K. V., and Markowitz, S. (1995). Cancer Res. 55, 5101– 5105. Ward, J. M. (1974). Lab. Invest. 30, 505–513. Williams, G. M., Iatropoulos, M. J., Wang, C. X., Ali, N., Rivenson, A., Peterson, L. A., Schultz, C., and Gebhardt, R. (1996). Carcinogenesis (London) 17, 2253–2258. Winaver, S. J., Fletcher, R. H., Miller, L., Godlee, F., Stolar, M. H., Mulrow, C. D., Woolf, S. H., Glick, S. N., Ganiats, T. G., Bond, J. H., Rosen, L., Zapka, J. G., Olsen, S. J., Giardello, F. M., Sisk, J. E., van Antwerp, R., Brown-Davis, C., Marciniak, D. A., and Mayer, R. J. (1997). Gastroenterology 112, 594–642. Yamamoto, N., Nomura, K., Kayano, T., and Kitagawa, T. (1994). Cancer Lett. 83, 59–68.

This Page Intentionally Left Blank

Index

A Adenomatous polyposis coli characterization, 11–12 function, 12, 14 function loss, 196–197 mutations in apoptosis, 204 CRC role, 180, 205 frequency, in CRC, 200 in RER tumors, 193–194 Tcf /-catenin signaling, 17–21 Aﬂatoxins, 115–116 AKT kinase, 149 APC, see Adenomatous polyposis coli Apoptosis impairment, 203–204 p53, 97–98 Aromatic dyes, 117–118 Ataxia telangiectasia, 92 A to T mutations, 118

B BAX gene, 204 Bladder cancers environmental factors, 117–118 ErbB-2 expression, 41 Bone marrow cells, 160 Brain tumors, 41–42 Breast cancers associated mutations, 118–119 ErbB-2 expression antibody response, 35–36 DCIS, 32–33 IDC, 33–34 lymph node status, 34–35 male, 33 molecular markers, 34–35 survival, 34–35 treatment response, 36–37

immunotoxin therapy, 60 MSP-Ron regulation, 153–154 prognosis, 121 Tcf--catenin association, 17 therapy, ErbB directed, 58–59

C CAD, see Context-dependent activation domain E-cadherin, 34 CAF, see 5-Fluorouracil–doxorubicin– cyclophosphamide Calcium, 213 Cancers, see also speciﬁc types CpG dinucleotide frequency, 109–110 detection, 120–121 ErbB-directed therapy, 55–61 ErbB-1 expression, 30 ErbB-3 expression, 31 ErbB-4 expression, 31–32 follow-up, 120–121 MSP-Ron association, 153–154 MSP-Ron signal role, 151–152 p53 mutations, frequency, 104–106 prognosis, 121–122 Tcf--catenin association, 17–20 Casein kinases, 93 -Catenin –Tcf complexes cancer association, 17–21 Wingless/Wnt pathways signaling, 7–8 target gene activation, 15–16 target gene repression, 16–17 –Tcf-1 complexes Wingless/Wnt signaling regulation, 10–12, 14–15 transcription factors, 8–10

223

224 Cathepsin D, 34 CC to TT transitions, 116–117 CD44 glycoproteins CRC role, 178–181 as ECM component receptors, 172–173 heparan sulfate decorated, 173–174 structure, 170, 172 tumor progression, 174–175, 178 CDKN2/INK4, 105 Cell development, p53 role, 98–99 Cell differentiation, p53 role, 98–99 Cervical cancer, 38–39 Chemotherapy ErbB-2 directed, 59–60 ErbB-2 overexpression, 36 Cigarettes, see Tobacco smoke Cisplatin sensitivity, 59–60 Clonality markers, 120 CMF, see Cyclophosphamide–methotrexate–5-ﬂuorouracil c-Myc, 20 Colitis, ulcerative, 208 Colorectal cancers carcinogenesis, 190–191 cause, 169–170 CD44 role, 178–181 classiﬁcation, 209–210 diet and, 212–213 ErbB-2 expression, 39–40 Lynch syndrome risks, 198–200 mortality, 169–170 RER model application, 195–198 assumptions, 203–205 epidemiological ﬁndings, 212–213 experimental cancers, 209–212 general features, 193–195 genetic predisposition, 214–217 historical reference, 217–218 hyperproliferative tissues, 208–209 supporting evidence, 198–203 telomerase reactivation, 206–208 testable hypotheses, 205–206 Tcf--catenin association, 17–20 tumor progression, 169 Context-dependent activation domain, 6 COX2 inhibition, 180 CpG dinucleotides associated cancers bladder cancer, 118 breast cancer, 119 general, 111

Index

lung cancer, 114–115 characterization, 108–109 frequency, 109 CRC, see Colorectal cancers CRMet, 198 Cyclophosphamide–methotrexate–5ﬂuorouracil, 36–37, 59 Cytoplasmic -catenin, 12

D DCIS, see Ductal carcinoma in situ Decapentaplegic gene, 15 Diet, CRC and, 212–213 Digestive track tumors, 195–198 DNA damage, p53 role, 92–93 HMG box, 3–4 Dpp, see Decapentaplegic gene Ductal carcinoma in situ, 32–33 Dyes, aromatic, 117–118

E ECM, see Extracellular matrix EGF, see Epidermal growth factors Endocytosis, 54–55 Endometrial cancer, 38 Endothelial cells, vascular, 160 Endotoxemia, 158–159 Environment CRC evolution, 210–211 p53 mutations factors, 106–107 patterns, 111–113 Epidermal growth factors ErbB binding, 43–44, 51 ErbB signaling, 54 receptor, identiﬁcation, 26 ErbB receptors characterization, 26–29 clinical aspects, 29–32 ErbB-1, 30–31 ErbB-2 bladder cancer association, 41 brain tumor association, 41–42 breast cancer association, 32–37 cancer induction, 42–46 cervical cancer association, 38–39 colon cancer association, 39–40 endometrial cancer association, 38 esophageal cancer association, 40

225

Index

gastric cancer association, 40 head and neck cancer association, 41 ligands activating, 42–44 direct, 44–45 independent receptor dimerization, 45–46 signaling function, 51–52 lung cancer association, 40 ovarian cancer association, 37–38 pancreatic cancer association, 40 prostate cancer association, 39 vulvar cancer association, 38 ErbB-3, 31 ErbB-4, 31–32 evolution, 46–48 gene therapy, 58–59 immunotherapy, 54–55 immunotoxin therapy, 60–61 signaling network endocytosis tuning, 54–55 heterodimer transformation, 49–50 inter-receptor interactions, 48–49 intracellular, 52–53 ligand function, 51–52 tyrosine kinase inhibitor therapy, 61 Esophageal cancer, 40 Evolution, 46–48 Extracellular matrix, 172–173

F FAK, see Focal adhesion kinase 5-Fluorouracile–doxorubicin– cyclophosphamide, 37 Focal adhesion kinase MSP-Ron signaling, 150 tumor progression, 152

G Gastric cancers, 40 GC to AT transitions breast cancer, 119 lung cancer, 115–116 G:C to C:G transitions, 117–118 Gene therapy, 58–59 Germline mutations, 103–104 GGF, see Glial growth factor Glial growth factor, 44 Glycogen synthase kinase 3, 12, 14–15 Gro, see Groucho

Groucho, 16–17 G to C transversions, 117–118 G to T transversions breast cancer, 119 lung cancer, 115

H Head and neck cancers ErbB-2 expression, 41 prognosis, 122 Heparan sulfate, 173–174 Hepatitis B virus, 115–116 Hepatocyte growth factors MSP comparison, 140, 142–143 receptor dimerization, 143–144 –scatter factor, CD44 binding, 174 HGF, see Hepatocyte growth factors HGH, see Human growth hormone High mobility group, Tcf family, 3–4 HMG, see High mobility group Hormonal therapy, 36 Host defense response, 153 Human growth hormone, 143–144 Hyaluronic acid, 172–173 Hypoxia, 93

I IDC, see Invasive ductal carcinomas Immunization, 122 Immunotherapy, 54–55 Immunotoxins therapy, 60–61 Infections, 117–118 Interleukin-6, 158, 160–161 Invasive ductal carcinomas, 33–34 Ionizing radiation, 92–93

J Jun N-terminal kinases MSP-Ron signaling, 151 p53 activation, 93

L Lef, see Lymphoid-enhancing factor-1 Li-Fraumeni syndrome, 103–104, 119 Lung cancers ErbB-2 expression, 40 prognosis, 122 smoking and, 113–115

226 Lymphoid-enhancing factor-1 characterization, 2–3 context-dependent transactivation, 4, 6 functions, 6–7 –-catenin complexes, 8–10 HMG box, 3–4 Wingless/Wnt pathways description, 7–8 target gene, 15–17 Lymphoproliferative disorders, 209 Lynch syndrome aging and, 202 CRC risks, 198–200 description, 189, 191–192 digestive track tumors, 195–198 risks factors, 192–193 lifetime, 201 telomerase reactivation, 206–207 tumor spectrum, 200–202

M Macrophage stimulating protein characterization, 139–140 HGF comparison, 140, 142–143 pro-MSP cleavage bovine serum, 145 human serum, 145 wound exudates, 146 mechanism, 144–143 murine peritoneal macrophages, 145–146 receptor binding, 142–143 characterization, 140–141 dimerization, 143–144 regulation modes abnormal cells, 153–154 host defense response, 153 signaling FAK pathway, 150 JNK pathways, 151 mechanism, 147–148 P13K pathways, 148 Ras-dependent pathways, 148 Src kinase pathways, 150–151 tumor progression, 151–152 structure, 141–142 target cells bone marrow cells, 160–161

Index

ectodermal, 159–160 macrophages, 157–159 message expression, 161 osteoclasts, 159 vascular endothelial cells, 160 Male breast cancer, 33 MAPK, see Mitogen-activated protein kinase MDM-2, 105 Melanoma, 20 Metastasis, CD44 role CRC, 178–181 general observations, 174–175, 178 MGF, see Myxoma virus growth factor Microsatellite instability, 190–191 Mismatch repair deﬁciency in CRCs, 190–191 in normal somatic cells, 202–203 Missense mutations, 101 Mitogen-activated protein kinase, 53–54 MMR, see Mismatch repair MSP, see Macrophage stimulating protein Mutations adenomatous polyposis coli, 17–21 APC, 180 p53 bladder cancer, 117–118 breast cancer, 118–119 clinical applications as clonality markers, 120 lesion detection, 120–121 therapies, 122 tumor prognosis, 121–122 environmental factors, 106–107, 111–113 germline, 103–104 hepatocellular carcinoma, 115–116 location, 107–111 prevalence, 104–106 skin cancer, 116–117 Ron, 154–156 somatic, 101–103 Myxoma virus growth factor, 45

N NAF, see Neu activating factor Neu activating factor, 45 Neuregulins characterization, 44 ErbB2 heterodimer induction, 49 Nijmegen breakage syndrome, 92–93 NRG, see Neuregulins

227

Index

O Oligomerization, 86 Ovarian cancers, 37–38

P p53 anatomy, 84–85 C terminus description, 87 induction role, 95 database, 82 DNA-binding domain description, 85 induction role, 95 homology, 84 location, 83 mutations bladder cancer, 117–118 breast cancer, 118–119 clinical applications as clonality markers, 120 lesion detection, 120–121 therapies, 122 tumor prognosis, 121–122 environmental factors, 106–107, 111–113 germline, 103–104 hepatocellular carcinoma, 115– 116 location, 107–111 lung cancer association, 113–115 prevalence, 104–106 skin cancer association, 116–117 somatic, 101–103 types, 107–111 oligomerization domain, 86 polymorphisms, 100–101 proline-rich domain, 85 research history, 81–82 signaling downstream apoptosis, 97–98 cell-cycle checkpoints, 96–97 cell development, 98–99 cell differentiation, 98–99 genomic stability, 99 mediation, 95 repair, 98 replication, 98 senescence, 99 transcription, 98

upstream activation, 87, 92–93 induction modiﬁcations, 93–95 as stress sensor, 83 structure, 84–85 N-terminus description, 85 induction role, 93–95 Pancreatic cancer, 40 PARP, see Poly(ADP-ribose) polymerase Peritoneal macrophages, 145–146 Phosphatidylinositol 3′-kinase ErbB receptor signaling, 52–53 MSP-Ron signaling, 149 p53 activation, 92 Phospholipase C, 52–53 P13K, see -kinase; Phosphatidylinositol 3′ Poly(ADP-ribose) polymerase, 92 Polymorphisms, 100–101 Proline-rich domain, 85 Prostate cancer, 39 Pyroxicam, 211

R Ras proteins MSP-Ron signaling, 148 tumor progression, 151–152 Replication errors characteristics, 190 tumor carcinogenesis model application, 195–198 assumptions, 203–205 epidemiological ﬁndings, 212–213 experimental cancers, 209–212 general features, 193–195 genetic predisposition, 214–217 historical reference, 217–218 hyperproliferative tissues, 208–209 supporting evidence, 198–203 telomerase reactivation, 206–208 testable hypotheses, 205–206 RER, see Replication errors Ron characterization, 140–141 dimerization, MSP-induced, 143–144 ectodermal expression, 160 macrophage expression, 157–159 MSP binding, 142–143 MSP regulation modes abnormal cells, 153–154 host defense response, 153

228 Ron (continued) MSP signaling FAK pathway, 150 JNK pathways, 151 mechanism, 147–148 P13K pathways, 149 Ras-dependent pathways, 148 Src kinase pathways, 150–151 tumor progression, 151–152 mutations, 154–156

S Senescence p53 trigger, 99 rescue, 206–208 Silent mutations, 101–102 Skin cancers, see also Melanoma de novo, 212 sunlight and, 116–117 Smoking, see Tobacco smoke Somatic cells, MMR deﬁciency, 202–203 Src kinase pathways, 150–151 Stk characterization, 140–141 macrophage expression, 157–159 mRNA expression, 161 MSP binding, 142–143 Strand bias mutations, 112 Stress p53 activation, 92–93 p53 response, 83 Sunlight,, 116–117

T Taxol, 59 T-cell factor-1 characterization, 2–3 context-dependent transactivation, 4, 6 functions mammalian, 6–7 –-catenin complexes Wingless/Wnt pathways cancer association, 17–21 regulation, 10–12, 14–15 signaling, 7–8 target gene, 15–17 transcription factors, 8–10 HMG box, 3–4 Tcf, see T-cell factor-1

Index

Telomerase reactivation, 206–208 Tobacco smoke bladder cancer and, 117–118 exposure, p53 mutations, 106–107 lung cancer and, 113–115 Transcription p53 role, 98 Wingless/Wnt signaling, 8–10 Transforming growth factor- ErbB binding, 43–44 ErbB signaling, 54 Transforming growth factor- in apoptosis, 204 Wingless/Wnt signaling, 16 Tumors detection, 120–121 mutation patterns bladder cancer, 117–118 breast cancer, 118–119 factors, 111–113 hepatocellular carcinoma, 115–116 lung cancer, 113–115 skin cancers, 116–117 point mutations, 104–106 prognosis, 121–122 progression CD44 role CRC, 178–181 general observations, 174–175, 178 MSP-Ron signal, 151–152 Tyrosine kinase inhibitor therapy, 61 Tyrphostins, 61

U Ubx, see Ultrabithorax Ulcerative colitis, 208 Ultrabithorax gene, 15

V Vaccinia virus growth factor, 45 Vascular endothelial cells, 160 VGF, see Vaccinia virus growth factor Vulvar cancer, 38

W Wingless/Wnt signaling cancer association, 17–21 characterization, 7–8

229

Index regulation, 10–12, 14–15 target genes Tcf activation, 15–16 Tcf repression, 16–17 transcription factors, 8–10

Wound exudates, 146

X Xenopus nodal-related 3 gene, 16 Xeroderma pigmentosum, 116

This Page Intentionally Left Blank

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

No title

Recommend Documents