The EGF Receptor Family: Biologic Mechanisms and Role in Cancer

The EGF Receptor Family Biologic Mechanisms and Role in Cancer The EGF Receptor Family Biologic Mechanisms and Role i...

Author: Graham Carpenter (Editor)

315 downloads 832 Views 3MB 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

The EGF Receptor Family Biologic Mechanisms and Role in Cancer

The EGF Receptor Family Biologic Mechanisms and Role in Cancer Edited by

Graham Carpenter Department of Biochemistry Vanderbilt University School of Medicine Nashville, TN 37232, USA

AMSTERDAM • BOSTON • HEIDELBERG • LONDON NEW YORK • OXFORD • PARIS • SAN DIEGO SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO

Elsevier Academic Press 525 B Street, Suite 1900, San Diego, California 92101-4495, USA 84 Theobald’s Road, London WC1X 8RR, UK • This book is printed on acid-free paper. Copyright 2004, Elsevier, Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail: [email protected]. You may also complete your request on-line via the Elsevier Science homepage (http://elsevier.com), by selecting “Customer Support” and then “Obtaining Permissions.” Library of Congress Cataloging-in-Publication Data Application has been submitted. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 0-12-160281-8 For all information on all Academic Press publications visit our website at www.academicpress.com Printed in China 03 04 05 06 07 08 9 8 7 6 5 4 3 2 1

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

Raymond C. Harris, Eunkyung Chung, and Robert J. Coffey — EGF receptor ligands . . . . . .

3

Douglas L. Falls — Neuregulins: functions, forms, and signaling strategies . . . . . . . . . . . . . . . . .

15

Robert N. Jorissen, Francesca Walker, Normand Pouliot, Thomas P.J. Garrett, Colin W. Ward, and Antony W. Burgess — Epidermal growth factor receptor: mechanisms of activation and signalling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

33

Ami Citri, Kochupurakkal Bose Skaria, and Yosef Yarden — The deaf and dumb: the biology of ErbB-2 and ErbB-3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

57

Graham Carpenter — ErbB-4: mechanism of action and biology . . . . . . . . . . . . . . . . . . . . . . . .

69

H. Steven Wiley — Trafficking of the ErbB receptors and its influence on signaling . . . . . . . . . . .

81

David F. Stern — ErbBs in mammary development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

93

Thomas Holbro, Gianluca Civenni, and Nancy E. Hynes — The ErbB receptors and their role in cancer progression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Julia L. Boerner, Andrew Danielsen, and Nita J. Maihle — Ligand-independent oncogenic signaling by the epidermal growth factor receptor: v-ErbB as a paradigm . . . . . . . . . . . . . . . . . . 115 Carlos L. Arteaga — ErbB-targeted therapeutic approaches in human cancer . . . . . . . . . . . . . . . . 127 David W. Fry — Mechanism of action of erbB tyrosine kinase inhibitors . . . . . . . . . . . . . . . . . . . 137 Ben-Zion Shilo — Signaling by the Drosophila epidermal growth factor receptor pathway during development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 Nadeem Moghal and Paul W. Sternberg — The epidermal growth factor system in Caenorhabaditis elegans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Contributors

Ruth H. Allen, Extramural Division of Cancer Control and Population Sciences. National Cancer Institute, Bethesda, Maryland 20892 Hani K. Atrash, Pregnancy and Infant Health Branch, Division of Reproductive Health, National Centers for Disease Control, Atlanta, Georgia 30341 Donna Day Baird, Epidemiology Branch, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina 27709 Carol M. Baldwin, Department of Medicine, Respiratory Sciences Center, College of Medicine, University of Arizona, Tucson, Arizona 85723 Robert L. Barbleri, Department of Obstetrics and Gynecology and Reproductive Biology, Brigham and Women’s Hospital, Boston, Massachusetts 02115 Richard Beasley, Wellington Asthma Research Group (WARG), Department of Medicine, Wellington School of Medicine, Wellington South, New Zealand Iris R. Bell, Tucson Veterans Affairs Medical Center, Tucson, Arizona 85723 Gertrud S. Berkowitz, Departments of Community and Preventive Medicine and Obstetrics and Gynecology and Reproductive Sciences, The Mount Sinai Hospital, Mount Sinai School of Medicine, New York, New York 10029 Leslie Bernstein, Department of Preventative Medicine, Norris Comprehensive Cancer Center, University of Southern California, Los Angeles, California 90033 F. Xavier Bosch, Catalan Institute of Oncology, L’Hospitalet del Llobregat, Barcelona, Spain Judith Bradford, Virginia Commonwealth University, Center for Public Policy, Richmond, Virginia 23284 Naomi Breslau, Departments of Psychiatry, Biostatistics, and Epidemiology, Henry Ford Health Systems, Detroit, Michigan 48202 Robin L. Brey, Department of Medicine, Division of Neurology, University of Texas Health Sciences Center, San Antonio, Texas 78204

Louise A. Brinton, Environmental Epidemiology Branch, National Cancer Institute, Bethesda, Maryland 20892 Evelyn J. Bromet, Department of Psychiatry and Behavioral Science, State University of New York at Stony Brook, Stony Brook, New York 11794 Deborah Brooks-Nelson, Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, Pennsylvania 19104 Joelle M. Brown, Departments of Biostatistics and Epidemiology, University of California, Berkeley, California 94720 Dedra Buchwald, Department of Medicine, University of Washington School of Medicine, Seattle, Washington 98117 Diana S. M. Buist, Center for Health Studies, Group Health Cooperative of Puget Sound, University of Washington, Seattle, Washington 98101 Gale Burstein, Division of Adolescent and School Health Center for Disease Control and Prevention, Johns Hopkins University School of Medicine, Hyattsville, Maryland 20782 Willard Cates, Jr., Family Health International, Research Triangle Park, North Carolina 27709 Jane A. Cauley, University of Pittsburgh, Pittsburgh, Pennsylvania 15261 Connie L. Celum, Department of Medicine, University of Washington, and King County Department of Public Health, Harborview STD Clinic, Seattle, Washington 98104 David C. Christiani, Harvard School of Public Health, Harvard Medical School, Massachusetts General Hospital, Boston, Massachusetts 02115 Carolyn M. Clancy, Center for Outcomes and Effectiveness Research, Agency for Health Care Policy and Research, Department of Health and Human Services, Rockville, Maryland 20852 Karen Scott Collins, The Commonwealth Fund, New York, New York 10021

Introduction In 1962 Stanley Cohen published an article entitled “Isolation of a Mouse Submaxillary Gland Protein Accelerating Incisor Eruption and Eyelid Opening in the Newborn Animal” (J. Biol. Chem. 237 (1962) 1555–1562). This report described the purification of a “tooth–lid factor,” which was subsequently renamed epidermal growth factor (EGF) by Cohen based on its capacity to stimulate the proliferation of epidermal cells. In the following 40 years, research based on this tooth–lid factor has expanded at a prodigious rate to include the identification of other similar growth factors, a family of receptors (termed ErbB) that mediates the action of these ligands, the relationship of the ligands and receptors to human cancers, the development of clinical cancer therapies based on inhibition of these receptors, and genetic analyses of this highly conserved signaling system in lower organisms. It is obvious that the biologic significance of EGF extends far beyond the tooth and eyelid. At this point the enormity of the literature on this growth factor system plus the breadth of biological disciplines and technical expertise involved prohibits a comprehensive review by one or even a small group of authors. If such a review were to be written, its scope would be a forbidding undertaking for all but the most dedicated readers. To provide an alternative that is feasible for authors and digestible for readers, this volume has been compiled. Each

chapter covers a limited aspect of the ErbB/EGF field, but together they constitute nearly complete coverage of the topic. Included in this volume are chapters on growth factor ligands, including the EGF-like growth factors plus the neureulins or heregulins, and the individual receptors—the EGF receptor or ErbB-1, ErbB-2, ErbB-3, and ErbB-4. These are followed by a chapter on the intracellular trafficking of these receptors. The second phase of this volume concentrates on the biological context of ErbB receptors, particularly in mammary development and in cancer. These include the development of therapeutic kinase inhibitors and the use of ErbB antagonists in the treatment of human cancers in the clinic. Finally, the genetic systems that have enabled significant advances are explored. It is hoped that interested readers will find the subject of their most direct interest and useful sources of related information.

Graham Carpenter Vanderbilt University School of Medicine Nashville, TN USA [email protected]

EGF receptor ligands Raymond C. Harris, Eunkyung Chung, and Robert J. Coffey Department of Medicine, Vanderbilt University, and Nashville Veterans Association, Nashville, TN 37232, USA

Introduction The mammalian ligands that bind the EGF receptor (EGFR [HER1, Erb-B1]) include EGF, transforming growth factor-a (TGFa), heparin-binding EGF-like growth factor (HB-EGF), amphiregulin (AR), betacellulin (BTC), epiregulin (EPR), and epigen [1–4,24]. Each of the mature peptide growth factors is characterized by a consensus sequence consisting of six spatially conserved cysteine residues (CX7 CX4–5 CX10–13 CXCX8 C) that form three intramolecular disulfide bonds, with the following interactions: C1–C3, C2–C4, C5–C6. This consensus sequence is known as the EGF motif and is crucial for binding members of the HER receptor tyrosine kinase family. In addition to binding EGFR, HB-EGF, BTC, and EPR are reported to also bind HER4 (see below). Mature HB-EGF and AR also contain an amino-terminal heparin-binding domain (HBD). All of the ligands are made as type I transmembrane proteins that are inserted into the plasma membrane and are then cleaved by cell surface proteases to release mature growth factor that binds EGFR (Fig. 1). The structures of the fulllength EGFR ligands with the residues that undergo glycosylation are shown in Fig. 3, and the putative cleavage sites are indicated in Table 2. The chromosomal topography of EGFR ligands is depicted in Fig. 2A, along with their chromosomal location in mouse and human (Fig. 2B). TGFa, HB-EGF, AR, and BTC are similar, consisting of six exons, in which exon 1 encodes the 5¢UTR and signal peptide, exon 2 is the N¢ terminal precursor, exon 3 is the mature peptide through first two disulfide loops of EGF motif, exon 4 is the third disulfide loop and transmembrane domain, exon 5 is the cytoplasmic domain, and exon 6 is the 3¢UTR. The close proximity of AR, BTC, and EPR in syntenic regions of human chromosome 4 and mouse chromosome 5 suggests a primordial duplication event prior to divergence of these two species. In this highly selective review, we will highlight advances made in the EGFR ligand field by considering TGFa and HB-EGF as representing different ends of the spectrum in terms of their fates upon arrival at the cell

surface (Fig. 4). Emphasis will be placed on delivery, processing and action of the different ligands.

EGFR ligand delivery in polarized epithelial cells We have examined the trafficking of constitutive and inducible-expressing forms of wild-type and mutant TGFa, EGF, and AR in polarized MDCK cells stably transfected with these cDNAs, as well as trafficking of endogenous TGFa and AR in polarized HCA-7 cells, a line that was generated from a patient with a well-differentiated rectal cancer [5–9]. These HCA-7 cells, like MDCK cells, form a uniform polarizing monolayer when cultured on polycarbonatecoated Transwell filters. Both of these cell lines contain 20,000–40,000 EGFRs that are found predominantly at the basolateral surface like they are in all polarized epithelial cells. In both systems, TGFa and AR are delivered preferentially to the basolateral surface and cleaved by TACE/ADAM17, which is restricted to this compartment. HB-EGF is also found preferentially on basolateral membranes of polarized cells, associated with proteins that localize at the adherens junctions of cell-cell contact [10]. In contrast, EGF is delivered equally to both the apical and basolateral surfaces, but the proteolytic machinery is active preferentially in the basolateral compartment [8]. Thus, under steady-state conditions, EGF immunoreactivity is observed only at the apical surface.

TGFa as an autocrine growth factor TGFa is sorted to the basolateral surface of polarized epithelial cells, where it is cleaved by TNFa-converting enzyme/disintegrin and metalloproteinase 17 (TACE/ ADAM17), whereupon mature soluble growth factor is avidly bound to basolaterally restricted EGFRs. We were initially frustrated by our inability to detect immunoreactive TGFa in the medium of polarized MDCK cells that had been engineered to overproduce full-length human TGFa [9]. One possible explanation was that TGFa was rapidly

4

The EGF Receptor Family

Fig. 1. Mammalian family of epidermal growth factor receptors (EGFRs) and ligands that bind EGFR. See introductory section for definitions of abbreviations.

consumed by the approximately 40,000 EGFRs that are confined to the basolateral surface (Fig. 5A). In support of this hypothesis, basolateral administration of a monoclonal antibody (mAb) that blocks EGFR binding (mAb 225) resulted in increased TGFa levels in the basolateral medium. We

recently observed that this phenomenon also appears to occur in vivo. One day following C225 administration to a Mémétrier’s disease patient, there was a three-fold increase in immunoreactive TGFa in his serum and gastric juice [11]. Moreover, there was a dramatic increase in EGFR protein

Fig. 2. Chromosomal topography of epidermal growth factor receptor (EGFR) ligands. (A) Overall genomic organization of human EGFR ligands. (B) Human and mouse chromosomal localization of EGFR ligands. See introductory section for definitions of abbreviations.

EGF Receptor Ligands

5

Table 1 EGF ligand cytoplasmic tailsa Human EGFR ligand

Cytoplasmic domain sequences

Length

EGF

AHYYRTQKLLSKNPKNPYEESSRDVRSRRPA DTEDGMSSCPQPWFVVIKEHQDLKNGGQPVA GEDGQAADGSMQPTSWRQEPQLCGMGTEQGC WIPVSSDKGSCPQVMERSFHMPSYGTQTLEGGV EKPHSLLSANPLWQQRALDPPHQMELTQ HCCQVRKHCEWCRALICRHEKPSALLKGRTACCHSETVV QLRRQYVRKYEGEAEERKKLRQENGNVHAIA RYHRRGGYDVENEEKVKLGMTNSH TCCHPLRKRRKRKKKEEEMETLGKDITPINEDIEETNIA CRWYRNRKSKEPKKEYERVTSGDPELPQV RCINLKSPYIICSGGSPL

154

TGFa AR HB-EGF BTC EPR Epigen

39 31 24 39 29 18

a

EGF, epidermal growth factor; EGFR, EGF receptor; TGFa, transforming growth factor-a; AR, amphiregulin; HB-EGF, heparin-binding EGF-like growth factor; BTC, betacellulin; EPR, epiregulin.

by western blot analysis of lysates from his gastric mucosa, a finding consistent with overcoming the receptor downregulation that one observes with active ligand receptor engagement. This rapid local capture of TGFa by the EGFR has fundamental biological importance and is a phenomenon that is critical in such diverse biological processes as vulva development in the nematode Caenorhabditis elegans (C. elegans) and hair follicle organization in mammals (Fig. 5B and C). In C. elegans vulva development, Lin-3, a soluble TGFa-like ligand, is released from the anchor cell and is taken up avidly by Let-23, the EGFR homologue, on the closest vulva precursor cell, P6.p; this step orchestrates the orderly development of the vulva [12,13]. Laser ablation of P6.p, as well as genetic mosaic analysis in which worms with reduced LET-23 activity in P6.p have been selected [14], results in diffusion of Lin-3 and its consumption by basolateral Let-23 on neighboring precursor cells (designated P5.p and P7.p). This leads to adoption of a primary cell fate by P5.p and P7.p, and subsequently results in an aberrant multivulva phenotype. Thus, rapid consumption of Lin-3 (TGFa) by basolateral Let-23 (EGFR) is critical in normal vulva development. Rapid consumption of TGFa by EGFR also appears to be important in mammalian hair follicle development (Fig. 5C). If this signaling is disrupted, as in the case of TGFa Table 2 EGF ligand cleavage sitesa Protein

EGF TGFa AR HB-EGF BTC EPR a b

Extracellular cleavage site Proximal

Distal

YSVR / NSDS VAAA / VVSH IVDD / SVRVb RKVR / DLQEb CVVA / DGNSb RVAQ / VSITb

WELR / HAGH [116] DLLA / VVAA [116] CGEK / SMKT [116] GLSL / PVEN [24] DLFY / LRGD [94] HFFL / TVHQ [3]

Abbreviations as for Table 1. Most N-terminal proximal site / cleavage site.

and EGFR null mice, the hair follicle is disorganized, resulting in an overt phenotype of wavy whiskers and lack of proper coat development [15–20]. Studies of EGFR null chimeric mice, in which the outer root sheath does not express EGFR but adjacent dermal cells do, have demonstrated that when TGFa is not rapidly consumed by EGFRs in the outer root sheath, the hair follicle becomes disorganized (David Threadgill, personal communication, 2002). In addition, these studies have suggested that when TGFa avoids local capture by EGFRs in the outer root sheath, it can diffuse and potentially act as a chemotactic factor for adjacent dermal cells. This may account for the acneiform eruptions that are observed with monoclonal antibody blockade of the EGFR where inflammatory cells are found at the base of the hair follicle. It appears that a rate-limiting step in the action of TGFa may be its delivery to the cell surface. Basolateral sorting information resides in TGFa’s 39-amino acid (39-aa) cytoplasmic tail, which includes a di-leucine basolateral sorting motif and a C-terminal PDZ target TVV. The cytoplasmic tails of the human EGFR ligands are shown in Table 1. Two PDZ proteins, Syntenin and GRASP 55, have been reported to bind TGFa in the ER and cis-Golgi apparatus, respectively [21,22]. To identify proteins that interact with TGFa’s cytoplasmic tail and facilitate basolateral sorting of TGFa, we used the full-length tail as bait to screen yeast two-hybrid libraries generated from a mouse embryo library and a polarizing human colorectal cancer cell line, HCA-7, that forms a uniform polarized monolayer when grown on Transwell filters. We have identified two proteins that bind TGFa’s cytoplasmic tail; membrane associated guanylate kinase inverted-3 (MAGI-3) binds the PDZ target TVV and Naked2 (NKD2) binds the di-leucine repeat. Deletion of the TVV motif impairs efficiency of cell surface delivery, but the fidelity of basolateral targeting is maintained. NKD2 coats a subset of basolaterally targeted vesicles that contain TGFa as cargo and is myristoylated, a process that facilitates docking of these TGFa-containing vesicles to the basolateral surface.

6


Fig. 3. Protein structure of epidermal growth factor receptor (EGFR) ligands. (A) Signal peptide, pro region, mature EGF domain, juxtamembrane, transmembrane, and cytoplasmic tail. (B) Amino acid residues that make up these domains in the individual EGFR ligands are listed. EGF consists of 9 EGF-like repeats. Asterisks indicate proximal and distal sites of cleavage. Glycosylation sites are shown on the right. See introductory section for definitions of abbreviations.

HB-EGF as both a regulated autocrine/paracrine and a juxtacrine growth factor HB-EGF is expressed in a wide variety of hematopoietic cells, endothelial cells, vascular smooth muscle cells, and epithelial cells [23–29]. HB-EGF has been implicated in wound healing, blastocyst implantation, SMC hyperplasia, atherosclerosis, and tumor growth [25,30–32]. The precursor for HB-EGF (proHB-EGF) is a 206-aa transmembrane protein that undergoes regulated metalloproteinase-dependent processing to an 86-aa secreted protein [23,24,33]. In

Fig. 4. Comparison of transforming growth factor-a (TGFa) and heparinbinding epidermal growth factor (EGF)-like growth factor (HB-EGF).

contradistinction to TGFa, the majority of protein delivered to the plasma membrane remains as proHB-EGF; in fact, proHB-EGF serves as the “diphtheria toxin receptor,” with diphtheria toxin entering cells via binding to and endocytosis with proHB-EGF [34]. Although mouse and rat cells express HB-EGF, they are resistant to diphtheria toxin because the toxin does not bind to the extracellular domain of the murine proHB-EGF [35,36]. The identity of the metalloproteinase(s) involved in cleaving HB-EGF is an area of current interest. Although there have been studies suggesting a role for matrix metalloproteinases (MMPs) [37,38], recent evidence suggests another class of metalloproteinases, the ADAMs (a disintegrin and metalloproteinase), which are integral membrane proteins with extracellular metalloproteinase and integrinbinding sites. This expanding family of proteins includes at least four candidates for the HB-EGF “sheddase”: ADAM9, ADAM10, ADAM12, and ADAM17 (TACE). ADAM9 was originally proposed to be the enzyme involved in HB-EGF cleavage, based on studies using dominant negative constructs [39], but recent studies of an ADAM9 knockout indicate unimpaired HB-EGF shedding, suggesting that other enzymes may be involved [40]. TACE/ADAM17 was originally identified as the sheddase involved in tumor


Fig. 5. Transforming growth factor-a (TGFa) is a locally captured growth factor. (A) TGFa is delivered to the basolateral surface of polarized epithelial cells, cleaved by TACE/ADAM17 and taken up avidly by epidermal growth factor receptor (EGFR). (B) In C. elegans, the anchor cell is dorsal to six vulva precursor cells, which are designated P3.p through P8.p. Each of the vulva precursor cells express EGFR (Let-23) on their basolateral surface. The anchor cell produces a soluble TGFa-like ligand (Lin-3) that is avidly taken up by P6.p, which is the vulva precursor cell in closest proximity to the anchor cell. Lin-3 engagement of Let-23 in P6.p activates a Ras-MAPK (Let-60-Mpk-1/Sur-1) signaling cascade, which induces P6.p to adopt a primary cell fate, and subsequently results in more laterally positioned precursor cells adopting secondary and tertiary fates. (C) The hair follicle is composed of two juxtaposed epithelial cell monolayers, the inner and outer root sheaths. The inner root sheath produces TGFa and does not express the EGFR, whereas the outer root sheath expresses EGFR, but does not produce TGFa.

necrosis factor (TNF) processing, but it has subsequently been identified as essential for TGFa processing (see above); its role in HB-EGF (and amphiregulin) processing are less well defined, with some studies suggesting a role [41]. Recently, ADAM12 has been implicated in HB-EGF shedding in heart and possibly in other tissues, given the wide distribution of ADAM12 [42]. Furthermore, ADAM10 (Kuzbanian), which has previously been implicated in Delta/Notch signaling, has been recently shown to mediate HB-EGF shedding [43]. Therefore, these results indicate that multiple ADAMs possess the capability of processing proHB-EGF, and the specific metalloproteinase involved may be dependent upon cellular/tissue specificity in ADAM distribution. The mechanism(s) by which the ADAMs are activated is still under investigation. HB-EGF shedding can be induced by protein kinase C (PKC) activation. Increasing intracellular calcium also leads to HB-EGF shedding; for both of these stimuli, however, activation of an ADAM has not been demonstrated, although there is a suggestion that ADAM9 is phosphorylated and activated by PKCd [39]. It is also of interest that ADAM12 has an SH3-binding site in its cytoplasmic tail and src kinases associate with it [44]. In addition to receptor activation by soluble HB-EGF, it has been suggested that the membrane-associated, uncleaved form of HB-EGF can activate EGFR in adjacent cells and stimulate proliferation, so-called juxtacrine stimu-

7

lation [45–48] (Fig. 6). Membrane associated HB-EGF forms a complex with CD9, a member of the tetraspanin membrane protein superfamily. Although CD9 interacts with HB-EGF’s extracellular heparin-binding domain [49], it does not interact with soluble HB-EGF but only associates with membrane-associated proHB-EGF. Tetraspanins are accessory molecules that stabilize and/or facilitate interactions of other proteins [50]. CD9 is expressed in hematopoietic cells but is also expressed in mesenchymal and epithelial cells. CD9 expression in carcinoma cell lines inhibits motility and metastasis [51,52], and decreased CD9 expression is correlated with poor prognosis in breast cancer [53]. In addition to CD9, other tetraspanins, including CD63, CD81, and CD82, also associate with HB-EGF, but only CD9 upregulates mitogenic and diphtheria toxin binding activities of HB-EGF [54,55]. The interaction of CD9 with membrane-associated HBEGF has suggested a possible role in juxtacrine interactions. CD9 association significantly increases the juxtacrine activity of formalin-fixed membrane-bound proHB-EGF [46], and coexpression of CD9 in renal epithelial cells increases proHB-EGF’s cytoprotective capacity [48]. CD9 interacts with selected integrins, especially b1 integrins [50], which are important in cell-cell and cell-ECM (extracellular matrix) adhesive properties [56,57]. In Vero cells expressing HB-EGF and CD9, HB-EGF-CD9 complexes localize to the cell-cell contact interface in association with a-catenin and vinculin [10,58], and CD9 interacts specifically with a3b1 integrins at the adherens junctions [10]. In polarized cells, EGFR is also localized to sites of cell-cell contact at the zona adherens in association with E-cadherin and a- and b-catenin [59–61]. b1 integrins appear to be important for epithelial cell-cell interactions [62], including renal epithelial cells, in which a2b1 and a3b1 integrins have been localized to sites of cell-cell contact [63,64].

Fig. 6. Heparin-binding epidermal growth factor (EGF)-like growth factor (HB-EGF) as a juxtacrine growth factor proposed model of HB-EGF interaction with EGF receptor (EGFR) on lateral membranes of polarized epithelial cells. As indicated in the text, there is evidence that membraneassociated proHB-EGF complexes with CD9, integrins, HSPG-expressing CD44, the ADAM family of metalloproteinases, and the anti-apoptotic protein, BAG-1.

8


The mitogenic actions of HB-EGF increase when the ligand associates with heparan sulfate proteoglycans (HSPG), presumably because interaction with HB-EGF’s amino terminal heparin-binding site stabilizes the interaction of the EGF domain with EGFR [28,65,66]. CD44 glycoprotein is a cell surface receptor for several extracellular matrix components, including hyaluronic acid, fibronectin, osteopontin, and collagens. It mediates cell adhesion and migration [67]. There is variable exon splicing of extracellular domains of CD44 that contain different GAG attachment sites [68]; one variant, v3, contains an HSPG attachment site, and CD44v3 has been shown to interact with heparin-binding growth factors, including HB-EGF [68,69]. The cytoplasmic tail of CD44 is known to associate with the actin cytoskeleton through its interactions with ERM (ezrin, radixin, and moesin) proteins, which are scaffolding proteins involved in anchoring actin filaments to the plasma membrane. The extracellular domain of CD44 is cleaved by a PKC and calcium-activated metalloproteinasedependent process [70,71], which releases ERMs and is a crucial step in CD44-mediated tumor cell metastasis. The small GTPase, Rac, mediates CD44 cleavage [71]. Rac is known to induce formation of lamellipodia and membrane ruffling [72], and cleavage of CD44 induces redistribution of CD44 and ERM proteins to areas of membrane ruffling [71]. Ezrin dephosphorylation is also an early event in renal microvillar breakdown in anoxic injury [73]. Phosphorylated ezrin interacts with the p85 subunit of PI-3 kinase (PI3K) and mediates activation of an Akt-mediated survival pathway in LLC-PK1 cells [74]. Yeast two-hybrid screening has identified two other proteins that may interact with the proHB-EGF extracellular domain, latent TGF-b binding protein 3 and fibulin-1, a calcium-binding extracellular matrix glycoprotein, but the physiological significance of these interactions has yet to be determined [75]. Besides protein-protein interactions with its extracellular domain, HB-EGF also interacts with the cochaperone, BAG-1, through its cytoplasmic domain, and this interaction may be involved in HB-EGF’s cytoprotective effects [47,76]. The cytoplasmic tails of the EGFR ligands is indicated in Table 2; identification of proteins that interact with the cytoplasmic tails of the other EGFR ligands may be a fruitful area of future investigation.

GPCR transactivation of EGFR G protein-coupled receptor (GPCR) transactivation of EGFR was thought initially to be an intracellular ligandindependent process due to the rapid onset (within minutes) of EGFR activation and the absence of detectable EGFR ligands in conditioned medium. However, Prenzel et al. [77] have provided convincing evidence for a metalloproteasedependent cleavage of HB-EGF upon GPCR activation, with the soluble HB-EGF then activating EGFR [77]. They

have termed this process triple membrane passing signal event (TMPS). In rat 1 cells transfected with chimeras consisting of EGFR ectodomain and platelet-derived growth factor (PDGF) receptor transmembrane and cytoplasmic domain, they demonstrated that GPCR agonists could activate the chimeras via a metalloprotease-mediated cleavage of HB-EGF. Furthermore, they demonstrated that bombesin and 12-0-tetradecamoylphorbol-13-acetate (TPA)-mediated tyrosine phosphorylation of endogenous EGFR in PC-3 cells could be blocked by batimastat, a broad-spectrum metalloprotease inhibitor that presumably prevents EGFR ligand release. The identity of the proteolytic activity was not identified. ADAM9 did not appear to mediate HB-EGF cleavage, as a dominant negative ADAM9 mutant transfected into COS7 cells and HEK293 cells did not block EGFR transactivation. These initial reports identified HB-EGF as the ligand mediating GPCR transactivation of EGFR. These studies took advantage of the previously mentioned observation that cell surface HB-EGF can act as the receptor for diphtheria toxin and utilized CRM197, a nontoxic analog of diphtheria toxin that specifically blocks the action of cell surface and soluble HB-EGF. Subsequent studies have confirmed involvement of metalloproteinase cleavage of proHB-EGF in EGFR transactivation by other GPCRs, as well as by other intracellular signals [77–81]. It is likely that other EGFR ligands in addition to HBEGF may also mediate GPCR transactivation of EGFR. McCole et al. [82] have recently shown that carbachol can transactivate the EGF receptor via release of TGFa in T84 cells, a human colorectal cancer cell line. TGFa was detected within minutes in the conditioned media of carbachol-treated cells and its release correlated with EGF receptor tyrosine phosphorylation. This work further underscores that TGFa is a rapidly released and rapidly consumed growth factor.

Other EGFR ligands Epidermal growth factor (EGF) is a 53-aa polypeptide, with a molecular weight of 6045, that is derived by proteolytic processing from the transmembrane precursor (prepro-EGF) of 1207 aa in humans or 1217 aa in rodents [83,84]. EGF was originally detected and isolated by Stanley Cohen because of its abilities to stimulate precocious tooth eruption and eyelid opening in newborn mice [85]. EGF was originally detected in mouse submaxillary gland, the most abundant source of production in the rodent. Immunohistochemical localization in rodents has shown EGF to be expressed in submaxillary glands, in the apical membrane of the thick ascending limb of the kidney, in exocrine glands of the gastrointestinal tract, and in serous acini of the nasal cavity. The metalloproteinase(s) involved in preproEGF cleavage have not been identified. Both calcium ionophores and


tyrosine phosphatase inhibitors induced cleavage, although PKC activation did not. Furthermore, proHB-EGF and preproEGF exhibit differing sensitivities to hydroxamic acidbased metalloproteinase inhibitors, suggesting that they may be cleaved by different metalloproteinases [86]. As mentioned above, in studies in polarized MDCK cells, transfected preproEGF was found predominantly at the apical membrane; delivery to both membranes was equivalent, but there was preferential constitutive metalloproteinasedependent cleavage of the basolateral protein [8]. Amphiregulin (AR) was originally identified as an EGFR ligand from a phorbol ester-treated human breast adenocarcinoma cell line, MCF-7 [2]. Although the growth inhibitory properties of AR have not been subsequently validated, Shoyab and coworkers [2] named it “amphi” because of its growth inhibitory effects for A431 and several other cancer cell lines and growth stimulation of many other cell lines, normal and transformed. Mature AR is initially synthesized as a 252-aa transmembrane precursor that is proteolytically cleaved to give the mature peptide of 78–84 aa peptide. Posttranslational modification results in several different cell surface and soluble isoforms [87]. Although some of these isoforms of AR have been shown to possess biological activity, it is not known whether all of these soluble AR forms are biological active, and, if so, whether they may differ in their biological activities. We have shown that in polarized epithelial cells, AR is preferentially delivered to the basolateral surface of both MDCK II and HCA-7 cells, where it is cleaved by TACE/ADAM17 [87]. The basolateral sorting information resides in the cytoplasmic tail as AR cDNA constructs that lack the tail are missorted in polarized MDCK II cells. However, cell surface processing of AR appears unaffected, suggesting that information in the tail is not required for cell surface proteolytic cleavage [5]. Similar with HB-EGF, AR is thought to associated with extracellular and cell-associated HSPGs as well as the tetraspanin, CD9 [88,89]. Interaction with these molecules may facilitate a depot form of AR and promote productive interaction between AR and cell surface EGFRs. In this regard, under certain conditions, the biological properties of AR may be distinct from that of either TGFa or EGF [90–93]. Betacellulin (BTC) was originally identified in conditioned media from a pancreatic b cell tumor line [94,95] but has since been shown to be expressed in a variety of mesenchymal and epithelial cell lines and in many tissues in the adult, with highest expression in pancreas, liver, kidney, and small intestine [94–97], and lesser expression in heart, lung, colon, testis, and ovary [97]. BTC has also been detected in a variety of tumor-derived cell lines and tumors in situ, with especially high expression in pancreatic cancer. There is one study suggesting that pro-BTC may signal in a juxtacrine manner, but this observation has not yet been confirmed in other systems [98]. As with the other members of this family, membrane-associated BTC is processed in a metalloproteinase-dependent process to release the soluble form of the peptide. There is evidence for constitutive

9

processing, but unlike HB-EGF, AR, and TGFa, there is no evidence for a phorbol ester-activated process, but addition of a calcium ionophore did lead to moderate increase in BTC processing [98]. In this regard, BTC behaves similarly to EGF [86]. Epiregulin (EPR) was purified from conditioned media of the mouse fibroblast-derived tumor cell line NIH 3T3/clone T7 as a 46-aa peptide that was cleaved from a 162-aa transmembane precursor [3]. EPR is expressed predominantly in the placenta and peripheral blood leukocytes of normal adults as well as carcinomas of bladder, lung, kidney, pancreas, and colon [99,100]. The structural organization is similar to that of TGFa. However, EPR is in close proximity to BTC and AR in human chromosome 4 and mouse chromosome 5. Mature EPR exhibits a stronger stimulation of DNA synthesis than EGF in rat primary hepatocytes in spite of its weaker binding affinity to EGFR [3,101], and EPR is an autocrine growth factor in normal human keratinocytes [102] and a potent pan-erbB ligand that preferentially activates heterodimeric receptor complexes [103]. To date, sorting and processing of EPR has not been studied. Although EGFR gene deletion produces significant abnormalities in brain, skin, kidney, and other organs [19], there is still uncertainty about the contribution of individual ligands to these processes. Gene deletions of EGF, TGFa, and AR have produced relatively minor phenotypes [90]. AR has recently been reported to be a downstream gene of WT1 in metanephric development [104], but no renal developmental abnormalities have been reported in AR knockouts, and triple knockouts of EGF, TGFa, and AR were viable and healthy, although there were abnormalities in mammary gland development [90]. Auto- and cross-induction of EGFR ligands is known to occur. This was shown initially for autoinduction of TGFa in human keratinocytes, but now has been shown for most of the EGFR ligands [27,66].

EGFR ligands and Ras transformation Ras signaling is complex and involves multiple, diverse effector pathways [105]. Although activation of the EGFR can initiate Ras signaling and thus is considered upstream of Ras, there is accumulating evidence that the mutant Ras phenotype is modulated by the ability of mutant Ras to upregulate EGFR signaling. Sibilia and colleagues [106] have shown this genetically in mice that develop skin papillomas as a result of engineered overexpression in the skin of dominant active SOS (that in turn activates Ras). The number of papillomas was significantly reduced when these mice were crossed to wa-2 mice, which have a spontaneous mutation in the EGFR tyrosine kinase domain and exhibit up to a 90% reduction in EGFR tyrosine kinase activity. One way that mutant Ras may enhance EGFR signaling is by upregulating expression of EGFR ligands that then activate EGFR signaling. In this model, it is thought that EGFR would further enhance Ras signaling, as well as activate

10


other signaling pathways. We have shown that activated Hand K-Ras, but not Raf, causes transformation of a rat intestinal cell line, RIE-1 [107]. Conditioned medium from these Ras-transformed cells is able to cause morphological transformation of parental RIE-1 cells and this effect can be abrogated by EGFR blockade. Moreover, expression of AR, TGFa, and HB-EGF is upregulated in Ras-transformed RIE1 cells. Administration of a farnesyltransferase inhibitor FTI L744,832 results in a marked downregulation of AR and TGFa that precedes FTI-mediated growth reduction and can be overcome with exogenous TGFa [108]. In addition, Chin and co-workers [109] have developed a mouse model in which they can turn on and off melanoma formation by doxycyline-regulated expression of mutant H-Ras in melanocytes. By microarray analysis, they have observed that AR and EPR are rapidly and significantly downregulated in the regressing melanomas when mutant Ras expression is silenced.

EGFR ligands and activation of other receptors In addition to HER1, HB-EGF, BTC, and EPR bind to and activate HER4 [66,110,111], as does another class of ligands, the neuregulins, which bind both HER3 and HER4. BTC is the only EGF family member that binds HER2/HER3 heterodimers, although it does not bind HER3/HER3 homodimers. In NIH 3T3 cells overexpressing HER4, HB-EGF activation of HER4 does not stimulate mitosis but does induce chemotaxis, an effect apparently mediated by PI-3K [111]. In addition, HB-EGF’s regulation of blastocyst implantation may be mediated by HER4 [112], and uterine HB-EGF interaction with HER4 is dependent upon CD44 [38]. Some differences in intracellular signaling activation between EGFR and HER4 have been described; specifically, activation of EGFR but not HER4 leads to association of Cbl, which may be a mechanism by which PI-3K can be associated with and activated by the EGFR [113]. In contrast, HER4 activation of PI-3K is mediated by direct binding at a consensus PI-3K binding site (YTPM) in the cytoplasmic tail of the receptor [113,114]; the EGFR does not contain such a binding motif. Alternative splice variants of human and mouse HER4 have been identified with or without a 48-bp exon encoding the region containing this PI3K binding site. Neuregulin 1 did not activate PI-3K in the HER4 splice form without the PI-3K binding sequence, and this splice variant mediated neuregulin 1-dependent proliferation but not chemotaxis or cytoprotection [113,115]; similar studies have not yet been performed for HB-EGF. A third HB-EGF “receptor” has recently been identified; N-arginine dibasic convertase (NRDC) functions as a metalloproteinase but does not appear to be involved in proHBEGF cleavage. It has been suggested that NABC may be involved in modulating ECM and mediating HB-EGF effects on cell motility; however, its metalloproteinase

activity was not required. HB-EGF interacts with NRDC through its extracellular heparin-binding domain, and this interaction appears to be relatively specific for HB-EGF, since other EGFR ligands do not bind or compete with HBEGF binding, with the exception of AR, which will partially compete by interaction with NRDC through its heparin. Acknowledgments This work was supported by NIH grants CA46413 (R.J.C.) and DK51265 (R.C.H.) and Mouse Models of Human Cancers Consortium and GI Special Program of Research Excellence to R.J.C. References [1] J. Massague, Transforming growth factors-a, J. Biol. Chem. 265 (1990) 21393–21396. [2] M. Shoyab, V.L. McDonald, J.G. Bradley, G.J. Todaro, Amphiregulin: a bifunctional growth-modulating glycoprotein produced by the phorbol 12-myristate 13-acetate-treated human breast adenocarcinoma cell line MCF-7, Proc. Natl. Acad. Sci. USA 85 (1988) 6528–6532. [3] H. Toyoda, T. Komurasaki, D. Uchida, Y. Takayama, T. Isobe, T. Okuyama, K. Hanada, Epiregulin. A novel epidermal growth factor with mitogenic activity for rat primary hepatocytes, J. Biol. Chem. 270 (1995) 7495–7500. [4] L. Strachan, J.G. Murison, R.L. Prestidge, M.A. Sleeman, J.D. Watson, K.D. Kumble, Cloning and biological activity of epigen, a novel member of the epidermal growth factor superfamily, J. Biol. Chem. 276 (2001) 18265–18271. [5] C.L. Brown, R.J. Coffey, P.J. Dempsey, The proamphiregulin cytoplasmic domain is required for basolateral sorting, but is not essential for constitutive or stimulus-induced processing in polarized Madin-Darby canine kidney cells, J. Biol. Chem. 276 (2001) 29538–29549. [6] S.K. Kuwada, X.F. Li, L. Damstrup, P.J. Dempsey, R.J. Coffey, H.S. Wiley, The dynamic expression of the epidermal growth factor receptor and epidermal growth factor ligand family in a differentiating intestinal epithelial cell line, Growth Factors 17 (1999) 139–153. [7] L. Damstrup, S.K. Kuwada, P.J. Dempsey, C.L. Brown, C.J. Hawkey, H.S. Poulsen, H.S. Wiley, R.J. Coffey Jr., Amphiregulin acts as an autocrine growth factor in two human polarizing colon cancer lines that exhibit domain selective EGF receptor mitogenesis, Br. J. Cancer 80 (1999) 1012–1019. [8] P.J. Dempsey, K.S. Meise, Y. Yoshitake, K. Nishikawa, R.J. Coffey, Apical enrichment of human EGF precursor in Madin-Darby canine kidney cells involves preferential basolateral ectodomain cleavage sensitive to a metalloprotease inhibitor, J. Cell Biol. 138 (1997) 747–758. [9] P.J. Dempsey, R.J. Coffey, Basolateral targeting and efficient consumption of transforming growth factor-alpha when expressed in Madin-Darby canine kidney cells, J. Biol. Chem. 269 (1994) 16878–16889. [10] K. Nakamura, R. Iwamoto, E. Mekada, Membrane-anchored heparin-binding EGF-like growth factor (HB-EGF) and diphtheria toxin receptor-associated protein (DRAP27)/CD9 form a complex with integrin alpha 3 beta 1 at cell-cell contact sites, J. Cell Biol. 129 (1995) 1691–1705. [11] J.S. Burdick, E. Chung, G. Tanner, M. Sun, J.E. Paciga, J.Q. Cheng, K. Washington, J.R. Goldenring, R.J. Coffey, Treatment of


[12] [13] [14]

[15]

[16]

[17]

[18] [19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28] [29]

[30]

Menetrier’s disease with a monoclonal antibody against the epidermal growth factor receptor, N. Engl. J. Med. 343 (2000) 1697–1701. P.W. Sternberg, H.R. Horvitz, Pattern formation during vulval development in C. elegans, Cell 44 (1986) 761–772. K. Kornfeld, Vulval development in Caenorhabditis elegans, Trends Genet. 13 (1997) 55–61. A. Hajnal, C.W. Whitfield, S.K. Kim, Inhibition of Caenorhabditis elegans vulval induction by gap-1 and by let-23 receptor tyrosine kinase, Genes Dev. 11 (1997) 2715–2728. N.C. Luetteke, T.H. Qiu, R.L. Peiffer, P. Oliver, O. Smithies, D.C. Lee, TGF alpha deficiency results in hair follicle and eye abnormalities in targeted and waved-1 mice, Cell 73 (1993) 263–278. G.B. Mann, K.J. Fowler, A. Gabriel, E.C. Nice, R.L. Williams, A.R. Dunn, Mice with a null mutation of the TGF alpha gene have abnormal skin architecture, wavy hair, and curly whiskers and often develop corneal inflammation, Cell 73 (1993) 249–261. P.J. Miettinen, J.E. Berger, J. Meneses, Y. Phung, R.A. Pedersen, Z. Werb, R. Derynck, Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor, Nature 376 (1995) 337–341. M.F. Sibilia, E. Wagner, Strain-dependent epithelial defects in mice lacking the EGF receptor, Science 269 (1995) 234–238. D.W. Threadgill, A.A. Dlugosz, L.A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. LaMantia, T. Mourton, K.C. Herrup, R. Harris, J.A. Barnard, S.H. Yuspa, R.J. Coffey, T. Magnuson, Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype, Science 269 (1995) 230–234. E.A. Berkowitz, K.B. Seroogy, J.A. Schroeder, W.E. Russell, E.P. Evans, R.F. Riedel, H.K. Phillips, C.A. Harrison, D.C. Lee, N.C. Luetteke, Characterization of the mouse transforming growth factor alpha gene: its expression during eyelid development and in waved 1 tissues, Cell Growth Differ. 7 (1996) 1271–1282. J. Fernandez-Larrea, A.M. Merlos-Suarez, J. Urena, J. Baselga, J. Arribas, A role for a PDZ protein in the early secretory pathway for the targeting of proTGF-alpha to the cell surface, Mol. Cell. 3 (1999) 423–433. F.A. Barr, C. Preisinger, R. Kopajtich, R. Korner, Golgi matrix proteins interact with p24 cargo receptors and aid their efficient retention in the Golgi apparatus, J. Cell Biol. 155 (2001) 885–891. S. Higashiyama, J.A. Abraham, J. Miller, J.C. Fiddes, M. Klagsbrun, A heparin-binding growth factor secreted by macrophage-like cells that is related to EGF, Science 251 (1991) 936–993. S. Higashiyama, K.E. Lau, G.A. Besner, J. Abraham, M. Klagsbrun, Structure of heparin-binding EGF-like growth factor. Multiple forms, primary structure, and glycosylation of the mature protein, J. Biol. Chem. 267 (1992) 6205–6212. G.E. Peoples, S. Blotnick, K. Takahashi, M.R. Freeman, M. Klagsbrun, T.J. Eberlein, T lymphocytes that infiltrate tumors and atherosclerotic plaques produce heparin-binding epidermal growth factorlike growth factor and basic fibroblast growth factor: a potential pathologic role, Proc. Natl. Acad. Sci. USA 92 (1995) 6547–6551. P.P. Powell, M. Klagsbrun, J.A. Abraham, R.C. Jones, Eosinophils expressing heparin-binding EGF-like growth factor mRNA localize around lung microvessels in pulmonary hypertension, Am. J. Pathol. 143 (1993) 784–793. J.A. Barnard, R. Graves-Deal, M.R. Pittelkow, R. DuBois, P. Cook, G.W. Ramsey, P.R. Bishop, L. Damstrup, R.J. Coffey, Auto- and cross-induction within the mammalian epidermal growth factorrelated peptide family, J. Biol. Chem. 269 (1994) 22817–22822. G. Raab, M. Klagsbrun, Heparin-binding EGF-like growth factor, Biochim. Biophys. Acta 1333 (1997) F179–F199. T. Homma, M. Sakai, H.F. Cheng, T. Yasuda, R.J. Coffey Jr., R.C. Harris, Induction of heparin-binding epidermal growth factor-like growth factor mRNA in rat kidney after acute injury, J. Clin. Invest. 96 (1995) 1018–1025. M. Marikovsky, K. Breuing, P.Y. Liu, E. Eriksson, S. Higashiyama, P. Farber, J. Abraham, M. Klagsbrun, Appearance of heparin-binding

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

11

EGF-like growth factor in wound fluid as a response to injury, Proc. Nat. Acad. Sci. USA 90 (1993) 3889–3893. G. Raab, K. Kover, B.C. Paria, S.K. Dey, R.M. Ezzell, M. Klagsbrun, Mouse preimplantation blastocysts adhere to cells expressing the transmembrane form of heparin-binding EGF-like growth factor, Development 122 (1996) 637–645. J. Miyagawa, S. Higashiyama, S. Kawata, Y. Inui, S. Tamura, K. Yamamoto, M. Nishida, T. Nakamura, S. Yamashita, Y. Matsuzawa, et al., Localization of heparin-binding EGF-like growth factor in the smooth muscle cells and macrophages of human atherosclerotic plaques, J. Clin. Invest. 95 (1995) 404–411. G. Raab, S. Higashiyama, S. Hetelekidis, J.A. Abraham, D. Damm, M. Ono, M. Klagsbrun, Biosynthesis and processing by phorbol ester of the cells surface-associated precursor form of heparin-binding EGF-like growth factor, Biochem. Biophys. Res. Commun. 204 (1994) 592–597. J.G. Naglich, J.E. Metherall, D.W. Russell, L. Eidels, Expression cloning of a diphtheria toxin receptor: identity with a heparinbinding EGF-like growth factor precursor, Cell 69 (1992) 1051– 1061. R. Iwamoto, S. Higashiyama, T. Mitamura, N. Taniguchi, M. Klagsbrun, E. Mekada, Heparin-binding EGF-like growth factor, which acts as the diphtheria toxin receptor, forms a complex with membrane protein DRAP27/CD9, which up-regulates functional receptors and diphtheria toxin sensitivity, EMBO J. 13 (1994) 2322–2330. T. Mitamura, S. Higashiyama, N. Taniguchi, M. Klagsbrun, E. Mekada, Diphtheria toxin binds to the epidermal growth factor (EGF)-like domain of human heparin-binding EGF-like growth factor/diphtheria toxin receptor and inhibits specifically its mitogenic activity, J. Biol. Chem. 270 (1995) 1015–1019. M. Suzuki, G. Raab, M.A. Moses, C.A. Fernandez, M. Klagsbrun, Matrix metalloproteinase-3 releases active heparin-binding EGF-like growth factor by cleavage at a specific juxtamembrane site, J. Biol. Chem. 272 (1997) 31730–31737. W.H. Yu, J.F. Woessner Jr., J.D. McNeish, I. Stamenkovic, CD44 anchors the assembly of matrilysin/MMP-7 with heparin-binding epidermal growth factor precursor and ErbB4 and regulates female reproductive organ remodeling, Genes Dev. 16 (2002) 307–323. Y. Izumi, M. Hirata, H. Hasuwa, R. Iwamoto, T. Umata, K. Miyado, Y. Tamai, T. Kurisaki, A. Sehara-Fujisawa, S. Ohno, E. Mekada, A metalloprotease-disintegrin, MDC9/meltrin-gamma/ADAM9 and PKCdelta are involved in TPA-induced ectodomain shedding of membrane-anchored heparin-binding EGF-like growth factor, EMBO J. 17 (1998) 7260–7272. G. Weskamp, H. Cai, T.A. Brodie, S. Higashyama, K. Manova, T. Ludwig, C.P. Blobel, Mice lacking the metalloprotease-disintegrin MDC9 (ADAM9) have no evident major abnormalities during development or adult life, Mol. Cell. Biol. 22 (2002) 1537–1544. S.W. Sunnarborg, C.L. Hinkle, M. Stevenson, W.E. Russell, C.S. Raska, J.J. Peschon, B.J. Castner, M.J. Gerhart, R.J. Paxton, R.A. Black, D.C. Lee, Tumor necrosis factor-alpha converting enzyme (TACE) regulates epidermal growth factor receptor ligand availability, J. Biol. Chem. 277 (2002) 12838–12845. M. Asakura, M. Kitakaze, S. Takashima, Y. Liao, F. Ishikura, T. Yoshinaka, H. Ohmoto, K. Node, K. Yoshino, H. Ishiguro, H. Asanuma, S. Sanadam, Y. Matsumura, H. Takeda, S. Beppu, M. Tada, M. Hori, S. Higashiyama, Cardiac hypertrophy is inhibited by antagonism of ADAM12 processing of HB-EGF: metalloproteinase inhibitors as a new therapy, Nat. Med. 8 (2002) 35–40. Y. Yan, K. Shirakabe, Z. Werb, The metalloprotease Kuzbanian (ADAM10) mediates the transactivation of EGF receptor by G protein-coupled receptors, J. Cell Biol. 158 (2002) 221–226. A. Suzuki, N. Kadota, T. Hara, Y. Nakagami, T. Izumi, T. Takenawa, H. Sabe, T. Endo, Meltrin alpha cytoplasmic domain interacts with SH3 domains of Src and Grb2 and is phosphorylated by v-Src, Oncogene 19 (2000) 5842–5850.

12


[45] M. Ono, G. Raab, K. Lau, J.A. Abraham, M. Klagsbrun, Purification and characterization of transmembrane forms of heparinbinding EGF-like growth factor, J. Biol. Chem. 269 (1994) 31315– 31321. [46] S. Higashiyama, R. Iwamoto, K. Goishi, G. Raab, N. Taniguchi, M. Klagsbrun, E. Mekada, The membrane protein CD9/DRAP 27 potentiates the juxtacrine growth factor activity of the membraneanchored heparin-binding EGF-like growth factor, J. Cell Biol. 128 (1995) 929–938. [47] T. Takemura, S. Kondo, T. Homma, M. Sakai, R.C. Harris, The membrane-bound form of heparin-binding epidermal growth factorlike growth factor promotes survival of cultured renal epithelial cells, J. Biol. Chem. 272 (1997) 31036–31042. [48] T. Takemura, S. Hino, Y. Murata, H. Yanagida, M. Okada, K. Yoshioka, R.C. Harris, Coexpression of CD9 augments the ability of membrane-bound heparin-binding epidermal growth factor-like growth factor (proHB-EGF) to preserve renal epithelial cell viability, Kidney Int. 55 (1999) 71–81. [49] T. Sakuma, S. Higashiyama, S. Hosoe, S. Hayashi, N. Taniguchi, CD9 antigen interacts with heparin-binding EGF-like growth factor through its heparin-binding domain, J. Biochem. (Tokyo) 122 (1997) 474–480. [50] H.T. Maecker, S.C. Todd, S. Levy, The tetraspanin superfamily: molecular facilitators, FASEB J. 11 (1997) 428–442. [51] S. Ikeyama, M. Koyama, M. Yamaoko, R. Sasada, M. Miyake, Suppression of cell motility and metastasis by transfection with human motility-related protein (MRP-1/CD9) DNA, J. Exp. Med. 177 (1993) 1231–1237. [52] J.F. Cajot, I. Sordat, T. Silvestre, B. Sordat, Differential display cloning identifies motility-related protein (MRP1/CD9) as highly expressed in primary compared to metastatic human colon carcinoma cells, Cancer Res. 57 (1997) 2593–2597. [53] M. Miyake, K. Nakano, S.I. Itoi, T. Koh, T. Taki, Motility-related protein-1 (MRP-1/CD9) reduction as a factor of poor prognosis in breast cancer, Cancer Res. 56 (1996) 1244–1249. [54] C. Lagaudriere-Gesbert, F. Le Naour, S. Lebel-Binay, M. Billard, E. Lemichez, P. Boquet, C. Boucheix, H. Conjeaud, E. Rubinstein, Functional analysis of four tetraspans, CD9, CD53, CD81, and CD82, suggests a common role in costimulation, cell adhesion, and migration: only CD9 upregulates HB-EGF activity, Cell Immunol. 182 (1997) 105–112. [55] K. Nakamura, T. Mitamura, T. Takahashi, T. Kobayashi, E. Mekada, Importance of the major extracellular domain of CD9 and the epidermal growth factor (EGF)-like domain of heparin-binding EGF-like growth factor for up-regulation of binding and activity, J. Biol. Chem. 275 (2000) 18284–18290. [56] B.E. Symington, Y. Takada, W.G. Carter, Interaction of integrins alpha 3 beta 1 and alpha 2 beta 1: potential role in keratinocyte intercellular adhesion, J. Cell Biol. 120 (1993) 523–535. [57] B.E. Symington, W.G. Carter, Modulation of epidermal differentiation by epiligrin and integrin alpha 3 beta 1, J. Cell Sci. 108 (1995) 831–838. [58] K. Goishi, S. Higashiyama, M. Klagsbrun, N. Nakano, T. Umata, M. Ishikawa, E. Mekada, N. Taniguchi, Phorbol ester induces the rapid processing of cell surface heparin-binding EGF-like growth factor: conversion from juxtacrine to paracrine growth factor activity, Mol. Biol. Cell 6 (1995) 967–980. [59] R. Fukuyama, N. Shimizu, Detection of epidermal growth factor receptors and E-cadherins in the basolateral membrane of A431 cells by laser scanning fluorescence microscopy, Jpn. Cancer Res. 82 (1991) 8–11. [60] D.F. Woods, P.J. Bryant, Apical junctions and cell signalling in epithelia, J. Cell Sci. Suppl. 17 (1993) 171–181. [61] H. Hoschuetzky, H. Aberle, R. Kemler, Beta-catenin mediates the interaction of the cadherin-catenin complex with epidermal growth factor receptor, J. Cell Biol. 127 (1994) 1375–1380.

[62] H. Larjava, J. Peltonen, S.K. Akiyama, S.S. Yamada, H.R. Gralnick, J. Uitto, K.M. Yamada, Novel function for beta 1 integrins in keratinocyte cell-cell interactions, J. Cell Biol. 110 (1990) 803–815. [63] E.E. Simon, C.H. Liu, M. Das, S. Nigam, T.J. Broekelmann, J.A. McDonald, Characterization of integrins in cultured human renal cortical tubule epithelial cells, Am. J. Physiol. 267 (1994) F612– F623. [64] W.G. Carter, E.A. Wayner, T.S. Bouchard, P. Kaur, The role of integrins alpha 2 beta 1 and alpha 3 beta 1 in cell-cell and cell-substrate adhesion of human epidermal cells, J. Cell Biol. 110 (1990) 1387–1404. [65] S. Higashiyama, J.A. Abraham, M. Klagsbrun, Heparin-binding EGF-like growth factor stimulation of smooth muscle cell migration: dependence on interactions with cell surface heparan sulfate, J. Cell Biol. 122 (1993) 933–940. [66] M. Piepkorn, M.R. Pittelkow, P.W. Cook, Autocrine regulation of keratinocytes: the emerging role of heparin-binding, epidermal growth factor-related growth factors, J. Invest. Dermatol. 111 (1998) 715–721. [67] T. St. John, J. Meyer, R. Idzerda, W.M. Gallatin, Expression of CD44 confers a new adhesive phenotype on transfected cells, Cell 60 (1990) 45–52. [68] K.L. Bennett, D.G. Jackson, J.C. Simon, E. Tanczos, R. Peach, B. Modrell, I. Stamenkovic, G. Plowman, A. Aruffo, CD44 isoforms containing exon V3 are responsible for the presentation of heparinbinding growth factor, J. Cell Biol. 128 (1995) 687–698. [69] R. van der Voort, T.E. Taher, V.J. Wielenga, M. Spaargaren, R. Prevo, L. Smit, G. David, G. Hartmann, E. Gherardi, S.T. Pals, Heparan sulfate-modified CD44 promotes hepatocyte growth factor/scatter factor-induced signal transduction through the receptor tyrosine kinase c-Met, J. Biol. Chem. 274 (1999) 6499–6506. [70] I. Okamoto, Y. Kawano, H. Tsuiki, J. Sasaki, M. Nakao, M. Matsumoto, M. Suga, M. Ando, M. Nakajima, H. Saya, CD44 cleavage induced by a membrane-associated metalloprotease plays a critical role in tumor cell migration, Oncogene 18 (1999) 1435–1446. [71] I. Okamoto, Y. Kawano, M. Matsumoto, M. Suga, K. Kaibuchi, M. Ando, H. Saya, Regulated CD44 cleavage under the control of protein kinase C, calcium influx, and the Rho family of small G proteins, J. Biol. Chem. 274 (1999) 25525–25534. [72] A.J. Ridley, Rho family proteins and regulation of the actin cytoskeleton, Prog Mol. Subcell. Biol. 22 (1999) 1–22. [73] J. Chen, J.A. Cohn, L.J. Mandel, Dephosphorylation of ezrin as an early event in renal microvillar breakdown and anoxic injury, Proc. Natl. Acad. Sci. USA 92 (1995) 7495–7499. [74] A. Gautreau, P. Poullet, D. Louvard, M. Arpin, Ezrin, a plasma membrane-microfilament linker, signals cell survival through the phosphatidylinositol 3-kinase/Akt pathway, Proc. Natl. Acad. Sci. USA 96 (1999) 7300–7305. [75] J.S. Brooke, J.H. Cha, L. Eidels, Latent transforming growth factor beta-binding protein-3 and fibulin-1C interact with the extracellular domain of the heparin-binding EGF-like growth factor precursor, BMC Cell Biol. 3 (2002) 2. [76] J. Lin, L. Hutchinson, S.M. Gaston, G. Raab, M.R. Freeman, BAG-1 is a novel cytoplasmic binding partner of the membrane form of heparin-binding EGF-like growth factor: a unique role for proHBEGF in cell survival regulation, J. Biol. Chem. 276 (2001) 30127–30132. [77] N. Prenzel, E. Zwick, H. Daub, M. Leserer, R. Abraham, C. Wallasch, A. Ullrich, EGF receptor transactivation by G-protein-coupled receptors requires metalloproteinase cleavage of proHB-EGF, Nature 402 (1999) 884–888. [78] S. Eguchi, P.J. Dempsey, G.D. Frank, E.D. Motley, T. Inagami, Activation of MAPKs by angiotensin II in vascular smooth muscle cells. Metalloprotease-dependent EGF receptor activation is required for activation of ERK and p38 MAPK but not for JNK, J. Biol. Chem. 276 (2001) 7957–7962.

EGF Receptor Ligands [79] K.L. Pierce, A. Tohgo, S. Ahn, M.E. Field, L.M. Luttrell, R.J. Lefkowitz, Epidermal growth factor (EGF) receptor-dependent ERK activation by G protein-coupled receptors: a co-culture system for identifying intermediates upstream and downstream of heparinbinding EGF shedding, J. Biol. Chem. 276 (2001) 23155–23160. [80] J.K. Chen, J. Capdevila, R.C. Harris, Heparin-binding EGF-like growth factor mediates the biological effects of P450 arachidonate epoxygenase metabolites in epithelial cells, Proc. Natl. Acad. Sci. USA 99 (2002) 6029–6034. [81] A. Kalmes, G. Daum, A.W. Clowes, EGFR transactivation in the regulation of SMC function, Ann. NY Acad. Sci. 947 (2001) 42–54. [82] D.F. McCole, S.J. Keely, R.J. Coffey, K.E. Barrett, Transactivation of the epidermal growth factor receptor in colonic epithelial cells by carbachol requires extracellular release of transforming growth factor-alpha, J. Biol. Chem. 277 (2002) 42603–42612. [83] G. Carpenter, Epidermal growth factor, Annu. Rev. Biochem. 48 (1979) 193–230. [84] G. Carpenter, Receptors for epidermal growth factor and other polypeptide mitogens, Annu. Rev. Biochem. 56 (1987) 881–934. [85] S. Cohen, Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the new-born animal, J. Biol. Chem. 237 (1962) 1555–1562. [86] J. Dong, H.S. Wiley, Trafficking and proteolytic release of epidermal growth factor receptor ligands are modulated by their membrane-anchoring domains, J. Biol. Chem. 275 (2000) 557–564. [87] C.L. Brown, K.S. Meise, G.D. Plowman, R.J. Coffey, P.J. Dempsey, Cell surface ectodomain cleavage of human amphiregulin precursor is sensitive to a metalloprotease inhibitor. Release of a predominant N-glycosylated 43-kDa soluble form, J. Biol. Chem. 273 (1998) 17258–17268. [88] N. Nylander, L.T. Smith, R.A. Underwood, M. Piepkorn, Topography of amphiregulin expression in cultured human keratinocytes: colocalization with the epidermal growth factor receptor and CD44, In Vitro Cell. Dev. Biol. Anim. 34 (1998) 182–188. [89] S. Inui, S. Higashiyama, K. Hashimoto, M. Higashiyama, K. Yoshikawa, N. Taniguchi, Possible role of coexpression of CD9 with membrane-anchored heparin-binding EGF-like growth factor and amphiregulin in cultured human keratinocyte growth, J. Cell. Physiol. 171 (1997) 291–298. [90] N.C. Luetteke, T.H. Qiu, S.E. Fenton, K.L. Troyer, R.F. Riedel, A. Chang, D.C. Lee, Targeted inactivation of the EGF and amphiregulin genes reveals distinct roles for EGF receptor ligands in mouse mammary gland development, Development 126 (1999) 2739–2750. [91] K.L. Troyer, N.C. Luetteke, M.L. Saxon, T.H. Qiu, C.J. Xian, D.C. Lee, Growth retardation, duodenal lesions, and aberrant ileum architecture in triple null mice lacking EGF, amphiregulin, and TGFalpha, Gastroenterology 121 (2001) 68–78. [92] P.W. Cook, M. Piepkorn, C.H. Clegg, G.D. Plowman, J.M. DeMay, J.R. Brown, M.R. Pittelkow, Transgenic expression of the human amphiregulin gene induces a psoriasis-like phenotype, J. Clin. Invest. 100 (1997) 2286–2294. [93] P.W. Cook, N.M. Ashton, C.E. Karkaria, D.C. Siess, G.D. Shipley, Differential effects of a heparin antagonist (hexadimethrine) or chlorate on amphiregulin, basic fibroblast growth factor, and heparin-binding EGF-like growth factor activity, J. Cell. Physiol. 163 (1995) 418–429. [94] R. Sasada, Y. Ono, Y. Taniyama, Y. Shing, J. Folkman, K. Igarashi, Cloning and expression of cDNA encoding human betacellulin, a new member of the EGF family, Biochem. Biophys. Res. Commun. 190 (1993) 1173–1179. [95] A.J. Dunbar, C. Goddard, Structure-function and biological role of betacellulin, Int. J. Biochem. Cell. Biol. 32 (2000) 805–815. [96] A.J. Dunbar, I.K. Priebe, D.A. Belford, C. Goddard, Identification of betacellulin as a major peptide growth factor in milk: purification, characterization and molecular cloning of bovine betacellulin, Biochem. J. 344 Pt 3 (1999) 713–721.

13

[97] M. Seno, H. Tada, M. Kosaka, R. Sasada, K. Igarashi, Y. Shing, J. Folkman, M. Ueda, H. Yamada, Human betacellulin, a member of the EGF family dominantly expressed in pancreas and small intestine, is fully active in a monomeric form, Growth Factors 13 (1996) 181–191. [98] H. Tada, R. Sasada, Y. Kawaguchi, I. Kojima, W.J. Gullick, D.S. Salomon, K. Igarashi, M. Seno, H. Yamada, Processing and juxtacrine activity of membrane-anchored betacellulin, J. Cell. Biochem. 72 (1999) 423–434. [99] H. Toyoda, T. Komurasaki, D. Uchida, S. Morimoto, Distribution of mRNA for human epiregulin, a differentially expressed member of the epidermal growth factor family, Biochem. J. 326 (1997) 69–75. [100] Z. Zhu, J. Kleeff, H. Friess, L. Wang, A. Zimmermann, Y. Yarden, M.W. Buchler, M. Korc, Epiregulin is Up-regulated in pancreatic cancer and stimulates pancreatic cancer cell growth, Biochem. Biophys. Res. Commun. 273 (2000) 1019–1024. [101] T. Komurasaki, H. Toyoda, D. Uchida, N. Nemoto, Mechanism of growth promoting activity of epiregulin in primary cultures of rat hepatocytes, Growth Factors 20 (2002) 61–69. [102] Y. Shirakata, T. Komurasaki, H. Toyoda, Y. Hanakawa, K. Yamasaki, S. Tokumaru, K. Sayama, K. Hashimoto, Epiregulin, a novel member of the epidermal growth factor family, is an autocrine growth factor in normal human keratinocytes, J. Biol. Chem. 275 (2000) 5748–5753. [103] M. Shelly, R. Pinkas-Kramarski, B.C. Guarino, H. Waterman, L.M. Wang, L. Lyass, M. Alimandi, A. Kuo, S.S. Bacus, J.H. Pierce, G.C. Andrews, Y. Yarden, Epiregulin is a potent pan-ErbB ligand that preferentially activates heterodimeric receptor complexes, J. Biol. Chem. 273 (1998) 10496–10505. [104] S.B. Lee, K. Huang, R. Palmer, V.B. Truong, D. Herzlinger, K.A. Kolquist, J. Wong, C. Paulding, S.K. Yoon, W. Gerald, J.D. Oliner, D.A. Haber, The Wilms tumor suppressor WT1 encodes a transcriptional activator of amphiregulin, Cell 98 (1999) 663–673. [105] S.L. Campbell, R. Khosravi-Far, K.L. Rossman, G.J. Clark, C.J. Der, Increasing complexity of Ras signaling, Oncogene 17 (1998) 1395–1413. [106] M. Sibilia, A. Fleischmann, A. Behrens, L. Stingl, J. Carroll, F.M. Watt, J. Schlessinger, E.F. Wagner, The EGF receptor provides an essential survival signal for SOS- dependent skin tumor development, Cell 102 (2000) 211–220. [107] L.M. Gangarosa, N. Sizemore, R. Graves-Deal, S.M. Oldham, C.J. Der, R.J. Coffey, A raf-independent epidermal growth factor receptor autocrine loop is necessary for Ras transformation of rat intestinal epithelial cells, J. Biol. Chem. 272 (1997) 18926–18931. [108] N. Sizemore, A.D. Cox, J.A. Barnard, S.M. Oldham, E.R. Reynolds, C.J. Der, R.J. Coffey, Pharmacological inhibition of Ras-transformed epithelial cell growth is linked to down-regulation of epidermal growth factor-related peptides, Gastroenterology 117 (1999) 567–576. [109] L. Chin, J. Pomerantz, D. Polsky, M. Jacobson, C. Cohen, C. Cordon-Cardo, J.W. Horner 2nd, R.A. DePinho, Cooperative effects of INK4a and ras in melanoma susceptibility in vivo, Genes Dev. 11 (1997) 2822–2834. [110] R.R. Beerli, N.E. Hynes, Epidermal growth factor-related peptides activate distinct subsets of ErbB receptors and differ in their biological activities, J. Biol. Chem. 271 (1996) 6071–6076. [111] K. Elenius, S. Paul, G. Allison, J. Sun, M. Klagsbrun, Activation of HER4 by heparin-binding EGF-like growth factor stimulates chemotaxis but not proliferation, EMBO J. 16 (1997) 1268–1278. [112] B.C. Paria, K. Elenius, M. Klagsbrun, S.K. Dey, Heparin-binding EGF-like growth factor interacts with mouse blastocysts independently of ErbB1: a possible role for heparan sulfate proteoglycans and ErbB4 in blastocyst implantation, Development 126 (1999) 1997–2005. [113] K. Elenius, C.J. Choi, S. Paul, E. Santiestevan, E. Nishi, M. Klagsbrun, Characterization of a naturally occurring ErbB4 isoform that

14


does not bind or activate phosphatidyl inositol 3-kinase, Oncogene 18 (1999) 2607–2615. [114] C. Sawyer, I. Hiles, M. Page, M. Crompton, C. Dean, Two erbB-4 transcripts are expressed in normal breast and in most breast cancers, Oncogene 17 (1998) 919–924. [115] V. Kainulainen, M. Sundvall, J.A. Maatta, E. Santiestevan, M. Klagsbrun, K. Elenius, A natural ErbB4 isoform that does not

activate phosphoinositide 3-kinase mediates proliferation but not survival or chemotaxis, J. Biol. Chem. 275 (2000) 8641–8649. [116] J. Massague, A. Pandiella, Membrane-anchored growth factors, Annu. Rev. Biochem. 62 (1993) 515–541.

Neuregulins: functions, forms, and signaling strategies Douglas L. Falls Center for Neurodegenerative Disease, Department of Neurology, Emory University, Atlanta, GA 30322, USA

Abstract The neuregulins (NRGs) are cell-cell signaling proteins that are ligands for receptor tyrosine kinases of the ErbB family. The neuregulin family of genes has four members: NRG1, NRG2, NRG3, and NRG4. Relatively little is known about the biological functions of the NRG2, 3, and 4 proteins, and they are considered in this review only briefly. The NRG1 proteins play essential roles in the nervous system, heart, and breast. There is also evidence for involvement of NRG signaling in the development and function of several other organ systems, and in human disease, including the pathogenesis of schizophrenia and breast cancer. There are many NRG1 isoforms, raising the question “Why so many neuregulins?” Study of mice with targeted mutations (“knockout mice”) has demonstrated that isoforms differing in their N-terminal region or in their epidermal growth factor (EGF)-like domain differ in their in vivo functions. These differences in function might arise because of differences in expression pattern or might reflect differences in intrinsic biological characteristics. While differences in expression pattern certainly contribute to the observed differences in in vivo functions, there are also marked differences in intrinsic characteristics that may tailor isoforms for specific signaling requirements, a theme that will be emphasized in this review. Keywords: Neuregulin; Acetylcholine receptor-inducing activity; Glial growth factor; Heregulin; Neu differentiation factor; Sensory and motor neuronderived factor; ErbB receptor tyrosine kinase; Schizophrenia; Neuromuscular synapse; Cell-cell signaling proteins; Juxtacrine signaling; Paracrine signaling; Transmembrane ligands; Proteolytic process; Shredding

The discovery of neuregulins (NRGs); NRG family genes; the focus of this review Neuregulins (NRGs) are signaling proteins that mediate cell-cell interactions in the nervous system, heart, breast, and other organ systems. “Forward” signaling by NRGs— i.e., signaling from a NRG-producing cell to a NRG-responsive cell—involves binding of NRG to the extracellular domain of the receptor tyrosine kinases ErbB3 or ErbB4, which leads to formation of ErbB homo- or heterodimers (often including ErbB2), which in turn activates intracellular signaling pathways leading to cellular responses that include stimulation or inhibition of proliferation, apoptosis (programmed cell death), migration, differentiation, and adhesion [1]. The first identifications of NRGs were reported in 1992–1993 by four groups. Two of these groups sought a ligand for the oncogene ErbB2 (a.k.a. neu, HER2) [2–4]; the third sought a factor that stimulated the proliferation of Schwann cells [5,6], and the fourth sought a factor that stimulated the synthesis by muscle of receptors for acetylcholine, the major neurotransmitter at developing neuromuscular synapses [7]. The neuregulin proteins isolated by each of these groups are encoded by the gene that would

now be referred to as NRG1. It should be noted that though one approach leading to the identification of NRGs was a search for ErbB2 ligands, in fact, it appears that NRG proteins interact with ErbB2 only after binding ErbB3 or ErbB4 [1]. Subsequent to the identification of the NRG1 gene, three other genes encoding related proteins were discovered. These “other” NRGs are referred to as NRG2 (a.k.a. Don1, NTAK [8–11]), NRG3 [12], and NRG4 [13]. The NRG1 proteins effectively bind to both ErbB3 and ErbB4; the protein products of these other NRG genes effectively bind one or the other or both of these ErbBs [1,14,15]. Very little is yet known about the functions of the NRG2, 3, and 4 proteins. One intriguing recent finding is evidence of NRG4 involvement in the differentiation of the somatostatin-expressing delta cells of pancreatic islets of islets [16]. Buonanno and Fischbach [17] compare the structure and expression patterns of NRG2, 3, and 4 proteins to NRG1 proteins and discuss the intriguing discovery that NRG2 activation of a specific ErbB receptor combination (i.e., ErbB4 homodimers) can elicit different patterns of receptor phosphorylation and downstream consequences than activation of the same receptor combination in the same cell type by NRG1 [18,19].

16


This review will focus on the NRG1 proteins, and unless explicitly indicated otherwise, the terms “neuregulin” and “NRG” here refer to NRG1 proteins. Since the discovery of NRGs 10 years ago, the field has grown rapidly. A search of Pub Med for “neuregulin” or various names by which these proteins have also been known (see below) in early December 2002 returned a list of over 800 publications. This rapid growth in the literature attests to the importance of NRGs, but also means that that this review cannot be in any sense comprehensive. Thus, I apologize in advance for the many important papers not referenced and for interesting areas of NRG research omitted or mentioned only in passing. A number of excellent reviews of NRGs and NRG-mediated cell-cell interactions have appeared over the years. Selected reviews published in the last 6 years include those focused specifically on NRGs [17,20–25], as well as reviews of neuromuscular synapse development [26–28], neuron-glial interactions [24,29–33] and cell interactions regulating heart development and function [169]. Companion reviews in this issue discuss the NRG receptors ErbB2, ErbB3, and ErbB4 and intracellular signaling pathways activated by these receptors (see reviews in this issue by Yarden, Carpenter, and Wiley; see also [1,17]). Other companion reviews in this issue describe the roles of ErbB family receptors and ligands in breast development, in cancer, and as therapeutic targets (reviews by Stern, Hynes, Arteaga, Fry, and Maihle). While I will begin with an overview of NRG biology, the availability of these other reviews allows me to emphasize recent literature and to focus on the normal biological functions of the NRG1 proteins during development and in the adult, evidence for the involvement of NRG1 signaling in neuropathology (other than cancer), and mechanisms regulating NRG signal production. Readers of the companion reviews on the EGFR and its ligands (see reviews by Coffey and Burgess) and the ErbB receptor/ligand homologues in invertebrates (see reviews by Shilo and Sternberg) will note both marked similarities and striking differences between

the biology of NRG1 proteins and these related signaling systems of vertebrates and invertebrates.

The NRG1 gene; NRG1 isoforms and nomenclature An important recent advance is the sequencing and assembly of the entire human NRG1 gene (Fig. 1A, [34]). The gene is ª 1.4 megabases long (ª 1/2000th of the genome); less than 0.3% of this span encodes protein. As a consequence of rich alternative splicing and multiple promoters, at least 15 different NRG isoforms are produced from the single NRG1 gene Fig. 1B and C [17,20]). The three structural characteristics we know to importantly differentiate isoforms with respect to in vivo functions and cell biological properties are the type of EGF-like domain (a or b), the N-terminal sequence (type I, II, or III), and whether the isoform is initially synthesized as a transmembrane or nonmembrane protein; the import of these differences will be discussed below. The EGF-like domain contained in all bioactive NRG isoforms is alone sufficient for activation of ErbB receptor-tyrosine kinases (see [17] for comparison of NRG1 EGF-like domain sequences to the EGF-like domain in NRG2, 3, and 4 and other EGF family members). Together the types I and II NRGs are sometimes referred to as “Ig-NRGs,” and the type III NRGs are sometimes referred to as “CRD-NRGs.” The names first used in the literature to refer to various NRG isoforms—acetylcholine receptorinducing activity (ARIA [7]), glial growth factor (GGF [5,6]), heregulin (HRG [2]), neu differentiation factor (NDF [3,4]), and sensory and motor neuron-derived factor (SMDF [35]),—cannot be taken to indicate specific biological functions of the isoforms to which these names have been applied. For example, it now seems likely that the major NRG isoforms that act as “glial growth factors” in vivo are type III NRGs, not the type II NRGs originally called glial growth factor (GGF).

Fig. 1. Æ NRG1 gene and isoform structure. (A) Human NRG1 gene structure (Genbank accession no. BK000383). The NRG1 gene is on the short arm of chromosome 8. On the expanded illustration of this region, the position of each exon included in reported NRG1 isoforms is indicated by a vertical line. Lines descending along the edge of the green box delineate the boundaries of the core at-risk haplotype for schizophrenia. Only the exon encoding the type II-specific N-terminal region lies within these bounds. Exon naming: Exons are named here for the structural region of the NRG1 protein they encode. Abbreviations used closely correspond to names of NRG protein structural regions indicated in panel B. EGFc refers to the exon encoding the portion of the EGFlike domain sequence shared by NRGs with an a-type and NRGs with a b-type EGF-like domain. The exon labeled TMc also includes adjacent extracellular juxtamembrane sequence and cytoplasmic tail sequence. (B) Illustration of NRG “coding segments.” Isoforms differ in their coding segment composition due to initiation of transcription from different NRG1 gene promoters and alternative splicing. The EGF-like domain alone is sufficient for high potency activation of the cognate ErbB receptor tyrosine kinases. Available evidence indicates that the NRGs most commonly expressed in the nervous system are transmembrane NRGs with a b-type EGF-like domain and the 374-amino acid a-type tail. Not all potential combinations of coding segments have been reported. CRD = cysteine-rich domain; EGF = epidermal growth factor-like domain; Ig = immunoglobulin-like domain; CTc and TMc = cytoplasmic tail and TM domain C-terminal of the EGF-like domain; CTn and TMn = cytoplasmic tail and TM domain N-terminal of the EGF-like domain. Only type III NRGs have the CTn and TMn. * = stop codon. (C) Structural regions of the I-b1a, II-b1a, III-b1a, and III-b3 proproteins. The I-b1a, II-b1a, and III-b1a isoforms differ only in their N-terminal region; their sequence is identical from the EGF-like domain through the carboxy-terminus. The sequence of III-b1a and III-b2a is identical except that in III-b2a the last eight amino acids of the “1” subtype are absent (illustrated under III-b1a sequence). The sequence of III-b1a and IIIb3 is identical from the N-terminus to just beyond the final cysteine of the EGF-like domain. Hydrophobic regions that serve as transmembrane domains are indicated by black boxes. A hydrophobic sequence in the type II N-terminal region is indicated by a hatched box. This is believed to serve as a noncleaved internal signal sequence but not as a transmembrane domain.

Neuregulins: Functions, Forms, and Signaling Strategies

17

18



NRG1 signaling in health; isoforms differing in their N-terminal region or EGF-like domain differ in their in vivo functions Without NRGs life is not possible. However, even I—a confirmed neuregulin fanatic—was surprised as I surveyed the literature in preparation for writing this review, at how pervasive NRG signaling appears to be Table 1). While I will here introduce the in vivo functions of NRGs by describing the dramatic phenotypes of the knockouts, it must be emphasized that a number of other experimental approaches have made important contributions to our current understanding of NRG functions, and it seems likely that many NRG functions remain to be discovered. Studies of mice with targeted mutations of the NRG1 gene have been very valuable in elucidating functions of NRG1 proteins [34,36–47]. Analysis of pan-NRG1 knockout (KO) mice (mice in which all NRG isoforms are unable to bind to and activate ErbB receptors due to disruption of the EGF-like domain) revealed an essential role of NRGs in cardiac morphogenesis [36], a role unsuspected from previous studies of NRG bioactivities. Due to the defect in cardiogenesis, these mice die midway through embryogenesis (E10.5), the time at which mouse embryos switch from dependence on the maternal circulation to dependence on their own circulation. The pan-NRG1 KO mice also have a severe reduction in several neural crest-derived cell populations including Schwann cells, the glia of the peripheral nervous system, which—among other things—form the myelin sheaths of peripheral nerves; neural crest-derived cranial sensory neurons; and sympathetic neurons [39,40]. Having a knockout with a severe phenotype is both a boon and a bane: a boon because it provides comforting reassurance that the gene of interest has essential roles and a bane because death or early developmental disruptions in the mutants preclude analysis of later developmental events. Such is the case with analysis of nervous system development in the pan-NRG1 knockout mice: at E10.5 the devel-

19

opment of the nervous system is only beginning to unfold; and the role of NRGs in processes such as neuromuscular synaptogenesis, which begins around E14, and the development of oligodendrocytes, the cells that myelinate axons in the central nervous system, is inaccessible for direct in vivo examination in these mice. In some cases, clever strategies can allow this limitation to be partially circumvented. Thus, through analysis of spinal cord slices harvested from E9.5 pan-NRG1 embryos and maintained in organ culture for up to 11 days, a strong case has been made for an essential role of NRGs in oligodendrocyte lineage development [48]. Mice with targeted mutations that inactivate only certain classes of NRG isoforms have revealed differential in vivo functions of NRG proteins Table 2). Ig-NRG1-/- mice: Mice with all Ig-NRG isoforms inactivated (types I and II NRGs inactivated)—but in which CRD-NRG (Type III NRG) production is presumably normal—die at E10.5 and have defects in cardiac, cranial sensory neuron, and sympathetic development similar to those of the pan-NRG knockouts [37,39,40]. However, unlike the pan-NRG KOs, the Ig-NRG KO mice have normal development of Schwann cell precursors [39]. CRD-NRG1-/- mice: In contrast to the pan- and Ig-NRG KO mice, mice with all CRD-NRG isoforms (type III NRGs) inactivated—but with normal expression of IgNRGs—do not have defects in heart development [42]. The CRD-NRG KO embryos survive to birth. They die at birth because they cannot breathe; they cannot breathe because they do not have functional neuromuscular synapses. Unlike the Ig-NRG knockouts, the CRD-NRG knockouts have a marked reduction in Schwann cell precursors. Other prominent phenotypic characteristics of the CRD-NRG KO mice include degeneration of peripheral and cranial nerves and an ª 50% reduction in the number of spinal motor and sensory neurons. The reduction in motor and sensory neuron number appears to be due to abnormal neuron death, as the initial number of these (postmitotic) neurons is normal. Like the spinal motor and sensory neurons, in the CRD-NRG1 KO mice, cranial motor and sensory neurons (which together

Fig. 2. Membrane orientation and proteolytic processing proposed for the NRG I-b1a isoform (A) and III-b1a isoform (B) based on studies in transfected fibroblastic cell lines. (A) Proteolytic cleavage of the I-b1a proprotein in the “stalk” region (arrow no. 1) produces an N-terminal fragment (NTF) containing the bioactive EGF-like domain and a C-terminal fragment (CTF), also referred to as the “a-tail remnant.” The NTF is efficiently released into the medium. The protease(s) catalyzing stalk cleavage—at least in fibroblasts—is likely to be a metalloprotease. All Type I and Type II proproteins with a transmembrane domain are expected to have similar topology and processing (see Fig. 3). (B) Available evidence indicates that the III-b1a proprotein has two transmembrane domains and that proteolytic cleavage in the stalk region (arrow no. 1) produces a transmembrane N-terminal fragment (NTFm) that accumulates at the cell surface. Cleavage of the III-b1a NTFm near the membrane (arrow no. 2) can release a fragment (NTFs) containing the EGF-like domain into the medium, but for NRGs expressed in fibroblasts, the amount of this released Type III NTFs is very small compared to the amount of released Type I NTFs. All Type III proproteins with a transmembrane domain C-terminal of the EGF-like domain are expected to have topology and processing similar to that illustrated here for III-b1a (see Fig. 3). Fig. 3. Proposed topology and mode of presentation to receptor for selected NRG1 proprotein isoforms. The data and reasoning supporting the assigned topology and mode of presentation (paracrine, shed; paracrine, secreted; juxtacrine) for the types I, II, and III isoforms are described in the text. The proteins encoded by two other mRNA sequences are illustrated to facilitate their comparison with the full-length NRGs shown. HRG-g is a truncated Type I sequence. It is unlikely to be bioactive for two reasons: first, the EGF-like domain, which is necessary for activating ErbBs, is incomplete; and second, since it is a truncated version of I-b3, it is unlikely to be released. The rightmost diagram illustrates a protein named g-HRG (π HRG-g). This protein is encoded by transcripts produced by the breast cancer cell line MDA-MB-175 [166]. It has now been shown that g-HRG is a fusion protein with N-terminal sequence from the human homologue of transcription factor DOC-4 [167]. This transcription factor is a member of the Oz/ten M family. On the basis of sequence similarity to the N-terminal portion of g-HRG, some sequences encoding Ten-m/Odz family members have been erroneously labeled as NRGs or “NRGlike.” There is as yet no evidence for a physiological role of the HRG-g sequences or g-HRG [168], and they are not further discussed in the text.

20


Table 1 Selected proposed functions of NRGsa Organ/cell type/structure Nervous system Schwann cells Oligodendrocytes Neuromuscular synapse

Muscle spindle Cranial sensory neurons

Motor and sensory neurons Peripheral and cranial nerves Sympathetic neurons/adrenal medulla Cerebellum Cortical neuron precursors/ cerebellar granule cells Hypothalamus Parasympathetic Hippocampus

Effect

Reference(s)

Survival, proliferation, migration, differentiation, myelination Proliferation, survival, differentiation, myelination Nerve–muscle interaction controlling initial formation, acetylcholine receptor synthesis during development and in the adult, nerve terminal interactions with “terminal Schwann cells” Muscle spindle development (Muscle spindles are muscle length/stretch sensors.) Initial population of cranial sensory ganglia (ganglia of cranial nerves) with neural-crest derived sensory neurons (However, the initial population of cranial sensory ganglia with placode-derived sensory neurons appears to be unaffected by NRG1 mutations.) Survival (spinal and probably also cranial) Fasciculation (bundling) of axons and/or integrity of nerves

[24,29,31–33] [48,129–134,153] [17,20,26–28,146]

Migration of sympathetic neuron/adrenal chromaffin precursors to the anlage of sympathetic ganglia/adrenal medullas Production of cerebellar neuron precursors Migration of CNS neuronal precursors along radial glia

Various neurons of CNS and PNS Heart Heart Heart Blood vessels Breast

Hypothalamic control of mammalian female sexual maturation Enteric ganglia development Inhibition of long-term potentiation (LTP) induction (LTP is a model for studying the neurophysiological basis of learning and memory.) Regulation of neuronal neurotransmitter receptors (NMDA, GABA, neuronal nicotinic acetylcholine receptors) and other neuronal ion channels Development of ventricular wall trabeculae, AV-septum, and cardiac valves Development of cardiac conduction system Growth, repair, survival of adult cardiomyocytes; response to increased work load Angiogenesis Breast development during pregnancy and lactation

Lung Muscle Muscle Muscle Gonads Stomach

Development of pulmonary epithelium (autocrine effect?) Myogenesis (autocrine effect?) Muscle fiber survival in neonatal period Glucose uptake Gonadogenesis Proliferation of gastric epithelium; regulation of parietal and chief cell population size

[172] [36–39,44]

[42] [36–39,42,44] [40] [38] [71,72] [135,173] [136,174,175] [97] [64,74–76,176] [36–38,44,137] [138] [89–91,139] [144] [45]; Review in this issue by D. Stern [142] [140] [141] [144] [145,177] [178,179]

a This list is not intended to be comprehensive, and further investigation will be required to confirm the physiological significance of many of the proposed roles. Some proposed functions have been inferred principally from the effects of exogenously supplied recombinant NRG1, ErbB blockade, or ErbB knockout; in these cases the physiological signal may actually be NRG2, 3, or 4 or another ErbB ligand. Caveat emptor. Functions proposed solely on the basis of mRNA/protein expression data are not included. Reviews have been cited for proposed functions of NRG1s in Schwann cell and neuromuscular synapse development due to the large body of relevant literature.

contribute most of the axons that make up the cranial nerves) appear to be reduced in number. However, unlike the IgNRG1 KO mice, both placode- and neural crest-derived cranial sensory neurons are affected. In the Ig-NRG1 KOs, the reduction in neural crest-derived cranial sensory neurons is due to a defect in initial accumulation of these neurons in the nascent ganglia (perhaps, like the defect in sympathetic neuron accumulation, caused by abnormal neuronal precursor migration?), but in the CRD-NRG1 KO mice it seems likely that the reduction in cranial sensory neurons results from abnormal death of neurons, and, as for spinal sensory neurons, this increased death is probably a consequence of disruption in signaling between these neurons and their supporting Schwann cells or targets. NRG1-/- mice: Mice with a targeted mutation that inactivates all NRG isoforms with

an a-type EGF-like domain (NRGas)—but in which production of isoforms with b-type EGF-like domain (NRGbs) is presumably normal—have not been reported to have abnormalities in nervous system or cardiac development, but have marked defects in breast development [45]. (Aside: In most assays, NRG1s with a b-type EGF-like domain are 10–100 times more potent than NRG1s with an a-type EGF-like domain. It is a puzzle as to why NRG1s with both a- and b-type EGF-like domains exist and why the a-type was selected by evolution for a critical role in breast development.) Comparison of the phenotypes of the ErbB knockouts [34,38,40,41,43,45,49–57] to the NRG1 knockouts has provided insight into the ErbB receptor combinations mediating early essential actions of NRGs. Furthermore, most all


21

Table 2 Comparison of NRG1 knockout mice with respect to selected characteristics Development of:

Genotype and isoforms INactivated NRG1-/All (Pan-NRG1 KO)

Ig-NRG1-/Ig-NRG1 (Type I and II)

Heart Schwann cell precursors Neuromuscular synapses Breast (during pregnancy) Homozygotes die at Cause of death:

E10.5 Heart failure

E10.5 Heart failure

References

[36,38,39]

[37]

CRD-NRG1-/CRD-NRG1 (Type III)

NRG1a-/NRG1a

NRG1DCT/DCT Type I NRG1 (and others?)** ?? (Not described)

Birth Respiratory (neuromuscular) failure [42]

Normal lifespan Old age

E10.5 Heart failure

[45]

[44]

a

The major neurotransmitter receptor at neuromuscular synapses is the muscle nicotinic acetylcholine receptor. Although neuromuscular synapse development (which begins around E14) cannot be assessed in the Ig-NRG KOs, adult Ig-NRG+/- mice (i.e., heterozygous for the mutation inactivating Ig-NRGs) have a 50% reduction in the number of acetylcholine receptors at neuromuscular synapses [47], indicating that Ig-NRGs do function as an “acetylcholine receptor-inducing activity” (ARIA). This “postsynaptic phenotype”—i.e., the effect on AChRs, which are concentrated in the postsynaptic muscle membrane—is different than the “presynaptic” and “Schwann cell phenotype” of the CRD-NRG KO mice (for review of neuromuscular synapse development and NRG functions at the neuromuscular synapse, see refs. [20,146]). b The DCT allele of the NRG1 gene has an in-frame stop codon within the sequence encoding the cytoplasmic tail. This mutation causes all transmembrane NRG1 proteins produced from this allele to have a cytoplasmic tail length of only 3 amino acids. See text (“A tale of the heart . . .”) for interpretation of the NRG1DCT/DCT phenotype. Black = abnormal; blank = normal; shaded = not accessible for analysis because mice die prior to occurrence of this developmental event. Mice are normally born at embryonic day 21 or 22 (E21 or E22).

reported characteristics of the various NRG1 knockouts are shared by one or more of the ErbB knockouts and vice versa, which suggests that interactions of NRG2, 3, and 4 with ErbBs are unlikely to play a prominent role in the developmental events that dominate the ErbB KO phenotypes; rather these developmental events are likely to be mediated principally by NRG1-ErbB interactions. There is one clear exception to this generalization; ErbB4 knockout mice have a defect in hindbrain segmentation not seen in the NRG1 knockout mice [49], indicating that a non-NRG1 ligand interacts with ErbB4 to guide this developmental event. One further note, many of the phenotypic characteristics of the NRG1 and ErbB KO mice center on abnormalities of ErbB expressing cell populations, suggesting that these defects result from disruption of forward signaling (NRGproducing cell signaling to ErbB-expressing cell). However, it has been proposed that NRG-ErbB signaling is bidirectional [58–60], similar to what has been demonstrated for Eph-Ephrin signaling (see [61] and references therein). In this model, it is proposed that there is not only the conventional “forward signaling” from NRG-producing cells to ErbB-expressing cells, but that there is also reverse signaling (or “back-signaling”) from ErbB-expressing cells to NRG-expressing cells. In the latter case, NRG would serve as the receptor and ErbB the ligand. Some phenotypic characteristics of the NRG1 and ErbB1 knockouts involve NRGproducing cells, raising the possibility that these defects result from disruption of NRG Æ ErbB reverse signaling. For example, motor neurons, which produce NRGs and which communicate with ErbB expressing Schwann and skeletal muscle cells, are reduced in number in the CRD-

NRG KO mice compared to wild-type mice. This might be due to interruption of ErbB Æ NRG reverse signaling, though alternatively, this abnormality could also result from interruption of forward signaling secondarily disrupting a separate “reciprocal” signaling pathway back to motor neurons from Schwann cells and/or muscle. While the hypothesis of NRG1-ErbB reverse signaling is very attractive, definitive evidence that this occurs has yet to be published.

NRG1 signaling in disease: evidence for involvement in pathophysiology and potential therapeutic uses The many functions of neuregulins revealed through knockout and other studies (Tables 1 and 2) attest to the importance of neuregulin signaling during development and in the adult. Are disorders of neuregulin signaling involved in the pathogenesis of disease, and what are the prospects for disease therapy based on modulating neuregulin signaling? Table 3 summarizes currently investigated pathological and therapeutic considerations with respect to NRG1. Here I will briefly describe only one: the recent evidence for NRG involvement in schizophrenia. Schizophrenia is a disabling neuropsychological disorder with strong familial characteristics suggestive of a genetic component [62]. A genome wide survey of patients with familial schizophrenia in Iceland, employing both linkage and association methodologies, uncovered the NRG1 gene as a candidate susceptibility gene for schizophrenia, and this association was confirmed in a Scottish population [34,63].

22


Table 3 Diseases/injuries in which the pathophysiology may involve perturbations in NRG1 signaling and/or in which NRG1s may be of therapeutic usea Organ/organ system

Effect

Type of evidence with references

Nervous system Nervous Nervous

Schizophrenia Multiple sclerosis (pathology and treatment) Promotion of neural regeneration/proliferation of olfactory ensheathing glia for therapeutic use in neural regeneration Protection against neuropathy induced by cancer chemotherapeutic drug cisplatin Response to traumatic brain injury Trastuzumab (Herceptin) cardiotoxicity Angiogenesis (in tumor growth) Breast tumor formation Paget’s disease (a-NRG as motility factor stimulating spread of neoplastic cells) Wound Healing (a-NRG as motility factor for keratinocytes) Hirschsprung’s disease Regeneration (newt limb)

[34,62,63,66] [147–153] [154–157]

Nervous system Nervous Heart Vascular Breast Breast Skin Gut Limb

[158] [159] [87,88,92] [160,161] [162] [163] [164] [136] [165]

a This compilation includes diseases/injuries for which NRG involvement is supported by genetic evidence and/or by manipulation of NRG1 signaling in animal models. The one exception is response to traumatic brain injury and stroke, for which involvement of NRG1 signaling has been proposed solely on the basis of changes in NRG1 expression.

The known activities of NRGs fit well with current hypotheses regarding the neurobiological basis of schizophrenia. One theory proposes that schizophrenia results from a deficiency of glutamatergic innervation relative to dopaminergic innervation. Consistent with the idea that impairment of NRG1 signaling contributes to the pathology of schizophrenia, mice heterozygous for two different mutations in the NRG1 gene or a null mutation of the ErbB4 gene display hyperactivity in behavioral tests similar to hyperactivity observed in mice treated with the psychogenic drug phencyclidine (PCP) or with mutations that impair glutamatergic neurotransmission or enhance dopaminergic neurotransmission [34,41,49]. The NRG1 mutant mice examined in these behavioral studies had an in-frame stop codon introduced within the sequence encoding the NRG1 EGF-like domain [41] (inactivating all NRG1 products of the mutated allele) or introduced within the sequence encoding the transmembrane domain [34]. The phenotype of this latter strain has not yet been reported, but I suspect it will be similar to mice in which the NRG1 cytoplasmic tail has been severely truncated by targeted mutagenesis (see “A tale of the heart . . .” below). Response to the antipsychotic drug clozapine and levels of glutamate receptors were also studied in the mice heterozygous for mutation of the NRG1 transmembrane domain [34]. Treatment with clozapine reversed the hyperactivity of these mice, and they had reduced levels of the NMDA type of glutamate receptors, as assessed by binding of the NMDA receptor ligand MK801. Furthermore, application of soluble NRG1 to cultured neurons stimulates transcription of the NMDA receptor subunit NR2C [64]. Another theory proposes that abnormalities in glial biology contribute to the pathology of schizophrenia [65]. Neuregulins are required for initial differentiation of oligodendrocyte precursors and for their survival [43,48]. A variant of this idea is that a deficiency of glial growth

factors—such as NRG—predisposes to synaptic destabilization [66]. It is clear that NRG signaling is required for the stabilization of neuromuscular synapses [67,68], and evidence for NRG involvement in astrocyte biology might implicate NRGs in formation or stabilization of central synapses [69] (see also [70]). A third idea is that schizophrenia results from abnormalities in cortical wiring, and NRGs have been shown to regulate migration of neuronal precursors in culture [71,72]. A fourth hypothesis is that schizophrenia results from abnormalities in synaptic plasticity, and NRG1s inhibit induction of long-term potentiation, a form of synaptic plasticity studied as a model for the neurophysiological substrates of learning and memory [97]. Fig. 1A depicts the boundaries of the genetic haplotype associated with schizophrenia. The only exon of reported NRG isoforms within these bounds is the exon encoding the type II (“GGF2”) N-terminal sequence. While widely expressed in the nervous system during development and postnatally [39,73], no in vivo functions of type II NRGs are yet known, and no mutation within this exon that segregates with schizophrenia susceptibility has yet been detected. This raises the possibility that if alterations in NRG signaling are indeed involved in the pathogenesis of schizophrenia, the causative mutation may be in intronic or upstream sequence that regulates transcription or splicing.

Sometimes a kiss sent from a distance may be sufficient, but in other situations a kiss on the lips may be required: NRG1 paracrine signaling by shedding and secretion and NRG1 juxtacrine signaling The ErbB family of receptors and their ligands has been described as a “signaling network” with an input layer comprised of ligands, receptors, and transactivators; a signal processing layer comprised of adapters, cascades, and


transcription factors; and an output layer comprised of the biological consequences of ligand-ErbB interaction, such as stimulation of proliferation, inhibition of apoptosis, and differentiation ([1]; see also [98–100] and companion reviews in this issue). In moving from level to level of the network, there is both convergence and divergence, and there are horizontal (lateral) interactions within each level. While we have learned much about the network, we are only beginning to understand how the components of the network are selected, arranged, modified, and modulated in individual cells to achieve physiologically adaptive outcomes of cell-cell interactions. Much of the recent excitement in the study of cell-cell interactions derives from advances in defining the intracellular signal transduction pathways that couple (translate) receptor activation to cellular responses. However, equally important to understanding cell-cell interactions is defining the mechanisms that regulate the presentation of signals to receptors; that is, understanding the events upstream of receptor activation. Here the role of isoform topology and proteolytic processing in governing NRG-mediated cell-cell interactions will be considered.

Paracrine signaling by Ig-NRGs Paracrine signaling refers to short distance cell-cell communication mediated by diffusible signaling molecules. Communication mediated by such diffusible signals allows cells not in direct contact to “talk to” each other. Proteins that serve as “paracrine signals” are commonly synthesized as soluble proteins, which—following processing in the ER-Golgi system and transport—are released by secretion, the spilling out of the trafficking vesicle’s contents when it fuses with the cell’s plasma membrane. Pre-1992, when NRGs were still molecularly unidentified “factors,” it was assumed they would turn out to be such typical paracrine signaling proteins, for they were being purified as soluble proteins from medium conditioned by cultured transformed cells and from aqueous (nondetergent) extracts of brain and pituitary and were bioassayed by dissolving the partially purified protein preparations in medium bathing responsive cells [2–7]. Indeed, one NRG isoform—the NRG II-b3 isoform, commonly referred to as GGF2 or simply GGF— is believed to conform to this model. However, most NRGs are synthesized as transmembrane proteins. So, (1) are these transmembrane NRGs released to serve as paracrine signals? And (2) if so, how? The answers are (1) yes, the Type I and (probably) the Type II transmembrane NRGs do generate paracrine signals and (2) the ectodomain is “shed” from the membrane by proteolytic processing [101–107]. The topology and processing of NRGs has been studied principally in cultures of fibroblastic cells in which NRG isoforms have been expressed by transfection [77,82,108–113]. Through such studies, it has been shown that type I TMc-NRGs are expressed as Nout/Cin single-

23

pass transmembrane proteins (type I membrane proteins, not to be confused with a type I NRG) that are cleaved in the “stalk” region to produce a paracrine signal. Type I NRGs (and likely type II NRGs, the other class of Ig-NRGs) appear to act principally as paracrine Type I (short distance, diffusible) signals. Since most type I NRGs in the nervous system are synthesized as transmembrane proteins, paracrine signaling requires proteolytic cleavage of the TMc-NRG proprotein in the stalk region to release the bioactive ectodomain fragment (NTF; see Fig. 2). One example of a cell-cell interaction that appears quite clearly to be mediated by a soluble bioactive NRG1 fragment produced by shedding is the communication between endocardium and myocardium that was discussed below (“A tale of the heart . . .”). Though space precludes more than a passing remark, it must be noted that shedding can produce an autocrine signal (signal-producing cell talks to itself or cells of the same type), as well as a paracrine signal (signalproducing cell talks to cells of a different type). Thus, reported cases of autocrine signaling by NRGs (i.e., [114,115,140,142]) might involve shed TMc-NRGs. Similar to Ig-NRGs, EGF, TGFa, and most other ligands for the EGFR are synthesized as transmembrane proproteins that are shed to serve as paracrine signals. Knockout of ADAM17 (a.k.a. TACE) has demonstrated that this metalloprotease is essential for the processing of TGFa and likely one or more other ligands of the EGFR [103,116]. ADAM17 and ADAM19 (a.k.a. meltrin-b) have been shown capable of mediating shedding of NRGs from cultured cells [117,118], but their role in governing NRG signal production in vivo remains unknown. Since inhibiting or stimulating metalloproteases is being considered in the therapy of various diseases, including Alzheimer’s disease and cancer it will be important to determine the effects of candidate drugs on NRG signaling.

Are CRD-NRGs (type III NRGs) specialized to serve as juxtacrine signals? Initially it was assumed that like type I NRGs, the type III NRGs with a transmembrane domain C-terminal of the EGF-like domain (TMc-NRGs), such as III-b1a, would be single pass transmembrane proteins and that stalk cleavage of Type III NRGs would shed a bioactive ectodomain fragment that includes both the “cysteine-rich domain” (CRD) and the EGF-like domain. However, a direct test of this model in which type I and type III NRGs were expressed by transfection in fibroblastic cell lines [77] yielded surprising results: the topology of NRG III-b1a is unlike the topology of NRG Ib1a, and, in fact, unlike the topology of any previously reported RTK ligand (however, see [119,120] for discussion of an RTK ligand with another interesting topology). The sequences of NRG III-b1a and I-b1a differ only in their N-terminal regions; their sequence from the EGFlike domain through the C-terminus—including the

24


sequence of the TM domain and juxtamembrane segments is identical. However, instead of being an Nout/Cin singlepass transmembrane protein like type I TMc-NRGs, the type III NRGs with a TM-domain C-terminal of the EGF-like domain are Nin/Cin two-pass transmembrane proteins, with a hydrophobic segment within the CRD serving as a second transmembrane domain (see Fig. 2). This has two major consequences: (1) Instead of being an extracellular proteinprotein interaction domain as originally suspected, the CRD domain is mostly intramembrane and intracellular. (2) Stalk cleavage of Type III TMc-NRGs does not shed a bioactive ectodomain fragment, but instead creates a transmembrane N-terminal fragment. As would be predicted from the topological differences, when type III and type I NRGs are expressed in parallel cultures, the amount of type III NRG released into the medium is much less than the amount of type I NRG, but the amount of type III NRG exposed at the cell surface—most of which is the transmembrane Nterminal fragment—is much more than the amount of type I NRG [77]. It should be noted also that NRG III-b3, a form lacking the TM-domain C-terminal of the EGF-like domain also accumulates on the cell surface ([79]; J. Wang and D. Falls, unpublished data). These results raise the possibility that type III NRGs are specialized for juxtacrine (directcontact) signaling, whereas type I NRGs are specialized for paracrine signaling. There is evidence that type III NRGs do in fact serve as juxtacrine signals in vivo. Schwann cells are the glia of the peripheral nerves. One well-known function of Schwann cells is to myelinate the axons of sensory and motor neurons, thereby dramatically speeding conduction of action potentials. In sensory neuron-Schwann cell cocultures, Schwann cells in contact with sensory neuron axons have a higher proliferation rate than Schwann cells not contacting axons and this growth-promoting activity is blocked by antibodies inhibiting NRG signaling [121,122]. That type III NRGs are an essential component of this contact-dependent signal is suggested by the profound depletion of Schwann cell populations in the type III NRG KO mice ([42]; see Table 2 and discussion above). A study of mechanisms regulating differentiation of Schwann cells from neural crest progenitors in a cell culture model has both demonstrated juxtacrine signaling by type III NRGs and shown that the consequences of signaling by membrane-bound type III can differ from the effects of signaling by soluble NRG [123]. In these experiments, NRG III-b3 was expressed in a small proportion of the cultured cells using a retroviral vector. As a control, green fluorescent protein (GFP) was expressed using the same vector in parallel cultures. Cells contacting the NRG expressing cells were positive for Schwann cell markers at a significantly higher frequency than cells contacting the GFP-expressing cells, demonstrating the juxtacrine signaling capability of type III NRG. Intriguingly, in similar cultures, soluble (recombinant) type III or type II NRG applied at high concentration was incapable of inducing expression of Schwann

cell markers. The proposed proprotein topology and mode of signaling for various isoforms is summarized in Fig. 3. Taken together, the current evidence argues for juxtacrine signaling by type III NRGs and paracrine signaling by types I and II.

Even for a kiss sent from a distance, the tingle can linger: prolongation of Ig-NRG’s effect by heparin The retention of type III NRGs in the membrane of type III expressing cells may not only limit the range of signaling, but also effectively concentrate the signal by confining it to the two-dimensional plane of the membrane. There is recent evidence for an alternative strategy of signal enhancement employed by Ig-NRGs. Each of the protein purification schemes by which NRGs were initially isolated employed a step of heparin chromatography, and each purified an Ig-NRG. In retrospect this is not surprising, as it was subsequently shown the Ig-like domain binds heparin and other highly charged glycosaminoglycans [124]. In contrast, CRD-NRGs do not bind heparin [79]. Glycosaminoglycans are the carbohydrate side chains of proteoglycans, proteins found in the extracellular matrix and on the surface of cells. The affinity of the NRG Ig-like domain for cell-surface and extracellular matrix proteoglycans may provide a mechanism for limiting diffusion of Ig-NRGs and/or creating extracellular reservoirs of NRGs. Ig-NRGs are deposited in the basal lamina of the neuromuscular synapse [125,126]. It has been proposed that Ig-NRGs become bound to the basal lamina by interaction of the Ig-like domain with basal lamina proteoglycans and that the bioactive EGF-like domain is subsequently freed from the matrix by a protease that cleaves the matrix bound N-terminal fragment between the EGF-like domain and the Ig-like domain [124,127]. A new twist to this story is evidence that binding of Ig-NRGs to the surface of cultured muscle through interaction with cell-surface proteoglycans enhances the potency of the Ig-NRG in inducing receptor phosphorylation compared to a recombinant form consisting only of the EGF-like domain. Furthermore, compared to the EGF-like domainonly form, the Ig-NRG induced a longer period of ErbB receptor phosphorylation and more effectively stimulated synthesis of acetylcholine receptors [128].

A tale of the heart (and of paracrine signaling, Ig-NRGs, the NRG cytoplasmic tail, and NRG trafficking) As noted above, mice genetically altered so that they produce no bioactive Ig-NRGs (Ig-NRG KOs) have the same cardiac phenotype as the pan-NRG KOs. Mice homozygous for NRG1 mutation that causes all transmembrane NRG1s (TMc-NRG1s) to have their tail truncated to


a length of only three amino acids (NRG1DCT/DCT mice) also have the same cardiac phenotype ([44]; see Table 2). However mice that produce no bioactive CRD-NRGs have not been reported to have cardiac defects. What is the underlying cell biology responsible for these results? Several lines of evidence can be woven together to construct an explanation. (1) The geometry of cardiac development is such that for normal cardiac morphogenesis, the endocardium must signal to cells in the presumptive myocardium with which it is not in direct contact. This requires a paracrine (diffusible) type of signal. As discussed above, the Ig-NRGs (Types I and II) may be specialized for paracrine signaling; whereas, the type III/CRD-NRGs may be poorly released from CRD-NRG producing cells and instead specialized for juxtacrine (direct-contact) signaling [77]. (2) Type I NRGs and low levels of type III NRGs are expressed by the endocardium during embryogenesis, but type II NRGs are not expressed by the embryonic myocardium ([39]; see also [78]). (3) While most released and transmembrane proteins have a classic N-terminal signal sequence that targets the nascent protein to the endoplasmic reticulum, none of the NRGs do. Instead various other sequences in NRG isoforms appear to serve as “cryptic”, noncleaved internal signals targeting nascent NRG proteins to the ER-Golgi-export pathway. The b3-NRGs lack the transmembrane domain C-terminal of the EGF-like domain (see Fig. 1). When expressed by transfection in fibroblastic cells, NRG II-b3 is effectively released into the medium [6] and NRG III-b3 is effectively trafficked to the cell surface [79], but NRG Ib3 is not released [2,6]. This suggests that the types II and III N-terminal sequences each contain an ER targeting signal, but that the type I N-terminal sequence does not. Both the type II and III N-terminal regions include a hydrophobic stretch of amino acids, and it is likely that this hydrophobic stretch is all or part of the signal. So does this mean that type I NRGs are a “dead” class of NRG isoforms? Indeed not! Unlike NRG I-b3, the transmembrane type I NRGs (i.e., I-b1a, I-b2a, and I-b4a) are effectively released. Since these type I NRGs with a transmembrane domain carboxyterminal of the EGF-like domain (“TMc-NRGs”) are identical with NRG I-b3 from the N-terminus through the EGF-like domain, the TM-NRGs must contain an export pathway targeting sequence downstream of the EGF-like domain. Likely the transmembrane domain serves as a signal-anchor sequence [80,81]. But it turns out that a substantial length of the NRG cytoplasmic tail is also required for trafficking of type I TMc-NRG to the cell surface and their release ([77]; see also [82]). Whether the cytoplasmic tail in conjunction with the TM domain is required for initial targeting type I NRGs to the ER, or—as is the case for TGFa [83–85]—for transport along the export pathway, is unclear. Now we have sufficient information on the table to synthesize an explanation for the similarity in the cardiac phenotypes of mice with all NRG1s inactivated, only Ig-NRG1s inactivated, and tail-truncated TMc-NRG1s. A paracrine NRG signal is required for endocardial induction of myo-

25

cardial differentiation, and type III/CRD-NRGs—though expressed at low levels by the myocardium—may not be suitable for paracrine signaling. Type II NRGs may be suitable, but they are not expressed. Type I NRGs are expressed, but type I NRGs without a cytoplasmic tail are not released: i.e., the NRG1DCT/DCT mice would—from today’s perspective—be expected to have exactly the same phenotype as a type I NRG null. In summary, what we have learned of NRG’s cell biology provides a satisfying explanation of the fact that both the Ig-NRG KO and NRG1DCT/DCT mice have the same cardiac phenotype as the pan-NRG1 KO. Do NRGs also have functions in the heart beyond early development? The cardiac toxicity of trastuzumab (Herceptin) suggests they do. Trastuzumab is used in the treatment of metastatic breast cancer. It is a humanized monoclonal antibody that binds to the NRG receptor ErbB2 (HER2) which is overexpressed in many breast cancers, and it not only reduces the ligand-independent activation of ErbB2 that occurs in cells highly expressing this protein or expressing mutated ErbB2 by down-regulating ErbB2 [1,86], but also blocks NRG activation of ErbB2/4 and ErbB3/4 heterodimers [1]. A fraction of patients treated with trastuzumab develop dilated cardiomyopathy, reflecting weakening of cardiac muscle contractility [87,88]. Most of the trastuzumab-treated patients that develop cardiomyopathy are also being treated with the chemotherapeutic agent anthracycline. Mouse models with a targeted mutation of ErbB2 affecting only the ventricular muscle [89,90] develop dilated cardiomyopathy closely resembling the pathology in trastuzumab-treated patients. Furthermore, in cell culture neuregulins promote survival and growth of cardiac myocytes, and protect them from anthracycline toxicity [91,92]. Thus, NRGs mediate critical signaling in the adult heart, as well as in the developing heart. In the adult, the endothelium of the cardiac microvasculature may be a source of the (paracrine) NRG signal [91]. Just as NRGs appear to mediate signaling in the adult heart, so NRGs are likely to mediate critical signaling events in the adult nervous system. For example, NRGs and their receptors are widely expressed in the postnatal nervous system [73,93–95], NRG expression in the brain is upregulated by activity [96], and NRGs can inhibit long-term potentiation (LTP), a model of learning [97]. Thus therapeutic strategies that involve perturbing NRG signaling, such as the use of trastuzumab, must take careful cognizance of the normal functions of NRGs in adult, as well as embryonic, organ systems.

Conclusion We certainly are just at the beginning of deciphering the functions of NRGs and the mechanisms by which the NRG signaling is shaped and modulated to achieve physiologically adaptive outcomes, but already it is clear the NRGs play critical roles in the functioning of a number of organ

26


systems, both during embryonic development and postnatally. Evidence that aberrations in NRG signaling contribute to the pathology of diseases such as schizophrenia and multiple sclerosis lend additional urgency to expanding our understanding of NRG biology. An appreciation of diversification of signaling through employment of combinations of receptors and variations in intracellular signaling cascades has grown over recent years. Now it seems clear that structural differences in NRG isoforms tailor them for different signaling strategies and requirements, providing considerable additional diversification upstream of receptor activation.

[7]

[8]

[9]

Acknowledgment [10]

Preparation of this article was supported by a grant to D.L.F. from the National Institutes of Health (GM56337). Note added in proof. A good entry point for access to the wealth of NRG isoform information freely available via the World Wide Web is LocusLink [170,171] (http:www. ncbi.nlm.nih.gov/LocusLink/). To begin accessing the LocusLink neuregulin information, on the LocusLink home page enter “neuregulin” in the query box (without quotation marks), and then click “Go.” This will take you to a page listing neuregulin loci. Note that not all species are included in the LocusLink database. For example, as of December 2002, the frog (Xenopus) and chicken (Gallus) NRG1 sequences are not referenced in LocusLink. The Mouse Genome Informatics database (MGI) can be accessed from LocusLink or directly (www.informatics.jax.org). Among many other things, this valuable compilation of NRG data includes a listing of each reported targeted mutations of the NRG1 gene, along with the official nomenclature for each mutation. References [1] Y. Yarden, M.X. Sliwkowski, Untangling the ErbB signaling network, Nat. Rev. Mol. Cell. Biol. 2 (2001) 127–137. [2] W.E. Holmes, M.X. Sliwkowski, R.W. Akita, W.J. Henzel, J. Lee, J.W. Park, D. Yansura, N. Abadi, H. Raab, G.D. Lewis, H.M. Shepard, W.-J. Kuang, W.I. Wood, D.V. Goeddel, R.L. Vandlen, Identification of heregulin, a specific activator of p185erbB2, Science 256 (1992) 1205–1210. [3] E. Peles, S.S. Bacus, R.A. Koski, H.S. Lu, D. Wen, S.G. Ogden, R.B. Levy, Y. Yarden, Isolation of the neu/HER-2 stimulatory ligand: a 44 kd glycoprotein that induces differentiation of mammary tumor cells, Cell 69 (1992) 205–216. [4] D. Wen, E. Peles, R. Cupples, S.V. Suggs, S.S. Bacus, Y. Luo, G. Trail, S. Hu, S.M. Silbiger, R.B. Levy, R.A. Koski, H.S. Lu, Y. Yarden, Neu differentiation factor: a transmembrane glycoprotein containing an EGF domain and an immunoglobulin homology unit, Cell 69 (1992) 559–572. [5] A.D.J. Goodearl, J.B. Davis, K. Mistry, L. Minghetti, M. Otsu, M.D. Waterfield, P. Stroobant, Purification of multiple forms of glial growth factor, J. Biol. Chem. 268 (1993) 18095–18102. [6] M.A. Marchionni, A.D.J. Goodearl, M.S. Chen, O. Bermingham-

[11]

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20] [21] [22]

[23] [24]

McDonogh, C. Kirk, M. Hendricks, F. Danehy, D. Misumi, J. Sudhalter, K. Kobayashi, D. Wroblewski, C. Lynch, M. Baldassare, I. Hiles, J.B. Davis, J.J. Hsuan, N.F. Totty, M. Otsu, R.N. McBurney, M.D. Waterfield, P. Stroobant, D. Gwynne, Glial growth factors are alternatively spliced erbB2 ligands expressed in the nervous system, Nature 362 (1993) 312–318. D.L. Falls, K.M. Rosen, G. Corfas, W.S. Lane, G.D. Fischbach, ARIA, a protein that stimulates acetylcholine receptor synthesis, is a member of the neu ligand family, Cell 72 (1993) 801–815. S.J. Busfield, D.A. Michnick, T.W. Chickering, T.L. Revett, J.Y. Ma, E.A. Woolf, C.A. Comrack, B.J. Dussault, J. Woolf, A.D.J. Goodearl, D.P. Gearing, Characterization of a neuregulin-related gene, don-1, that is highly expressed in restricted regions of the cerebellum and hippocampus, Mol. Cell. Biol. 17 (1997) 4007–4014. K.L. Carraway, J.L. Weber, M.J. Unger, J. Ledesma, N. Yu, M. Gassmann, C. Lai, Neuregulin-2, a new ligand of erbB3/erbB4receptor tyrosine kinases, Nature 387 (1997) 512–516. H. Chang, D.J. Riese, W. Gilbert, D.F. Stern, U.J. McMahan, Ligands for erbB-family receptors encoded by a neuregulin-like gene, Nature 387 (1997) 509–512. S. Higashiyama, M. Horikawa, K. Yamada, N. Ichino, N. Nakano, T. Nakagawa, J. Miyagawa, N. Matsushita, T. Nagatsu, N. Taniguchi, H. Ishiguro, A novel brain-derived member of the epidermal growth factor family that interacts with ErbB3 and ErbB4, J. Biochem. (Tokyo) 122 (1997) 675–680. D. Zhang, M.X. Sliwkowski, M. Mark, G. Frantz, R. Akita, Y. Sun, K. Hillan, C. Crowley, J. Brush, P.J. Godowski, Neuregulin-3 (NRG3): a novel neural tissue-enriched protein that binds and activates ErbB4, Proc. Natl. Acad. Sci. USA 94 (1997) 9562–9567. D. Harari, E. Tzahar, J. Romano, M. Shelly, J.H. Pierce, G.C. Andrews, Y. Yarden, Neuregulin-4: a novel growth factor that acts through the ErbB-4 receptor tyrosine kinase, Oncogene 18 (1999) 2681–2689. J.T. Jones, R.W. Akita, M.X. Sliwkowski, Binding specificities and affinities of egf domains for ErbB receptors, FEBS Lett. 447 (1999) 227–231. S.S. Hobbs, S.L. Coffing, A.T. Le, E.M. Cameron, E.E. Williams, M. Andrew, E.N. Blommel, R.P. Hammer, H. Chang, D.J. Riese 2nd, Neuregulin isoforms exhibit distinct patterns of ErbB family receptor activation, Oncogene 21 (2002) 8442–8452. M.A. Huotari, P.J. Miettinen, J. Palgi, T. Koivisto, J. Ustinov, D. Harari, Y. Yarden, T. Otonkoski, ErbB signaling regulates lineage determination of developing pancreatic islet cells in embryonic organ culture, Endocrinology 143 (2002) 4437–4446. A. Buonanno, G.D. Fischbach, Neuregulin and ErbB receptor signaling pathways in the nervous system, Curr. Opin. Neurobiol. 11 (2001) 287–296. C.S. Crovello, C. Lai, L.C. Cantley, K.L. Carraway 3rd, Differential signaling by the epidermal growth factor-like growth factors neuregulin-1 and neuregulin-2, J. Biol. Chem. 273 (1998) 26954–26961. C. Sweeney, C. Lai, D.J. Riese II, J. Diamonti, L.C. Cantley, K.L. Carraway III, Ligand discrimination in signaling through an ErbB4 receptor homodimer, J. Biol. Chem. 275 (2000) 19803–19807. G.D. Fischbach, K.M. Rosen, ARIA—a neuromuscular junction neuregulin, Annu. Rev. Neurosci. 20 (1997) 429–458. M. Gassmann, G. Lemke, Neuregulins and neuregulin receptors in neural development, Curr. Opin. Neurobiol. 7 (1997) 87–92. S. Burden, Y. Yarden, Neuregulins and their receptors: a versatile signaling module in organogenesis and oncogenesis, Neuron 18 (1997) 847–855. K. Adlkofer, C. Lai, Role of neuregulins in glial cell development, Glia 29 (2000) 104–111. A.N. Garratt, S. Britsch, C. Birchmeier, Neuregulin, a factor with many functions in the life of a schwann cell, Bioessays 22 (2000) 987–996.

Neuregulins: Functions, Forms, and Signaling Strategies [25] A.M. Davies, Neuronal survival: early dependence on Schwann cells, Curr. Biol. 8 (1998) R15–18. [26] S.J. Burden, The formation of neuromuscular synapses, Genes Dev. 12 (1998) 133–148. [27] J.R. Sanes, J.W. Lichtman, Induction, assembly, maturation and maintenance of a postsynaptic apparatus, Nat. Rev. Neurosci. 2 (2001) 791–805. [28] L. Schaeffer, A. de Kerchove d’Exaerde, J.P. Changeux, Targeting transcription to the neuromuscular synapse, Neuron 31 (2001) 15–22. [29] K.R. Jessen, R. Mirsky, Origin and early development of Schwann cells, Microsc. Res. Tech. 41 (1998) 393–402. [30] B.A. Barres, M.C. Raff, Axonal control of oligodendrocyte development, J. Cell. Biol. 147 (1998) 1123–1128. [31] K.R. Jessen, R. Mirsky, Schwann cells and their precursors emerge as major regulators of nerve development, Trends Neurosci. 22 (1999) 402–410. [32] G. Lemke, Glial control of neuronal development, Annu. Rev. Neurosci. 24 (2001) 87–105. [33] R. Mirsky, K.R. Jessen, A. Brennan, D. Parkinson, Z. Dong, C. Meier, E. Parmantier, D. Lawson, Schwann cells as regulators of nerve development, J. Physiol. Paris 96 (2002) 17–24. [34] H. Stefansson, E. Sigurdsson, V. Steinthorsdottir, S. Bjornsdottir, T. Sigmundsson, S. Ghosh, J. Brynjolfsson, S. Gunnarsdottir, O. Ivarsson, T.T. Chou, O. Hjaltason, B. Birgisdottir, H. Jonsson, V.G. Gudnadottir, E. Gudmundsdottir, A. Bjornsson, B. Ingvarsson, A. Ingason, S. Sigfusson, H. Hardardottir, R.P. Harvey, D. Lai, M. Zhou, D. Brunner, V. Mutel, A. Gonzalo, G. Lemke, J. Sainz, G. Johannesson, T. Andresson, D. Gudbjartsson, A. Manolescu, M.L. Frigge, M.E. Gurney, A. Kong, J.R. Gulcher, H. Petursson, K. Stefansson, Neuregulin 1 and susceptibility to schizophrenia, Am. J. Hum. Genet. 71 (2002) 877–892. [35] W.H. Ho, M.P. Armanini, A. Nuijens, H.S. Phillips, P.L. Osheroff, Sensory and motor neuron-derived factor—a novel heregulin variant highly expressed in sensory and motor neurons, J. Biol. Chem. 270 (1995) 14523–14532. [36] D. Meyer, C. Birchmeier, Multiple essential functions of neuregulin in development, Nature 378 (1995) 386–390. [37] R. Kramer, N. Bucay, D.J. Kane, L.E. Martin, J.E. Tarpley, L.E. Theill, Neuregulins with an Ig-like domain are essential for mouse myocardial and neuronal development, Proc. Natl. Acad. Sci. USA 93 (1996) 4833–4838. [38] S. Erickson, K. Shea, N. Ghaboosi, L. Loverro, G. Frantz, M. Bauer, L. Lu, M. Moore, ErbB3 is required for normal cerebellar and cardiac developmenta comparison with ErbB2- and heregulindeficient mice, Development 124 (1997) 4999–5011. [39] D. Meyer, T. Yamaai, A. Garratt, E. Riethmacher-Sonnenberg, D. Kane, L.E. Theill, C. Birchmeier, Isoform-specific expression and function of neuregulin, Development 124 (1997) 3575–3586. [40] S. Britsch, L. Li, S. Kirchhoff, F. Theuring, V. Brinkmann, C. Birchmeier, D. Riethmacher, The ErbB2 and ErbB3 receptors and their ligand, neuregulin-1, are essential for development of the sympathetic nervous system, Genes Dev. 12 (1998) 1825–1836. [41] R. Gerlai, P. Pisacane, S. Erickson, Heregulin, but not ErbB2 or ErbB3, heterozygous mutant mice exhibit hyperactivity in multiple behavioral tasks, Behav. Brain Res. 109 (2000) 219–227. [42] D. Wolpowitz, T.B. Mason, P. Dietrich, M. Mendelsohn, D.A. Talmage, L.W. Role, Cysteine-rich domain isoforms of the neuregulin-1 gene are required for maintenance of peripheral synapses, Neuron 25 (2000) 79–91. [43] S.K. Park, R. Miller, I. Krane, T. Vartanian, The erbB2 gene is required for the development of terminally differentiated spinal cord oligodendrocytes, J. Cell. Biol. 154 (2001) 1245–1258. [44] X. Liu, H. Hwang, L. Cao, M. Buckland, A. Cunningham, J. Chen, K.R. Chien, R.M. Graham, M. Zhou, Domain-specific gene disruption reveals critical regulation of neuregulin signaling by

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61] [62] [63]

27

its cytoplasmic tail, Proc. Natl. Acad. Sci. USA 95 (1998) 13024–13029. L. Li, S. Cleary, M.A. Mandarano, W. Long, C. Birchmeier, F.E. Jones, The breast proto-oncogene, HRGalpha regulates epithelial proliferation and lobuloalveolar development in the mouse mammary gland, Oncogene 21 (2002) 4900–4907. X. Yang, S. Arber, C. William, L. Li, Y. Tanabe, T.M. Jessell, C. Birchmeier, S.J. Burden, Patterning of muscle acetylcholine receptor gene expression in the absence of motor innervation, Neuron 30 (2001) 399–410. A.W. Sandrock, S.E. Dryer, K.M. Rosen, S.N. Gozani, R. Kramer, L.E. Theill, G.D. Fischbach, Maintenance of acetylcholine receptor number by neuregulins at the neuromuscular junction in vivo, Science 276 (1997) 599–603. T. Vartanian, G. Fischbach, R. Miller, Failure of spinal cord oligodendrocyte development in mice lacking neuregulin, Proc. Natl. Acad. Sci. USA 96 (1999) 731–735. M. Gassmann, F. Casagranda, D. Orioli, H. Simon, C. Lai, R. Klein, G. Lemke, Aberrant neural and cardiac development in mice lacking the erbb4 neuregulin receptor, Nature 378 (1995) 390–394. K.F. Lee, H. Simon, H. Chen, B. Bates, M.C. Hung, C. Hauser, Requirement for neuregulin receptor erbB2 in neural and cardiac development, Nature 378 (1995) 394–398. D. Riethmacher, E. Sonnenberg-Riethmacher, V. Brinkmann, T. Yamaai, G.R. Lewin, C. Birchmeier, Severe neuropathies in mice with targeted mutations in the ErbB3 receptor, Nature 389 (1997) 725–730. J.K. Morris, W. Lin, C. Hauser, Y. Marchuk, D. Getman, K.F. Lee, Rescue of the cardiac defect in ErbB2 mutant mice reveals essential roles of ErbB2 in peripheral nervous system development, Neuron 23 (1990) 273–283. M.T. Woldeyesus, S. Britsch, D. Riethmacher, L. Xu, E. SonnenbergRiethmacher, F. Abou-Rebyeh, R. Harvey, P. Caroni, C. Birchmeier, Peripheral nervous system defects in erbB2 mutants following genetic rescue of heart development, Genes Dev. 13 (1999) 2538–2548. A.N. Garratt, O. Voiculescu, P. Topilko, P. Charnay, C. Birchmeier, A dual role of erbB2 in myelination and in expansion of the Schwann cell precursor pool, J. Cell Biol. 148 (2000) 1035–1046. J.P. Golding, P. Trainor, R. Krumlauf, M. Gassmann, Defects in pathfinding by cranial neural crest cells in mice lacking the neuregulin receptor ErbB4, Nat. Cell. Biol. 2 (2000) 103–109. W. Lin, H.B. Sanchez, T. Deerinck, J.K. Morris, M. Ellisman, K.F. Lee, Aberrant development of motor axons and neuromuscular synapses in erbB2-deficient mice, Proc. Natl. Acad. Sci. USA 97 (2000) 1299–1304. R. Chan, W.R. Hardy, M.A. Laing, S.E. Hardy, W.J. Muller, The catalytic activity of the ErbB-2 receptor tyrosine kinase is essential for embryonic development, Mol. Cell. Biol. 22 (2002) 1073–1078. J.Y. Wang, K.E. Frenzel, D. Wen, D.L. Falls, Transmembrane neuregulins interact with LIM kinase 1, a cytoplasmic protein kinase implicated in development of visuospatial cognition, J. Biol. Chem. 273 (1998) 20525–20534. J. Bao, J. Gautier, L. Role, D. Talmage, Novel functions of the cytoplasmic domain of neuregulin, Soc. Neurosci. Abstr. 25 (1999) 303.9. M.L. Hancock, L. Role, D. Talmage, Neuregulin 1 intracellular domain-dependent signaling: requirements for intramembranous proteolysis and nuclear targeting, Soc. Neurosci. Abstr. 28 (2002) 822.11. D.G. Wilkinson, Multiple roles of EPH receptors and ephrins in neural development, Nat. Rev. Neurosci. 2 (2001) 155–164. Cloninger, C.R. The discovery of susceptibility genes for mental disorders, Proc. Natl. Acad. Sci. USA 99 (2002) 13365–13367. Stefansson, H., Sarginson, J., Kong, A., Yates, P., Steinthorsdottir, V., Gudfinnsson, E., Gunnarsdottir, S., Walker, N., Petursson, H.,

28

[64]

[65]

[66]

[67]

[68]

[69]

[70] [71] [72]

[73]

[74]

[75]

[76]

[77]

[78]

[79]

[80] [81]

[82]

The EGF Receptor Family Crombie, C., Ingason, A., Gulcher, J.R., Stefansson, K., Clair, D.S., Association of neuregulin 1 with schizophrenia confirmed in a Scottish population, Am. J. Hum. Genet. 72 (2003) 83–87. M. Ozaki, M. Sasner, R. Yano, H.S. Lu, A. Buonanno, Neuregulinbeta induces expression of an NMDA-receptor subunit, Nature 390 (1997) 691–694. Y. Hakak, J.R. Walker, C. Li, W.H. Wong, K.L. Davis, J.D. Buxbaum, V. Haroutunian, A.A. Fienberg, Genome-wide expression analysis reveals dysregulation of myelination-related genes in chronic schizophrenia, Proc. Natl. Acad. Sci. USA 98 (2001) 4746–4751. H.W. Moises, T. Zoega, I.I. Gottesman, The glial growth factors deficiency and synaptic destabilization hypothesis of schizophrenia. BMC Psychiatry 2 (2002) 8. J.T. Trachtenberg, W.J. Thompson, Nerve terminal withdrawal from rat neuromuscular junctions induced by neuregulin and Schwann cells, J. Neurosci. 17 (1997) 6243–6255. J.T. Trachtenberg, W.J. Thompson, Schwann cell apoptosis at developing neuromuscular junctions is regulated by glial growth factor, Nature 379 (1996) 174–177. R. Pinkas-Kramarski, R. Eilam, O. Spiegler, S. Lavi, N. Liu, D. Chang, D. Wen, M. Schwartz, Y. Yarden, Brain neurons and glial cells express Neu differentiation factor/heregulin: a survival factor for astrocytes, Proc. Natl. Acad. Sci. USA 91 (1994) 9387– 9391. E.M. Ullian, S.K. Sapperstein, K.S. Christopherson, B.A. Barres, Control of synapse number by glia, Science 291 (2001) 657–661. C. Rio, H.I. Rieff, P. Qi, G. Corfas, Neuregulin and erbB receptors play a critical role in neuronal migration, Neuron 19 (1997) 39–50. E.S. Anton, M.A. Marchionni, K.F. Lee, P. Rakic, Role of GGF/neuregulin signaling in interactions between migrating neurons and radial glia in the developing cerebral cortex, Development 124 (1997) 3501–3510. M.S. Chen, O. Bermingham-McDonogh, F.T. Danehy, C. Nolan, S.S. Scherer, J. Lucas, D. Gwynne, M.A. Marchionni, Expression of multiple neuregulin transcripts in postnatal rat brains, J. Comp. Neurol. 349 (1994) 389–400. H.I. Rieff, L.T. Raetzman, D.W. Sapp, H.H. Yeh, R.E. Siegel, G. Corfas, Neuregulin induces GABA(A) receptor subunit expression and neurite outgrowth in cerebellar granule cells, J. Neurosci. 19 (1999) 10757–10766. Y. Liu, B. Ford, M.A. Mann, G.D. Fischbach, Neuregulins increase alpha7 nicotinic acetylcholine receptors and enhance excitatory synaptic transmission in GABAergic interneurons of the hippocampus, J. Neurosci. 21 (2001) 5660–5669. X. Yang, Y. Kuo, P. Devay, C. Yu, L. Role, A cysteine-rich isoform of neuregulin controls the level of expression of neuronal nicotinic receptor channels during synaptogenesis, Neuron 20 (1998) 255–270. J.Y. Wang, S.J. Miller, D.L. Falls, The N-terminal region of neuregulin isoforms determines the accumulation of cell-surface and released neuregulin ectodomain, J. Biol. Chem. 276 (2001) 2841–2851. G. Corfas, K.M. Rosen, H. Aratake, R. Krauss, G.D. Fischbach, Differential expression of ARIA isoforms in the rat brain, Neuron 14 (1995) 103–115. A. Schroering, D.J. Carey, Sensory and motor neuron-derived factor is a transmembrane heregulin that is expressed on the plasma membrane with the active domain exposed to the extracellular environment, J. Biol. Chem. 273 (1998) 30643–30650. K.E. Matlack, W. Mothes, T.A. Rapoport, Protein translocation: tunnel vision, Cell 92 (1998) 381–390. K. Ota, M. Sakaguchi, G. von Heijne, N. Hamasaki, K. Mihara, Forced transmembrane orientation of hydrophilic polypeptide segments in multispanning membrane proteins, Mol. Cell. 2 (1998) 495–503. X. Liu, H. Hwang, L. Cao, D. Wen, N. Liu, R.M. Graham, M. Zhou,

[83]

[84]

[85]

[86]

[87]

[88]

[89]

[90]

[91]

[92]

[93]

[94]

[95]

[96]

[97]

[98]

[99]

Release of the neuregulin functional polypeptide requires its cytoplasmic tail, J. Biol. Chem. 273 (1998) 34335–34340. J.M. Urena, A. Merlos-Suarez, J. Baselga, J. Arribas, The cytoplasmic carboxy-terminal amino acid determines the subcellular localization of proTGF-(alpha) and membrane type matrix metalloprotease (MT1-MMP), J. Cell. Sci. 112 (1999) 773–784. A. Kuo, C. Zhong, W.S. Lane, R. Derynck, Transmembrane transforming growth factor-alpha tethers to the PDZ domain-containing, golgi membrane-associated protein, p59/GRASP55, EMBO J. 19 (2000) 6427–6439. J. Fernandez-Larrea, A. Merlos-Suarez, J.M. Urena, J. Baselga, J. Arribas, A role for a PDZ protein in the early secretory pathway for the targeting of proTGF-alpha to the cell surface, Mol. Cell. 3 (1999) 423–433. E. Penuel, R.W. Akita, M.X. Sliwkowski, Identification of a region within the ErbB2/HER2 intracellular domain that is necessary for ligand-independent association, J. Biol. Chem. 277 (2002) 28468–28473. J.W. Schneider, A.Y. Chang, A. Garratt, Trastuzumab cardiotoxicity: Speculations regarding pathophysiology and targets for further study, Semin. Oncol. 29 (2002) 22–28. J.W. Schneider, A.Y. Chang, T.P. Rocco, Cardiotoxicity in signal transduction therapeutics: erbB2 antibodies and the heart, Semin. Oncol. 28 (2001) 18–26. C. Ozcelik, B. Erdmann, B. Pilz, N. Wettschureck, S. Britsch, N. Hubner, K.R. Chen, C. Birchmeier, A.N. Garratt, Conditional mutation of the ErbB2 (HER2) receptor in cardiomyocytes leads to dilated cardiomyopathy, Proc. Natl. Acad. Sci. USA 99 (2002) 8880–8885. S.A. Crone, Y.Y. Zhao, L. Fan, Y. Gu, S. Minamisawa, Y. Liu, K.L. Peterson, J. Chen, R. Kahn, G. Condorelli, J. Ross Jr., K.R. Chien, K.F. Lee, ErbB2 is essential in the prevention of dilated cardiomyopathy, Nat. Med. 8 (2002) 459–465. Y. Zhao, D.R. Sawyer, R.R. Baliga, D.J. Opel, X. Han, M.A. Marchionni, R.A. Kelly, Neuregulins promote survival and growth of cardiac myocytes. Persistence of erbb2 and erbb4 expression in neonatal and adult ventricular myocytes, J. Biol. Chem. 273 (1998) 10261–10269. D.B. Sawyer, C. Zuppinger, T.A. Miller, H.M. Eppenberger, T.M. Suter, Modulation of anthracycline-induced myofibrillar disarray in rat ventricular myocytes by neuregulin-1beta and anti-erbB2: potential mechanism for trastuzumab-induced cardiotoxicity, Circulation 105 (2002) 1551–1554. O. Bermingham-McDonogh, Y.T. Xu, M.A. Marchionni, S.S. Scherer, Neuregulin expression in PNS neurons: isoforms and regulation by target interactions, Mol. Cell. Neurosci. 10 (1997) 184–195. S.L. Carroll, M.L. Miller, P.W. Frohnert, S.S. Kim, J.A. Corbett, Expression of neuregulins and their putative receptors, erbb2 and erbb3, is induced during wallerian degeneration, J. Neurosci. 17 (1997) 1642–1659. K.M. Gerecke, J.M. Wyss, I. Karavanova, A. Buonanno, S.L. Carroll, Erbb transmembrane tyrosine kinase receptors are differentially expressed throughout the adult rat central nervous system, J. Comp. Neurol. 433 (2001) 86–100. R. Eilam, R. Pinkas-Kramarski, B.J. Ratzkin, M. Segal, Y. Yarden, Activity-dependent regulation of neu differentiation factor/ neuregulin expression in rat brain, Proc. Natl. Acad. Sci. USA 95 (1998) 1888–1893. Y.Z. Huang, S. Won, D.W. Ali, Q. Wang, M. Tanowitz, Q.S. Du, K.A. Pelkey, D.J. Yang, W.C. Xiong, M.W. Salter, L. Mei, Regulation of neuregulin signaling by PSD-95 interacting with ErbB4 at CNS synapses, Neuron 26 (2000) 443–455. M.A. Olayioye, R.M. Neve, H.A. Lane, N.E. Hynes, The ErbB signaling network: receptor heterodimerization in development and cancer, EMBO J. 19 (2000) 3159–3167. R. Pinkas-Kramarski, M. Shelly, B.C. Guarino, L.M. Wang, L. Lyass, I. Alroy, M. Alimandi, A. Kuo, J.D. Moyer, S. Lavi, M.


[100]

[101]

[102] [103]

[104] [105]

[106] [107]

[108]

[109]

[110]

[111]

[112]

[113]

[114]

[115]

[116]

Eisenstein, B.J. Ratzkin, R. Seger, S.S. Bacus, J.H. Pierce, G.C. Andrews, Y. Yarden, M. Alamandi, ErbB tyrosine kinases and the two neuregulin families constitute a ligand-receptor network [published errata, including important error re switched labeling of NRG2-a and NRG2-b, appear in Mol. Cell. Biol. 1998 Dec; 18(12): 7602 and 1999 Dec;19(12):8695], Mol. Cell. Biol. 18 (1998) 6090–6101. R. Pinkas-Kramarski, L. Soussan, H. Waterman, G. Levkowitz, I. Alroy, L. Klapper, S. Lavi, R. Seger, B.J. Ratzkin, M. Sela, Y. Yarden, Diversification of neu differentiation factor and epidermal growth factor signaling by combinatorial receptor interactions, EMBO J. 15 (1996) 2452–2467. J. Massagué, Transforming Growth Factor-a: a model for membrane-anchored growth factors, J. Biol. Chem. 265 (1990) 21393–21396. R.A. Black, J.M. White, ADAMs: focus on the protease domain, Curr. Opin. Cell. Biol. 10 (1998) 654–659. J.J. Peschon, J.L. Slack, P. Reddy, K.L. Stocking, S.W. Sunnarborg, D.C. Lee, W.E. Russell, B.J. Castner, R.S. Johnson, J.N. Fitzner, R.W. Boyce, N. Nelson, C.J. Kozlosky, M.F. Wolfson, C.T. Rauch, D.P. Cerretti, R.J. Paxton, C.J. March, R.A. Black, An essential role for ectodomain shedding in mammalian development, Science 282 (1998) 1281–1284. Z. Werb, Y. Yan, A cellular striptease act [comment], Science 282 (1998) 1279–1280. A.J. Turner, N.M. Hooper, Role for ADAM-family proteinases as membrane protein secretases, Biochem. Soc. Trans. 27 (1999) 255–259. C.P. Blobel, Remarkable roles of proteolysis on and beyond the cell surface, Curr. Opin. Cell. Biol. 12 (2000) 606–612. P. Primakoff, D.G. Myles, The ADAM gene family: surface proteins with adhesion and protease activity, Trends Genet. 16 (2000) 83–87. D.Z. Wen, S.V. Suggs, D. Karunagaran, N.L. Liu, R.L. Cupples, Y. Luo, A.M. Janssen, N. Benbaruch, D.B. Trollinger, V.L. Jacobsen, S.Y. Meng, H.S. Lu, S. Hu, D. Chang, W.N. Yang, D. Yanigahara, R.A. Koski, Y. Yarden, Structural and functional aspects of the multiplicity of neu differentiation factors, Mol. Cell. Biol. 14 (1994) 1909–1919. T.L. Burgess, S.L. Ross, Y.X. Qian, D. Brankow, S. Hu, Biosynthetic processing of neu differentiation factor—glycosylation, trafficking, and regulated cleavage from the cell surface, J. Biol. Chem. 270 (1995) 19188–19196. J.A. Loeb, E.T. Susanto, G.D. Fischbach, The neuregulin precursor proARIA is processed to ARIA after expression on the cell surface by a protein kinase C-enhanced mechanism, Mol. Cell. Neurosci. 11 (1998) 77–91. B. Han, G.D. Fischbach, The release of acetylcholine receptor inducing activity (ARIA) from its transmembrane precursor in transfected fibroblasts, J. Biol. Chem. 274 (1999) 26407–26415. E. Diaz-Rodriguez, A. Esparis-Ogando, J.C. Montero, L. Yuste, A. Pandiella, Stimulation of cleavage of membrane proteins by calmodulin inhibitors, Biochem. J. 346 (Pt 2) (2000) 359–367. H.S. Lu, S. Hara, L. Wong, M.D. Jones, V. Katta, G. Trail, A.H. Zou, D. Brankow, S. Cole, S. Hu, D.Z. Wen, Post-translational processing of membrane-associated neu differentiation factor proisoforms expressed in mammalian cells, J. Biol. Chem. 270 (1995) 4775–4783. M.A. Avila, J.A. Velasco, C. Cho, R. Lupu, D.Z. Wen, V. Notario, Hyperactive autocrine loop mediated by a Ndf-related factor in neoplastic hamster embryo fibroblasts expressing an activated Cph oncogene, Oncogene 10 (1995) 963–971. M. Alimandi, A. Romano, M.C. Curia, R. Muraro, P. Fedi, S.A. Aaronson, P.P. Difiore, M.H. Kraus, Cooperative signaling of Erbb3 and Erbb2 in neoplastic transformation and human mammary carcinomas, Oncogene 10 (1995) 1813–1821. S.W. Sunnarborg, C.L. Hinkle, M. Stevenson, W.E. Russell, C.S.

[117]

[118]

[119]

[120]

[121]

[122]

[123]

[124]

[125]

[126]

[127]

[128]

[129]

[130]

[131]

[132]

[133]

[134]

29

Raska, J.J. Peschon, B.J. Castner, M.J. Gerhart, R.J. Paxton, R.A. Black, D.C. Lee, Tumor necrosis factor-alpha converting enzyme (TACE) regulates epidermal growth factor receptor ligand availability, J. Biol. Chem. 277 (2002) 12838–12845. K. Shirakabe, S. Wakatsuki, T. Kurisaki, A. Fujisawa-Sehara, Roles of meltrin beta/ADAM19 in the processing of neuregulin, J. Biol. Chem. 276 (2001) 9352. J.C. Montero, L. Yuste, E. Diaz-Rodriguez, A. Esparis-Ogando, A. Pandiella, Differential shedding of transmembrane neuregulin isoforms by the tumor necrosis factor-alpha-converting enzyme, Mol. Cell. Neurosci. 16 (2000) 631–648. R.L. Cagan, H. Kramer, A.C. Hart, S.L. Zipursky, The bride of sevenless and sevenless interaction: internalization of a transmembrane ligand, Cell 69 (1992) 393–399. Hart, A.C., Kramer, H., Van, D.L. Vactor, Jr, Paidhungat, M. Zipursky, S.L. Induction of cell fate in the Drosophila retina: the bride of sevenless protein is predicted to contain a large extracellular domain and seven transmembrane segments. Genes Dev. 4 (1990) 1835–1847. J.L. Salzer, R.P. Bunge, L. Glaser, Studies of Schwann cell proliferation. III. Evidence for the surface localization of the neurite mitogen, J. Cell Biol. 84 (1980) 767–778. T.K. Morrissey, A. Levi, A. Nuijens, M.X. Sliwkowski, R.P. Bunge, Axon-induced mitogenesis of human schwann cells involves heregulin and p185 (erbb2), Proc. Natl. Acad. Sci. USA 92 (1995) 1431–1435. R. Leimeroth, C. Lobsiger, A. Lussi, V. Taylor, U. Suter, L. Sommer, Membrane-bound neuregulin 1 type III actively promotes Schwann cell differentiation of multipotent progenitor cells, Dev. Biol. 246 (2002) 245–258. J.A. Loeb, G.D. Fischbach, Aria can be released from extracellular matrix through cleavage of heparin-binding domain, J. Cell Biol. 130 (1995) 127–135. A.D.J. Goodearl, A.G. Yee, A.W. Sandrock, G. Corfas, G.D. Fischbach, Aria is concentrated in the synaptic basal lamina of the developing chick neuromuscular junction, J. Cell Biol. 130 (1995) 1423–1434. S.A. Jo, X.J. Zhu, M.A. Marchionni, S.J. Burden, Neuregulins are concentrated at nerve-muscle synapses and activate Ach-receptor gene expression, Nature 373 (1995) 158–161. J.A. Loeb, T.S. Khurana, J.T. Robbins, A.G. Yee, G.D. Fischbach, Expression patterns of transmembrane and released forms of neuregulin during spinal cord and neuromuscular synapse development, Development 126 (1999) 781–791. Q. Li, J.A. Loeb, Neuregulin-heparan-sulfate proteoglycan interactions produce sustained erbB receptor activation required for the induction of acetylcholine receptors in muscle, J. Biol. Chem. 276 (2001) 38068–38075. Colognato, H., Baron, W., Avellana-Adalid, V., Relvas, J.B., Evercooren, A.B., Georges-Labouesse, E., Ffrench-Constant, C. CNS integrins switch growth factor signalling to promote targetdependent survival, Nat. Cell Biol. 4 (2002) 833–841. A.I. Flores, B.S. Mallon, T. Matsul, W. Ogawa, A. Rosenzweig, T. Okamoto, W.B. Macklin, Akt-mediated survival of oligodendrocytes induced by neuregulins, J. Neurosci. 20 (2000) 7622–7630. P.A. Fernandez, D.G. Tang, L. Cheng, A. Prochiantz, A.W. Mudge, M.C. Raff, Evidence that axon-derived neuregulin promotes oligodendrocyte survival in the developing rat optic nerve [in process citation], Neuron 28 (2000) 81–90. P.D. Canoll, R. Kraemer, K.K. Teng, M.A. Marchionni, J.L. Salzer, GGF/neuregulin induces a phenotypic reversion of oligodendrocytes, Mol. Cell. Neurosci. 13 (1999) 79–94. P.D. Canoll, J.M. Musacchio, R. Hardy, R. Reynolds, M.A. Marchionni, J.L. Salzer, Ggf/neuregulin is a neuronal signal that promotes the proliferation and survival and inhibits the differentiation of oligodendrocyte progenitors, Neuron 17 (1996) 229–243. V. Calaora, B. Rogister, K. Bismuth, K. Murray, H. Brandt, P.

30

[135]

[136]

[137]

[138]

[139]

[140]

[141] [142]

[143]

[144]

[145]

[146]

[147]

[148] [149]

[150]

[151]

The EGF Receptor Family Leprince, M. Marchionni, M. Dubois-Dalcq, Neuregulin signaling regulates neural precursor growth and the generation of oligodendrocytes in vitro, J. Neurosci. 21 (2001) 4740–4751. Y.J. Ma, D.F. Hill, K.E. Creswick, M.E. Costa, A. Cornea, M.N. Lioubin, G.D. Plowman, S.R. Ojeda, Neuregulins signaling via a glial erbB-2-erbB-4 receptor complex contribute to the neuroendocrine control of mammalian sexual development, J. Neurosci. 19 (1999) 9913–9927. C. Paratore, D.E. Goerich, U. Suter, M. Wegner, L. Sommer, Survival and glial fate acquisition of neural crest cells are regulated by an interplay between the transcription factor Sox10 and extrinsic combinatorial signaling, Development 128 (2001) 3949–3961. T.D. Camenisch, J.A. Schroeder, J. Bradley, S.E. Klewer, J.A. McDonald, Heart-valve mesenchyme formation is dependent on hyaluronan-augmented activation of ErbB2 ErbB3 receptors, Nat. Med. 8 (2002) 850–855. S. Rentschler, J. Zander, K. Meyers, D. France, R. Levine, G. Porter, S.A. Rivkees, G.E. Morley, G.I. Fishman, Neuregulin-1 promotes formation of the murine cardiac conduction system, Proc. Natl. Acad. Sci. USA 99 (2002) 10464–10469. S. Rohrbach, X. Yan, E.O. Weinberg, F. Hasan, J. Bartunek, M.A. Marchionni, B.H. Lorell, Neuregulin in cardiac hypertrophy in rats with aortic stenosis. Differential expression of erbB2 and erbB4 receptors, Circulation 100 (1999) 407–412. D. Kim, S. Chi, K.H. Lee, S. Rhee, Y.K. Kwon, C.H. Chung, H. Kwon, M.S. Kang, Neuregulin stimulates myogenic differentiation in an autocrine manner, J. Biol. Chem. 274 (1999) 15395– 15400. J.T. Trachtenberg, Fiber apoptosis in developing rat muscles is regulated by activity, neuregulin, Dev. Biol. 196 (1998) 193–203. N.V. Patel, M.J. Acarregui, J.M. Snyder, J.M. Klein, M.X. Sliwkowski, J.A. Kern, Neuregulin-1 and human epidermal growth factor receptors 2 and 3 play a role in human lung development in vitro, Am. J. Respir. Cell. Mol. Biol. 22 (2000) 432–440. K.S. Russell, D.F. Stern, P.J. Polverini, J.R. Bender, Neuregulin activation of ErbB receptors in vascular endothelium leads to angiogenesis, Am. J. Physiol. 277 (1999) H2205–H2211. E. Suarez, D. Bach, J. Cadefau, M. Palacin, A. Zorzano, A. Guma, A novel role of neuregulin in skeletal muscle. Neuregulin stimulates glucose uptake, glucose transporter translocation, and transporter expression in muscle cells, J. Biol. Chem. 276 (2001) 18257–18264. A.L. Kierszenbaum, L.L. Tres, Primordial germ cell-somatic cell partnership: A balancing cell signaling act, Mol. Reprod. Dev. 60 (2001) 277–280. D.L. Falls, Neuregulin-ErbB signaling: roles in regulating neurotransmitter receptor synthesis and development of the neuromuscular junction. In G. Adelman, B.H. Smith, (Eds.), Encyclopedia of Neuroscience, 3rd Edition. (2003) Elsevier. A. Viehover, R.H. Miller, S.K. Park, G. Fischbach, T. Vartanian, Neuregulin: an oligodendrocyte growth factor absent in active multiple sclerosis lesions, Dev. Neurosci. 23 (2001) 377–386. Deleted in proof. B. Cannella, C.J. Hoban, Y.L. Gao, R. Garcia-Arenas, D. Lawson, M. Marchionni, D. Gwynne, C.S. Raine, The neuregulin, glial growth factor 2, diminishes autoimmune demyelination and enhances remyelination in a chronic relapsing model for multiple sclerosis, Proc. Natl. Acad. Sci. USA 95 (1998) 10100–10105. B. Cannella, D. Pitt, M. Marchionni, C.S. Raine, Neuregulin and erbB receptor expression in normal and diseased human white matter, J. Neuroimmunol. 100 (1999) 233–242. M.A. Marchionni, B. Cannella, C. Hoban, Y.L. Gao, R. GarciaArenas, D. Lawson, E. Happel, F. Noel, P. Tofilon, D. Gwynne, C.S. Raine, Neuregulin in neuron/glial interactions in the central nervous system. GGF2 diminishes autoimmune demyelination, promotes oligodendrocyte progenitor expansion, and enhances remyelination, Adv. Exp. Med. Biol. 468 (199••) 283–295.

[152] M. Dubois-Dalcq, K. Murray, Why are growth factors important in oligodendrocyte physiology?, Pathol. Biol. (Paris) 48 (2000) 80–86. [153] S.K. Park, D. Solomon, T. Vartanian, Growth factor control of CNS myelination, Dev. Neurosci. 23 (2001) 327–337. [154] H. Yan, M.B. Bunge, P.M. Wood, G.W. Plant, Mitogenic response of adult rat olfactory ensheathing glia to four growth factors, Glia 33 (2001) 334–342. [155] D.J. Bryan, A.H. Holway, K.K. Wang, A.E. Silva, D.J. Trantolo, D. Wise, I.C. Summerhayes, Influence of glial growth factor and Schwann cells in a bioresorbable guidance channel on peripheral nerve regeneration, Tissue Eng. 6 (2000) 129–138. [156] A.V. Boruch, J.J. Conners, M. Pipitone, G. Deadwyler, P.D. Storer, G.H. Devries, K.J. Jones, Neurotrophic and migratory properties of an olfactory ensheathing cell line, Glia 33 (2001) 225–229. [157] J.L. Zheng, G. Frantz, A.K. Lewis, M. Sliwkowski, W.Q. Gao, Heregulin enhances regenerative proliferation in postnatal rat utricular sensory epithelium after ototoxic damage, J. Neurocytol. 28 (1999) 901–912. [158] M.P. ter Laak, F.P. Hamers, C.J. Kirk, W.H. Gispen, rhGGF2 protects against cisplatin-induced neuropathy in the rat, J. Neurosci. Res. 60 (2000) 237–244. [159] Y. Tokita, H. Keino, F. Matsui, S. Aono, H. Ishiguro, S. Higashiyama, A. Oohira, Regulation of neuregulin expression in the injured rat brain and cultured astrocytes, J. Neurosci. 21 (2001) 1257–1264. [160] R. Bagheri-Yarmand, R.K. Vadlamudi, R.A. Wang, J. Mendelsohn, R. Kumar, Vascular endothelial growth factor up-regulation via p21-activated kinase-1 signaling regulates heregulin-beta1-mediated angiogenesis, J. Biol. Chem. 275 (2000) 39451–39457. [161] L. Yen, X.L. You, A.E. Al Moustafa, G. Batist, N.E. Hynes, S. Mader, S. Meloche, M.A. Alaoui-Jamali, Heregulin selectively upregulates vascular endothelial growth factor secretion in cancer cells and stimulates angiogenesis, Oncogene 19 (2000) 3460–3469. [162] I.M. Krane, P. Leder, Ndf/heregulin induces persistence of terminal end buds and adenocarcinomas in the mammary glands of transgenic mice, Oncogene 12 (1996) 1781–1788. [163] V.R. Schelfhout, E.D. Coene, B. Delaey, S. Thys, D.L. Page, C.R. De Potter, Pathogenesis of Paget’s disease: epidermal heregulinalpha, motility factor, and the HER receptor family, J. Natl. Cancer Inst. 92 (2000) 622–628. [164] V.R. Schelfhout, E.D. Coene, B. Delaey, A.A. Waeytens, L. De Rycke, M. Deleu, C.R. De Potter, The role of heregulin-alpha as a motility factor and amphiregulin as a growth factor in wound healing, J. Pathol. 198 (2002) 523–533. [165] L. Wang, M.A. Marchionni, R.A. Tassava, Cloning and neuronal expression of a type III newt neuregulin and rescue of denervated, nerve-dependent newt limb blastemas by rhGGF2 [in process citation], J. Neurobiol. 43 (2000) 150–158. [166] G. Schaefer, V.D. Fitzpatrick, M.X. Sliwkowski, Gamma-heregulin: a novel heregulin isoform that is an autocrine growth factor for the human breast cancer cell line, MDA-MB-175, 175, Oncogene 15, (1997) 1385–1394. [167] X. Liu, E. Baker, H.J. Eyre, G.R. Sutherland, M. Zhou, Gammaheregulin: a fusion gene of DOC-4 and neuregulin-1 derived from a chromosome translocation, Oncogene 18 (1999) 7110–7114. [168] E.A. Sanchez-Valdivieso, J.J. Cruz, R. Salazar, M. del Mar Abad, A. Gomez-Alonso, A. Gomez, R. Gonzalez-Sarmiento, Gamma-heregulin has no biological significance in primary breast cancer, Br. J. Cancer 86 (2002) 1362–1363. [169] D.L. Brutsaert, Cardiac endothelial-myocardial signaling: its role in cardiac growth, contractile performance, and rhythmicity, Physiol. Rev. 83 (2003) 59–115. [170] K.D. Pruitt, K.S. Katz, H. Sicotte, D.R. Maglott, Introducing RefSeq and LocusLink: curated human genome resources at the NCBI, Trends Genet. 16 (2000) 44–47.

Neuregulins: Functions, Forms, and Signaling Strategies [171] K.D. Pruitt, D.R. Maglott, RefSeq and LocusLink: NCBI genecentered resources, Nucleic Acids Res. 29 (2001) 137–140. [172] S. Hippenmeyer, N.A. Schneider, C. Birchmeier, S.J. Burden, T.M. Jessell, S. Arber, A role for neuregulin1 signaling in muscle spindle differentiation, Neuron 36 (2002) 1035–1049. [173] V. Prevot, C. Rio, G.J. Cho, A. Lomniczi, S. Heger, C.M. Neville, N.A. Rosenthal, S.R. Ojeda, G. Corfas, Normal female sexual development requires neuregulin-erbB receptor signaling in hypothalamic astrocytes, J. Neurosci. 23 (2003) 230–239. [174] S. Britsch, D.E. Goerich, D. Riethmacher, R.I. Peirano, M. Rossner, K.A. Nave, C. Birchmeier, M. Wegner, The transcription factor Sox10 is a key regulator of peripheral glial development, Genes Dev. 15 (2001) 66–78. [175] S.A. Crone, A. Negro, A. Trumpp, M. Giovannini, K.F. Lee, Colonic epithelial expression of ErbB2 is required for postnatal maintenance of the enteric nervous system, Neuron 37 (2003) 29–40.

31

[176] J.S. Cameron, L. Dryer, S.E. Dryer, beta-Neuregulin-1 is required for the in vivo development of functional Ca2+-activated K+ channels in parasympathetic neurons, Proc. Natl. Acad. Sci. USA 98 (2001) 2832–2836. [177] H. Toyoda-Ohno, M. Obinata, Y. Matsui, Members of the ErbB receptor tyrosine kinases are involved in germ cell development in fetal mouse gonads, Dev. Biol. 215 (1999) 399–406. [178] H. Noguchi, C. Sakamoto, K. Wada, T. Akamatsu, T. Uchida, A. Tatsuguchi, H. Matsui, H. Fukui, T. Fujimori, M. Kasuga, Expression of heregulin alpha, erbB2, and erbB3 and their influences on proliferation of gastric epithelial cells, Gastroenterology 117 (1999) 1119–1127. [179] S. O’Shea, K. Johnson, R. Clark, M.X. Sliwkowski, S.L. Erickson, Effects of in vivo heregulin beta1 treatment in wild-type and ErbB gene-targeted mice depend on receptor levels and pregnancy, Am. J. Pathol. 158 (2001) 1871–1880.

Epidermal growth factor receptor: mechanisms of activation and signalling Robert N. Jorissen,a,c Francesca Walker,a,c Normand Pouliot,a Thomas P.J. Garrett,b,c Colin W. Ward,c,d and Antony W. Burgessa,c a

Ludwig Institute for Cancer Research, Parkville, Victoria, Australia Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia c Cooperative Research Centre for Cellular Growth Factors, Post Office Royal Melbourne Hospital, Parkville, Victoria 3050, Australia d CSIRO Health Sciences and Nutrition, 343 Royal Parade, Parkville, Victoria 3052, Australia b

Abstract The epidermal growth factor (EGF) receptor (EGFR) is one of four homologous transmembrane proteins that mediate the actions of a family of growth factors including EGF, transforming growth factor-a, and the neuregulins. We review the structure and function of the EGFR, from ligand binding to the initiation of intracellular signalling pathways that lead to changes in the biochemical state of the cell. The recent crystal structures of different domains from several members of the EGFR family have challenged our concepts of these processes.

Introduction The epidermal growth factor receptor (EGFR) regulates the intracellular effects of ligands such as EGF and transforming growth factor-a (TGFa) [1–3]. For many years it has been known that upon ligand binding to the EGFR extracellular domains (collectively called the ectodomain), there is an increase in the proportion of dimerized receptor and the enzymatic activity of its intracellular tyrosine kinase domain increases greatly [4–6]. The EGFR kinase catalyses the transfer of the g-phosphate of bound ATP to the tyrosine residues of exogenous substrates and the C-terminal domains of the EGFR, the latter in a trans manner [7,8]. After the induction of tyrosine phosphorylation, some signalling pathways appear to start with the recognition of the C-terminal phosphotyrosines by appropriate adaptor or signalling molecules [9,10]. The binding of ligand and activation of the EGFR kinase also induces the migration of EGFR from the caveolae/raft component of the cell membrane to the bulk membrane component [11] and the clustering of EGFR complexes into clathrin-coated pits that are subsequently internalized ([12–14]; see also Wiley et al., this issue) in a kinasedependent manner [15,16]. Abnormal expression and/or mutation of the EGFR has been implicated in the progression of some classes of solid tumours (see Hynes article in this issue).

The EGFR also interacts with its three known homologues, ErbB2 (also called Neu or HER2), ErbB3 (HER3), and ErbB4 (HER4), in a ligand-dependent fashion to form heterodimers [3,17]. Differences in the C-terminal domains of these proteins results in changes to the repertoire of signalling molecules that interact with the heterodimers, thus leading to an expansion in the number of possible signalling pathways stimulated by a single ligand. The strength and duration of intracellular signalling from the EGFR are also controlled by internalization and recycling of the receptor, which can be modulated by heterodimerization at the cell surface and by association with intracellular signalling molecules; these aspects of EGFR behavior, relating to its trafficking in cells, are reviewed elsewhere in this issue (Wiley et al.). The mechanism of the activation of the EGFR has been studied for many years; however, much remains to be determined. Significant progress has recently occurred; the crystal structures of extracellular portions of two ErbB family members and of the EGFR kinase domain have been reported [18–21]. Some of these structures reveal the modes of ligand binding, ectodomain dimerization, and the conformation of the apo-kinase domain. Full elucidation of the mechanisms of behaviour of both wild-type and oncogenic mutants of the EGFR should help with the design of new molecules to antagonize the action of the mutant or overexpressed receptor in cancer.

34


Fig. 1. Schematic representation of domains of the epidermal growth factor receptor sequence. The abbreviations used: L and CR, for the ligandbinding and the cysteine-rich domains [also known as I(L1), II(CR1), III(L2), and IV(CR2) or S1(CR1) and S2(CR1), where L and S refer to large and small]; JM and CT, juxtamembrane domain and carboxy-terminal terminus. The transmembrane domain (residues 622–644) is between the CR2 and the juxtamembrane domains.

Architecture of the EGFR The EGFR is synthesized from a 1210-residue polypeptide precursor; after cleavage of the N-terminal sequence, an 1186-residue protein is inserted into the cell membrane [22]. Over 20% of the receptor’s 170-kDa mass is N-linked glycosylation and this is required for translocation of the EGFR to the cell surface and subsequent acquisition of function [23]; overexpression of the EGFR or altered glycosylation can reveal peptide epitopes suitable for antibody therapies [24]. The sequence can be categorized into a number of domains as shown in Fig. 1. The sequence identity of the EGFR family varies from 37% (53% similarity) for the EGFR and ErbB3 to 49% (64% similarity) for the EGFR and ErbB2. The amino acids identities can also vary significantly among the domains with the tyrosine kinase domains having the highest sequence identities (average 59–81% identity) and the carboxy-terminal domains having the lowest (average 12–30% identity). The three-dimensional folds of corresponding domains of the different EGFR homologues are expected to be similar with the possible exception of the heavily divergent C-terminal domains. The EGFR extracellular portion (or ectodomain) consists of four domains that we refer to as the L1, CR1, L2, and CR2 domains (Fig. 1). The structure determinations of ectodomain fragments of the EGFR and ErbB3 show the L1 and L2 domains to consist of so-called b-solenoid or b-helix folds, which resemble the corresponding domains of the IGF-1 receptor [25]. Ligand binds between the L1 and L2 domains of the EGFR [18,19]. The orientation of the L1 and L2 domains of the unligated ErbB3 structure have been well defined and clearly must reorient to create the ligandbinding pocket. The CR1 and CR2 domains consist of a number of small modules, each appearing to be held together by one or two disulfide bonds. A large loop that protrudes from the back of the CR1 domain makes contact with the CR1 domain of the other receptor in the dimer [18,19]. The first module of the CR1 and CR2 domains contain conserved tryptophan residues (Trp 176 and Trp 492) that intercalate between the fourth and fifth helical turns of the b-helical L domain and sit in a hydrophobic environment that includes other conserved tryptophan residues (Trp 140 and Trp 453). An EGFR construct consisting of residues 1–476 lacks this second tryptophan interaction and does not

bind ligand with high affinity [26]. Residues 557–617 in the CR2 domain are considered sufficient to target more than half of EGFR to the caveolae/raft component of the cell membrane prior to ligand binding [27]. The original assignment of the transmembrane domain as residues 622–644 was performed by visual analysis of the EGFR sequence [22], but other prediction methods indicate variation in the assignment of the boundaries of the transmembrane domains [28]. Nuclear magnetic resonance analysis of a peptide corresponding to the EGFR transmembrane and beginning of the cytoplasmic domain indicate that residues 626–647 are a-helical [28], suggesting that the transmembrane a-helix continues into the juxtamembrane domain. The juxtamembrane region appears to have a number of regulatory functions, i.e., downregulation and ligand-dependent internalization events [29], basolateral sorting of the EGFR in polarized cells [30], and association with proteins such as eps8 [31] and calmodulin (which is competitive with protein kinase C [PKC] association) [32,33]. The experimental three-dimensional structure of the EGFR kinase domain is similar to other tyrosine kinases [21]. By analogy with protein kinase structures, the ATP sits between the N-terminal lobe (dominated by a b-sheet) and the larger C-terminal lobe (mainly a-helical). The structure of the insulin receptor kinase with an ATP-analogue and small substrate peptide bound shows that the g-phosphate group is positioned to be transferred to the acceptor tyrosine residue of the substrate [34]. The carboxy-terminal domain of the EGFR contains tyrosine residues where phosphorylation modulates EGFR-mediated signal transduction. There are also several serine/threonine residues (and another tyrosine residue) where phosphorylation has been inferred to be important for the receptor downregulation processes and sequences thought to be necessary for endocytosis. Residues 984–996 in the C-terminus have been identified as a binding site for actin [35] and may well be involved in the formation of higher order receptor oligomers and/or receptor clustering after ligand activation of the kinase domain. A number of EGFR mutants have been observed in tumours where gene amplification has occurred (Table 1; reviewed by Kuan et al. [36]). The best characterized EGFR mutant is the D2–7 truncation (or vIII), in which amino acids encoded by exons 2–7 of the receptor (residues 6–273) are missing. This receptor mutant is constituently active and has defective downregulation behaviour [37]. Other EGFR mutants have deletions, regions of sequence duplication (summarized in Table 1) or defective kinase regulatory signals. A soluble 105-kDa ectodomain fragment of the EGFR is produced by the A431 carcinoma cell line [38]. A secreted 80-kDa EGFR fragment that corresponds to part of the ectodomain of the full-length receptor has been observed to be produced in the placenta and in ovarian cancer [39,40]. Transcripts from other ectodomain fragments of human,

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling Table 1 Mutations of the EGFR detected in tumour cells [36]; novel residues that occur at the splice sites are not showna Type

Alteration in sequence

EGFR vI EGFR vII EGFR vIII EGFR vIII/D12–13 EGFR vIV EGFR vV EGFR.TDM/2–7 EGFR.TDM/18–25 EGFR.TDM/18–26

Translation starts at aa 543 Deletion of aa 521–603 Deletion of aa 6–273 Deletions of aa 6–273 and 409–520 Deletion of aa 959–1030 Truncation at residue 958 Tandem duplication of 6–273 Tandem duplication of 664–1030 Tandem duplication of 664–1014

a

EGFR, epidermal growth factor receptor; aa, amino acid(s).

chicken, and rat EGFRs have also been detected ([41] and references therein).

Ligand binding to the EGFR The three-dimensional structures of the EGF- and TGFabound EGFR ectodomain fragments show that EGF and TGFa bind to the EGFR in the same mode [18,19]. Each bound ligand interacts with the L1 and L2 domains of a given EGFR molecule Fig. 2). The conserved EGF residue Arg 41 (Arg 42 in TGFa) makes bidentate hydrogen bonds with Asp 355. Arg 41 is surrounded by Tyr 13 and Leu 15 (Phe 15 and Phe 17, respectively, in TGFa), orienting the arginine residue and shielding the salt bridge interaction from water molecules. Tyr 13 also interacts with Phe 357 of the receptor (Fig. 2A). The sidechain of Gln 384 of the EGFR makes two hydrogen bonds to the EGF mainchain atoms Gln43 O and Arg45 N (Glu 44 O and Ala 44 N, respectively, of TGFa) (Fig. 2A). The sidechain of Leu 47 (Leu 48 in TGFa) projects into a hydrophobic pocket consisting of Leu 382, Phe 412, and Ile 438 with the sidechain of Ala 415 at its base (Fig. 2A). The EGFR L1 residues Gln 16 and Gly 18 contribute three mainchain-to-mainchain hydrogen bonds to Cys 31 and Cys 33 of EGF (Cys 32 and Cys 34 in TGFa), thus extending the larger of the two ligand b-sheets into the receptor (colored green in Fig. 2A). The sidechain of the EGFR residue Asn 12 also makes a hydrogen bond with the mainchain nitrogen atom of Gly 40 of TGFa [18] (Fig. 2A). The aliphatic sidechains of Ile 23 of EGF and Leu 24 of TGFa interact with the receptor sidechain of Leu 14 (Fig. 2A). There are some compensating differences that distinguish the binding of individual ligands. A salt bridge between TGFa residue Glu 27 and Arg 125 of the receptor is not replicated in the EGF-bound receptor [18] (Fig. 2B). The corresponding EGF residue is Leu 26, which sits in a similar position and interacts with Leu 14, Leu 69, Leu 98, and Ser

35

99 [19]. Clearly, both acidic or aliphatic residues can be accommodated in this ligand position, as is the case for the other known ligands of the EGFR [42,43]. Examination of the ligand-bound EGF and TGFa structures suggests that in addition to the conserved cysteine residues, Gly 18, Gly 39, and Tyr 38 (that can be replaced by Phe) are required to form or maintain the ligand conformation. Other ligand residues are also conserved or semiconserved across the EGF family and support the notion that all ligands adopt the same folding and mode of binding as that for EGF and TGFa. EGF residue Arg 41 is completely conserved and its proximal residues Tyr 13 and Tyr 15 can be replaced with residues with aromatic and aromatic/ aliphatic residues, respectively. Ile 23 of corresponds to aliphatic residues in all other ErbB ligands except for the weakly binding ligand, epigen [44]. EGFR residues Leu 14, Glu 355, and Phe 357 are conserved in all four of the ErbBs and residues Gln 384 and Asn 12 are conserved in ErbB3 and ErbB4. Conservation of these residues also supports the notion that ligands for ErbB3 and ErbB4 have the same mode of binding as that employed by EGF and TGFa to the EGFR. The EGF residue Leu 47 is conserved among the ligands of the ErbB family except for amphiregulin, where the corresponding residue is methionine. The binding affinity of amphiregulin is several orders of magnitude less than that of EGF [45,46]. The predicted binding sites in ErbB3 and ErbB4 for the neuregulin residues that correspond to EGF residue Leu 47 are hydrophobic in nature, but appear to lack the defined pocket present in the EGFR due to the substitution of EGFR residue Ala 415 with Leu 412 and Leu 415 for ErbB3 and ErbB4, respectively. Interestingly, the bforms of the neuregulins have an aliphatic residue equivalent to Leu 47 in EGF, whereas the corresponding residue for the neuregulin a-forms is proline [47,48]. As there are no other consistent sequence trends among the a- and bforms of the neuregulins, the identity of this residue appears to be a major determinant of affinity. The major ligand binding domain of the EGFR appears to be the L2 domain. The proteolytically generated fragment of the EGFR that contains the L2 domain and small regions of the CR1 and CR2 domains bind EGF and TGFa with submicromolar affinity [49,50]. Chimeras of the EGFR and ErbB4 show that the L2 domain is the major binding determinant [51]. In contrast to the EGFR, the L1 domain of ErbB4 appears to confer the preference for NRG1b over EGF [51]. A proteolytically generated fragment of ErbB3 that binds NRG1b with 68 nM affinity consists of the L1 and most of the CR1 domain (residues 1–270). Ligand binding protects further cleavage at position 50 in the L1 domain [52]. The identity of the residues in the N-termini of the ErbB ligands’ EGF domains appear to determine whether the ligand is able to bind to ErbB3 and ErbB4 [47,53–55]. The N-terminus of EGF and TGFa bind to the L1 domain of the EGFR [18,19]. Comparison of the chemical shift data from nuclear magnetic resonance experiments performed on

36


the NRG1a EGF domain and an NRG1a chimera in which its N-terminus is substituted with that of TGFa shows that the chimera has altered sidechain packing in the region of the mutation, possibly rendering this chimera unable to bind to ErbB4 [54,56]. The contributions of the different regions of the ligand for binding to the EGFR, ErbB3, or ErbB4 accounts for the ability of betacellulin to bind to both the EGFR and the ErbB4 with high affinity [47,57]. EGF binds to both low affinity (KD = 1–2 nM) and high affinity (KD = 10–50 pM) sites on cells that express the EGFR [58]. The precise nature of the origin of the two affinities has yet to be determined; however, many studies have linked the apparent high affinity sites to the presence of receptor dimers [59–62]. Truncation of the long CR1 domain loop that mediates dimerization abolishes the apparent high affinity binding population [18]. Examination of the ligand-bound EGFR ectodomain dimer indicates that opening up the receptor binding pocket to release bound ligand is less likely to occur in the dimer than in the monomer. The high affinity binding of full-length EGFR on cells appears to be also modulated by interactions between its intracellular domains and other intracellular proteins or mutations of the tyrosine kinase domain [63–67]. The affinity of the EGFR is not necessarily a property of the receptor alone. It has been proposed that an intracellular protein mediates formation of the high affinity binding site of the EGFR [68,69]. The soluble ectodomain of the EGFR (residues 1–621) binds ligand and dimerizes to form a 2 : 2 complex [50,70,71]. The affinity for binding of EGF and TGFa to the soluble ectodomain of the EGFR is 100–500 nM ([50] and references therein). This affinity is comparable to the affinity of ligand binding to a proteolytically generated EGFR fragment that contains all of the L2 domain and only small portions of the surrounding CR1 and CR2 domains [49,50]. Surprisingly, removal of most of the CR2 domain from the EGFR 1–621, to produce EGFR 1–501, increases the binding affinity to 13–21 and 35–40 nM for EGF and TGFa, respectively [26]. A three-dimensional structure for the ErbB3 ectodomain reveals a conformation that excludes the possibility of ligand interacting with the L1 and L2 domains simultaneously [20]. High affinity binding (1–20 nM) has been detected for the EGFR ectodomain [71], but this is yet to be fully explained; while this could be due to a small proportion of preformed dimer, it is also possible that a small proportion of the receptor fails to form the “inactive conformation” seen in the crystal structure of ErbB3.

Ligand-induced EGFR oligomerization The 2 : 2 ligand-EGFR complex forms on the cell surface [72]. Ligated EGFR ectodomain fragments undergo a novel mode of receptor dimerization [18,19]; a loop from the back of the CR1 domain from one receptor molecule interacts with a pocket at the base of the CR1 loop in the partner

EGFR (Fig. 3). There are also some minor contacts between the CR1 loop and the L1 and L2 domains of the partner receptor. This interface participates in the formation of the physiological active dimer on the cell surface [18,19]. Key residues in the interface are conserved or substituted with residues that are expected to retain the dimer interactions (e.g., Phe Æ Tyr substitution where the phenol hydroxyl group is not used), indicating that the mode of binding is plausible for all of the ErbB proteins. Superimposition of the L2 and CR2 domains from the ErbB3 ectodomain [20] onto the L2 and CR2 domains of the ligated sEGFR501 dimer indicates that the CR2 domains are likely to project to the same position of the cell surface (Fig. 3). The closeness of the C-terminal ends of the CR2 domains is consistent with ability to cross-link the receptors by the addition of a cysteine residue in the CR2 domain close to the transmembrane domain [61,73]. The positions of the superimposed CR2 domains are consistent with the concept that the long loops of the two CR2 domains (residues 572–582) in the EGFR dimer interact [74], but definitive evidence for this interaction has yet to be reported. Localization of the EGFR to the cell membrane increases the effective concentration of the receptor, thus enhancing receptor dimerization relative to the soluble receptor ectodomain [50,75]. More than half of the unstimulated EGFRs on the cell surface are considered to be concentrated in caveolae, which account for approximately 5–10% of the membrane [11], thus further facilitating dimerization. A recombinant form of the EGFR, consisting of only the transmembrane and kinase domains, is capable of selfassociation [76]; thus, the transmembrane and kinase domains have active roles in stabilizing the dimer. Indeed the presence of the transmembrane domain enhances ligandinduced dimer formation in solution [77]. The formation of heterodimers of the ErbB family in solution is less well characterized than the formation of the EGFR homodimer. Most notably, the stoichiometry of the ligands and receptors in the heterodimer complexes is unknown. ErbB2 is the preferred interacting partner for the EGFR [78,79]. This interaction has been reported to reduce the rate of EGFR degradation [80]. It has been suggested that so-called heterodimers may actually be heterotetramers, possibly organized around a nucleating ErbB homodimer [81,82]. The formation of secondary hetero-oligomers can be induced by a ligand for a third ErbB protein. For example, EGF stimulates the formation of ErbB2–ErbB3 heterooligomers in cells that also express the EGFR [81,83]. Johannessen et al. [84] have reported constitutive EGFRErbB2 association, although exposure to EGF increased the phosphorylation of the ErbB2 residue Tyr 1248. Heterooligomers involving the EGFR and cell surface receptors outside of the ErbB family, such as complexes involving the EGFR and the platelet-derived growth factor (PDGF) receptor, have also been reported [85]. Such heterocomplexes may be mediated by interactions with intracellular adaptors and/or scaffolding systems [68].

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling

Fig. 2. Interactions between the epidermal growth factor (EGF) receptor (yellow) L1 and L2 domains with bound transforming growth factor-a (TGFa) (metallic blue). Residues that contribute to a b-sheet that involves both the receptor and the ligand are colored green. The sidechains of selected ligand and receptor residues are shown as sticks. Selected hydrogen bonds are represented as dotted lines and mainchain atoms involved in these interactions are not rendered. (A) Ca worm representation of TGFa bound to the EGF receptor residue 1–501 [18] with a number of key interacting residues displayed. Carbon atoms of the sidechains of the EGF receptor are colored grey to increase their visibility. (B) Detail of interaction between TGFa residue Glu 27 with the EGF receptor L1 domain. For comparison, the EGF receptor L1 domain (colored cyan) and EGF (dark pink) [19] are shown as superimposed on the TGFa-bound EGF receptor structure. The sidechain of EGF residue corresponding to TGFa Glu 27, Leu 26, is also rendered to illustrate how this residue can be accommodated by the receptor. This figure and Fig. 3 were created by using the programs Molscript [292] and Raster3D [293].

37

kinase configuration [73,86]. At present, no other regions of the ectodomain have been directly implicated to control receptor reorientation on ligand binding. Comparison of the structures of the first three domains of the EGFR and ErbB3 show that significant rearrangements of the ectodomain can occur as changes in the CR1 domain, altering the relative positions of the L1 and L2 domains from those of the ligand-bound structure [18–20]. Most notably, a change of the conformation of the C-terminal end of the CR1 domain alters the juxtaposition of the L2 domain with respect to the preceding two domains. From Fig. 3, it can be envisaged that changing the angle between the CR1 and L2 domains, while conserving the CR1-CR1 interface, completely alters the L1–L2 juxtaposition and consequently the juxtaposition of the succeeding domains. Thus, the role of ligand binding may be to appropriately orient the L1, CR1, and L2 domains, which, in turn, position the CR2 domains and the intracellular domains of the EGFR. In this scenario, the tyrosine kinase domains are correctly positioned to enable their activation. Removal of all of the ectodomain, or residues 6–273 of the ectodomain (e.g., the D2–7 mutant [88–90]), results in a constitutively active complex [37,91–93]. Indeed, the D2–7 mutant of the EGFR has been reported to be constitutively dimerized and to have tyrosine kinase activity similar to the ligand-bound wild-type receptor [93]. While this indicates that the ectodomain, and in particular its first two domains, plays an active role in preventing kinase activation, it is unclear as to the mechanism of this inhibition.

Ligand-induced activation of the EGFR In the absence of ligand binding, the EGFR exists on cells as both monomers and dimers [72,73,86,87]. Yet ligand binding to the EGFR kinase is required to elevate the receptor’s tyrosine kinase activity. The position-dependent effects of adding a cysteine residue in the membrane-proximal part of the EGFR’s extracellular region suggests that a ligand-associated orientation of the EGF dimer is required for activation of the tyrosine kinase domains [73]. Clearly, dimerization of the EGFR, while necessary, is not sufficient to activate the intracellular kinase. Moriki et al. [73] demonstrated that it was possible to form cross-linked dimers of the EGFR by adding cysteinecontaining insertions of nine residues to the membraneproximal region of the EGFR (at position 618). These EGFR dimers form in both the absence and the presence of ligand. There are position-dependent preferences for formation of dimers cross-linked by these cysteine residues. The results of this study are consistent with a previously proposed model, referred to as the rotation-twist model, in which ligand-binding induces the predimerized EGFR to twist about a pivot point near or in the transmembrane domain and reorients the intracellular domains to form an active

Fig. 3. Transforming growth factor-a (TGFa)-bound EGFR501 dimer with superimposed CR2 domains of ErbB3 [18–20]. Each of the proteins and protein domains are rendered as Ca worms. The EGFR501 molecules are colored red and blue with their CR2 domains colored with darker tones; the bound TGFa molecules are colored yellow and dark purple. The CR2 domains of ErbB3 were superimposed onto the epidermal growth factor (EGF) receptor fragments by using the Ca atoms of the first module of the CR2 domains of each. The two ErbB3 CR2 domains are colored orange and green except for residues 572–582, which are colored in darker tones.

38


The structure of the ErbB3 ectodomain [20] offers a view of the unliganded EGFR in its monomeric form. In the ErbB3 structure, there is an intriguing twist that allows Tyr 246 in the CR1 domain to make hydrogen bonds to the sidechains of Asp 562 and Lys 583 in the CR2 domain and hydrophobic interactions with Pro 571, Val 574, and Ile 581 also in the CR2 domain. (The equivalent residues in the EGFR are Tyr 246, Asp 563, Lys 585, Val 575, and Leu 582, respectively.) This conformation would prevent the ligand binding to both the L1 and L2 domains simultaneously and the presence of the CR1–CR2 interaction would prevent formation of the back-to-back dimer. The ErbB3 conformation is likely to place the second module of the CR1 domain (Cys 191–Cys 207 of the EGFR) close to the cell membrane and so its orientation would be completely different to that observed in the ligand-bound EGFR structure. The minimum requirement for dimerization has been shown to be membrane-bound kinase domain itself [76]. In the absence of the ectodomain, the transmembrane-kinase form of the receptor is constitutively active. The binding of ligand to the ectodomain releases the extracellular restraints on the formation of an active kinase dimer configuration. The EGFR does not require the tyrosine kinase domain to be catalytically competent in order to dimerize [76]. The monomeric EGFR has much reduced kinase activity compared to the dimerized receptor [59,61,94,95]; it is assumed that in the absence of dimerization, the kinase is in an inactive conformation. Interestingly, deletion studies have identified the tyrosine kinase domain residues 835–918 as being necessary for formation of the dimer in the absence of ligand [87]. While many tyrosine kinases require phosphorylation of the activation loop for full enzymatic activity [96], the EGFR does not appear to be regulated at this level. Mutation of Tyr 845, the only tyrosine residue in the EGFR’s activation loop, to phenylalanine does not alter the protein’s kinase or autophosphorylation activities [97]. In the crystal structure, the conformation of the activation loop of the EGFR kinase in its apo-state (and also with an ATPcompetitive inhibitor bound) exhibits some similarity to the phosphorylated Lck and insulin receptor kinases in the conformations of the activation loops, catalytic residues, and relative orientation of the two lobes [21,96,98]. The apo- and inhibitor-bound EGFR kinase crystal structures also show that the orientations of the two subdomains of the kinase resemble those of crystal structures of the two active tyrosine kinases. The binding of 4-anilinoquinazolines to the EGFR can induce receptor dimerization independent of ligand [99–101]. Therefore, the crystal structure conformation of the EGFR kinase complexed with the inhibitor should resemble the conformation of the kinase in the ligand-bound EGFR dimer. Analysis of the sets of crystal contacts in the EGFR kinase structures [21] shows that the crystallographic interface with the most protein-protein contacts is between a region in the N-terminal subdomain of one copy of the

EGFR kinase, including the C-helix, and a C-terminal subdomain region of a second kinase molecule, which includes its H-helix (results not shown). Interestingly, this arrangement of EGFR kinase molecules is similar to a previously proposed model of the EGFR kinase dimer [65]. This model may explain the reduction in the kinase activity of the EGFR mutant Tyr 740 Æ Phe; Tyr 740 is a solvent-exposed residue in the C-helix of the kinase [65,66]. The interface contains a cluster of hydrophobic residues that are largely conserved across the EGFR family and also the conserved residue Gln 911 whose sidechain makes two hydrogen bonds to the mainchain of the other kinase molecule. Although the structures of the EGFR kinase do not suggest how the EGFR is inactivated, movement of the C-helix is thought to feature in the regulation of activity of a number of protein kinases [102]. Measurements of the kinetics of stimulated and unstimulated EGFR showed that ligand binding doubles the Vmax parameter and decreases the Km parameter for ATP by 10-fold [103]. We propose a variation on the mechanism of activation suggested by Ge et al. [103], i.e., ligand binding increases the proportion of dimerized EGFR and the reorientation of the kinase domains in a way that increases the affinity for ATP binding, probably due to conformational change, thereby enhancing the kinase activity.

Molecular targets perturbed by the activation of the EGFR The EGFR exerts its function in the cellular environment mainly, if not exclusively, via its tyrosine kinase activity. Tyrosine phosphorylation of cellular substrates is thus the first and crucial step in transducing EGFR-mediated signals. It is often difficult to determine whether a protein, phosphorylated in response to cellular stimulation with EGF, is a direct substrate of the EGFR kinase or it is phosphorylated following EGFR-dependent activation of other cellular kinases. Given the propensity of EGFR to heterodimerize with, and activate, other members of the EGFR family [104], even direct phosphorylation in in vitro kinase assays can be confounded by the presence of heterodimers, making it difficult to unequivocally assign substrates of the EGFR kinase. For many of the proteins identified as belonging to EGFinitiated signal transduction pathways, the question of direct or indirect phosphorylation is still unresolved. One of the few phosphoproteins that are undoubtedly direct substrates of the EGFR kinase is the EGFR itself, although in a cellular context, EGFR phosphorylation and signalling can also occur through ErbB dimerization partners (see, for example, Deb et al. [105] and Ewald et al. [106]), or by activation of intracellular tyrosine kinases such as Src and JAK-2. The EGFR is autophosphorylated on five C-terminal tyrosines, most likely in an intermolecular reaction through a dimerization partner. The putative role of autophosphorylation in the maintenance of the activated state is described elsewhere; here, we will address the role of EGFR

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling Table 2 Signalling proteins that associate directly with the EGFR, their function, and preferred docking sites on the EGFRa Protein

Function

Docking sites on EGFR

Reference

GRB-2 Nck Crk Shc Dok-R PLC-g

Adaptor Adaptor Adaptor Adaptor Adaptor Phospholipase

[284] [285] [286] [123] [287] [173]

P120RasGAP PTB-1B SHP-1 Src Abl

Ras attenuator Phosphatase Phosphatase Tyrosine kinase Tyrosine kinase

pY1068, pY1086 ND ND pY1148, pY1173 pY1086, pY1148 pY1173 (N-SH2) pY992 (C-SH2) ND pY992, pY1148 pY1173 pY891, pY920 pY1086

a

[288] [289] [290] [119] [291]

EGFR, epidermal growth factor receptor; ND, not determined.

phosphorylation sites in the formation of signalling complexes, and the phosphorylation-dependent activation of major intracellular signalling pathways.

Physical association between EGFR and signalling proteins Phosphorylation of the EGFR’s C-terminus, be it autophosphorylation or transphosphorylation by other kinases such as Src and Jak-2 [97,107], provides specific docking sites for the SH2 or PTB domains of intracellular signal transducers and adaptors, leading to their colocalization and to the assembly of multicomponent signalling “particles.” Signalling proteins that associate directly with the EGFR in this manner, and the EGFR tyrosines that mediate the association, are listed in Table 2. The association of other proteins with the phosphorylated EGFR is thought to be indirect (e.g., Cbl [108], PI3K-C2b [109], and Stat5b [110]), while the mode of EGFR binding for proteins such as Eps8 and Eps-15 is still unclear. A third mode of recruitment to the EGFR occurs via the C-terminal phosphorylation sites of heterodimer partners; the sequence divergence between EGFR family members at the C-terminus allows different proteins to preferentially associate with specific EGFR heterodimer complexes, greatly enhancing the multiplicity of signals that can emanate from the set of EGFR homo- and heterodimers. This is exemplified by the p85 subunit of PI3-K, which preferentially associates with the YXXM motifs in the ErbB3 C-terminus rather than with the EGFR itself [111]. SH2 or PTB domain-mediated association of intracellular proteins with the EGFR, whether direct or indirect, is inducible and determined by the phosphorylation state of key tyrosine residues on the receptor. However, there are some proteins that are associated with the EGFR in its resting state, only to be activated or translocated to other

39

cellular locations when ligand binds. This is the case for the zinc-binding protein ZPR-1 [112] and STAT transcription factors [113,114]. The physical association of EGFR and signalling or adaptor proteins greatly increases the efficiency of substrate phosphorylation, as well as aiding in the assembly of spatially organized multicomponent signalling complexes. It must be emphasized, however, that autophosphorylation of the EGFR is not a prerequisite for EGFR signalling; apparently normal signalling is stimulated by C-terminally truncated EGFRs expressed alone [115] or in combination with other EGFR family members [116–118]. Interestingly, an intact EGFR C-terminus has been reported to be critical for signalling initiated by amphiregulin but not EGF [118]. The ability to dispense with EGFR C-terminal phosphorylation sites is still puzzling, in view of the abundance and specificity of receptor-protein interactions mediated by the Cterminus. One could speculate that the assembly of signalling complexes can still occur in the absence of the EGFR C-terminal scaffold by association with other molecules (which in turn provide the docking sites) or by using alternative modules for binding to signalling proteins. The binding of the p85 subunit of PI3-K to the EGFR provides an example of both these modes of action. Normally EGFdependent association of p85 with the heterodimeric complex EGFR/ErbB3 occurs via ErbB3. In this way p85 is brought in close contact with the kinase domain of the EGFR and is phosphorylated. However, direct association between p85 and the EGFR can also occur via the EGFR pY920, which is located within the receptor’s kinase domain and is phosphorylated by the cytosolic kinase Src [119], thus bypassing the requirement for either ErbB3 or EGFR C-terminal association sites. Similarly, phosphorylation of the transcription factor STAT5b appears to be dependent on phosphorylation of the EGFR by Src on Y845 [110].

Signalling pathways activated by the EGFR Given the functional diversity of proteins that complex with, or are phosphorylated by, the EGFR, it is hardly surprising that EGF stimulation of a cell results in the simultaneous activation of multiple pathways. These pathways are often functionally interlinked and ideally should not be considered in isolation; however, for the sake of simplicity we will discuss them individually and in particular attempt to describe the earliest steps of their EGFR-mediated activation. Shc, Grb2, and the Ras/MAPK pathway The cascade of biochemical events that leads from the EGFR to the activation of the proto-oncogene Ras and, eventually, of the serine/threonine kinase MAPK has been analyzed extensively. The key player in EGF-dependent Ras activation is the adaptor protein Grb2 [120]. Grb2 is

40


constitutively bound to the Ras exchange factor Sos and is normally localized to the cytosol. Following activation of the EGFR kinase and autophosphorylation, the SH2 domain of Grb2 can bind to the EGFR. It must be noted that Grb2 can associate with the receptor either directly (via Y1068 and Y1086 [121]) or indirectly, by binding to EGFR-associated, tyrosine phosphorylated Shc [122]. It has been suggested that association of Shc to EGFR via its PTB domain, leading to its tyrosine phosphorylation and to the recruitment of Grb2, is the main step in EGF-dependent induction of the Ras/MAPK pathway [123]. However, in a different cellular system, Hashimoto et al. [124] have shown that Shc is not necessary for Ras activation by the EGFR; it is therefore still unclear whether the two modes of recruitment of Grb2 to the receptor have different functional roles or whether the predominance of one over the other is cell-type specific. In either case, relocation of the Grb2/Sos complex to the receptor at the plasma membrane facilitates the interaction of membrane-associated Ras with Sos, resulting in the exchange of Ras-bound GDP for GTP and hence in Ras activation. Activated Ras in turn activates the serine/threonine kinase Raf-1 [125]. Raf-1 activation, through a series of intermediate kinases, leads to the phosphorylation, activation, and nuclear translocation of Erk-1 and Erk-2, which catalyze the phosphorylation of nuclear transcription factors [126]. Activation of the MAP kinases also provides a negative feedback loop for this pathway since the GrB2-Sos complex is dissociated following MAPK phosphorylation of Sos [127]. This very simple outline of signalling downstream of Ras hides an incredible complexity of cross-talk between signalling pathways, feedback loops, protein relocalization, and signalling complex formation, which are beyond the scope of this article, but that have been addressed in recent reviews [128–130]. Both Grb2 and Shc play important roles in the activation of other EGFR-dependent pathways. This is due to their “modular” construction. Grb2 contains two SH2 domains and one SH3 domain, which enables it to interact with tyrosine-phosphorylated motifs as well as with proline-rich regions of other proteins (see, for example, Meisner and Czech [131]). Shc can associate with specific tyrosinephosphorylated sequences via its SH2 and PTB domain, and, being itself phosphorylated on tyrosine by activated receptors and cytosolic tyrosine kinases, serves in turn as a binding partner for SH2-containing proteins. SH2 and SH3 domains recognize specific sequences preferentially, but not exclusively; thus, they can bind to many proteins with different affinity. For example, Grb2 has been shown to complex with proteins involved in cytoskeletal reorganization, such as FAK and dynamin [132] with negative regulators of growth factor action such as Cb1 [108], Dab-2 [133], and SOCS-1 [134], and with the inositol phosphatase SHIP [135]. Shc has also has been detected in complexes with many other proteins [136], including MEKK-1, which links it to JNK pathway activation [137], and cadherin, implying a role for Shc in cell-cell adhesion [138].

The existence of interactions between Shc and Grb2 with the EGFR, with each other, and with a subset of cellular proteins raises the questions of how interactions are controlled: Do all possible interactions occur in a single cell and, if so, does the activation of one pathway influence the activation of alternative pathways? SH2- and SH3-mediated protein interactions are dependent on both the affinity and the relative concentration of the binding partner; high affinity interactions will be favoured at low concentrations of the target molecule, but could be displaced by low affinity interactions driven by high concentrations of alternative partners. There are many ways in which protein association patterns can vary between cell types, or within the same cell depending on the stimulus and the timing of the stimulation. Analyses of EGFR-associated signalling pathways often utilize different cell lines, and the cell type-specific levels of expression of binding partners for Grb2 or Shc may bias the detectable associations. Furthermore, when analyzing transformed cell lines it must be expected that the complexes will be different from those detected in resting cells or in cells being stimulated (e.g., during wound healing or antigenic responses). These caveats also apply to overexpression experiments, which must be interpreted with caution. It is also important to recognize that the “local” abundance of a protein may determine its availability for binding; colocalization of binding partners (e.g., at the plasma membrane) will favour interactions even when the affinity is low. It is well established that many signalling proteins relocalize within the cell following stimulation with growth factors; presumably this relocalization plays a significant role in controlling the timing and compartmentalization of protein-protein interactions. Finally, posttranslational modifications of proteins may alter the affinity of specific interactions (as is the case for Sos and Grb2), and allow alternative complexes to form. Recently, interactions between proteins have been studied directly in cells using GFP/YFP fusion proteins and FRET analysis [139]. Provided the levels of the transfected proteins remains in the physiological range, this technique offers considerable promise for studying the formation and localization of protein complexes following stimulation of a cell with ligands such as EGF. The Src family of kinases c-Src and other members of this family of cytosolic tyrosine kinases have long been implicated in signal transduction from polypeptide growth factor receptors such as the EGFR (reviewed by Belsches et al. [140]). In the case of EGFR signalling, the evidence for an involvement of members of the Src family of kinases is overwhelming. Overexpression of Src proteins strongly enhances EGFmediated proliferation and transformation in fibroblasts and epithelial cells [141,142]. Conversely, inhibition of Src activity by microinjection of antibodies, by dominant negative Src kinase constructs or by exposure of the cells to


Src-specific pharmacological inhibitors, can block EGFdependent DNA synthesis [143,144] and reverses the transformed phenotype of EGFR- or ErbB2-overexpressing cells [145]. However, it is still not clear whether Src is a signal transducer downstream of the EGFR or a contributor to EGFR activation. There is evidence to support both models. In A431 cells and in colon carcinoma cell lines, endogenous c-Src has constitutively elevated kinase activity, which is reduced to basal levels by the EGFR-specific kinase inhibitor AG1478 [146,147]. EGF-dependent Src kinase activation is observed in cells that overexpress the EGFR, such as A431; in these cells it is also possible to detect the association of Src and the receptor, while in cells that do not overexpress either EGFR or Src, association between the two proteins has been difficult to prove. The association is most likely direct, and mediated via the Src SH2 domain, although the exact binding site on the EGFR is still unclear. Using in vitro assays with the two purified kinases. Stover et al. [119] have shown that Src does not bind to the major autophosphorylation sites of the EGFR, but phosphorylates novel sites within the kinase domain of the receptor (Y891) and (Y920). These two phosphotyrosines bind the SH2 domain of Src, and Y920 may also provide a docking site for the p85 subunit of PI3-K. In these experiments, Srcphosphorylated EGFR, but not autophosphorylated EGFR, caused a marked activation of Csk-inactivated Src kinase activity. Two other Src-dependent phosphorylation sites have recently been identified within the EGFR, i.e., Y845 and Y1101 [148]. pY1101 is a potential Src-binding site [149], while pY845 appears necessary for STAT5b activation [110]. The physiological relevance of Src providing its own EGFR phosphorylation sites for docking has been put in doubt by recent results, in which binding of Src to the EGFR was found to be independent of Src kinase activity [97]. The possibility that phosphorylation by Src may contribute significantly to the activation of EGFR is of great interest. Src phosphorylates the EGFR on Y845 both in vitro and in vivo [148,150]; this residue, located on the activation loop of the EGFR kinase domain, is highly conserved in tyrosine kinases, and plays a crucial role in the activation of receptor kinases such as KDR [151]. Trk [152], and insulin receptor [34]. The role of this phosphorylation site in EGFmitogenic signalling is controversial; Gotoh et al. [153] found no effect of Y845F mutation on EGF-mediated proliferation and transformation in fibroblasts, while Tice et al. [97] found that the same mutations abolishes not only EGFdependent, but also serum-dependent, stimulation of DNA synthesis. The ability of Y845F-EGFR to autophosphorylate or phosphorylate. She did not appear to be affected, suggesting that the mutant receptor maintains tyrosine kinase activity. However, both C-terminal phosphorylation of the EGFR and She phosphorylation have been shown to occur in cells expressing kinase-negative EGFR, without concomitant stimulation of DNA synthesis [154]. In this case, the phosphorylation appears to be caused by heterodimerization between the mutant EGFR and ErbB2. It is therefore

41

difficult to address the question of Src-mediated EGFR kinase activation in any cell that coexpresses EGFR and one of its heterodimer partners, ErbB2 or ErbB4. The Src and EGFR tyrosine kinases share many substrates, again making it difficult to discriminate between Src-mediated and EGFR-mediated signalling following stimulation with EGF (reviewed by Belsches et al. [140]). EGFR and the She protein are phosphorylated by both kinases, but the phosphorylation occurs at different sites [155], potentially enhancing the spectrum of She-mediated responses from the EGFR. p120RasGAP is also a substrate for both kinases, while p190RhoGAP has been shown to be selectively phosphorylated by c-Src [156]. Since the association between p120RasGAP and p190RhoGAP is implicated in EGF-mediated cytoskeletal rearrangement, and is mostly dependent on phosphorylation of RhoGAP by Src [157], c-Src appears to act as a downstream signal transducer from EGFR. There is indeed circumstantial evidence that most of the cytoskeletal reorganization that follows stimulation of cells by EGF is mediated by preferential substrates of the c-Src kinase; these include FAK [132], p130Cas [158], cortactin [159], EAST [160], and Eps-8 [161]. Activated c-Src is also intimately linked to the activation of PI3-K. As mentioned above, Src-dependent phosphorylation of the EGFR molecule can provide a docking site for p85, presumably facilitating its phosphorylation by EGFR and the consequent activation of PI3-K. Src also directly phosphorylates and activates PI3-K [162], once again pointing to the large overlap in signal activation by Src and EGFR and to the difficulties of unequivocally assigning specific pathway activation to either kinase. The development of truly specific tyrosine kinase inhibitors will be of great help in dissecting the relative roles of c-Src and EGFR kinases in a cellular context. The JAKs and STATs pathways STATs were first identified as signal transducers downstream of cytokine receptors (reviewed by Darnell [163] and Ihle et al. [164]). In mammals, seven STAT genes have been identified (STAT 1 to 4, 5a, 5b, and STAT6). STAT proteins are inactive transcription factors, which are activated and translocated to the nucleus upon specific receptor stimulation. Classically, STATs are recruited to the intracellular domain of the cytokine receptors through specific binding between STAT SH2 domains and receptor phosphotyrosine residues. Homo- and heterodimerization of STAT proteins is a prerequisite for activation and translocation to the nucleus, and is mediated by tyrosine phosphorylation of critical residues (Y699 in STAT5b, Y694 in STAT5a, and Y701 in STAT1); further residues have also been implicated in the activation of STAT5b (see Kloth et al. [110]). In cytokine signalling, activation is mediated by the JAK family of kinases (reviewed by Leonard [165]). STAT proteins, in particular STAT-1, 3, and 5, have also been implicated in EGFR signalling; however, the mode of activation appears to be

42


significantly different from that used by cytokine receptors. First, the ligand-dependent phosphorylation of STATs by EGFR does not require JAK kinases [166–168]. Second, STATs do not bind to the C-terminal phosphotyrosines of the EGFR; indeed it appears that STATs are constitutively associated with the EGFR [113,114]. However, as in JAK kinase signalling, activation of STAT transcriptional activity is strictly dependent upon the EGFR tyrosine kinase activity [167]. More recent reports have implicated the Src kinase in EGF-dependent STAT activation [110,114], but it is unclear whether Src acts upstream or downstream of EGFR activation in this case. Phospholipid metabolism: PLD, PLCg, and PI3-K EGF stimulation of a cell has marked effects on its phospholipid metabolism, including phosphatidylinositol turnover and production of phosphatidic acid (PA) and arachidonic acid (AA). Of the enzymes involved in these pathways, at least three can be activated directly by the EGFR, i.e., phospholipase C-g (PLCg), phosphatidylinositol-3-kinase (PI3-K), and phospholipase D (PLD), while others, such as phospholipase A2, are regulated indirectly by EGF-mediated activation of other pathways. PLD hydrolyses phosphatidylcholine to generate choline and the second messenger PA (reviewed by Houle and Bourgoin [169]). PLD activity is stimulated in whole cells by EGF treatment, but until recently the stimulation was thought to be indirect and mediated by cofactors such as PKC, Rho, and phosphotidylinositol biphosphate (PIP2). While this may indeed be the case for PLD1, PLD2 has now been shown to be associated with, and activated by, the EGFR [170]. The mechanism of activation, while still obscure, appears to require the physical association of PLD with EGFR but not necessarily tyrosine phosphorylation; although Y11 of PLD2 has been identified as the major site of phosphorylation by the EGFR kinase, mutations at this residue do not abolish activation [170]. Activation of PLD may require a conformational change that is stabilized, but not induced, by tyrosine phosphorylation; a similar mode of activation (dependent on complex formation but independent of tyrosine phosphorylation) has been proposed for PLC g [171], suggesting a common mechanism of activation for this class of molecules. PLCg (reviewed by Kamat and Carpenter [172]) binds directly to the autophosphorylated EGFR via Y1173 and Y992 [173] and is phosphorylated by the EGFR kinase on Y771 and Y1254 [174]. The exact mode of PLCg activation by the EGFR is not clear; reportedly it requires direct association with the receptor but not necessarily tyrosine phosphorylation [171]. Once activated, PLCg catalyzes the hydrolysis of PtdIns(4,5)-P2 to yield the important second messengers 1,2diacylglycerol (DAG) and inositol 1,3,5-trisphosphate (IP3). IP3 mediates calcium release from intracellular stores, affecting a host of Ca2+-dependent enzymes, while DAG is a cofactor for the activation of the serine/threonine kinase

PKC. Thus, through IP3, EGFR can activate Ca2+-dependent pathways such as RaI [175] and NFkB [176], and through PKC multiple signalling components, including the MAPK and JNK pathways [177,178] and possibly the Na+/H+ exchanger [179]. Phosphoinositide-3-kinases are major players in cellular functions, where they contribute to a variety of cellular processes including proliferation, survival, adhesion, and migration (reviewed by Cantley [180]). PI3-kinases catalyse phosphorylation on the 3¢ position of phosphatidylinositols (PtdIns) and are assigned to three classes according to their subunit structure and their preferred lipid substrate (reviewed by Djordjevic and Driscoll [181]). Of the three classes of typical PI3-kinases, only class Ia is activated by tyrosine kinase receptors. Interaction between PI3-kinase and the ErbB receptors is required for activation, and is mediated by association of the phosphorylated receptors with the p85 subunit of PI3-K via the latter’s SH2 domain [182]. As mentioned previously, the major binding partner of p85 is not the EGFR, but ErbB3 [183,184]; however, activation of PI3-K is observed in response to EGFR ligands through formation of ErbB1/ErbB3 heterodimers, as well as potentially by Src phosphorylation of the EGFR, so it is relevant to this discussion of EGF-mediated signalling pathways. PI3-K Ia generates phosphatidylinositol-3,4,5-trisphosphate (PIP3). One of the best characterized targets of this second messenger is the Ser/Thr kinase Akt (PKB [185]), which binds to the lipid and is translocated to the plasma membrane where it is phosphorylated and activated by phosphoinositide-dependent kinase-1 (PDK-1) and possibly other kinases (reviewed by Nicholson and Anderson [186]). PKB/Akt is a major mediator of PI3-K action in survival and proliferation, and may well be the major mediator of the antiapoptotic effects of EGFR activation. Recently, the crystal structure of the catalytic subunit (p110) of PI3-Kg in complex with Ras has been solved [187]. The structure shows a change in conformation of the catalytic p110 upon binding to Ras, consistent with a Ras-mediated activation model. Since activated Ras is one of the major downstream effectors of EGFR signalling, this mechanism may represent yet another way in which activated EGFR regulates PI3-kinase activity.

The role of the EGF family of ligands and EGFR in mammalian physiology and pathology A vast body of knowledge has been accumulating in recent years on the role of the EGF family of ligands and receptors in embryonic development, physiology, and pathology. Thanks to the power of genetic screens, much of the progress on the developmental role of the EGF/EGFR system has come from studies on invertebrates, such as Drosophila and C. elegans. The developmental aspects of EGF/EGFR signalling, both in invertebrates and in


mammals, are covered elsewhere in this issue (Shilo and Sternberg). In this article we will concentrate on the role of EGF ligands and receptors in newborn and adult mammals. Murine EGF and its human equivalent, b-urogastrone, were first isolated and identified because of their effects on tooth eruption and eyelid opening [188] or inhibition of gastric acid secretion [189], respectively. Another family member, TGFa, was identified as a component of “sarcoma growth factor,” produced by retrovirally transformed fibroblasts [190]. Attempts to determine the physiological role of EGF and TGFa in in vivo studies date back to the early 1980s. Initially, the ligands were injected in neonatal mice, and physiological changes were monitored. The most striking effects of EGF were precocious eyelid opening and tooth eruption [191,192], although more subtle effects on neurobehavioural development [193] and, unexpectedly, a reduction in growth rates [193,194] were also observed. With the analysis of natural mouse mutants, the development of transgene technology, and the advent of gene targeting in murine ES cells, the study of gain-of-function or loss-of-function in the EGF/EGFR axis became much easier and led to a clearer understanding of the role of these ligands and receptors in mammalian physiology and pathology. It must be emphasized, however, that the effects of altering the EGF/EGFR axis may be indirect; for example, EGF and TGFa modulate hormonal responses such as the release of luteinizing hormone and thyroid hormone, and the EGFR is required in mediating many of the effects of estrogen.

Gain-of-function: EGFR and its ligands Apart from the in vitro data, suggesting a role of EGF/EGFR in cell proliferation, evidence has been accumulating that overexpression of the ligands and/or receptors, as well as ligand-independent receptor activation, occurs in many epithelial cancers, most notably gliomas and breast, pancreas, and liver carcinoma. What is not clear is whether this overexpression/activation is indeed causative for the formation of tumours or occurs during tumour progression. The use of transgenic animals has allowed the role of these proteins to be addressed. Of the EGF family ligands, TGFa has been the most studied by using this technology. Targeting of TGFa to the skin by means of keratin promoters results in hypertrophy and hyperkeratosis accompanied by alopecia or stunted hair growth. The scaly skin and localized leukocyte infiltration are reminiscent of psoriasis [195,196]. The psoriasis-like lesions and hyperkeratosis are even more prominent in mice expressing a K14amphiregulin transgene [197], strengthening the case for involvement of EGFR activation in this skin condition. Interestingly, TGFa transgene expression is linked to the appearance of papillomas following irritation or wounding, but without progression to carcinomas [195,196]. Inducible expression of the TGFa transgene in the kidney, as a model for polycystic kidney disease (PKD), has been linked to the

43

formation of renal cysts and accelerated progression of the disease in a strain of mice predisposed to PKD, but not to the onset of polycystic kidneys [198,199]. Targeted overexpression of TGFa in the mammary gland results in hyperplasia, cystic expansion, and papillary adenomas following multiple pregnancies and lactation. Expression of TGFa in these mice inhibits involution of the mammary gland after pregnancy and lactation, resulting in hyperplastic alveoli in multiparous females; however, the incidence of tumour formation, while variable, is generally low (see for example, Davies et al. [200] and Sandgren et al. [201]). It appears therefore that overexpression of TGFa is linked to hyperproliferative responses but does not generally lead to tumours in rodents. Even in transgene models where TGFaassociated tumours are observed, the latency period tends to be long and the incidence low, suggesting that the TGFa/EGFR system provides only one of the steps in multistage carcinogenesis, and neoplastic transformation only occurs when other genes within the target tissue are also mutated. Apart from the ras oncogene and the components of the EGF/EGFR family of ligands and receptors, the most commonly amplified oncogene in breast cancer is c-myc [202–204]. Studies using double transgenic mice have addressed the significance of EGF system/c-myc interactions in the mammary gland [201,205]. Transgenic mice that overexpress both TGFa and c-myc develop mammary tumors irrespective of the sex, while in transgenic mice expressing c-myc alone, tumor formation only occurs in females, and even then with much reduced frequency and with a long latency periods. The cooperativity between c-myc and TGFa has been attributed to an antiapoptotic effect exerted by TGFa, coupled with increased proliferative responses associated with c-myc overexpression [205]. A strong antiapoptotic effect of TGFa could also help explain the lack of involution of the mammary gland described in mice overexpressing TGFa [200]. Even more striking is the cooperativity between TGFa and c-myc in the induction of hepatocarcinomas. TGFa transgenic mice develop hepatic tumors at low frequency and with very long latency. Overexpression of c-myc in these mice dramatically shortens the latency period and accelerates tumor growth [201,206,207]. The hepatocellular carcinomas from c-myc/TGFa transgenic mice display a very low apoptotic index compared to hepatocarcinomas from TGFa or c-myc single transgenic mice, present abnormalities in cell-cycle protein expression [208], and overt tumor formation is often preceded by aneuploidy and chromosomal breakages [209]. The commonly observed features in these models of TGFa overexpression are an enhancement of cellular proliferation, as evidenced by increased mitotic index, and a reduction in the rate of apoptosis, resulting in hyperplasia of skin, mammary glands, and liver. Thus, TGFa appears to be capable of altering the balance between cellular proliferation and death. Stimulation of the EGFR pathways may

44


override the DNA damage check points, resulting in decreased apoptosis and in the accumulation of secondary mutations. Confirmation of this link between overactivation of the EGFR and transformation should come from experiments where the EGFR is either overexpressed or constitutively activated. There is ample documentary evidence of EGFR overexpression or activation in spontaneous tumours, as well as numerous studies on the tumorigenicity of cells lines expressing activated EGFR in mouse models. Of particular interest in this context is a naturally occurring deletion mutant of the EGFR, D2–7 (also called EGFRvIII). EGFRvIII is the most common EGFR mutation in human cancers, having been detected in 40–50% of grade VI glioblastomas [210] and in up to 70% of medulloblastomas and a small proportion of breast and ovarian carcinomas [211]. EGFRvIII arises from a genomic deletion of exons 2–7 [212,213], resulting in a protein that lacks most of the extracellular domain. As a result, EGFRvIII is not activated by ligand; however, it is constitutively activated and is not internalized, which results in constitutive long-term signalling [37]. Transfection of glioma cell lines with EGFRvIII dramatically increases their tumorigenicity in nude mice [214]. The tumorigenic potential of cells expressing EGFRvIII has been linked to upregulation of Bcl-XL and resulting inhibition of apoptosis [215]. However, constitutive activation of the EGFR may not be sufficient to initiate and maintain the transformed phenotype. The experiments described above were performed by transfecting the activated receptor in cells adapted to continuous growth in culture, which are likely to harbour complementing mutations. When the expression of an activated EGFR is directly targeted to glial progenitor cells in mouse models, there is no evidence of increased rates of tumour formation. However, expression of the activated receptor in mice genetically defective in cell-cycle inhibitor proteins (such as INK4a) does lead to the development of gliomas [216]. These findings have been confirmed and extended by Bachoo et al. [217]; EGFR activation in INK4a/Arfdeficient mice leads to dedifferentiation of astrocytes and is instrumental in gliomagenesis. Effects of the loss of EGFR function While the gain-of-function experiments address mainly the role of the EGF/EGFR system in abnormal proliferation, loss-of-function mice have shed some light on the developmental and physiological role of the system. Since EGF is produced by the submaxillary glands, sialoadenectomy was initially used as a tool to investigate the effects of reduced EGF levels in vivo. In these studies the organs most affected were the mammary gland [218] and the epidermis [219]. In both organs there was a reduction in size and thickness, which could be reversed by administration of EGF. In these early experiments, however, the levels of EGF were reduced but not abolished completely. Genetically null animals have

been created by targeted inactivation of the EGF, amphiregulin, and TGFa genes. Surprisingly, in view of the pleiotropic role of TGFa, the phenotype of the TGFa-null animals is very mild. The most striking abnormalities are found in the skin architecture and in the development of the eyes; the hair follicles are deformed resulting in wavy fur and whiskers, and the eyes of the TGFa-null mice are open at birth and are opaque [220,221]. More recently, a significant reduction in the number of dopaminergic neurons in the substantia nigra of TGFa knockout mice has also been described [222]. Recently, Luetteke and colleagues [223] have published the results of targeted inactivation of the genes coding for other ligand in the EGF family. Their analysis covers single, double, and triple deletions of the EGF, amphiregulin (AR), and TGFa genes. EGF-null or AR-null mice display no obvious phenotype, not even the wavy coat characteristic of the TGFa-null mice. The lack of phenotype contrasts with the results obtained by reducing EGF levels via sialoadenectomy, and may be in part due to compensatory mechanisms, such as upregulation of other EGF family ligands, in animals chronically deficient for the growth factor and/or nutritional problems with the sialoadonectomized mice. Double- and triple-null mice were generated by intercrosses. Compared to the original TGFa knockouts, triple-null mice have an increased penetrance of eye defects, dermatitis, and skin ulcerations with aging. However, only the absence of AR, combined with absence of either EGF or TGFa, results in impaired mammary gland development. Interestingly, the defects in duct formation and lobuloalveolar development were not attributable to decreased proliferation or increased apoptosis, but it has been suggested that the defects were probably the result of abnormal epithelial cell migration. In these experiments, three of the six known EGFR ligands have been ablated, and yet the phenotype is relatively mild. Again, it is possible that upregulation of the other EGF family members (e.g., HB-EGF or betacellulin) could still regulate the physiological activation of the EGFR in most tissues. Inactivation of the EGFR, the common signalling partner for EGF, TGFa, HB-EGF, AR, betacellulin, and epiregulin, should prevent compensatory mechanisms involving ligand overexpression from masking the effects of ligand deficiency. Indeed, EGFR knockout mice are severely affected, although to varying degrees depending on their genetic background [224–226]. At its most severe, the lack of EGFR causes peri-implantation or mid-gestational death. In some strains, however, the mice survive up to 3 weeks post birth; these animals show severe abnormalities of skin, lungs, the gastrointestinal tract, brain, and liver [224], confirming the importance of the EGF/EGFR system in epithelial cell regulation. EGFR-null mice show disorganized hair follicles and curly coats. This phenotype also occurs in mice expressing a dominant-negative EGFR construct targeted to the skin, and in the naturally occurring mouse strain wa-2, which carries an inactivating mutation in the EGFR gene


[64,227]. This waviness of fur and whiskers is characteristic of the TGFa-null mice, but is not present in EGF knockouts, implying that TGFa is a major physiological ligand regulating the activation of the EGFR in the skin. Given the very short life span of the EGFR-null mice, a study of mammary gland development in these mice is not possible. The effects of EGFR loss-of-function in this tissue have been studied instead by targeting a dominant-negative EGFR to the mammary gland. Expression of the dominantnegative EGFR construct inhibits ductal branching and outgrowth in virgin mice, although postpartum lactation still occurs, probably following upregulation of the endogenous wild-type EGFR [228]. The mammary glands of mice expressing wa-2 EGFR mutation are reduced in size and have underdeveloped ducts ([64]; K.J. Fowler, personal communication). Using transplants of neonatal mammary glands from EGFR-null mice into normal mice, and in tissue recombination experiments. Wiesen and coworkers [229] concluded that EGFR presence in the stroma, rather than the epithelium, is essential for ductal growth and branching. The impairment of ductal morphogenesis is not apparent in EGF or TGFa-null mice but is characteristic of the AR knockout mice; this implicates amphiregulin as a key EGFR ligand in mammary tissue. The tissue specificity of these phenotypes suggests that individual EGFR ligands are important in particular tissues, or at particular stages of development. Alternatively, the specificity may be conferred by selective coexpression of EGFR family members in a given tissue. In the latter case, two further alternatives are possible, i.e., the same ligand, binding to different receptor combinations, triggers tissuespecific responses, or different receptor combinations preferentially bind to, and are activated by, selected EGF family ligands. More information is needed, from both in vivo and in vitro models, before these challenges can be answered and the complexities of signalling from the EGF/EGFR family of ligands and receptors in a physiological setting can be understood in detail.

Cell motility: EGF receptor–integrin cooperativity Cell migration is a complex, coordinated process that allows cells to reach specific destinations during embryonic development, to maintain the cellular architecture of selfrenewing tissues, repair wounds, and to defend against infectious agents [230–232]. Signals from several classes of receptors play critical roles in the regulation of cell movement; integrins, through their ability to signal and form adhesive contacts linking the extracellular matrix (ECM) and the actin cytoskeleton [233–235], growth factor receptors activated through either autocrine or paracrine pathways, also regulating the actin cytoskeleton [236], and chemotactic receptors [237]. The involvement of EGF receptor signaling in various normal physiological processes requiring cell movement and deregulation of the motility

45

response in pathological conditions such as tumour invasion is well documented and has been reviewed by others [238–241]. Our understanding of the mechanisms by which signals from the EGF receptor modulate cell locomotion has progressed significantly over the past decade and will be briefly summarized below. Where appropriate we have referred to other excellent recent reviews for more in-depth discussions of particular topics. Although EGF receptor stimulation can lead to both cell proliferation and migration [2], these responses are separable and mediated via different signalling pathways [242]. A series of reports by Wells and coworkers ([242–244] and reviewed by Wells [2]) linked a PLCg-dependent pathway to cell motility triggered by activation of the EGF receptor. Using cells expressing various receptor mutants that differ in their ability to induce a motility or mitogenic response, they found that the motility response elicited by EGF receptor stimulation requires kinase activity of the receptor and the presence of at least one autophosphorylation site, tyrosine992 [243]. The ability of EGF to induce cell movement correlates with the activation of PLCg and movement can be inhibited by blocking the function of this enzyme [242]. The observation that EGF stimulates MAP kinase activation in both motogenic and nonmotogenic EGF receptor-expressing cells is consistent with the current view that the motility and mitogenic responses elicited by the EGF receptor diverge at the immediate postreceptor level and that MAP kinase activation alone is not sufficient to induce a motility response [242]. Further work by Chen and colleagues [244] provided evidence that EGF-induced activation of PLCg stimulates cell motility by releasing PIP2-bound gelsolin (and possibly other actin-modifying proteins, profilin, cofilin, and CapG) from the membrane, thereby restoring its ability to bind, sever, and cap polymerized actin filaments, a process required for filopodia/ lamellipodia extension and retraction in motile cells [245]. Thus, EGF receptor-mediated activation of PLCg is believed to be critical for the reorganization of the actin cytoskeleton and contribute to the initiation of the asymmetric motile phenotype (reviewed by Wells et al. [238]). Although not sufficient to stimulate motility by itself, MAP kinase may in part regulate cell motility by modulating integrin adhesive functions. Expression of activated mutants of H-Ras or its kinase effector Raf-1 in CHO cells expressing chimeric integrins suppresses integrin activation (ligand binding affinity). The suppression of integrin activity correlates with MAP kinase activation and appears to result in loss of cell spreading [246]. The downstream elements of this pathway have yet to be identified and it is still unclear how MAP kinase activation relates specifically to EGF receptor-mediated cell motility. One possibility is that MAP kinase stimulates the disassembly of focal adhesions [247]. Of course, disassembly of the focal adhesions results in decreased cell adhesion to the substratum. The effect of EGF in fibroblasts can be inhibited by blocking MAP kinase activation with the MEK inhibitor, PD98059. Together these

46


studies implicate the MAP kinase pathway in both weakening of adhesion associated with initial motility shape change [246] and as an effector of deadhesion of the uropod [230,248]. The family of intracellular calcium-dependent proteases, calpains, is also important for focal deadhesion necessary during retraction of the trailing edge of migrating cells. Inhibition of calpains results in an elongated but ultimately immobile cell [249,250]. Calpain activation in EGFstimulated dermal fibroblasts coincides with cell compaction, detachment, and enhanced motility [250]. The enzyme could be critical for cell movement under conditions where detachment of the trailing edge of motile cells becomes rate limiting, e.g., in situations where there is high substrate adhesiveness [249,251]. The events leading to activation of calpain are still incompletely understood and may differ between isoforms. At least two isoforms of calpain, including m-calpain and M-calpain, have been implicated in cell motility through their ability to cleave several proteins found in adhesion complexes (reviewed by Glading et al. [252]). The former may be critical for integrin-mediated motility (haptokinesis [249]) and appears to be more sensitive to activation by Ca2+ fluxes, possibly triggered through stretch-activated calcium channels as demonstrated in fish keratocytes [253]. M-calpain is the more likely isoform downstream of growth factor receptor-mediated cell motility but its activation in vitro requires Ca2+ concentrations that appear unlikely to be attained under physiological conditions. Hence, a number of alternative or complementary mechanisms have been proposed for activation of M-calpain by growth factor receptors (reviewed by Glading et al. [252]). Using a combination of antisense oligonucleotides, inactive mutants, activated mutants, or specific inhibitors of MEK, ERK1, ERK2, and myosin light chain kinase (MLCK), Klemke et al. [254] linked the EGF receptor, MAP kinase activation, and myosin-mediated contraction forces. The model suggests that basal and EGF-directed haptotactic responses involve direct phosphorylation of MLCK by MAP kinase. This phosphorylation event enhances MLCK’s ability to phosphorylate the myosin regulatory light chain, thereby promoting myosin ATPase activity and polymerization of actin cables, a well-established role for myosin-II in the modulation cell contraction [255]. In contrast to the study by Hughes et al. [246], the MAP kinase-MLCK pathway appears to be independent of initial adhesion and spreading of cells on coated ECMs [254]. Although the authors did not examine whether the increased motility was a result of cell : substratum detachment or cell body translocation, recent evidence showing that the kinase inhibitor H-7 or the specific MLCK inhibitor ML-7 induces dissolution of focal adhesion plaques in REF52 fibroblasts [256]. These results suggest that the primary role of the EGF-induced MAP kinase-MLCK pathway in cell locomotion would be in the translocation of the cell body and/or

retraction (rather than deadhesion) of the trailing edge of migrating cells. The role of PI3-K in EGF-mediated cell movement is less clear. Treatment of human breast cancer cells and NIH 3T3 fibroblasts overexpressing the EGF receptor with PI3-K inhibitors was reported to enhance EGF-stimulated haptokinesis. The increased rates of cell movement were postulated to be due to increased adhesiveness [257]. In contrast, wortmannin partially inhibits EGF-directed chemokinesis of bladder carcinoma cells [258]. Similarly, attenuation EGF receptor-dependent PI3-K activation by overexpression of SIRPa1 in U87MG glioblastoma cells blocks their haptotactic migration [259]. The discrepancies between these studies may reflect differences in the signalling networks of the three cell types. Alternatively, the difference may depend on the contribution of PI3-K to integrin- and growth factormediated motility (haptokinesis vs. chemokinesis) and the linkage of these pathways to the Rho-like GTPases. The role of Rho-like GTPases Rho, Rac, and Cdc42 in integrinmediated actin cytoskeleton remodelling and cell movement has been studied extensively. Rho is believed to regulate stress fibers and focal adhesion formation while Rac and Cdc42 control the formation of lamellipodia and filopodia/cell polarity, respectively [260–263]. Rac and Rho appear to have antagonistic roles on cell morphology by controlling the actinomyosin cytoskeleton via phosphorylation of myosin heavy chain and regulatory light chain, respectively [245]. In A431 cells, EGF stimulates motility and this response in associated with Rhodependent cell rounding and cortical actin polymerization and Rac-dependent membrane ruffling and lamellipodia formation [264]. Further evidence suggests that integrin influences on motility require PI3-K downstream of Rac and Cdc42 [235,265,266]. In contrast, PI3-K acts upstream of Rac in growth factor-mediated motility, presumably by activating GTPase exchange factors through 3¢phosphoinositides [267–270]. Lipid products of PI3-K may also regulate the adhesive strength and contribute to EGFmediated cell motility by two alternative mechanisms, both involving remodeling of the actin cytoskeleton. First, activation of PI3-K upon PDGF stimulation was found to induce restructuring of focal adhesion plaques in rat embryonic fibroblasts and this effect could be mimicked by PtdIns (3,4,5)-P3 [271]. The authors provided evidence that PtdIns(3,4,5)-P3 mediates its effect by disrupting the interaction between a-actinin and integrin b-subunit. It is thus conceivable that PI3-K lipid products may act in an analogous way in response to EGF stimulation although this remains to be demonstrated. More recently, Piccolo et al. [272] reported that EGF-mediated activation of PI3-K leads to translocation of PLCg1 to the leading edge of migrating cells in a wound-healing assay. Migration could be inhibited by expression of the pleckstrin homology domain of PLCg1, providing evidence that this effect was mediated by direct interaction between PLCg PH domain and PtdIns(3,4,5)-P3


as reported by others. EGF/PI3-K pathways are likely to be critical for the establishment of cell polarity during EGFRmediated chemotaxis [273]. Clearly, more work is required to elucidate the precise contribution of PI3-K to EGF or other growth factor effects on cell motility. It is noteworthy, however, that the 3¢phosphoinositide products of PI3-K activate the protooncogene Akt, which in turn phosphorylates GSK3, preventing the formation of GSK3/APC/b-catenin complexes [274,275]. Thus, EGF receptor-activated PI3-K could regulate the function of both b-catenin and APC. The latter has been implicated in the migration of normal epithelial cells in the intestinal crypts and localizes at the tip of microtubules in actively migrating regions of epithelial cell membranes [276]. Although its precise function remains elusive, the potential role of APC in EGF-mediated cell migration is supported by the recent observation that EGF induces migration of murine colon epithelial cells in an APC genotypedependent manner [277]. Although it is clear that signals from growth factors and integrins are intimately linked and often overlapping [233–235], the complex relationship between these classes of receptors and how they cooperate to control cell movement is still the subject of intensive investigation. Recent studies have shed some light on this issue. The chemokinetic response of fibroblasts to EGF depends on the substratum concentration, which influences both the speed and directional persistence of migration [278]. A recent study by Gu et al. [279] addressed the role of the lipid/protein tyrosine phosphatase PTEN in integrin-mediated cell migration. Using various inhibitors and activators of integrin signaling, they defined two pathways that diverge at the level of Shc and FAK in glioblastomas. Their evidence indicates that the Shc-MAP kinase pathway regulates integrin-mediated random migration, whereas the FAK-p130Cas pathway regulates directional migration. Both these pathways are inhibited by PTEN, which results in reduced rate and persistence of cell migration. Interestingly, PTEN was shown to associate with Shc and to directly dephosphorylate Shc (Tyr239 and Tyr317). These Tyr residues are phosphorylated after the EGF receptor is activated. Studies in keratinocytes using galvanotaxis assay (cathodal migration of cells in an electric field) support the involvement of the EGF receptor in the regulation of directed migration. Interfering with MAP kinase activation with PD98059 completely abolishes the effect of EGF on the rate of migration but only partially blocks directional migration on laminin or fibronectin, indicating that other signals (from integrin and/or EGF receptor) are required to control directionality [280]. Most important, this and another study [281] have shown that directional migration of keratinocytes requires kinase activity and redistribution of the EGF receptor at the leading edge, resulting in asymmetric actin polymerization in migrating cells [282]. Specific inhibition of the EGF receptor kinase activity with low concentrations of PD158780 blocks directional migra-

47

tion. Higher concentrations are required to block the rate of random migration, indicating that the rate of migration is controlled by multiple kinase effectors. The importance of EGF receptor redistribution in directional migration is supported by the work of Li et al. [283]. Using chemotaxis chambers, these authors observed that EGF is required to be copositioned with ECM proteins (in the bottom chamber) to stimulate the haptotactic migration of B82L fibroblasts. EGF added directly to the cells (in the upper chamber) decreased their maximal migration toward the gradient of EGF/ECM. Most important, haptotactic migration toward fibronectin requires the presence of active EGF receptors even in the absence of apparent autocrine or exogenous ligand stimulation. Copresentation of EGF and ECM components enhances the polarization events required for directional migration and facilitates cross-talk between integrin and EGF receptors at the leading edge. In summary, multiple signalling pathways are generated by activation of the EGF receptor and these pathways control the distinct steps of the motility process. Further studies are required to clarify the precise contribution of PI3-K and APC to EGF-mediated cell motility and how the various pathways are temporally integrated. Most of the studies described above have been performed in fibroblasts or keratinocytes and data for several other cellular systems will be required for a more complete understanding of the molecular regulation of the movement of mammalian cells.

Concluding notes Much progress has been made in understanding the mechanism of EGFR activation upon ligand binding. However, there are many basic questions that must be answered about the nature of the EGFR on the cell surface, i.e., the nature of the inactive, unliganded EGFR monomer and dimer: how ligand induces the conformational transition in the ectodomain; how ligand binding stimulates the activation of the kinase; and whether mechanisms such as secondary dimerization or the formation of higher order complexes regulate the activation on EGFR homodimers and ErbB hetero-oligomers. The three-dimensional structures for parts of the EGFR have already led to a much clearer understanding of the activation processes. Mutation experiments will soon lead to a more detailed understanding of the mechanisms regulating the action of the EGFR. It is critical that these experiments are performed under conditions where the normal regulatory mechanisms of receptor activation, turnover, and signalling apply and that the effects of heterodimerization are controlled. The EGFR is a complex signalling system important in normal physiology and in the maintenance of the tumorigenic state. Studies of its biochemistry and biology have already made deep contributions to cell signalling and there are bound to be many more surprises in the near future.

48


References [1] G. Carpenter, S. Cohen, Epidermal growth factor, J. Biol. Chem. 265 (1990) 7709–7712. [2] A. Wells, EGF receptor, Int. J. Biochem. Cell. Biol. 31 (1999) 637–643. [3] Y. Yarden, M.X. Sliwkowski, Untangling the ErbB signalling network, Nat. Rev. Mol. Cell. Biol. 2 (2001) 127–137. [4] S. Cohen, H. Ushiro, C. Stoscheck, M. Chinkers, A native 170,000 epidermal growth factor receptor-kinase complex from shed plasma membrane vesicles, J. Biol. Chem. 257 (1982) 1523–1531. [5] Y. Yarden, J. Schlessinger, Epidermal growth factor induces rapid, reversible aggregation of the purified epidermal growth factor receptor, Biochemistry 26 (1987) 1443–1451. [6] Y. Yarden, J. Schlessinger, Self-phosphorylation of epidermal growth factor receptor: evidence for a model of intermolecular allosteric activation, Biochemistry 26 (1987) 1434–1442. [7] A.M. Honegger, R.M. Kris, A. Ullrich, J. Schlessinger, Evidence that autophosphorylation of solubilized receptors for epidermal growth factor is mediated by intermolecular cross-phosphorylation, Proc. Natl. Acad. Sci. USA 86 (1989) 925–929. [8] A.M. Honegger, A. Schmidt, A. Ullrich, J. Schlessinger, Evidence for epidermal growth factor (EGF)-induced intermolecular autophosphorylation of the EGF receptors in living cells, Mol. Cell. Biol. 10 (1990) 4035–4044. [9] S. Bogdan, C. Klambt, Epidermal growth factor receptor signaling, Curr. Biol. 11 (2001) R292–R295. [10] P.O. Hackel, E. Zwick, N. Prenzel, A. Ullrich, Epidermal growth factor receptors: critical mediators of multiple receptor pathways, Curr. Opin. Cell. Biol. 11 (1999) 184–189. [11] C. Mineo, G.N. Gill, R.G. Anderson, Regulated migration of epidermal growth factor receptor from caveolae, J. Biol. Chem. 274 (1999) 30636–30643. [12] G. Carpenter, S. Cohen, 125I-labeled human epidermal growth, factor. Binding, internalization, and degradation in human fibroblasts, J. Cell. Biol. 71 (1976) 159–171. [13] L. Beguinot, R.M. Lyall, M.C. Willingham, I. Pastan, Downregulation of the epidermal growth factor receptor in KB cells is due to receptor internalization and subsequent degradation in lysosomes, Proc. Natl. Acad. Sci. USA. 81 (1984) 2384–2388. [14] A. Sorkin, C.M. Waters, Endocytosis of growth factor receptors, Bioessays 15 (1993) 375–382. [15] H.S. Wiley, J.J. Herbst, B.J. Walsh, D.A. Lauffenburger, M.G. Rosenfeld, G.N. Gill, The role of tyrosine kinase activity in endocytosis, compartmentation, and down-regulation of the epidermal growth factor receptor, J. Biol. Chem. 266 (1991) 11083–11094. [16] C. Lamaze, S.L. Schmid, Recruitment of epidermal growth factor receptors into coated pits requires their activated tyrosine kinase, J. Cell. Biol. 129 (1995) 47–54. [17] M.A. Olayloye, R.M. Neve, H.A. Lane, N.E. Hyne, The ErbB signaling network: receptor heterodimerization in development and cancer, EMBO J. 19 (2000) 3159–3167. [18] T.P. Garrett, N.M. McKern, M. Lou, T.C. Elleman, T.E. Adams, G.O. Lovrecz, H.J. Zhu, F. Walker, M.J. Frenkel, P.A. Hoyne, R.N. Jorissen, E.C. Nice, A.W. Burgess, C.W. Ward, Crystal structure of a truncated epidermal growth factor receptor extracellular domain bound to transforming growth factor alpha, Cell 110 (2002) 763– 773. [19] H. Ogiso, R. Ishitani, O. Nureki, S. Fukai, M. Yamanaka, J.H. Kim, K. Saito, A. Sakamoto, M. Inoue, M. Shirouzu, S. Yokoyama, Crystal structure of the complex of human epidermal growth factor and receptor extracellular domains, Cell 110 (2002) 775–787. [20] H.S. Cho, D.J. Leahy, Structure of the extracellular region of HER3 reveals an interdomain tether, Science 297 (2002) 1330–1333. [21] J. Stamos, M.X. Sliwkowski, C. Eigenbrot, Structure of the EGF receptor kinase domain alone and in complex with a4-anilinoquinazoline inhibitor, J. Biol. Chem. 277 (2002) 46265–46272.

[22] A. Ullrich, L. Coussens, J.S. Hayflick, T.J. Dull, A. Gray, A.W. Tam, J. Lee, Y. Yarden, T.A. Libermann, J. Schlessinger, J. Downward, E.L.V. Mayes, N. Whittle, M.D. Waterfield, P.H. Seeburg, Human epidermal growth factor receptor cDNA sequence and aberrant expression of the amplified gene in A431 epidermoid carcinoma cells, Nature 309 (1984) 418–425. [23] L.J. Slieker, T.M. Martensen, M.D. Lane, Synthesis of epidermal growth factor receptor in human A431 cells. Glycosylationdependent acquisition of ligand binding activity occurs posttranslationally in the endoplasmic reticulum, J. Biol. Chem. 261 (1986) 15233–15241. [24] T.G. Johns, E. Stockert, G. Ritter, A.A. Jungbluth, H.J. Huang, W.K. Cavenee, F.E. Smyth, C.M. Hall, N. Watson, E.C. Nice, W.J. Gullick, L.J. Old, A.W. Burgess, A.M. Scott, Novel monoclonal antibody specific for the de2-7 epidermal growth factor receptor (EGFR) that also recognizes the EGFR expressed in cells containing amplification of the EGFR gene, Int. J. Cancer. 98 (2002) 398–408. [25] T.P. Garrett, N.M. McKern, M. Lou, M.J. Frenkel, J.D. Bentley, G.O. Lovreez, T.C. Elleman, L.J. Cosgrove, C.W. Ward, Crystal structure of the first three domains of the type-1 insulin-like growth factor receptor, Nature 394 (1998) 395–399. [26] T.C. Elleman, T. Domagala, N.M. McKern, M. Nerrie, B. Lonnqvist, T.E. Adams, J. Lewis, G.O. Lovrecz, P.A. Hoyne, K.M. Richards, G.J. Howlett, J. Rothacker, R.N. Jorissen, M. Lou, T.P. Garrett, A.W. Burgess, E.C. Nice, C.W. Ward, Identification of a determinant of epidermal growth factor receptor ligand-binding specificity using a truncated, high-affinity form of the ectodomain, Biochemistry 40 (2001) 8930–8939. [27] M. Yamabhai, R.G. Anderson, Second cysteine-rich region of epidermal growth factor receptor contains targeting information for caveolae/rafts, J. Biol. Chem. 277 (2002) 24843–24846. [28] A.C. Rigby, C.W. Grant, G.S. Shaw, Solution and solid state conformation of the human EGF receptor transmembrane region, Biochim. Biophys. Acta. 1371 (1998) 241–253. [29] S.J. Kil, C. Carlin, EGF receptor residues leu(679), leu(680) mediate selective sorting of ligand-receptor complexes in early endosomal compartments, J. Cell. Physiol. 185 (2000) 47–60. [30] C. He, M. Hobert, L. Friend, C. Carlin, The epidermal growth factor receptor juxtamembrane domain has multiple basolateral plasma membrane localization determinants, including a dominant signal with a polyproline core, J. Biol. Chem. 277 (2002) 38284–38293. [31] P. Castagnino, Z. Biesova, W.T. Wong, F. Fazioli, G.N. Gill, P.P. Di Fiore, Direct binding of eps8 to the juxtamembrane domain of EGFR is phosphotyrosine- and SH2-independent, Oncogene 10 (1995) 723–729. [32] J. Martin-Nieto, A. Villalobo, The human epidermal growth factor receptor contains a juxtamembrane calmodulin-binding site, Biochemistry 37 (1998) 227–236. [33] H. Li, A. Villalobo, Evidence for the direct interaction between calmodulin and the human epidermal growth factor receptor, Biochem. J. 362 (2002) 499–505. [34] S.R. Hubbard, Crystal structure of the activated insulin receptor tyrosine kinase in complex with peptide substrate and ATP analog, EMBO J. 16 (1997) 5572–5581. [35] J.C. den Hartigh, P.M. van Bergen en Henegouwen, A.J. Verkleij, J. Boonstra, The EGF receptor is an actin-binding protein, J. Cell. Biol. 119 (1992) 349–355. [36] C.T. Kuan, C.J. Wikstrand, D.D. Bigner, EGF mutant receptor vIII as a molecular target in cancer therapy, Endocr. Relat. Cancer 8 (2001) 83–96. [37] H.S. Huang, M. Nagane, C.K. Klingbeil, H. Lin, R. Nishikawa, X.D. Ji, C.M. Huang, G.N. Gill, H.S. Wiley, W.K. Cavenee, The enhanced tumorigenic activity of a mutant epidermal growth factor receptor common in human cancers is mediated by threshold levels of constitutive tyrosine phosphorylation and unattenuated signaling, J. Biol. Chem. 272 (1997) 2927–2935. [38] W. Weber, G.N. Gill, J. Spiess, Production of an epidermal growth factor receptor-related protein, Science 224 (1984) 294–297.

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling [39] J.V. Ilekis, B.C. Stark, B. Scoccia, Possible role of variant RNA transcripts in the regulation of epidermal growth factor receptor expression in human placenta, Mol. Reprod. Dev. 41 (1995) 149–156. [40] J.V. Ilekis, J. Gariti, C. Niederberger, B. Scoccia, Expression of a truncated epidermal growth factor receptor-like protein (TEGFR) in ovarian cancer, Gynecol. Oncol. 65 (1997) 36–41. [41] J.L. Reiter, N.J. Maihle, A 1.8 kb alternative transcript from the human epidermal growth factor receptor gene encodes a truncated form of the receptor, Nucleic Acids. Res. 24 (1996) 4050–4056. [42] L.C. Groenen, E.C. Nice, A.W. Burgess, Structure-function relationships for the EGF/TGF-alpha family of mitogens, Growth Factors 11 (1994) 235–257. [43] E.C. Nice, T. Domagala, L. Fabri, M. Nerriea, F. Walker, R.N. Jorissen, A.W. Burgess, D.F. Cui, Y.S. Zhang, Rapid microscale enzymic semisynthesis of epidermal growth factor (EGF) analogues, Growth Factors 20 (2002) 71–80. [44] L. Strachan, J.G. Murison, R.L. Prestidge, M.A. Sleeman, J.D. Watson, K.D. Kumble, Cloning and biological activity of epigen, a novel member of the epidermal growth factor superfamily, J. Biol. Chem. 276 (2001) 18265–18271. [45] S.A. Thompson, A. Harris, D. Hoang, M. Ferrer, G.R. Johnson, COOH-terminal extended recombinant amphiregulin with bioactivity comparable with naturally derived growth factor, J. Biol. Chem. 271 (1996) 17927–17931. [46] B. Neelam, A. Richter, S.G. Chamberlin, S.M. Puddicombe, L. Wood, M.B. Murray, K. Nandagopal, S.K. Niyogi, D.E. Davies, Structure-function studies of ligand-induced epidermal growth factor receptor dimerization, Biochemistry 37 (1998) 4884–4891. [47] J.T. Jones, R.W. Akita, M.X. Sliwkowski, Binding specificities and affinities of egf domains for ErbB receptors, FEBS Lett. 447 (1999) 227–231. [48] D. Harari, E. Tzahar, J. Romano, M. Shelly, J.H. Pierce, G.C. Andrews, Y. Yarden, Neuregulin-4: a novel growth factor that acts through the ErbB-4 receptor tyrosine kinase, Oncogene 18 (1999) 2681–2689. [49] D. Kohda, M. Odaka, I. Lax, H. Kawasaki, K. Suzuki, A. Ullrich, J. Schlessinger, F. Inagaki, A 40-kDa epidermal growth factor/transforming growth factor alpha-binding domain produced by limited proteolysis of the extracellular domain of the epidermal growth factor receptor, J. Biol. Chem. 268 (1993) 1976–1981. [50] M.A. Lemmon, Z. Bu, J.E. Ladbury, M. Zhou, D. Pinchasi, I. Lax, D.M. Engelman, J. Schlessinger, Two EGF molecules contribute additively to stabilization of the EGFR dimer, EMBO J. 16 (1997) 281–294. [51] J.H. Kim, K. Saito, S. Yokoyama, Chimeric receptor analyses of the interactions of the ectodomains of ErbB-1 with epidermal growth factor and of those of ErbB-4 with neuregulin, Eur. J. Biochem. 269 (2002) 2323–2329. [52] E. Singer, R. Landgraf, T. Horan, D. Slamon, D. Eisenberg, Identification of a heregulin binding site in HER3 extracellular domain, J. Biol. Chem. 276 (2001) 44266–44274. [53] E.G. Barbacci, B.C. Guarino, J.G. Stroh, D.H. Singleton, K.J. Rosnack, J.D. Moyer, G.C. Andrews, The structural basis for the specificity of epidermal growth factor and heregulin binding, J. Biol. Chem. 270 (1995) 9585–9589. [54] A. Harris, M. Adler, J. Brink, R. Lin, M. Foehr, M. Ferrer, B.C. Langton-Webster, R.N. Harkins, S.A. Thompson, Homologue scanning mutagenesis of heregulin reveals receptor specific binding epitopes, Biochem. Biophys. Res. Commun. 251 (1998) 220– 224. [55] C. Stortelers, C. Souriau, E. van Liempt, M.L. van de Poll, E.J. van Zoelen, Role of the N-terminus of epidermal growth factor in ErbB2/ErbB-3 binding studied by phage display, Biochemistry 41 (2002) 8732–8741. [56] M. Adler, S.A. Thompson, The solution structure of heregulin-alpha and a N-terminal mutant with suppressed activity, Biochem. Biophys. Res. Commun. 256 (1999) 156–161.

49

[57] R.R. Beerli, N.E. Hynes, Epidermal growth factor-related peptides activate distinct subsets of ErbB receptors and differ in their biological activities, J. Biol. Chem. 271 (1996) 6071–6076. [58] C.R. Lin, W.S. Chen, C.S. Lazar, C.D. Carpenter, G.N. Gill, R.M. Evans, M.G. Rosenfeld, Protein kinase C phosphorylation at Thr 654 of the unoccupied EGF receptor and EGF binding regulate functional receptor loss by independent mechanisms, Cell 44 (1986) 839–848. [59] M. Boni-Schnetzler, P.F. Pilch, Mechanism of epidermal growth factor receptor autophosphorylation and high-affinity binding, Proc. Natl. Acad. Sci. USA 84 (1987) 7832–7836. [60] D.R. Hurwitz, S.L. Emanuel, M.H. Nathan, N. Sarver, A. Ullrich, S. Felder, I. Lax, J. Schlessinger, EGF induces increased ligand binding affinity and dimerization of soluble epidermal growth factor (EGF) receptor extracellular domain, J. Biol. Chem. 266 (1991) 22035–22043. [61] A. Sorokin, M.A. Lemmon, A. Ullrich, J. Schlessinger, Stabilization of an active dimeric form of the epidermal growth factor receptor by introduction of an inter-receptor disulfide bond, J. Biol. Chem. 269 (1994) 9752–9759. [62] M. Martin-Fernandez, D.T. Clarke, M.J. Tobin, S.V. Jones, G.R. Jones, Preformed oligomeric epidermal growth factor receptors undergo an ectodomain structure change during signaling, Biophys. J. 82 (2002) 2415–2427. [63] R. Prywes, E. Livneh, A. Ullrich, J. Schlessinger, Mutations in the cytoplasmic domain of EGF receptor affect EGF binding and receptor internalization, EMBO J. 5 (1986) 2179–2190. [64] K.J. Fowler, F. Walker, W. Alexander, M.L. Hibbs, E.C. Nice, R.M. Bohmer, G.B. Mann, C. Thumwood, R. Maglitto, J.A. Danks, A mutation in the epidermal growth factor receptor in waved-2 mice has a profound effect on receptor biochemistry that results in impaired lactation, Proc. Natl. Acad. Sci. USA 92 (1995) 1465– 1469. [65] L.C. Groenen, F. Walker, A.W. Burgess, H.R. Treutlein, A model for the activation of the epidermal growth factor receptor kinase involvement of an asymmetric dimer? Biochemistry 36 (1997) 3826–3836. [66] F. Walker, M.L. Hibbs, H.H. Zhang, L.J. Gonez, A.W. Burgess, Biochemical characterization of mutant EGF receptors expressed in the hemopoietic cell line BaF/3, Growth Factors 16 (1998) 53–67. [67] M.A. Van der Heyden, M. Nievers, A.J. Verkleij, J. Boonstra, P.M. van Bergen en Henegouwen, Identification of an intracellular domain of the EGF receptor required for high-affinity binding of EGF, FEBS Lett. 410 (1997) 265–268. [68] F. Walker, A.W. Burgess, Reconstitution of the high affinity epidermal growth factor receptor on cell-free membranes after transmodulation by platelet-derived growth factor, J. Biol. Chem. 266 (1991) 2746–2752. [69] M.R. Holbrook, L.L. Slakey, D.J. Gross, Thermodynamic mixing of molecular states of the epidermal growth factor receptor modulates macroscopic ligand binding affinity, Biochem. J. 352 (Pt 1) (2000) 99–108. [70] M. Odaka, D. Kohda, I. Lax, J. Schlessinger, F. Inagaki, Ligandbinding enhances the affinity of dimerization of the extracellular domain of the epidermal growth factor receptor, J. Biochem. (Tokyo) 122 (1997) 116–121. [71] T. Domagala, N. Konstantopoulos, F. Smyth, R.N. Jorissen, L. Fabri, D. Geleick, I. Lax, J. Schlessinger, W. Sawyer, G.J. Howlett, A.W. Burgess, E.C. Nice, Stoichiometry, kinetic and binding analysis of the interaction between epidermal growth factor (EGF) and the extracellular domain of the EGF receptor, Growth Factors 18 (2000) 11–29. [72] Y. Sako, S. Minoghchi, T. Yanagida, Single-molecule imaging of EGFR signalling on the surface of living cells, Nat. Cell. Biol. 2 (2000) 168–172. [73] T. Moriki, H. Maruyama, I.N. Maruyama, Activation of preformed EGF receptor dimers by ligand-induced rotation of the transmembrane domain, J. Mol. Biol. 311 (2001) 1011–1026.

50


[74] A. Berezov, J. Chen, Q. Liu, H.T. Zhang, M.I. Greene, R. Murali, Disabling receptor ensembles with rationally designed interface peptidomimetics, J. Biol. Chem. 277 (2002) 28330–28339. [75] J. Schlessinger, Dynamics of hormone receptors on cell membrane, Ann. NY Acad. Sci. 366 (1981) 274–284. [76] A. Chantry, The kinase domain and membrane localization determine intracellular interactions between epidermal growth factor receptors, J. Biol. Chem. 270 (1995) 3068–3073. [77] K.G. Tanner, J. Kyte, Dimerization of the extracellular domain of the receptor for epidermal growth factor containing the membranespanning segment in response to treatment with epidermal growth factor, J. Biol. Chem. 274 (1999) 35985–35990. [78] D. Graus-Porta, R.R. Beerli, J.M. Daly, N.E. Hynes, ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling, EMBO J. 16 (1997) 1647–1655. [79] D. Karunagaran, E. Tzahar, R.R. Beerli, X. Chen, D. Graus-Porta, B.J. Ratzkin, R. Seger, N.E. Hynes, Y. Yarden, ErbB-2 is a common auxiliary subunit of NDF and EGF receptors: implications for breast cancer, EMBO J. 15 (1996) 254–264. [80] A.E. Lenferink, R. Pinkas-Kramarski, M.L. van de Poll, M.J. van Vugt, L.N. Klapper, E. Tzahar, H. Waterman, M. Sela, E.J. van Zoelen, Y. Yarden, Differential endocytic routing of homo- and hetero-dimeric ErbB tyrosine kinases confers signaling superiority to receptor heterodimers, EMBO J. 17 (1998) 3385–3397. [81] G.C. Huang, X. Ouyang, R.J. Epstein, Proxy activation of protein ErbB2 by heterologous ligands implies a heterotetrameric mode of receptor tyrosine kinase interaction, Biochem. J. 331 (Pt 1) (1998) 113–119. [82] K.M. Ferguson, P.J. Darling, M.J. Mohan, T.L. Macatee, M.A. Lemmon, Extracellular domains drive homo- but not heterodimerization of erbB receptors, EMBO J. 19 (2000) 4632–4643. [83] D.C. Gamett, G. Pearson, R.A. Cerione, I. Friedberg, Secondary dimerization between members of the epidermal growth factor receptor family, J. Biol. Chem. 272 (1997) 12052–12056. [84] L.E. Johannessen, K.E. Haugen, A.C. Ostvold, E. Stang, I.H. Madshus, Heterodimerization of the epidermal-growth-factor (EGF) receptor and ErbB2 and the affinity of EGF binding are regulated by different mechanisms, Biochem. J. 356 (2001) 87–96. [85] Y. Saito, J. Haendeler, Y. Hojo, K. Yamamoto, B.C. Berk, Receptor heterodimerization: essential mechanism for platelet-derived growth factor-induced epidermal growth factor receptor transactivation, Mol. Cell. Biol. 21 (2001) 6387–6394. [86] T.W. Gadella Jr., T.M. Jovin, Oligomerization of epidermal growth factor receptors on A431 cells studied by time-resolved fluorescence imaging microscopy. A stereochemical model for tyrosine kinase receptor activation, J. Cell. Biol. 129 (1995) 1543–1558. [87] X. Yu, K.D. Sharma, T. Takahashi, R. Iwamoto, E. Mekada, Ligandindependent dimer formation of epidermal growth factor receptor (EGFR) is a step separable from ligand-induced EGFR signaling, Mol. Biol. Cell. 13 (2002) 2547–2557. [88] P.A. Humphrey, A.J. Wong, B. Vogelstein, H.S. Friedman, M.H. Werner, D.D. Bigner, S.H. Bigner, Amplification and expression of the epidermal growth factor receptor gene in human glioma xenografts, Cancer Res. 48 (1988) 2231–2238. [89] H. Yamazaki, Y. Fukui, Y. Ueyama, N. Tamaoki, T. Kawamoto, S. Taniguchi, M. Shibuya, Amplification of the structurally and functionally altered epidermal growth factor receptor gene (c-erbB) in human brain tumors, Mol. Cell. Biol. 8 (1988) 1816–1820. [90] L.T. Malden, U. Novak, A.H. Kaye, A.W. Burgess, Selective amplification of the cytoplasmic domain of the epidermal growth factor receptor gene in glioblastoma multiforme, Cancer Res. 48 (1988) 2711–2714. [91] M.M. Kwatra, D.D. Bigner, J.A. Cohn, The ligand binding domain of the epidermal growth factor receptor is not required for receptor dimerization, Biochim. Biophys. Acta (1134) (1992) 178–181. [92] A.J. Ekstrand, N. Longo, M.L. Hamid, J.J. Olson, L. Liu, V.P. Collins, C.D. James, Functional characterization of an EGF receptor

[93]

[94]

[95]

[96]

[97]

[98]

[99]

[100]

[101]

[102] [103]

[104]

[105]

[106]

[107]

[108]

[109]

[110]

[111]

with a truncated extracellular domain expressed in glioblastomas with EGFR gene amplification Oncogene 9 (1994) 2313–2320. H. Fernandes, S. Cohen, S. Bishayee, Glycosylation-induced conformational modification positively regulates receptor-receptor association a study with an aberrant epidermal growth factor receptor (EGFRvIII/DeltaEGFR) expressed in cancer cells, J. Biol. Chem. 276 (2001) 5375–5383. F. Canals, Signal transmission by epidermal growth factor receptor: coincidence of activation and dimerization, Biochemistry 31 (1992) 4493–4501. J.M. Sherrill, Insufficiency of self-phosphorylation for the activation of epidermal growth factor receptor, Biochemistry 36 (1997) 5677–5684. S.R. Hubbard, M. Mohammadi, J. Schlessinger, Autoregulatory mechanisms in protein-tyrosine kinases, J. Biol. Chem. 273 (1998) 11987–11990. D.A. Tice, J.S. Biscardi, A.L. Nickles, S.J. Parsons, Mechanism of biological synergy between cellular Src and epidermal growth factor receptor, Proc. Natl. Acad. Sci. USA 96 (1999) 1415–1420. H. Yamaguchi, W.A. Hendrickson, Structural basis for activation of human lymphocyte kinase Lck upon tyrosine phosphorylation, Nature 384 (1996) 484–489. C.L. Arteaga, T.T. Ramsey, L.K. Shawver, C.A. Guyer, Unliganded epidermal growth factor receptor dimerization induced by direct interaction of quinazolines with the ATP binding site, J. Biol. Chem. 272 (1997) 23247–23254. R.B. Lichtner, A. Menrad, A. Sommer, U. Klar, M.R. Schneider, Signaling-inactive epidermal growth factor receptor/ligand complexes in intact carcinoma cells by quinazoline tyrosine kinase inhibitors, Cancer Res. 61 (2001) 5790–5795. R.B. Montgomery, Antagonistic and agonistic effects of quinazoline tyrosine kinase inhibitors on mutant EGF receptor function, Int. J. Cancer 101 (2002) 111–117. M. Huse, J. Kuriyan, The conformational plasticity of protein kinases, Cell 109 (2002) 275–282. G. Ge, J. Wu, Y. Wang, Q. Lin, Activation mechanism of solubilized epidermal growth factor receptor tyrosine kinase, Biochem. Biophys. Res. Commun. 290 (2002) 914–920. R.J. Daly, Take your partners, please–signal diversification by the erbB family of receptor tyrosine kinases, Growth Factors 16 (1999) 255–263. T.B. Deb, L. Su, L. Wong, E. Bonvini, A. Wells, M. David, G.R. Johnson, Epidermal growth factor (EGF) receptor kinaseindependent signaling by EGF, J. Biol. Chem. 276 (2001) 15554– 15560. J.A. Ewald, K.J. Coker, J.O. Price, J.V. Staros, C.A. Guyer, Stimulation of mitogenic pathways through kinase-impaired mutants of the epidermal growth factor receptor, Exp. Cell. Res. 268 (2001) 262–273. T. Yamauchi, K. Ueki, K. Tobe, H. Tamemoto, N. Sekine, M. Wada, M. Honjo, M. Takahashi, T. Takahashi, H. Hirai, T. Tushima, Y. Akanuma, T. Fujita, I. Komuro, Y. Yazaki, T. Kadowaki, Tyrosine phosphorylation of the EGF receptor by the kinase Jak2 is induced by growth hormone, Nature 390 (1997) 91–96. T. Fukazawa, S. Miyake, V. Band, H. Band, Tyrosine phosphorylation of Cb1 upon epidermal growth factor (EGF) stimulation and its association with EGF receptor and downstream signaling proteins, J. Biol. Chem. 271 (1996) 14554–14559. M. Wheeler, J. Domin, Recruitment of the class II phosphoinositide 3-kinase C2beta to the epidermal growth factor receptor: role of Grb2, Mol. Cell. Biol. 21 (2001) 6660–6667. M.T. Kloth, K.K. Laughlin, J.S. Biscardi, J.L. Boerner, S.J. Parsons, C.M. Silva, STAT5b, a mediator of synergism between c-Src and the EGF receptor, J. Biol. Chem. 278 (2003) 1671–1679. N.J. Hellyer, K. Cheng, J.G. Koland, ErbB3 (HER3) interaction with the p85 regulatory subunit of phosphoinositide 3-kinase, Biochem. J. 333 (Pt 3) (1998) 757–763.

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling [112] Z. Galcheva-Gargova, K.N. Konstantinov, I.H. Wu, F.G. Klier, T. Barrett, R.J. Davis, Binding of zinc finger protein ZPR1 to the epidermal growth factor receptor, Science 272 (1996) 1795– 1797. [113] L. Xia, L. Wang, A.S. Chung, S.S. Ivanov, M.Y. Ling, A.M. Dragoi, A. Platt, T.M. Gilmer, X.Y. Fu, Y.E. Chin, Identification of both positive and negative domains within the epidermal growth factor receptor COOH-terminal region for signal transducer and activator of transcription (STAT) activation, J. Biol. Chem. 277 (2002) 30716–30723. [114] M.A. Olayioye, I. Beuvink, K. Horsch, J.M. Daly, N.E. Hynes, ErbB receptor-induced activation of stat transcription factors is mediated by Src tyrosine kinases, J. Biol. Chem. 274 (1999) 17209–17218. [115] F. Walker, A. Kato, L.J. Gonez, M.L. Hibbs, N. Pouliot, A. Levitzki, A.W. Burgess, Activation of the Ras/mitogen-activated protein kinase pathway by kinase-defective epidermal growth factor receptors results in cell survival but not proliferation, Mol. Cell. Biol. 18 (1998) 7192–7204. [116] G.M. Walton, W.S. Chen, M.G. Rosenfeld, G.N. Gill, Analysis of deletions of the carboxyl terminus of the epidermal growth factor receptor reveals self-phosphorylation at tyrosine 992 and enhanced in vivo tyrosine phosphorylation of cell substrates, J. Biol. Chem. 265 (1990) 1750–1754. [117] N. Gotoh, A. Tojo, K. Muroya, Y. Hashimoto, S. Hattori, S. Nakamura, T. Takenawa, Y. Yazaki, M. Shibuya, Epidermal growth factor-receptor mutant lacking the autophosphorylation sites induces phosphorylation of Shc protein and Shc-Grb2/ASH association and retains mitogenic activity, Proc. Natl. Acad. Sci. USA 91 (1994) 167–171. [118] L. Wong, T.B. Deb, S.A. Thompson, A. Wells, G.R. Johnson, A differential requirement for the COOH-terminal region of the epidermal growth factor (EGF) receptor in amphiregulin and EGF mitogenic signaling, J. Biol. Chem. 274 (1999) 8900–8909. [119] D.R. Stover, M. Becker, J. Liebetanz, N.B. Lydon, Src phosphorylation of the epidermal growth factor receptor at novel sites mediates receptor interaction with Src and P85 alpha, J. Biol. Chem. 270 (1995) 15591–15597. [120] E.J. Lowenstein, R.J. Daly, A.G. Batzer, W. Li, B. Margolis, R. Lammers, A. Ullrich, E.Y. Skolnik, D. Bar-Sagi, J. Schlessinger, The SH2 and SH3 domain-containing protein GRB2 links receptor tyrosine kinases to ras signaling, Cell 70 (1992) 431–442. [121] A.G. Batzer, D. Rotin, J.M. Urena, E.Y. Skolnik, J. Schlessinger, Hierarchy of binding sites for Grb2 and Shc on the epidermal growth factor receptor, Mol. Cell. Biol. 14 (1994) 5192–5201. [122] T. Sasaoka, W.J. Langlois, J.W. Leitner, B. Draznin, J.M. Olefsky, The signaling pathway coupling epidermal growth factor receptors to activation of p21ras, J. Biol. Chem. 269 (1994) 32621–32625. [123] K. Sakaguchi, Y. Okabayashi, Y. Kido, S. Kimura, Y. Matsumura, K. Inushima, M. Kasuga, Shc phosphotyrosine-binding domain dominantly interacts with epidermal growth factor receptors and mediates Ras activation in intact cells, Mol. Endocrinol. 12 (1998) 536– 543. [124] A. Hashimoto, M. Kurosaki, N. Gotoh, M. Shibuya, T. Kurosaki, Shc regulates epidermal growth factor-induced activation of the JNK signaling pathway, J. Biol. Chem. 274 (1999) 20139–20143. [125] B. Hallberg, S.I. Rayter, J. Downward, Interaction of Ras and Raf in intact mammalian cells upon extracellular stimulation, J. Biol. Chem. 269 (1994) 3913–3916. [126] G.L. Johnson, R.R. Vaillancourt, Sequential protein kinase reactions controlling cell growth and differentiation, Curr. Opin. Cell. Biol. 6 (1994) 230–238. [127] W.J. Langlois, T. Sasaoka, A.R. Saltiel, J.M. Olefsky, Negative feedback regulation and desensitization of insulin- and epidermal growth factor-stimulated p21ras activation, J. Biol. Chem. 270 (1995) 25320–25323. [128] T. Jun, O. Gjoerup, T.M. Roberts, Tangled webs: evidence of crosstalk between c-Raf-1 and Akt, Sci. STKE. (1999) (1999) E1.

51

[129] C. Liebmann, Regulation of MAP kinase activity by peptide receptor signalling pathwayparadigms of multiplicity, Cell. Signal. 13 (2001) 777–785. [130] J. Pouyssegur, V. Volmat, P. Lenormand, Fidelity and spatiotemporal control in MAP kinase (ERKs) signalling, Biochem. Pharmacol. 64 (2002) 755–763. [131] H. Meisner, M.P. Czech, Coupling of the proto-oncogene product cCb1 to the epidermal growth factor receptor, J. Biol. Chem. 270 (1995) 25332–25335. [132] D.D. Schlaepfer, C.R. Hauck, D.J. Sieg, Signaling through focal adhesion kinase, Prog. Biophys. Mol. Biol. 71 (1999) 435–478. [133] X.X. Xu, T. Yi, B. Tang, J.D. Lambeth, Disabled-2 (Dab2) is an SH3 domain-binding partner of Grb2, Oncogene 16 (1998) 1561–1569. [134] P. De Sepulveda, K. Okkenhaug, J.L. Rose, R.G. Hawley, P. Dubreuil, R. Rottapel, Socs1 binds to multiple signalling proteins and suppresses steel factor-dependent proliferation, EMBO J. 18 (1999) 904–915. [135] S.L. Harmer, A.L. DeFranco, The src homology domain 2containing inositol phosphatase SHIP forms a ternary complex with Shc and Grb2 in antigen receptor-stimulated B lymphocytes, J. Biol. Chem. 274 (1999) 12183–12191. [136] G. Pelicci, L. Lanfrancone, A.E. Salcini, A. Romano, S. Mele, B.M. Grazia, O. Segatto, P.P. Di Fiore, P.G. Pelicci, Constitutive phosphorylation of Shc proteins in human tumors, Oncogene 11 (1995) 899–907. [137] M. Pomerance, M.C. Multon, F. Parker, C. Venot, J.P. Blondeau, B. Tocque, F. Schweighoffer, Grb2 interaction with MEK-kinase 1 is involved in regulation of Jun-kinase activities in response to epidermal growth factor, J. Biol. Chem. 273 (1998) 24301–24304. [138] Y. Xu, D.F. Guo, M. Davidson, T. Inagami, G. Carpenter, Interaction of the adaptor protein Shc and the adhesion molecule cadherin, J. Biol. Chem. 272 (1997) 13463–13466. [139] M. Damelin, P.A. Silver, Mapping interactions between nuclear transport factors in living cells reveals pathways through the nuclear pore complex, Mol. Cell. 5 (2000) 133–140. [140] A.P. Belsches, M.D. Haskell, S.J. Parsons, Role of c-Src tyrosine kinase in EGF-induced mitogenesis, Front. Biosci. 2 (1997) d501–d518. [141] D.K. Luttrell, L.M. Luttrell, S.J. Parsons, Augmented mitogenic responsiveness to epidermal growth factor in murine fibroblasts that overexpress pp60c-src, Mol. Cell. Biol. 8 (1988) 497–501. [142] M.C. Maa, T.H. Leu, D.J. McCarley, R.C. Schatzman, S.J. Parsons, Potentiation of epidermal growth factor receptor-mediated oncogenesis by c-Src: implications for the etiology of multiple human cancers, Proc. Natl. Acad. Sci. USA 92 (1995) 6981–6985. [143] S. Roche, M. Koegl, M.V. Barone, M.F. Roussel, S.A. Courtneidge, DNA synthesis induced by some but not all growth factors requires Src family protein tyrosine kinases, Mol. Cell. Biol. 15 (1995) 1102–1109. [144] M.V. Barone, S.A. Courtneidge, Myc but not Fos rescue of PDGF signalling block caused by kinase-inactive Src [see comments], Nature 378 (1995) 509–512. [145] R. Karni, R. Jove, A. Levitzki, Inhibition of pp60c-Src reduces Bcl-XL expression and reverses the transformed phenotype of cells overexpressing EGF and HER-2 receptors, Oncogene 18 (1999) 4654–4662. [146] N. Osherov, A. Levitzki, Epidermal-growth-factor-dependent activation of the src-family kinases, Eur. J. Biochem. 225 (1994) 1047–1053. [147] W. Mao, R. Irby, D. Coppola, L. Fu, M. Wloch, J. Turner, H. Yu, R. Garcia, R. Jove, T.J. Yeatman, Activation of c-Src by receptor tyrosine kinases in human colon cancer cells with high metastatic potential, Oncogene 15 (1997) 3083–3090. [148] J.S. Biscardi, M.C. Maa, D.A. Tice, M.E. Cox, T.H. Leu, S.J. Parsons, c-Src-mediated phosphorylation of the epidermal growth factor receptor on Tyr845 and Tyr1101 is associated with modulation of receptor function, J. Biol. Chem. 274 (1999) 8335–8343.

52


[149] C.R. Lombardo, T.G. Consler, D.B. Kassel, In vitro phosphorylation of the epidermal growth factor receptor autophosphorylation domain by c-src: identification of phosphorylation sites and c-src SH2 domain binding sites, Biochemistry 34 (1995) 16456–16466. [150] K. Sato, A. Sato, M. Aoto, Y. Fukami, c-Src phosphorylates epidermal growth factor receptor on tyrosine 845, Biochem. Biophys. Res. Commun. 215 (1995) 1078–1087. [151] R.L. Kendall, R.Z. Rutledge, X. Mao, A.J. Tebben, R.W. Hungate, K.A. Thomas, Vascular endothelial growth factor receptor KDR tyrosine kinase activity is increased by autophosphorylation of two activation loop tyrosine residues, J. Biol. Chem. 274 (1999) 6453–6460. [152] M.E. Cunningham, L.A. Greene, A function-structure model for NGF-activated TRK, EMBO J. 17 (1998) 7282–7293. [153] N. Gotoh, A. Tojo, M. Hino, Y. Yakazi, M. Shibuya, A highly conserved tyrosine residue at codon 845 within the kinase domain is not required for the transforming activity of human epidermal growth factor receptor, Biochem. Biophys. Res. Commun. 186 (1992) 768–774. [154] J.D. Wright, C.W. Reuter, M.J. Weber, An incomplete program of cellular tyrosine phosphorylations induced by kinase-defective epidermal growth factor receptors, J. Biol. Chem. 270 (1995) 12085–12093. [155] K. Sato, N. Gotoh, T. Otsuki, M. Kakumoto, M. Aoto, A.A. Tokmakov, M. Shibuya, Y. Fukami, Tyrosine residues 239 and 240 of Shc are phosphatidylinositol 4,5-bisphosphate-dependent phosphorylation sites by c-Src, Biochem. Biophys. Res. Commun. 240 (1997) 399–404. [156] J.H. Chang, S. Gill, J. Settleman, S.J. Parsons, c-Src regulates the simultaneous rearrangement of actin cytoskeleton, p190RhoGAP, and p120RasGAP following epidermal growth factor stimulation, J. Cell. Biol. 130 (1995) 355–368. [157] R.W. Roof, M.D. Haskell, B.D. Dukes, N. Sherman, M. Kinter, S.J. Parsons, Phosphotyrosine (p-Tyr)-dependent and -independent mechanisms of p190 RhoGAP-p120 RasGAP interaction: Tyr 1105 of p190, a substrate for c-Src, is the sole p-Tyr mediator of complex formation, Mol. Cell. Biol. 18 (1998) 7052–7063. [158] Y. Nojima, T. Mimura, N. Morino, K. Hamasaki, H. Furuya, R. Sakai, T. Nakamoto, Y. Yazaki, H. Hirai, Tyrosine phosphorylation of p130Cas in cell adhesion and transformation, Hum. Cell 9 (1996) 169–174. [159] C. Huang, J. Liu, C.C. Haudenschild, X. Zhan, The role of tyrosine phosphorylation of cortactin in the locomotion of endothelial cells, J. Biol. Chem. 273 (1998) 25770–25776. [160] O. Lohi, V.P. Lehto, EAST, a novel EGF receptor substrate, associates with focal adhesions and actin fibers, FEBS Lett. 436 (1998) 419–423. [161] C. Provenzano, R. Gallo, R. Carbone, P.P. Di Fiore, G. Falcone, L. Castellani, S. Alema, Eps8, a tyrosine kinase substrate, is recruited to the cell cortex and dynamic F-actin upon cytoskeleton remodeling, Exp. Cell. Res. 242 (1998) 186–200. [162] S.E. Shoelson, M. Sivaraja, K.P. Williams, P. Hu, J. Schlessinger, M.A. Weiss, Specific phosphopeptide binding regulates a conformational change in the PI 3-kinase SH2 domain associated with enzyme activation, EMBO J. 12 (1993) 795–802. [163] J.E. Damell Jr., STATs and gene regulation, Science 277 (1997) 1630–1635. [164] J.N. Ihle, T. Nosaka, W. Thierfelder, F.W. Quelle, K. Shimoda, Jaks and Stats in cytokine signaling, Stem Cells 15 Suppl 1 (1997) 105–111. [165] W.J. Leonard, Role of Jak kinases and STATs in cytokine signal transduction, Int. J. Hematol. 73 (2001) 271–277. [166] O.K. Park, T.S. Schaefer, D. Nathans, In vitro activation of Stat3 by epidermal growth factor receptor kinase, Proc. Natl. Acad. Sci. USA 93 (1996) 13704–13708. [167] M. David, L. Wong, R. Flavell, S.A. Thompson, A. Wells, A.C. Larner, G.R. Johnson, STAT activation by epidermal growth factor

[168]

[169]

[170]

[171]

[172]

[173]

[174]

[175] [176]

[177]

[178]

[179]

[180] [181]

[182]

[183]

[184]

[185]

(EGF) and amphiregulin. Requirement for the EGF receptor kinase but not for tyrosine phosphorylation sites or JAK1, J. Biol. Chem. 271 (1996) 9185–9188. D.W. Leaman, S. Pisharody, T.W. Flickinger, M.A. Commane, J. Schlessinger, I.M. Kerr, D.E. Levy, G.R. Stark, Roles of JAKs in activation of STATs and stimulation of c-fos gene expression by epidermal growth factor, Mol. Cell. Biol. 16 (1996) 369–375. M.G. Houle, S. Bourgoin, Regulation of phospholipase D by phosphorylation-dependent mechanisms, Biochim. Biophys. Acta 1439 (1999) 135–149. R. Slaaby, T. Jensen, H.S. Hansen, M.A. Frohman, K. Seedorf, PLD2 complexes with the EGF receptor and undergoes tyrosine phosphorylation at a single site upon agonist stimulation, J. Biol. Chem. 273 (1998) 33722–33727. S.M. Hernandez-Sotomayor, G. Carpenter, Non-catalytic activation of phospholipase C-gamma 1 in vitro by epidermal growth factor receptor, Biochem. J. 293 (Pt 2) (1993) 507–511. A. Kamat, G. Carpenter, Phospholipase C-gamma1: regulation of enzyme function and role in growth factor-dependent signal transduction, Cytokine Growth Factor Rev. 8 (1997) 109–117. A. Chattopadhyay, M. Vecchi, Q. Ji, R. Mernaugh, G. Carpenter, The role of individual SH2 domains in mediating association of phospholipase C-gamma 1 with the activated EGF receptor, J. Biol. Chem. 274 (1999) 26091–26097. M. Wahl, S. Nishibe, J.W. Kim, H. Kim, S.G. Rhee, G. Carpenter, Identification of two epidermal growth factor-sensitive tyrosine phosphorylation sites of phospholipase C-gamma in intact HSC-1 cells, J. Biol. Chem. 265 (1990) 3944–3948. F. Hofer, R. Berdeaux, G.S. Martin, Ras-independent activation of Ral by a Ca(2+)-dependent pathway, Curr. Biol. 8 (1998) 839–842. L. Sun, G. Carpenter, Epidermal growth factor activation of NFkappaB is mediated through IkappaBalpha degradation and intracellular free calcium, Oncogene 16 (1998) 2095–2102. R. Marais, Y. Light, C. Mason, H. Paterson, M.F. Olson, C.J. Marshall, Requirement of Ras-GTP-Raf complexes for activation of Raf-1 by protein kinase C [published erratum appears in Science 1998 May 15;280(5366):987], Science 280 (1998) 109–112. M. McClellan, P. Kievit, N. Auersperg, K. Rodland, Regulation of proliferation and apoptosis by epidermal growth factor and protein kinase C in human ovarian surface epithelial cells, Exp. Cell. Res. 246 (1999) 471–479. Y.S. Liaw, P.C. Yang, C.J. Yu, S.H. Kuo, K.T. Luh, Y.J. Lin, M.L. Wu, PKC activation is required by EGF-stimulated Na(+)-H+ exchanger in human pleural mesothelial cells, Am. J. Physiol. 274 (1998) L665–L672. L.C. Cantley, The phosphoinositide 3-kinase pathway, Science 296 (2002) 1655–1657. S. Djordjevic, P.C. Driscoll, Structural insight into substrate specificity and regulatory mechanisms of phosphoinositide 3-kinases, Trends. Biochem. Sci. 27 (2002) 426–432. C.L. Carpenter, K.R. Auger, M. Chanudhuri, M. Yoakim, B. Schaffhausen, S. Shoelson, L.C. Cantley, Phosphoinositide 3-kinase is activated by phosphopeptides that bind to the SH2 domains of the 85-kDa subunit, J. Biol. Chem. 268 (1993) 9478–9483. H.H. Kim, S.L. Sierke, J.G. Koland, Epidermal growth factordependent association of phosphatidylinositol 3-kinase with the erbB3 gene product, J. Biol. Chem. 269 (1994) 24747–24755. T.G. Ram, S.P. Ethier, Phosphatidylinositol 3-kinase recruitment by p185erbB-2 and erbB-3 is potently induced by neu differentiation factor/heregulin during mitogenesis and is constitutively elevated in growth factor-independent breast carcinoma cells with c-erbB-2 gene amplification, Cell. Growth. Differ. 7 (1996) 551– 561. P.F. Jones, T. Jakubowicz, F.J. Pitossi, F. Maurer, B.A. Hemmings, Molecular cloning and identification of a serine/threonine protein kinase of the second-messenger subfamily, Proc. Natl. Acad. Sci. USA 88 (1991) 4171–4175.

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling [186] K.M. Nicholson, N.G. Anderson, The protein kinase B/Akt signalling pathway in human malignancy, Cell. Signal. 14 (2002) 381–395. [187] M.E. Pacold, S. Suire, O. Perisic, S. Lara-Gonzalez, C.T. Davis, E.H. Walker, P.T. Hawkins, L. Stephens, J.F. Eccleston, R.L. Williams, Crystal structure and functional analysis of Ras binding to its effector phosphoinositide 3-kinase gamma, Cell 103 (2000) 931–943. [188] S. Cohen, Isolation of a mouse submaxillary gland protein accelerating incisor eruption and eyelid opening in the newborn animal, J. Biol. Chem. 237 (1962) 1555–1562. [189] H. Gregory, Isolation and structure of urogastrone and its relationship to epidermal growth factor, Nature 257 (1975) 325–327. [190] J.E. de Larco, G.J. Todaro, Growth factors from murine sarcoma virus-transformed cells, Proc. Natl. Acad. Sci. USA 75 (1978) 4001–4005. [191] J.M. Smith, M.B. Sporn, A.B. Roberts, R. Derynck, M.E. Winkler, H. Gregory, Human transforming growth factor-alpha causes precocious eyelid opening in newborn mice, Nature 315 (1985) 515–516. [192] R.T. Topham, D.J.J. Chiego, V.H. Gattone, D.A. Hinton, R.M. Klein, The effect of epidermal growth factor on neonatal incisor differentiation in the mouse, Dev. Biol. 124 (1987) 532–543. [193] G. Calamandrei, E. Alleva, Epidermal growth factor has both growth-promoting and growth-inhibiting effects on physical and neurobehavioral development of neonatal mice, Brain Res. 477 (1989) 1–6. [194] J.L. Smart, S.V. da, L.R. Malheiros, F.J. Paumgartten, R.F. Massey, Epidermal growth factor advances some aspects of development but retards others in both rats and hamsters, J. Dev. Physiol. 11 (1989) 153–158. [195] R. Vassar, E. Fuchs, Transgenic mice provide new insights into the role of TGF-alpha during epidermal development and differentiation, Genes Dev. 5 (1991) 714–727. [196] A.M. Dominey, X.J. Wang, L.E.J. King, L.B. Nanney, T.A. Gagne, K. Sellheyer, D.S. Bundman, M.A. Longley, J.A. Rothnagel, D.A. Greenhalgh, Targeted overexpression of transforming growth factor alpha in the epidermis of transgenic mice elicits hyperplasia, hyperkeratosis, and spontaneous, squamous papillomas, Cell Growth Differ. 4 (1993) 1071–1082. [197] P.W. Cook, M. Piepkorn, C.H. Clegg, G.D. Plowman, J.M. DeMay, J.R. Brown, M.R. Pittelkow, Transgenic expression of the human amphiregulin gene induces a psoriasis-like phenotype, J. Clin. Invest. 100 (1997) 2286–2294. [198] D.A. Lowden, G.W. Lindemann, G. Merlino, B.D. Barash, J.P. Calvet, V.H. Gattone, Renal cysts in transgenic mice expressing transforming growth factor-alpha [see comments], J. Lab. Clin. Med. 124 (1994) 386–394. [199] V.H. Gattone, K.A. Kuenstler, G.W. Lindemann, X. Lu, B.D.J. Cowley, C.A. Rankin, J.P. Calvet, Renal expression of a transforming growth factor-alpha transgene accelerates the progression of inherited, slowly progressive polycystic kidney disease in the mouse, J. Lab. Clin. Med. 127 (1996) 214–222. [200] B.R. Davies, A.M. Platt-Higgins, G. Schmidt, P.S. Rudland, Development of hyperplasias, preneoplasias, and mammary tumors in MMTV-c-erbB-2 and MMTV-TGF alpha transgenic rats, Am. J. Pathol. 155 (1999) 303–314. [201] E.P. Sandgren, J.A. Schroeder, T.H. Qui, R.D. Palmiter, R.L. Brinster, D.C. Lee, Inhibition of mammary gland involution is associated with transforming growth factor alpha but not c-mycinduced tumorigenesis in transgenic mice, Cancer Res. 55 (1995) 3915–3927. [202] D.J. Hehir, G. McGreal, W.O. Kirwan, W. Kealy, M.P. Brady, c-myc oncogene expressiona marker for females at risk of breast carcinoma, J. Surg. Oncol. 54 (1993) 207–209. [203] D.W. Visscher, T. Wallis, S. Awussah, A. Mohamed, J.D. Crissman, Evaluation of MYC and chromosome 8 copy number in breast carcinoma by interphase cytogenetics, Genes Chromosomes Cancer 18 (1997) 1–7.

53

[204] I. Bieche, R. Lidereau, Genetic alterations in breast cancer, Genes Chromosomes Cancer 14 (1995) 227–251. [205] L.T. Amundadottir, S.J. Nass, G.J. Berchem, M.D. Johnson, R.B. Dickson, Cooperation of TGF alpha and c-Myc in mouse mammary tumorigenesis: coordinated stimulation of growth and suppression of apoptosis, Oncogene 13 (1996) 757–765. [206] H. Murakami, N.D. Sanderson, P. Nagy, P.A. Marino, G. Merlino, S.S. Thorgeirsson, Transgenic mouse model for synergistic effects of nuclear oncogenes and growth factors in tumorigenesis: interaction of c-myc and transforming growth factor alpha in hepatic oncogenesis, Cancer Res. 53 (1993) 1719–1723. [207] S.S. Thorgeirsson, E. Santoni-Rugiu, Interaction of c-myc with transforming growth factor alpha and hepatocyte growth factor in hepatocarcinogenesis, Mutat. Res. 376 (1997) 221–234. [208] E. Santoni-Rugiu, M.R. Jensen, S.S. Thorgeirsson, Disruption of the pRb/E2F pathway and inhibition of apoptosis are major oncogenic events in liver constitutively expressing c-myc and transforming growth factor alpha, Cancer Res. 58 (1998) 123–134. [209] L.M. Sargent, N.D. Sanderson, S.S. Thorgeirsson, Ploidy and karyotypic alterations associated with early events in the development of hepatocarcinogenesis in transgenic mice harboring c-myc and transforming growth factor alpha transgenes, Cancer Res. 56 (1996) 2137–2142. [210] J. Schlegel, A. Merdes, G. Stumm, F.K. Albert, M. Forsting, N. Hynes, M. Kiessling, Amplification of the epidermal-growth-factorreceptor gene correlates with different growth behaviour in human glioblastoma, Int. J. Cancer 56 (1994) 72–77. [211] D.K. Moscatello, M. Holgado-Madruga, A.K. Godwin, G. Ramirez, G. Gunn, P.W. Zoltick, J.A. Biegel, R.L. Hayes, A.J. Wong, Frequent expression of a mutant epidermal growth factor receptor in multiple human tumors, Cancer Res. 55 (1995) 5536–5539. [212] N. Sugawa, A.J. Ekstrand, C.D. James, V.P. Collins, Identical splicing of aberrant epidermal growth factor receptor transcripts from amplified rearranged genes in human glioblastomas, Proc. Natl. Acad. Sci. USA 87 (1990) 8602–8606. [213] A.J. Ekstrand, N. Sugawa, C.D. James, V.P. Collins, Amplified and rearranged epidermal growth factor receptor genes in human glioblastomas reveal deletions of sequences encoding portions of the N- and/or C-terminal tails, Proc. Natl. Acad. Sci. USA 89 (1992) 4309–4313. [214] R. Nishikawa, X.D. Ji, R.C. Harmon, C.S. Lazar, G.N. Gill, W.K. Cavenee, H.J. Huang, A mutant epidermal growth factor receptor common in human glioma confers enhanced tumorigenicity, Proc. Natl. Acad. Sci. USA 91 (1994) 7727–7731. [215] M. Nagane, F. Coufal, H. Lin, O. Bogler, W.K. Cavenee, H.J. Huang, A common mutant epidermal growth factor receptor confers enhanced tumorigenicity on human glioblastoma cells by increasing proliferation and reducing apoptosis, Cancer Res. 56 (1996) 5079–5086. [216] E.C. Holland, W.P. Hively, R.A. DePinho, H.E. Varmus, A constitutively active epidermal growth factor receptor cooperates with disruption of G1 cell-cycle arrest pathways to induce glioma-like lesions in mice, Genes Dev. 12 (1998) 3675–3685. [217] R.M. Bachoo, E.A. Maher, K.L. Ligon, N.E. Sharpless, S.S. Chan, M.J. You, Y. Tang, J. DeFrances, E. Stover, R. Weissleder, D.H. Rowitch, D.N. Louis, R.A. DePinho, Epidermal growth factor receptor and Ink4a/Arf: convergent mechanisms governing terminal differentiation and transformation along the neural stem cell to astrocyte axis, Cancer Cell. 1 (2002) 269–277. [218] S. Okamoto, T. Oka, Evidence for physiological function of epidermal growth factor: pregestational sialoadenectomy of mice decreases milk production and increases offspring mortality during lactation period, Proc. Natl. Acad. Sci. USA 81 (1984) 6059–6063. [219] O. Tsutsumi, Y. Kubota, T. Oka, Effect of sialoadenectomy, treatment with epidermal growth factor (EGF) antiserum and replacement of EGF on the epidermis in mice, J. Endocrinol. 113 (1987) 193–197.

54


[220] N.C. Luetteke, T.H. Qiu, R.L. Peiffer, P. Oliver, O. Smithies, D.C. Lee, TGF alpha deficiency results in hair follicle and eye abnormalities in targeted and waved-1 mice, Cell 73 (1993) 263–278. [221] G.B. Mann, K.J. Fowler, A. Gabriel, E.C. Nice, R.L. Williams, A.R. Dunn, Mice with a null mutation of the TGF alpha gene have abnormal skin architecture, wavy hair, and curly whiskers and often develop corneal inflammation, Cell 73 (1993) 249–261. [222] M. Blum, A null mutation in TGF-alpha leads to a reduction in midbrain dopaminergic neurons in the substantia nigra, Nat. Neurosci. 1 (1998) 374–377. [223] N.C. Luetteke, T.H. Qiu, S.E. Fenton, K.L. Troyer, R.F. Riedel, A. Chang, D.C. Lee, Targeted inactivation of the EGF and amphiregulin genes reveals distinct roles for EGF receptor ligands in mouse mammary gland development, Development 126 (1999) 2739–2750. [224] D.W. Threadgill, A.A. Dlugosz, L.A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. LaMantia, T. Mourton, K. Herrup, R.C. Harris, Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype, Science 269 (1995) 230–234. [225] M. Sibilia, E.F. Wagner, Strain-dependent epithelial defects in mice lacking the EGF receptor [published erratum appears in Science 1995 Aug 18;269(5226):909], Science 269 (1995) 234–238. [226] P.J. Miettinen, J.E. Berger, J. Meneses, Y. Phung, R.A. Pedersen, Z. Werb, R. Derynck, Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor, Nature 376 (1995) 337–341. [227] N.C. Luetteke, H.K. Phillips, T.H. Qiu, N.G. Copeland, H.S. Earp, N.A. Jenkins, D.C. Lee, The mouse waved-2 phenotype results from a point mutation in the EGF receptor tyrosine kinase, Genes Dev. 8 (1994) 399–413. [228] W. Xie, A.J. Paterson, E. Chin, L.M. Nabell, J.E. Kudlow, Targeted expression of a dominant negative epidermal growth factor receptor in the mammary gland of transgenic mice inhibits pubertal mammary duct development, Mol. Endocrinol. 11 (1997) 1766–1781. [229] J.F. Wiesen, P. Young, Z. Werb, G.R. Cunha, Signaling through the stromal epidermal growth factor receptor is necessary for mammary ductal development, Development 126 (1999) 335–344. [230] D.A. Lauffenburger, A.F. Horwitz, Cell migration: a physically integrated molecular process, Cell 84 (1996) 359–369. [231] T.J. Mitchison, L.P. Cramer, Actin-based cell motility and cell locomotion, Cell 84 (1996) 371–379. [232] I. Kaverina, O. Krylyshkina, J.V. Small, Regulation of substrate adhesion dynamics during cell motility, Int. J. Biochem. Cell. Biol. 34 (2002) 746–761. [233] C.C. Kumar, Signaling by integrin receptors, Oncogene 17 (1998) 1365–1373. [234] S. Dedhar, Cell-substrate interactions and signaling through ILK, Curr. Opin. Cell. Biol. 12 (2000) 250–256. [235] F.G. Giancotti, E. Ruoslahti, Integrin signaling, Science 285 (1999) 1028–1032. [236] K. Matsumoto, B.L. Ziober, C.C. Yao, R.H. Kramer, Growth factor regulation of integrin-mediated cell motility, Cancer Metastasis Rev. 14 (1995) 205–217. [237] M. Iijima, Y.E. Huang, P. Devreotes, Temporal and spatial regulation of chemotaxis, Dev. Cell. 3 (2002) 469–478. [238] A. Wells, K. Gupta, P. Chang, S. Swindle, A. Glading, H. Shiraha, Epidermal growth factor receptor-mediated motility in fibroblasts, Microsc. Res. Tech. 43 (1998) 395–411. [239] L.G. Hudson, L.J. McCawley, Contributions of the epidermal growth factor receptor to keratinocyte motility, Microsc. Res. Tech. 43 (1998) 444–455. [240] A. Wells, Tumor invasion: role of growth factor-induced cell motility, Adv. Cancer Res. 78 (2000) 31–101. [241] J. Kassis, D.A. Lauffenburger, T. Turner, A. Wells, Tumor invasion as dysregulated cell motility, Semin. Cancer Biol. 11 (2001) 105–117. [242] P. Chen, H. Xie, M.C. Sekar, K. Gupta, A. Wells, Epidermal growth factor receptor-mediated cell motility: phospholipase C activity is

[243]

[244]

[245]

[246]

[247]

[248]

[249]

[250]

[251]

[252]

[253]

[254]

[255] [256]

[257]

[258]

[259]

[260] [261]

required, but mitogen-activated protein kinase activity is not sufficient for induced cell movement, J. Cell. Biol. 127 (1994) 847–857. P. Chen, K. Gupta, A. Wells, Cell movement elicited by epidermal growth factor receptor requires kinase and autophosphorylation but is separable from mitogenesis, J. Cell. Biol. 124 (1994) 547–555. P. Chen, J.E. Murphy-Ullrich, A. Wells, A role for gelsolin in actuating epidermal growth factor receptor-mediated cell motility, J. Cell. Biol. 134 (1996) 689–698. F.N. van Leeuwen, S. van Delft, H.E. Kain, R.A. van der Kammen, J.G. Collard, Rac regulates phosphorylation of the myosin-II heavy chain, actinomyosin disassembly and cell spreading, Nat. Cell. Biol. 1 (1999) 242–248. P.E. Hughes, M.W. Renshaw, M. Pfaff, J. Forsyth, V.M. Keivens, M.A. Schwartz, M.H. Ginsberg, Suppression of integrin activationa novel function of a Ras/Raf-initiated MAP kinase pathway, Cell 88 (1997) 521–530. H. Xie, M.A. Pallero, K. Gupta, P. Chang, M.F. Ware, W. Witke, D.J. Kwiatkowski, D.A. Lauffenburger, J.E. Murphy-Ullrich, A. Wells, EGF receptor regulation of cell motility: EGF induces disassembly of focal adhesions independently of the motility-associated PLC gamma signaling pathway, J. Cell. Sci. 111 (Pt 5) (1998) 615– 624. M.P. Sheetz, D.P. Felsenfeld, C.G. Galbraith, Cell migration: regulation of force on extracellular-matrix-integrin complexes, Trends Cell. Biol. 8 (1998) 51–54. A. Huttenlocher, S.P. Palecek, Q. Lu, W. Zhang, R.L. Mellgren, D.A. Lauffenburger, M.H. Ginsberg, A.F. Horwitz, Regulation of cell migration by the calcium-dependent protease calpain, J. Biol. Chem. 272 (1997) 32719–32722. H. Shiraha, A. Glading, K. Gupta, A. Wells, IP-10 inhibits epidermal growth factor-induced motility by decreasing epidermal growth factor receptor-mediated calpain activity, J. Cell. Biol. 146 (1999) 243–254. S.P. Palecek, J.C. Loftus, M.H. Ginsberg, D.A. Lauffenburger, A.F. Horwitz, Integrin-ligand binding properties govern cell migration speed through cell-substratum adhesiveness, Nature 385 (1997) 537–540. A. Glading, D.A. Lauffenburger, A. Wells, Cutting to the chase: calpain proteases in cell motility, Trends Cell. Biol. 12 (2002) 46– 54. J. Lee, A. Ishihara, G. Oxford, B. Johnson, K. Jacobson, Regulation of cell movement is mediated by stretch-activated calcium channels, Nature 400 (1999) 382–386. R.L. Klemke, S. Cai, A.L. Giannini, P.J. Gallagher, P. de Lanerolle, D.A. Cheresh, Regulation of cell motility by mitogen-activated protein kinase, J. Cell. Biol. 137 (1997) 481–492. A.R. Bresnick, Molecular mechanisms of nonmuscle myosin-II regulation, Curr. Opin. Cell. Biol. 11 (1999) 26–33. E. Zamir, B.Z. Katz, S. Aota, K.M. Yamada, B. Geiger, Z. Kam, Molecular diversity of cell-matrix adhesions, J. Cell. Sci. 112 (Pt 11) (1999) 1655–1669. B.S. Verbeek, S.S. Adriaansen-Slot, T.M. Vroom, T. Beckers, G. Rijksen, Overexpression of EGFR and c-erbB2 causes enhanced cell migration in human breast cancer cells and NIH3T3 fibroblasts, FEBS Lett. 425 (1998) 145–150. D. Theodorescu, K.R. Laderoute, K.M. Gulding, Epidermal growth factor receptor-regulated human bladder cancer motility is in part a phosphatidylinositol 3-kinase-mediated process, Cell. Growth Differ. 9 (1998) 919–928. C.J. Wu, Z. Chen, A. Ullrich, M.I. Greene, D.M. O’Rourke, Inhibition of EGFR-mediated phosphoinositide-3-OH kinase (PI3-K) signaling and glioblastoma phenotype by signal-regulatory proteins (SIRPs), Oncogene 19 (2000) 3999–4010. A. Hall, Rho GTPases and the actin cytoskeleton, Science 279 (1998) 509–514. P. Aspenstrom, Effectors for the Rho GTPases, Curr. Opin. Cell. Biol. 11 (1999) 95–102.

Epidermal Growth Factor Receptor: Mechanisms of Activation and Signalling [262] S.M. Schoenwaelder, K. Burridge, Bidirectional signaling between the cytoskeleton and integrins, Curr. Opin. Cell. Biol. 11 (1999) 274–286. [263] C.D. Nobes, A. Hall, Rho GTPases control polarity, protrusion, and adhesion during cell movement, J. Cell. Biol. 144 (1999) 1235–1244. [264] A. Malliri, M. Symons, R.F. Hennigan, A.F. Hurlstone, R.F. Lamb, T. Wheeler, B.W. Ozanne, The transcription factor AP-1 is required for EGF-induced activation of rho-like GTPases, cytoskeletal rearrangements, motility, and in vitro invasion of A43I cells, J. Cell. Biol. 143 (1998) 1087–1099. [265] P.J. Keely, J.K. Westwick, I.P. Whitehead, C.J. Der, L.V. Parise, Cdc42 and Rac1 induce integrin-mediated cell motility and invasiveness through PI(3)K, Nature 390 (1997) 632–636. [266] P. Keely, L. Parise, R. Juliano, Integrins and GTPases in tumour cell growth, motility and invasion, Trends Cell. Biol. 8 (1998) 101–106. [267] L.E. Rameh, L.C. Cantley, The role of phosphoinositide 3-kinase lipid products in cell function, J. Biol. Chem. 274 (1999) 8347–8350. [268] S.J. Leevers, B. Vanhaesebroeck, M.D. Waterfield, Signalling through phosphoinositide 3-kinases: the lipids take centre stage, Curr. Opin. Cell. Biol. 11 (1999) 219–225. [269] P.T. Hawkins, A. Eguinoa, R.G. Qiu, D. Stokoe, F.T. Cooke, R. Walters, S. Wennstrom, L. Claesson-Welsh, T. Evans, M. Symons, L. Stephens, PDGF stimulates an increase in GTP-Rac via activation of phosphoinositide 3-kinase, Curr. Biol. 5 (1995) 393–403. [270] J. Han, K. Luby-Phelps, B. Das, X. Shu, Y. Xia, R.D. Mosteller, U.M. Krishna, J.R. Falck, M.A. White, D. Broek, Role of substrates and products of PI 3-kinase in regulating activation of Rac-related guanosine triphosphatases by Vav, Science 279 (1998) 558–560. [271] J.A. Greenwood, A.B. Theibert, G.D. Prestwich, J.E. MurphyUllrich, Restructuring of focal adhesion plaques by PI 3-kinase. Regulation by PtdIns (3,4,5)-p(3) binding to alpha-actinin, J. Cell. Biol. 150 (2000) 627–642. [272] E. Piccolo, P.F. Innominato, M.A. Mariggio, T. Maffucci, S. Iacobelli, M. Falasca, The mechanism involved in the regulation of phospholipase Cgamma1 activity in cell migration, Oncogene 21 (2002) 6520–6529. [273] O.D. Weiner, Rac activationP-Rex1—a convergence point for PIP(3) and Gbetagamma? Curr. Biol. 12 (2002) R429–R431. [274] A.C. Porter, R.R. Vaillancourt, Tyrosine kinase receptor-activated signal transduction pathways which lead to oncogenesis, Oncogene 17 (1998) 1343–1352. [275] S. Corvera, M.P. Czech, Direct targets of phosphoinositide 3-kinase products in membrane traffic and signal transduction, Trends, Cell. Biol. 8 (1998) 442–446. [276] I.S. Nathke, C.L. Adams, P. Polakis, J.H. Sellin, W.J. Nelson, The adenomatous polyposis coli tumor suppressor protein localizes to plasma membrane sites involved in active cell migration, J. Cell. Biol. 134 (1996) 165–179. [277] J.I. Fenton, M.S. Wolff, M.W. Orth, N.G. Hord, Membrane-type matrix metalloproteinases mediate curcumin-induced cell migration in non-tumorigenic colon epithelial cells differing in Apc genotype, Carcinogenesis 23 (2002) 1065–1070. [278] M.F. Ware, A. Wells, D.A. Lauffenburger, Epidermal growth factor alters fibroblast migration speed and directional persistence reciprocally and in a matrix-dependent manner, J. Cell. Sci. 111 (Pt 16) (1998) 2423–2432. [279] J. Gu, M. Tamura, R. Pankov, E.H. Danen, T. Takino, K. Matsumoto, K.M. Yamada, Shc and FAK differentially regulate cell motility and

[280]

[281]

[282]

[283]

[284]

[285]

[286]

[287]

[288]

[289]

[290]

[291]

[292]

[293]

55

directionality modulated by PTEN, J. Cell. Biol. 146 (1999) 389–403. M. Zhao, A. Dick, J.V. Forrester, C.D. McCaig, Electric fielddirected cell motility involves up-regulated expression and asymmetric redistribution of the epidermal growth factor receptors and is enhanced by fibronectin and laminin, Mol. Biol. Cell. 10 (1999) 1259–1276. K.S. Fang, E. Ionides, G. Oster, R. Nuccitelli, R.R. Isseroff, Epidermal growth factor receptor relocalization and kinase activity are necessary for directional migration of keratinocytes in DC electric fields, J. Cell. Sci. 112 (Pt 12) (1999) 1967–1978. M. Zhao, J. Pu, J.V. Forrester, C.D. McCaig, Membrane lipids, EGF receptors, and intracellular signals colocalize and are polarized in epithelial cells moving directionally in a physiological electric field, FASEB J. 16 (2002) 857–859. J. Li, M.L. Lin, G.J. Wiepz, A.G. Guadarrama, P.J. Bertics, Integrinmediated migration of murine B82L fibroblasts is dependent on the expression of an intact epidermal growth factor receptor, J. Biol. Chem. 274 (1999) 11209–11219. T. Okutani, Y. Okabayashi, Y. Kido, Y. Sugimoto, K. Sakaguchi, K. Matuoka, T. Takenawa, M. Kasuga, Grb2/Ash binds directly to tyrosines 1068 and 1086 and indirectly to tyrosine 1148 of activated human epidermal growth factor receptors in intact cells, J. Biol. Chem. 269 (1994) 31310–31314. J.H. McCarty, The Nck SH2/SH3 adaptor protein: a regulator of multiple intracellular signal transduction events, Bioessays 20 (1998) 913–921. Y. Hashimoto, H. Katayama, E. Kiyokawa, S. Ota, T. Kurata, N. Gotoh, N. Otsuka, M. Shibata, M. Matsuda, Phosphorylation of CrkII adaptor protein at tyrosine 221 by epidermal growth factor receptor, J. Biol. Chem. 273 (1998) 17186–17191. N. Jones, D.J. Dumont, Recruitment of Dok-R to the EGF receptor through its PTB domain is required for attenuation of Erk MAP kinase activation, Curr. Biol. 9 (1999) 1057–1060. J. Serth, W. Weber, M. Frech, A. Wittinghofer, A. Pingoud, Binding of the H-ras p21 GTPase activating protein by the activated epidermal growth factor receptor leads to inhibition of the p21 GTPase activity in vitro, Biochemistry 31 (1992) 6361–6365. K.L. Milarski, G. Zhu, C.G. Pearl, D.J. McNamara, E.M. Dobrusin, D. MacLean, A. Thieme-Sefler, Z.Y. Zhang, T. Sawyer, S.J. Decker, Sequence specificity in recognition of the epidermal growth factor receptor by protein tyrosine phosphatase 1B, J. Biol. Chem. 268 (1993) 23634–23639. H. Keilhack, T. Tenev, E. Nyakatura, J. Godovac-Zimmermann, L. Nielsen, K. Seedorf, F.D. Bohmer, Phosphotyrosine 1173 mediates binding of the protein-tyrosine phosphatase SHP-1 to the epidermal growth factor receptor and attenuation of receptor signaling, J. Biol. Chem. 273 (1998) 24839–24846. G. Zhu, S.J. Decker, D. MacLean, D.J. McNamara, J. Singh, T.K. Sawyer, A.R. Saltiel, Sequence specificity in the recognition of the epidermal growth factor receptor by the ab1 Src homology 2 domain, Oncogene 9 (1994) 1379–1385. P.J. Kraulis, MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures, J. Appl. Crystallogr. 24 (1991) 946–950. E.A. Merritt, D.J. Bacon, Raster3D: photorealistic molecular graphics, Methods Enzymol. 277 (1997) 505–524.

The deaf and the dumb: the biology of ErbB-2 and ErbB-3 Ami Citri, Kochupurakkal Bose Skaria, and Yosef Yarden Department of Biological Regulation, The Weizmann Institute of Science, Rehovot 76100, Israel

Abstract ErbB-2 (also called HER2/neu) and ErbB-3 are closely related to the epidermal growth factor receptor (EGFR/ErbB-1), but unlike EGFR, ErbB2 is a ligandless receptor, whereas ErbB-3 lacks tyrosine kinase activity. Hence, both ErbB-2 and ErbB-3 are active only in the context of ErbB heterodimers, and ErbB-2 · ErbB-3 heterodimers, which are driven by neuregulin ligands, are the most prevalent and potent complexes. These stringently controlled heterodimers are repeatedly employed throughout embryonic development and dictate the establishment of several cell lineages through mesenchyme-epithelial inductive processes and the interactions of neurons with muscle, glia, and Schwann cells. Likewise, the potent combination of signaling pathways engaged by the heterodimers drives an aggressive phenotype of tumors of secretory epithelia, including breast and lung cancers. This review highlights recent structural insights into the mechanism of ligand-induced heterodimer formation, and concentrates on signaling pathways employed by ErbB-2 and ErbB-3 in normal and in malignant cells.

Introduction The ErbB-2 · ErbB-3 heterodimer constitutes the pinnacle of ErbB receptor evolution, demonstrating the capacity of evolution to form an extremely potent signaling module from a pair of singly inactive individual proteins. The diversification of the ErbB family during evolution, from one receptor/ligand in worms, through one receptor/multiple ligands in flies, to four receptors and multiple ligands in mammals, has created a network capable of precise signaling in a widely divergent fashion [1]. Thus, through utilization of defined receptor pairs, activated by specific ligands, a graded signaling potency can be obtained, leading to a precise cellular outcome. An additional level of signaling diversity is obtained through differential activation of distinct signaling molecules downstream of each receptor. The capacity to form precise signaling is best exemplified by the ErbB-2 · ErbB-3 dimer. ErbB-3 is an impaired kinase due to substitutions in critical residues in its catalytic domain [2], and thus can signal only in the context of a receptor heterodimer. In addition, it is now clear that ErbB-2 is devoid of an activating ligand [3] and can act only in the context of a heterodimer with a ligand-bound receptor. In stark contrast to their apparent disabilities, this receptor pair forms the most potent signaling module of the ErbB-receptor family in terms of cell growth and transformation [4,5]. That the most potent signaling module is formed by partners that are incapable of productively signaling in isolation suggests that evolutionary forces formed these mechanisms as a measure

to tightly control the output of the network. As will be described below, the basis for the potency of signaling by the ligand-activated ErbB-2 · ErbB-3 heterodimer lies in the fact that this dimer has the capacity to signal very potently, both through the Ras-Erk pathway for proliferation, and through the phosphatidylinositol-3¢-kinase (PI3K)-Akt pathway for survival. In addition, this receptor dimer evades downregulation mechanisms, leading to prolonged signaling [6,7].

ErbB-2 and ErbB-3 as determinants of cell lineages Organ morphogenesis is controlled, at least in part, by multiple polypeptide factors that transmit signals between neighboring cells. The ErbB-receptor family plays a pivotal role in cell lineage determination in a variety of tissues, including mesenchyme-epithelial inductive processes in epithelial organs (reviewed by Burden and Yarden [8]). ErbB-2 and ErbB-3, as well as ErbB-1, are expressed in most epithelial cell layers, and mesenchymal cells are a rich source of ErbB-ligands, both neuregulins (NRGs) and epidermal growth factor (EGF)-like ligands. During development of the human fetus, ErbB-2 is found in the nervous system, developing bone, muscle, skin, heart, lungs, and intestinal epithelium [9]. Likewise, ErbB-3 is found in cells of the gastrointestinal, reproductive, and urinary tracts, as well as the skin, endocrine, and nervous systems [10].

58


Cardiac development The roles of ErbB-2 and ErbB-3 in development are best delineated by the phenotypes of mice in which the erbB-2 and erbB-3 genes have been inactivated, as well as the phenotype of nrg-1-/- mice. Essentially, the cardiac phenotypes of NRG-1 null mice and ErbB-2 null mice appear similar, and represent the sum of the changes observed in ErbB-3 and ErbB-4 null mice [11]. The most striking phenotype of ErbB-2 null mice is early death at mid-gestation (E10.5) due to malformation of trabeculae in the heart [12], a phenotype shared with ErbB-4 and NRG-1-defective mice [13,14]. NRG-1 is expressed in the endocardium, an endothelial ventricular lining, while ErbB-2 and ErbB-4 are expressed in the myocardium, which is the underlying muscular portion of the ventricle and atrium. Mice expressing a kinasedefective mutant of ErbB-2 instead of wild-type receptor die at mid-gestation and express the same spectrum of embryonic defects seen in ErbB-2-defective mice, demonstrating that the catalytic activity of ErbB-2 is essential for its function in embryogenesis [15]. While the process of heart trabeculation does not appear to require ErbB-3, this receptor is essential for normal cardiac development [16]. Expression of ErbB-3 is restricted to the mesenchyme of the endocardial cushion, which develops into the heart valves. In erbB3-/- mouse embryos the trabeculation occurs in a delayed but otherwise normal fashion, but their atrioventricular valves are rudimentary and thinned, leading to death at E13.5. NRG-1 plays a crucial role in this inductive process [17], but ErbB-2 does not appear to be involved, since it is not expressed in the embryonic endocardial cushion. Targeted inactivation of ErbB-2 in ventricular cardiomyocytes led to a severe dilated cardiomyopathy, causing cardiac dysfunction by the second postnatal month, suggesting that ErbB-2 signaling is essential for adult heart function [18,19]. The cardiac phenotypes expressed by mice carrying mutations in erbB-2 and erbB-3 demonstrate the basic principle underlying the function of the encoded receptors, i.e., heterodimerization is essential for induction of receptor’s function.

ferentiation, using complex transgenic systems it has been established that ErbB-2 is important for late oligodendrocyte differentiation, and for the development of myelin [22,23]. In addition, it has also been demonstrated that a heterodimer of ErbB-2 and ErbB-3 is the active receptor in Schwann cell differentiation [24,25]. Consistent with observations made in vitro with cultured cells, in erbB-3-defective mice Schwann cells fail to develop, and most sensory and motor neurons subsequently die (reviewed by Davies [26]). Mammary gland development The mammary gland is one of the few organs in which major morphogenetic changes take place after birth. Two phases of morphogenesis occur, i.e., during puberty, and during pregnancy. In contrast to their ligands, which are expressed in defined time windows, all ErbB receptors are expressed throughout most developmental stages of the mammary gland (reviewed by Troyer and Lee [27]). In organ cultures of mammary glands, NRG-1 stimulated lobuloalveolar budding and the production of milk proteins [28]. In addition, branching morphogenesis and lobulo-alveolar differentiation of the mammary gland could be abolished by blocking expression of endogenous NRG. However, in transgenic animals with targeted expression of NRG-1 in the mammary gland, persistence of terminal end buds was observed, suggesting that NRG-1 inhibits signals that normally lead to the terminal differentiation of these structures [29]. Consistent with a role for ErbB-2 as a coreceptor, transgenic mice expressing a dominant-negative ErbB-2 in the mammary gland display normal ductal growth, but have defective lobuloalveoli and reduced milk protein secretion [30]. Due to coexpression of all ErbBs and many of their ligands, more experimental models will be needed to resolve the exact role of ErbB-2 · ErbB-3 heterodimers in mammary development.

Formation of ErbB-2 · ErbB-3 heterodimers: structural insights

Glial and neuronal cell development In addition to their roles in synapse formation and neuronal development, NRGs appear to act as major regulators in the development of myelinating cells in the peripheral (Schwann cells) and central (oligodendrocytes) nervous systems [20]. Mice individually mutant for erbB-2, erbB-3, or nrg-1 display a failure in neural crest development, leading to impaired formation of the sympathetic nervous system [21]. The action of NRG-1, secreted by neurons, is essential for both the proliferation and the maturation of Schwann cells and oligodendrocytes. Accordingly, ErbB-3 is expressed in the responding myelinating cells, and wherever investigated, the effect of NRG-1 appears to be mediated by ErbB-3, acting in the context of ErbB-1 or ErbB-2. While NRG-1 is important throughout oligodendrocyte dif-

While it is clear that dimerization of ErbB proteins is crucial for signaling (reviewed by Heldin and Ostman [31]), the underlying mechanism remained elusive until very recently. Three published structures of the extracellular domains of ErbB family receptors [32–34] have recently provided a framework for understanding the large amount of data that have accumulated over the years with respect to ligand binding and receptor activation. ErbB receptors share a high degree of primary sequence homology. Four subdomains have been identified in the extracellular domain; subdomains I (L1) and subdomain III (L2) mediate ligand binding [35,36], while according to the recently published structures, the cysteine-rich subdomains II (S1) and IV (S2) play a role in receptor dimerization. Most previous lines of evidence relate to the interaction of EGF with ErbB-1

The Deaf and the Dumb: the Biology of ErbB-2 and ErbB-3

(reviewed by Groenen et al. [37]). For example, several lines of evidence concluded that EGF binds ErbB-1 with a 1 : 1 stoichiometry, perhaps through a cleft formed by subdomains L1, L2, and S1 [38]. In addition, EGF binding to an isolated extracellular domain has been associated with a conformational change [39], but the exact mechanism of dimerization remained unknown until very recently. Earlier lines of evidence Mutagenesis of both NRG-1 [40] and EGF [36] concluded that these ligands bind their receptors in a bivalent manner. On the other hand, biophysical analyses of soluble ErbB-1 suggested dominance of a 2 : 2 ligand : receptor configuration [41]. Last, fluorescence imaging microscopy suggested that ErbB-1 exists in a predimerized state, but ligand binding induces a rotational rearrangement of the monomeric subunits [42]. This view is consistent with single molecule imaging of ErbB-1, which concluded that EGF first binds to predimerized receptors, and then a second EGF molecule binds to the 1 : 2 complex [43]. Insights from crystal structures The crystal structures of ligand-bound ErbB-1 confirmed both bivalent ligand binding and a final 2 : 2 complex [33,34], consistent with the implications of the structure of a nonliganded ErbB-3 [32]. In the crystal structures of ErbB1 and ErbB-3, L1 and L2 have a b-helical fold, while the S1 and S2 subdomains have an extended structure held together by disulfide bonds. The S1 subdomain traverses along one face of the b-helix of the L1 domain with a large interface conferring rigidity to the juxtaposition of the L1-S1 domains. Two structures of ErbB-1 have been described, one bound to EGF [33] and the other bound to transforming growth factor-a (cTGFa) [34]. In both cases, bivalent ligand binding to the L1 and L2 domains is observed and the ligand holds these domains in a rigid conformation. A long bhairpin, termed the “dimerization loop” extends out of S1 and was found to be the primary mediator of dimerization of two monomeric ErbB-1 molecules. In the dimer, the two ligand-bound monomers approach each other back-to-back and the dimerization loop of one receptor extends deep into the dimer partner. The tip of the loop contacts residues in the L1 and L2 domains, contributing to dimer stabilization.

59

the unliganded ErbB-3, an intramolecular interaction between the dimerization loop (from S1) and a b-hairpin protruding out from the S2 domains holds the receptor L domains far apart in an open conformation [32]. Taken together with the conformation of the ligand-bound ErbB-1, these lines of evidence raise the possibility that unengaged ErbB-1 and ErbB-4 may exist in a similar locked conformation on the cell surface. According to this model, ligand binding to domains L1 and L2 induces rotation of the rigid L1-S1 domains with respect to the L2 domain around the S1-L2 linker, and tightly bridges the L1 and L2 domains. Consequently, this conformational change exposes the dimerization loop of the receptor, rendering monomeric, ligand-bound receptors amenable to dimerization. Dimers containing ErbB-2 Although ErbB-2 binds no known ligand, when recruited into heterodimers it increases ligand binding affinity [46,47]. It is also the favored receptor for heterodimerization [48,49]. Interestingly, in the structure of ErbB-2, unlike the other receptors in the family, a strong interaction between L1 and L2 domains was observed, mimicking the ligand-bound form in the ErbB-1 structure (A.W. Burgess, personal communication). This interaction involves regions corresponding to ligand-binding sites in the L1 and L2 domains of ErbB-1, rendering ErbB-2 incapable of binding ligands. The consequence of the L1-L2 interaction in ErbB-2 is a constitutively extended conformation of the dimerization loop. Hence, the promiscuous behavior of ErbB-2 and its inability to bind EGF-like ligands seem inherent to its structure. Fig. 1A presents a model for the formation of ErbB-2 · ErbB-3 heterodimers. This model was conceived based on the known structures of ErbB-3 and ErbB-1, in combination with previously described lines of evidence. Accordingly, the ligandless ErbB-2 is predisposed for dimerization because its dimerization loop is preextended. On the other hand, ligand binding to ErbB-3 releases a locked conformation and extends the dimerization loop. Finally, within the dimer, both S1 and S2 domains of each receptor form two distinct interfaces, which stabilize the heterodimer. An alternative view would suggest that a preformed heterodimer assumes a twisted active conformation upon binding of a ligand [42]. Dimerization driven by receptor overexpression?

The role of the S2 domain The published structures of ErbB-1 did not resolve the structure of S2. However, short cyclic peptides identical in sequence to the C-terminus of the S2 domain of ErbB-2 were found to bind the receptor as well as inhibit receptor signaling [44], suggesting that the S2 domain may be involved in dimerization. In addition, S2 has been previously demonstrated to reduce ligand binding affinity, suggestive of an intramolecular inhibitory interaction [45]. In the structure of

Amplification of the erbB-1 and erbB-2 genes is a common theme in epithelial cancers, and breast cancer patients whose tumors overexpress ErbB-2 better benefit from treatment with anti-ErbB-2 antibodies (reviewed by Yarden and Sliwkowski [1]). In model systems, overexpression of ErbB-1 is oncogenic but only in the presence of a ligand, whereas overexpression of ErbB-2 is transforming even in the absence of a ligand [50,51]. Likewise, although ErbB-2 bearing an activating point mutation is transforming

60


Fig. 1. Schematic representation of ligand-induced receptor heterodimers. The proposed mechanism of neuregulin-induced ErbB-2 · ErbB-3 heterodimers is depicted. (A) The extracellular domains are represented by two cysteine-rich domains (S1 and S2) and two cysteine-free, ligand-binding domains (L1 and L2). ErbB-3 (blue) exists on the cell surface in an autoinhibited conformation resulting from the interactions between the S1 and S2 domains. Bivalent binding of neuregulin (NRG, represented here as a red dumbbell) to the L1 and L2 domains of ErbB-3 rearranges the conformation of the extracellular domain, leading to protrusion of the S1 dimerization loop. In the case of ErbB-2, the intramolecular interaction between L1 and L2 results in a preextended conformation of the S1 dimerization loop. Dimerization between a ligand-bound ErbB-3 and an ErbB-2 molecule is mediated primarily by the dimerization loop (dotted circle I), with additional possible contributions from the loop in the S2 domain (dotted circle II). Additional stabilizing interactions between the transmembrane and kinase domains may also play a role. While the simplest scenario is depicted, receptor trimers and tetramers have also been proposed. (B) Overexpression of ErbB-2 at the cell surface may spontaneously recruit an autoinhibited ErbB-3 into heterodimers. The formed dimers may assume the ligand-induced conformation, resulting in weak but prolonged receptor activation. Alternatively, spontaneous homodimers formed upon overexpression of ErbB-2 cannot be excluded.


61

Fig. 2. The major signaling pathways stimulated by ligand-activated ErbB-2 · ErbB-3 heterodimers. Ligand-induced formation of the ErbB-2 · ErbB-3 heterodimer (see Fig. 1) at the cell surface leads to activation of several major pathways of signal transduction. This process results in enhanced cell survival and mitogenicity, and its deregulation can lead to tumorigenesis. Erk activation by the Ras-Raf pathway leads to cell proliferation through the activation of a number of nuclear targets, including Elk1, PEA3, Sp1, AP1, and the c-Myc oncoprotein, which is a major transcription factor and regulator of cell cycle progression. Another pathway is the P13K-Akt pathway, activation of which results in enhanced antiapoptotic and prosurvival signals, through inhibition of the proapoptotic proteins Bad, GSK3, and the transcription factor FKHR-L1. In addition, the PLCg and the JAK-STAT pathways are indicated, with their resulting enhancement of transcription leading to cell proliferation. A major player acting downstream of ErbB-2 · ErbB-3 is cyclin D1. As indicated, a number of pathways lead from the receptors to enhanced activation of cyclin D1, thereby promoting cell cycle progression. Note that the outcome of activation of these different signaling pathways depends on the cellular context, and can vary from proliferation to differentiation, migration, and even induction of apoptosis.

in the absence of a ligand [52], the presence of another member of the ErbB family seems essential [53]. Hence, activation of ErbB-2 may not occur through formation of its own homodimers. Instead, ErbB-2-containing heterodimers may form when ErbB-2 is overexpressed. The model presented in Fig. 1B explains how heterodimerization of ErbB-2 with ErbB-3 may occur upon overexpression of ErbB-2, and a similar mechanism may underlie the oncogenic potential of coexpressed ErbB-1 and ErbB-2 [54].

between cell division and apoptosis is the crucial factor [55]. This principle is demonstrated by the oncogenic potential of the ErbB-2 · ErbB-3 dimer; this receptor signals through both the mitogen-activated protein kinase (MAPK) pathway, which drives cell proliferation and additional processes, and through the PI3K/Akt pathway, which primarily drives cellular survival and antiapoptotic signals (Fig. 2).

Signal transduction by ErbB-2 · ErbB-3 dimers

Comparative analyses of individual homodimers and heterodimers of ErbB proteins introduced into naive, ErbB-free cells, revealed that ErbB-3 is signaling defective, whereas ErbB-2 cannot be stimulated by any known ligand [4,56,57].

Cancers do not necessarily arise as a linear result of an increased rate of cellular proliferation, but rather the balance

The ErbB-2 · ErbB-3 complex is the most active ErbB dimer

62


Nevertheless, ErbB-2 can enhance and prolong signaling by many EGF-like ligands [46]. In addition, within the hierarchical dimerization network of ErbB receptors, ErbB-2 represents the preferred heterodimerization partner of all other ErbB receptors, and the preferred dimerization partner of ErbB-2 is ErbB-3 [49]. The remarkable signaling potency of ErbB-2 · ErbB-3 dimers is the outcome of several features outlined below: Slow rate of ligand dissociation ErbB-2 enhances the affinity of the direct receptors for their ligands [47,49,58,59]. This effect seems to be shared by all EGF-like ligands and may reflect the promiscuous behavior of ErbB-2 as a heterodimerization partner [3]. While ligand affinity is determined by both the rate of association with the receptor, as well as by the rate of complex dissociation, only the latter parameter is modulated by ErbB-2 [46]. Relaxed specificity to EGF-like ligands When ErbB-2 joins ErbB-3 it not only confers higher binding affinity, but also widens the spectrum of ligand binding toward neuregulins and EGF-like ligands [60,61]. This attribute is not a simple outcome of increased affinity, because some ligands, e.g., EGF and betacellulin, gain recognition, while others, e.g., TGFa, do not. Although the underlying mechanism remains unknown, it seems that this activity is due to an intrinsic capability of ErbB-2, because ErbB-2 · ErbB-4 heterodimers demonstrate a similar phenomenon [62]. Evasion of endocytosis The major pathway leading to inactivation of signals emanating from ligand-activated growth factor receptors is an endocytic process that sorts active receptors to degradation in lysosomes (reviewed by Waterman and Yarden [63]). In the case of ErbB-1 this process is robustly regulated by an E3 ubiquitin ligase, called c-Cbl, that binds a specific phosphotyrosine of ErbB-1, thereby enhancing receptor ubiquitination and subsequent sorting to endocytosis and degradation [64–68]. In contrast, ligand-induced endocytosis of other ErbBs is slower [4,6], probably because these receptors are only weakly coupled to c-Cbl [48,69,70]. Consequently, ligands of ErbB-3 undergo relatively slow degradation [71], whereas ErbB-3 itself evades downregulation and undergoes recycling to the plasma membrane [72]. Defective endocytosis and enhanced recycling characterize not only the ErbB-2 · ErbB-3 dimer, but also other ErbB-2containing heterodimers [73–75]. Coupling to potent signaling pathways Within a ligand-occupied ErbB-2 · ErbB-3 dimer, transphosphorylation takes place and as a result, several phosphotyrosine residues located in the carboxyl-terminus of each receptor undergo phosphorylation [5]. Remarkably, a single site of ErbB-2, which recruits Shc and couples to

the MAPK pathway, is sufficient for cell transformation [76], and multiple sites of ErbB-3 are able to recruit PI3K [77], thereby activating the Akt pathway. Normally ErbB-3 is restrained through its lack of kinase activity, which is compensated by heterodimerization. The strong signaling potential of this receptor is exemplified by an artificial fusion protein containing the kinase domain of ErbB-1 and the carboxyl-terminal tail of ErbB-3 [7]. This chimera is extremely mitogenic because it strongly couples to PI3K, avoids c-Cbl and endocytosis, and transmits prolonged signals through the Shc-MAPK pathway. In conclusion, several mechanisms allow ErbB-2 and ErbB-3 to escape normal constraints, and their combined dimer is characterized by ligand promiscuity and potent signaling. Signaling pathways activated by ErbB-2 and ErbB-3 Several major pathways are stimulated upon activation of ErbB-2 and ErbB-3. These are MAPK [76], PI3K [77,78], phospholipase-Cg (PLCg; [79,80]), protein kinase C, and the Janus kinase (Jak-STAT; [81]). Remarkably, it appears that the heterodimers avoid coupling to Grb2 and the Ras-specific GTPase-activating protein (Ras-GAP; [82]), effectors that can also negatively regulate mitogenic signals [83,84]. MAPK pathway Stimulation of Erk occurs upon ligand-induced activation of a receptor dimer, which binds Grb2 through a phosphorylated tyrosine-based consensus site, or indirectly, through interaction with Shc (reviewed by Marshall [85]). Grb2 is associated with Sos, a guanine nucleotide exchange factor specific for Ras, and Sos activates Ras by exchanging GDP for GTP. In the GTPase active state, Ras interacts with Raf and stimulates a linear kinase cascade culminating in activation of Erk/MAPK. Erk phosphorylates a variety of cytoplasmic and membranal substrates, and is rapidly translocated to the nucleus, where it activates a number of transcription factors including Sp1, PEA3, E2F, Elk1, and the AP1 transcription factor formed by Jun and Fos. PI3K/AKT Activation of PI3K occurs through binding of the regulatory p85 subunit of the lipid kinase to a phosphotyrosine consensus site on the receptor, leading to allosteric activation of the p110 catalytic subunit. p110 activation produces phosphatidylinositol-3,4,5-trisphosphate [PtdIns(3,4,5)P3] from PtdIns(4,5)P2 within seconds, and delayed production of PtdIns(3,4)P2 through the action of 5¢-inositol phosphatases. The effects of polyphosphinositides in the cell are mediated through the action of two lipid-binding domains, the FYVE domain, which binds to PtdIns(3)P, and the PH domain, which binds to PtdIns(3,4)P2 and PtdIns(3,4,5)P3. The PH domain-containing proteins PDK-1 (reviewed by Tokev and Newton [86]) and Akt/PKB are key mediators of PI3K signaling, and both are essential for the transforming effects of PI3K (reviewed by Blume-Jensen and Hunter


[87]). Upon production of PtdIns(3,4)P2 and PtdIns(3,4,5)P3 following activation of PI3K by the ErbB-2·ErbB-3 receptor dimer, Akt is recruited to the plasma membrane by its PH domain, and is phosphorylated by PDK-1. Akt phosphorylation causes its activation and translocation to the nucleus, where it acts upon its targets, which are either regulators of apoptosis or of cell growth (reviewed by Meier and Hemmings [88] and Cantley [89]). The tumor suppressor PTEN is a lipid phosphatase, which dephosphorylates the 3¢-OH position of PtdIns(3,4)P2 and PtdIns(3,4,5)P3, thereby reverting the activity of PI3K, and downregulating the activity of PDK-1 and Akt. PLCg Activation of PLCg; by ErbB-2, rather than by ErbB-3 [82], occurs through its SH2-mediated recruitment to phosphorylation-dependent docking sites on ErbB-2, as well as recruitment through its PH domain to the plasma membrane. In its phosphorylated active form, PLCg; hydrolyzes PIP2 (PI4,5P2; phosphatidylinositol 4,5 biphosphate) into IP3 (inositol 1,4,5-triphosphate), and diacylglycerol. IP3 activates the release of calcium from intracellular stores, and thereby activates calcium/calmodulin-dependent kinases, as well as additional pathways, and it collaborates with diacylglycerol to stimulate protein kinase C (reviewed by Karin and Hunter [90] and Hunter [91]). Cell adhesion molecules CD44, a surface glycoprotein implicated in cell adhesion and motility, has been found to complex with ErbB-2 in ovarian and Schwann cells. CD44 coimmunoprecipitates with ErbB-2 and ErbB-3, and may potentiate the response to NRG-1 by facilitating receptor heterodimerization [92,93]. An additional positive modulator is the MUC4 sialomucin, which forms a complex with ErbB-2 in a number of tissues. MUC4 has been suggested to act as a modulator of the signaling activity of ErbB-2, inducing specific phosphorylation of ErbB-2, and potentiating NRG signaling [94]. Cyclin-dependent kinases and cell cycle regulation Hyperactivated signaling through ErbB receptors results in deregulation of the cell cycle homeostatic machinery and upregulation of complexes containing cyclin D and cyclindependent kinases (CDKs), resulting in enhanced proliferation and malignant transformation. Cyclin D1 is a central effector of signaling by ErbB-2 and ErbB-3, and has been implicated as a major player in breast cancer acting to promote cell cycle progression, through activation of its catalytic partners CDK4 and CDK6 (reviewed by Harari and Yarden [95]). Both MAPK and PI3K can modulate the activity of cyclin D1 downstream of ErbB-2 [96]. The MAPK pathway has been implicated in transcriptional upregulation of cyclin D1 through the Sp1, AP1, and E2F transcription factors [97,98], and posttranslational stabilization of cyclin D1 can be conferred by its phosphorylation by Akt [99].

63

Additional key regulators of CDK function are p27KIP1 and p21Waf1, previously implicated as CDK inhibitors, but currently suggested to produce both activating and inhibiting functions, depending on their expression levels. Thus, ErbB2/Neu-induced, cyclin D1-dependent transformation is accelerated in p27-haplo-insufficient mammary epithelial cells, but impaired in p27 null cells [100], and ErbB-2 and ErbB-3 function together to stimulate mitogenic signaling networks by Akt- and c-Myc-dependent sequestration of p27KIP1, leading to deregulation of the G1-S cell cycle transition [101]. Negative regulators RALT is a feedback inhibitor of ErbB-2 signaling whose expression is induced upon stimulation of the MAPK pathway [102]. RALT binds to the kinase domain of ErbB2, and inhibits Erk activation and cellular transformation driven by ErbB-2 [103]. A different mode of ErbB-2 inhibition is exemplified by Herstatin, an alternatively spliced form consisting of a segment of the extracellular domain of ErbB-2 fused to a novel carboxyl-terminus. Herstatin binds to ErbB-2 and inhibits heterodimerization and activation of ErbB-3 [104]. A similar type of naturally secreted protein consisting of the extracellular domain of the receptor has been described for ErbB-3. This secreted protein (p85-s) inhibits NRG-stimulated activation of ErbB-2, ErbB-3, and ErbB-4 through sequestration of the ligand [105]. An additional mode of regulation of ErbB-2 is mediated by the action of the heat shock proteins Hsp90 and Hsp70, and their associated E3-ubiquitin ligase CHIP, which acts to promote the ubiquitylation of ErbB-2, and its subsequent degradation [106,107]. This mechanism of cellular regulation bears significant potential for pharmaceutical intervention in ErbB-2-dependent tumors.

Clinical implications of the cooperation between ErbB-2 and ErbB-3 Several lines of evidence derived from animal models and in vitro systems imply that ErbB-1 can transform naive cells only when one of its ligands, primarily TGFa, is available [50,108]. Consistent with this notion, analyses of human tumors from gastric, breast, pancreatic, and other origins indicated that autocrine loops underlie poor prognosis of the relevant ErbB-1-overexpressing tumors (reviewed by Salomon et al. [109]). Another critical partner of ErbB1 is ErbB-2, as their coexpression drives oncogenesis in model systems [54,57]. In analogy to ErbB-1, when singly expressed, ErbB-3 is nonmitogenic [4], but together with ErbB-2 and a neuregulin it transmits not only potent mitogenic signals [4,56,57], but also signals for tumorigenic growth [5,57,110]. As discussed earlier, whether or not ErbB-2 can transform cells when singly expressed is yet unclear; while the catalytic activity of this receptor is relatively high, even in the absence of a stimulating ligand [111],

64


a transforming mutant, whose catalytic activity is constitutively elevated [112], loses its oncogenic potential when expressed in an ErbB null cellular environment [53]. ErbB-2 is overexpressed in a large proportion of breast and ovarian tumors (20–30%), primarily due to gene amplification [113,114]. ErbB-2 appears not to be expressed in benign tumors before the onset of malignant disease [115], but overexpression is maintained in metastatic lesions. The prognostic significance of ErbB-2 overexpresion in human cancer has been extensively reviewed [116–118], and therefore we limit our discussion to the available clinical data related to coexpression of ErbB-2 and neuregulin receptors. Expression of the high affinity neuregulin receptor ErbB-4 is relatively variable in carcinomas, and it may associate with a differentiated phenotype and better prognosis of breast tumors [119]. In contrast, coexpression of ErbB-2 with ErbB-4 in childhood medulloblastoma predicts poor prognosis [120]. Significantly, when singly analyzed in brain tumors, expression of neither receptor was predictive. Coexpression of ErbB-2 with the low affinity neuregulin receptor ErbB-3 may be similarly relevant to epithelial tumors. However, detection of such an association is potentially blurred by the following two factors: First, ErbB-3 is expressed in the majority of tumors of the breast, skin, ovary, and gastrointestinal tract [121–124]. Second, the respective gene shows no amplification or rearrangements. Nevertheless, along with reports that failed detecting association between ErbB-3 and clinical outcome, several studies associated ErbB-3 expression with pathological parameters. Examples include advanced non-small cell lung carcinomas in which high ErbB-3 predicted shorter patient survival [125], early invasive ovarian lesions [126], hepatocellular carcinomas [127], oral squamous cell carcinomas [124], and bladder cancers in which coexpression correlated with patient survival [128]. In conclusion, there are clinical indications supporting the concept emerging from in vitro studies that neither ErbB2 nor ErbB-3 can be considered as stand-alone receptors. Future studies must also address the presence of neuregulins, because unlike ligands of EGFR, which seem to control autocrine loops in human cancers [129], neuregulins may form paracrine loops in breast [130] and prostate cancer [131]. Another variable is the occurrence in tumors of secreted ErbB-3 isoforms capable of neuregulin binding [105].

signal. This context is beginning to be extended to the interaction of the ErbBs with receptors of other families and to cross-talk between signaling pathways (reviewed by Carpenter [132]). The recently resolved structures of the ectodomains of ErbB receptors will be instructive not only for understanding how ligands promote receptor dimers, but also help develop peptidergic and other ErbB blockers. It is interesting that a naturally occurring antagonistic ligand exists in flies (i.e., Argos [133]), but attempts to generate a similar blocker of mammalian ErbBs have failed so far. Apart from spaciotemporal regulation of the expression of ligands, receptors, and downstream effectors, it is currently unclear how different neuregulins elicit unique responses, although they utilize similar receptor combinations. Differences in affinity of different ligands for the same receptor combination have been suggested to be a deciding factor in signal outcomes. Another possibility is the amplification of subtle differences in the conformations of the dimers induced by different ligands, when higher order oligomers of the receptors are formed [134]. Would low affinity viral ligands induce an open conformation of the dimerization loop, or do they exploit the small fraction of predimerized receptors by rearranging them to potentiate signaling [135]? If a ligand binds to only one receptor, how is the identity of the receptor dimer decided? What is the mechanism by which heterodimerization is preferred over homodimerization? And, how is the dimerization signal transferred through the membrane, leading to activation of the kinase domain? These and other questions will most likely await the resolution of the structure of a receptor heterodimer in the context of a ligand. Last, in terms of cancer therapy, our understanding of the basic mechanisms underlying ErbB-dependent tumorigenesis have led to a major focus on the receptors as targets for therapeutics. Given the perception of multilayered signaling through the ErbB family, this choice appears to be valid, but future targeting of additional components of the ErbB network is expected to increase clinical success in the future.

Acknowledgments We thank Sarah Bacus, Jorma Isola, and Mark Sliwkowski for insightful discussions. Our laboratory is supported by grants from the National Cancer Institute, the U.S. Army, the European Commission, and the Israel Academy of Sciences and Humanities.

Perspectives Peaking in the recent months, the past 15 years have been highly instructive as to the basic principles underlying the action of ErbB receptors, and in describing their relevance to tumorigenesis, thereby opening windows for therapeutic opportunities. Thus, the basic principle of receptor heterodimerization has taught us that the context in which a receptor functions is crucial for predicting the resulting

References [1] Y. Yarden, M.X. Sliwkowski, Untangling the ErbB signalling network, Nat. Rev. Mol. Cell. Biol. 2 (2001) 127–137. [2] P.M. Guy, J.V. Platko, L.C. Cantley, R.A. Cerione, K.L. Carraway 3rd, Insect cell-expressed p 180erbB3 possesses an impaired tyrosine kinase activity, Proc. Natl. Acad. Sci. USA 91 (1994) 8132– 8136.

The Deaf and the Dumb: the Biology of ErbB-2 and ErbB-3 [3] L.N. Klapper, S. Glathe, N. Vaisman, N.E. Hynes, G.C. Andrews, M. Sela, Y. Yarden, The ErbB-2/HER2 oncoprotein of human carcinomas may function solely as a shared coreceptor for multiple stroma-derived growth factors, Proc. Natl. Acad. Sci. USA 96 (1999) 4995–5000. [4] R. Pinkas-Kramarski, L. Soussan, H. Waterman, G. Levkowitz, I. Alroy, L. Klapper, S. Lavi, R. Seger, B.J. Ratzkin, M. Sela, Y. Yarden, Diversification of Neu differentiation factor and epidermal growth factor signaling by combinatorial receptor interactions, EMBO J. 15 (1996) 2452–2467. [5] C. Wallasch, F.U. Weiss, G. Niederfellner, B. Jallal, W. Issing, A. Ullrich, Heregulin-dependent regulation of HER2/neu oncogenic signaling by heterodimerization with HER3, EMBO J. 14 (1995) 4267–4275. [6] J. Baulida, M.H. Kraus, M. Alimandi, P.P. Di Fiore, G. Carpenter, All ErbB receptors other than the epidermal growth factor receptor are endocytosis impaired, J. Biol. Chem. 271 (1996) 5251–5257. [7] H. Waterman, I. Alroy, S. Strano, R. Seger, Y. Yarden, The C-terminus of the kinase-defective neuregulin receptor ErbB-3 confers mitogenic superiority and dictates endocytic routing, EMBO J. 18 (1999) 3348–3358. [8] S. Burden, Y. Yarden, Neuregulins and their receptors: a versatile signaling module in organogenesis and oncogenesis, Neuron 18 (1997) 847–855. [9] L. Coussens, T.I. Yang, Feng, Y.C. Liao, E. Chen, A. Gray, J. McGrath, P.H. Seeburg, T.A. Libermann, J. Schlessinger, U. Francke, A. Levinson, A. Ullrich, Tyrosine kinase receptor with extensive homology to EGF receptor shares chromosomal location with neu oncogene, Science 230 (1985) 1132–1139. [10] S.A. Prigent, N.R. Lemoine, C.M. Hughes, G.D. Plowman, C. Selden, W.J. Gullick, Expression of the c-erbB-3 protein in normal human adult and fetal tissues, Oncogene 7 (1992) 1273–1278. [11] A.N. Garratt, S. Britsch, C. Birchmeier, Neuregulin, a factor with many functions in the life of a schwann cell, Bioessays 22 (2000) 987–996. [12] K.F. Lee, H. Simon, H. Chen, B. Bates, M.C. Hung, C. Hauser, Requirement for neuregulin receptor erbB2 in neural and cardiac development, Nature 378 (1995) 394–398. [13] M. Gassmann, F. Casagranda, D. Orioli, H. Simon, C. Lai, R. Klein, G. Lemke, Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor, Nature 378 (1995) 390–394. [14] D. Meyer, C. Birchmeier, Multiple essential functions of neuregulin in development, Nature 378 (1995) 386–390. [15] R. Chan, W.R. Hardy, M.A. Laing, S.E. Hardy, W.J. Muller, The catalytic activity of the ErbB-2 receptor tyrosine kinase is essential for embryonic development, Mol. Cell. Biol. 22 (2002) 1073–1078. [16] S.L. Erickson, K.S. O’Shea, N. Ghaboosi, L. Loverro, G. Frantz, M. Bauer, L.H. Lu, M.W. Moore, ErbB3 is required for normal cerebellar and cardiac development: a comparison with ErbB2-and heregulin-deficient mice, Development 124 (1997) 4999–5011. [17] D. Riethmacher, E. Sonnenberg-Riethmacher, V. Brinkmann, T. Yamaai, G.R. Lewin, C. Birchmeier, Severe neuropathies in mice with targeted mutations in the ErbB3 receptor, Nature 389 (1997) 725–730. [18] S.A. Crone, Y.Y. Zhao, L. Fan, Y. Gu, S. Minamisawa, Y. Liu, K.L. Peterson, J. Chen, R. Kahn, G. Condorelli, J. Ross, Jr., K.R. Chien, K.F. Lee, ErbB2 is essential in the prevention of dilated cardiomyopathy, Nat. Med. 8 (2002) 459–465. [19] C. Ozcelik, B. Erdmann, B, Pilz, N. Wettschureck, S. Britsch, N. Hubner, K.R. Chien, C. Birchmeier, A.N. Garratt, Conditional mutation of the ErbB2 (HER2) receptor in cardiomyocytes leads to dilated cardiomyopathy, Proc. Natl. Acad. Sci. USA 99 (2002) 8880– 8885. [20] K. Adlkofer, C. Lai, Role of neuregulins in glial cell development, Glia 29 (2000) 104–111. [21] S. Britsch, L. Li, S. Kirchhoff, F. Theuring, V. Brinkmann, C. Birchmeier, D. Riethmacher, The ErbB2 and ErbB3 receptors and their

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

[32] [33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

65

ligand, neuregulin-1, are essential for development of the sympathetic nervous system, Genes Dev. 12 (1998) 1825–1836. J.K. Morris, W. Lin, C. Hauser, Y. Marchuk, D. Getman, K.F. Lee, Rescue of the cardiac defect in ErbB2 mutant mice reveals essential roles of ErbB2 in peripheral nervous system development, Neuron 23 (1999) 273–283. S.K. Park, R. Miller, I. Krane, T. Vartanian, The erbB2 gene is required for the development of terminally differentiated spinal cord oligodendrocytes, J. Cell Biol. 154 (2001) 1245–1258. J.B. Grinspan, M.A. Marchionni, M. Reeves, M. Coulaloglou, S.S. Scherer, Axonal interactions regulate Schwann cell apoptosis in developing peripheral nerve: neuregulin receptors and the role of neureguins, J. Neurosci. 16 (1996) 6107–6118. A.D.O. Levi, R.P. Bunge, J.A. Lofgren, L. Meima, F. Hefti, K. Nikolics, X.M. Sliwkowski, The influence of heregulins on human Schwann cell proliferation, J. Neurosci. 15 (1995) 1329–1340. A.M. Davies, Neuronal survival: early dependence on Schwann cells, Curr. Biol. 8 (1998) R15–R18. K.L. Troyer, D.C. Lee, Regulation of mouse mammary gland development and tumorigenesis by the ERBB signaling network, J. Mammary Gland Biol. Neoplasia 6 (2001) 7–21. Y. Yang, E. Spitzer, D. Meyer, M. Sachs, C. Niemann, G. Hartmann, K.M. Weidner, C. Birchmeier, W. Birchmeier, Sequential requirement of hepatocyte growth factor and neuregulin in the morphogenesis and differentiation of the mammary gland, J. Cell Biol. 131 (1995) 215–226. I.M. Krane, P. Leder, NDF/heregulin induces persistence of terminal end buds and adenocarcinomas in the mammary glands of transgenic mice, Oncogene 12 (1996) 1781–1788. F.E. Jones, D.F. Stern, Expression of dominant-negative ErbB2 in the mammary gland of transgenic mice reveals a role in lobuloalveolar development and lactation, Oncogene 18 (1999) 3481–3490. C.-H. Heldin, A. Ostman, Ligand-induced dimerization of growth factor receptors: variations on the theme, Cytokine Growth Factor Rev. 7 (1996) 33–40. H.S. Cho, D.J. Leahy, Structure of the extracellular region of HER3 reveals an interdomain tether, Science 297 (2002) 1330–1333. T.P. Garrett, N.M. McKern, M. Lou, T.C. Elleman, T.E. Adams, G.O. Lovrecz, H.J. Zhu, F. Walker, M.J. Frenkel, P.A. Hoyne, R.N. Jorissen, E.C. Nice, A.W. Burgess, C.W. Ward, Crystal structure of a truncated epidermal growth factor receptor extracellular domain bound to transforming growth factor alpha, Cell 110 (2002) 763–773. H. Ogiso, R. Ishitani, O. Nureki, S. Fukai, M. Yamanaka, J.H. Kim, K. Saito, A. Sakamoto, M. Inoue, M. Shirouzu, S. Yokoyama, Crystal structure of the complex of human epidermal growth factor and receptor extracellular domains, Cell 110 (2002) 775–787. I. Lax, F. Bellot, R. Howk, A. Ullrich, D. Givol, J. Schlessinger, Functional analysis of the ligand binding site of EGF-receptor utilizing chicken/human receptor molecules, EMBO J. 8 (1989) 421–427. A.E. Summerfield, A.K. Hudnall, T.J. Lukas, C.A. Guyer, J.V. Staros, Identification of residues of the epidermal growth factor receptor proximal to residue 45 of bound epidermal growth factor, J. Biol. Chem. 271 (1996) 19656–19659. L.C. Groenen, E.C. Nice, A.W. Burgess, Structure-function relationships for the EGF/TGF-a family of mitogens, Growth Factors 11 (1994) 235–257. R.N. Jorissen, V.C. Epa, H.R. Treutlein, T.P. Garrett, C.W. Ward, A.W. Burgess, Characterization of a comparative model of the extracellular domain of the epidermal growth factor receptor, Protein Sci. 9 (2000) 310–324. C. Greenfield, I. Hils, M.D. Waterfield, M. Federwisch, A. Wollmer, T.L. Blundell, N. McDonald, EGF binding induces conformational change in the external domain of its receptor, EMBO J. 8 (1989) 4115–4123. E. Tzahar, R. Pinkas Kramarski, J.D. Moyer, L.N. Klapper, I. Alroy, G. Levkowitz, M. Shelly, S. Henis, M. Eisenstein, B.J. Ratzkin,

66

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54]

[55] [56]

[57]

The EGF Receptor Family M. Sela, G.C. Andrews, Y. Yarden, Bivalence of EGF-like ligands drives the ErbB signaling network, EMBO J. 16 (1997) 4938–4950. M.A. Lemmon, Z. Bu, J.E. Ladbury, M. Zhou, D. Pinchasi, I. Lax, D.E. Engelman, J. Schlessinger, Two EGF molecules contribute additively to stabilization of the EGFR dimer, EMBO J. 16 (1997) 281–294. T.W. Gadella, T.M. Jovin, Oligomerization of epidermal growth factor receptors on A431 cells studied by time-resolved fluorescence imaging microscopy. A stereochemical model for tyrosine kinase receptor activation, J. Cell Biol. 129 (1995) 1543–1558. Y. Sako, S. Minoghchi, T. Yanagida, Single-molecule imaging of EGFR signalling on the surface of living cells, Nat. Cell Biol. 2 (2000) 168–172. A. Berezov, J. Chen, Q. Liu, H.T. Zhang, M.I. Greene, R. Murali, Disabling receptor ensembles with rationally designed interface peptidomimetics, J. Biol. Chem. 277 (2002) 28330–28339. T.C. Elleman, T. Domagala, N.M. McKern, M. Nerrie, B. Lonnqvist, T.E. Adams, J. Lewis, G.O. Lovrecz, P.A. Hoyne, K.M. Richards, G.J. Howlett, J. Rothacker, R.N. Jorissen, M. Lou, T.P. Garrett, A.W. Burgess, E.C. Nice, C.W. Ward, Identification of a determinant of epidermal growth factor receptor ligand-binding specificity using a truncated, high-affinity form of the ectodomain, Biochemistry 40 (2001) 8930–8939. D. Karunagaran, E. Tzahar, R.R. Beerli, X. Chen, D. Graus Porta, B.J. Ratzkin, R. Seger, N.E. Hynes, Y. Yarden, ErbB-2 is a common auxiliary subunit of NDF and EGF receptors: implications for breast cancer, EMBO J. 15 (1996) 254–264. M.X. Sliwkowski, G. Schaefer, R.W. Akita, J.A. Lofgren, V.D. Fitzpatrick, A. Nuijens, B.M. Fendly, R.A. Cerione, R.L. Vandlen, K.L. Carraway, Coexpression of erbB2 and erbB3 proteins reconstitutes a high affinity receptor for heregulin, J. Biol. Chem. 269 (1994) 14661–14665. D. Graus Porta, R.R. Beerli, J.M. Daly, N.E. Hynes, ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling, EMBO J. 16 (1997) 1647–1655. E. Tzahar, H. Waterman, X. Chen, G. Levkowitz, D. Karunagaran, S. Lavi, B.J. Ratzkin, Y. Yarden, A. hierarchical network of interreceptor interactions determines signal transduction by Neu differentiation factor/neuregulin and epidermal growth factor, Mol. Cell. Biol. 16 (1996) 5276–5287. P.P. Di Fiore, J.H. Pierce, M.H. Kraus, O. Segatto, C.R. King, S.A. Aaronson, erbB-2 is a potent oncogene when overexpressed in NIH/3T3 cells, Science 237 (1987) 178–182. R.M. Hudziak, J. Schlessinger, A. Ullrich, Increased expression of the putative growth factor receptor p185HER2 causes transformation and tumorigenesis of NIH 3T3 cells, Proc. Natl. Acad. Sci. USA 84 (1987) 7159–7163. C.I. Bargmann, M.C. Hung, R.A. Weinberg, Multiple independent activations of the neu oncogene by a point mutation altering the transmembrane domain of p185, Cell 45 (1986) 649–657. B.D. Cohen, P.A. Kiener, J.M. Green, L. Foy, H.P. Fell, K. Zhang, The relationship between human epidermal growth-like factor receptor expression and cellular transformation in NIH3T3 cells, J. Biol. Chem. 271 (1996) 30897–30903. Y. Kokai, J.N. Myers, T. Wada, V.I. Brown, C.M. LeVea, J.G. Davis, K. Dobashi, M.I. Greene, Synergistic interaction of p185c-neu and the EGF receptor leads to transformation of rodent fibroblasts, Cell 58 (1989) 287–292. J.C. Reed, Dysregulation of apoptosis in cancer, J. Clin. Oncol. 17 (1999) 2941–2953. D.J. Riese 2nd, T.M. van Raaij, G.D. Plowman, G.C. Andrews, D.F. Stern, The cellular response to neuregulins is governed by complex interactions of the erbB receptor family, Mol. Cell. Biol. 15 (1995) 5770–5776. K. Zhang, J. Sun, N. Liu, D. Wen, D. Chang, A. Thomason, S.K. Yoshinaga, Transformation of NIH 3T3 cells by HER3 or HER4 receptors requires the presence of HER1 or HER2, J. Biol. Chem. 271 (1996) 3884–3890.

[58] E. Peles, R. Ben-Levy, E. Tzahar, N. Liu, D. Wen, Y. Yarden, Celltype specific interaction of Neu differentiation factor (NDF/heregulin) with Neu/HER-2 suggests complex ligand-receptor relationships, EMBO J. 12 (1993) 961–971. [59] T. Wada, X. Qian, M.I. Greene, Intermolecular association of the p185neu protein and EGF receptor modulates EGF receptor function, Cell 61 (1990) 1339–1347. [60] M. Alimandi, L.M. Wang, D. Bottaro, C.C. Lee, A. Kuo, M. Frankel, P. Fedi, C. Tang, M. Lippman, J.H. Pierce, Epidermal growth factor and betacellulin mediate signal transduction through co-expressed ErbB2 and ErbB3 receptors, EMBO J. 16 (1997) 5608–5617. [61] R. Pinkas-Kramarski, A.E. Lenferink, S.S. Bacus, L. Lyass, M.L. van de Poll, L.N. Klapper, E. Tzahar, M. Sela, E.J. van Zoelen, Y. Yarden, The oncogenic ErbB-2/ErbB-3 heterodimer is a surrogate receptor of the epidermal growth factor and betacellulin, Oncogene 16 (1998) 1249–1258. [62] L.M. Wang, A. Kuo, M. Alimandi, M.C. Very, C.C. Lee, V. Kapoor, N. Ellmore, X.H. Chen, J.H. Pierce, ErbB-2 expression increases the spectrum and potency of ligand-mediated signal transduction through ErbB-4, Proc. Natl. Acad. Sci. USA 95 (1998) 6809–6814. [63] H. Waterman, Y. Yarden, Molecular mechanisms underlying endocytosis and sorting of ErbB receptor tyrosine kinases, FEBS Lett. 490 (2001) 142–152. [64] G. Levkowitz, H. Waterman, E. Zamir, Z. Kam, S. Oved, W.Y. Langdon, L. Beguinot, B. Geiger, Y. Yarden, c-Cb1/Sli-1 regulates endocytic sorting and ubiquitination of the epidermal growth factor receptor, Genes Dev. 12 (1998) 3663–3674. [65] G. Levkowitz, H. Waterman, S.A. Ettenberg, M. Katz, A.Y. Tsygankov, I. Alroy, S. Lavi, K. Iwai, Y. Reiss, A. Ciechanover, S. Lipkowitz, Y. Yarden, Ubiquitin ligase activity and tyrosine phosphorylation underlie suppression of growth factor signaling by cCb1/Sli-1, Mol. Cell. 4 (1999) 1029–1040. [66] N.L. Lill, P. Douillard, R.A. Awwad, S.M.L. Ota, S. Lupher Miyake, Jr., N. Meissner-Lula, V.W. Hsu, H. Band, The evolutionarily conserved N-terminal region of Cbl is sufficient to enhance downregulation of the epidermal growth factor receptor, J. Biol. Chem. 275 (2000) 367–377. [67] H. Waterman, G. Levkowitz, I. Alroy, Y. Yarden, The RING finger of c-Cb1 mediates desensitization of the epidermal growth factor receptor, J. Biol. Chem. 274 (1999) 22151–22154. [68] M. Yokouchi, T. Kondo, A. Houghton, M. Bartkiewicz, W.C. Horne, H. Zhang, A. Yoshimura, R. Baron, Ligand-induced ubiquitination of the epidermal growth factor receptor involves the interaction of the c-Cbl RING finger and UbeH7, J. Biol. Chem. 274 (1999) 31707–31712. [69] G. Levkowitz, L.N. Klapper, E. Tzahar, A. Freywald, M. Sela, Y. Yarden, Coupling of the c-Cbl protooncogene product to ErbB1/EGF-receptor but not to other ErbB proteins, Oncogene 12 (1996) 1117–1125. [70] S.K. Muthuswamy, M. Gilman, J.S. Brugge, Controlled dimerization of ErbB receptors provides evidence for differential signaling by homo- and heterodimers, Mol. Cell. Biol. 19 (1999) 6845–6857. [71] J. Baulida, G. Carpenter, Heregulin degradation in the absence of rapid receptor-mediated internalization, Exp. Cell. Res. 232 (1997) 167–172. [72] H. Waterman, I. Sabanai, B. Geiger, Y. Yarden, Alternative intracellular routing of ErbB receptors may determine signaling potency, J. Biol. Chem. 273 (1998) 13819–13827. [73] A.E. Lenferink, R. Pinkas Kramarski, M.L. van de Poll, M.J. van Vugt, L.N. Klapper, E. Tzahar, H. Waterman, M. Sela, E.J. van Zoelen, Y. Yarden, Differential endocytic routing of homo- and hetero-dimeric ErbB tyrosine kinases confers signaling superiority to receptor heterodimers, EMBO J. 17 (1998) 3385–3397. [74] A. Sorkin, P.P. Di Fiore, G. Carpenter, The carboxyl terminus of epidermal growth factor receptor/erbB-2 chimerae is internalization impaired, Oncogene 8 (1993) 3021–3028. [75] R. Worthylake, L.K. Opresko, H.S. Wiley, ErbB-2 amplification inhibits down-regulation and induces constitutive activation of both


[76]

[77]

[78]

[79]

[80]

[81]

[82]

[83] [84]

[85] [86] [87] [88] [89] [90]

[91] [92]

[93]

[94]

[95] [96]

ErbB-2 and epidermal growth factor receptors, J. Biol. Chem. 274 (1999) 8865–8874. R. Ben-Levy, H.F. Paterson, C.J. Marshall, Y. Yarden, A. single autophosphorylation site confers oncogenicity to the Neu/ErbB-2 receptor and enables coupling to the MAP kinase pathway, EMBO J. 13 (1994) 3302–3311. S.A. Prigent, W.J. Gullick, Identification of c-erbB-3 binding sites for phosphatidylinositol 3¢-kinase and SHC using an EGF receptor/cerbB-3 chimera, EMBO J. 13 (1994) 2831–2841. E. Peles, R. Lamprecht, R. Ben-Levy, E. Tzahar, Y. Yarden, Regulated coupling of the Neu receptor to phosphatidylinositol 3¢-kinase and its release by oncogenic activation, J. Biol. Chem. 267 (1992) 12266–12274. F. Fazioli, U.H. Kim, S.G. Rhee, C.J. Molloy, O. Segatto, P.P. Di Fiore, The erbB-2 mitogenic signaling pathway: tyrosine phosphorylation of phospholipase C-gamma and GTPase-activating protein does not correlate with erbB-2 mitogenic potency, Mol. Cell. Biol. 11 (1991) 2040–2048. E. Peles, R.B. Levy, E. Or, A. Ullrich, Y. Yarden, Oncogenic forms of the neu/HER2 tyrosine kinase are permanently coupled to phospholipase C gamma, EMBO J. 10 (1991) 2077–2086. J. Liu, J.A. Kern, Neuregulin-1 activates the JAK-STAT pathway and regulates lung epithelial cell proliferation, Am. J. Respir. Cell. Mol. Biol. 27 (2002) 306–313. P. Fedi, J.H. Pierce, P.P. di Fiore, M.H. Kraus, Efficient coupling with phosphatidylinositol 3-kinase, but not phospholipase C gamma or GTPase-activating protein, distinguishes ErbB-3 signaling from that of other ErbB/EGFR family members, Mol. Cell. Biol. 14 (1994) 492–500. D. Bar-Sagi, A. Hall, Ras and Rho GTPases: a family reunion, Cell 103 (2000) 227–238. H. Waterman, M. Katz, C. Rubin, K. Shtiegman, S. Lavi, A. Elson, T. Jovin, Y. Yarden, A mutant EGF-receptor defective in ubiquitylation and endocytosis unveils a role for Grb2 in negative signaling, EMBO J. 21 (2002) 303–313. C.J. Marshall, MAP kinase kinase kinase, MAP kinase kinase and MAP kinase, Curr. Opin. Gen. Dev. 4 (1994) 82–89. A. Toker, A.C. Newton, Cellular signaling: pivoting around PDK-1, Cell 103 (2000) 185–188. P. Blume-Jensen, T. Hunter, Oncogenic kinase signaling, Nature 411 (2001) 355–365. R. Meier, B.A. Hemmings, Regulation of protein kinase B, J. Recept. Signal Transduct. Res. 19 (1999) 121–128. L.C. Cantley, The phosphoinositide 3-kinase pathway, Science 296 (2002) 1655–1657. M. Karin, T. Hunter, Transcriptional control by protein phosphorylation: signal transmission from the cell surface to the nucleus, Curr. Biol. 5 (1995) 747–757. T. Hunter, Signaling—2000 and beyond, Cell 100 (2000) 113–127. L.Y. Bourguignon, H. Zhu, A. Chu, N. Iida, L. Zhang, M.C. Hung, Interaction between the adhesion receptor, CD44, and the oncogene product, p185HER2, promotes human ovarian tumor cell activation, J. Biol. Chem. 272 (1997) 27913–27918. L.S. Sherman, T.A. Rizvi, S. Karyala, N. Ratner, CD44 enhances neuregulin signaling by Schwann cells, J. Cell. Biol. 150 (2000) 1071–1084. S. Jepson, M. Komatsu, B. Haq, M.E. Arango, D. Huang, C.A. Carraway, K.L. Carraway, Muc4/sialomucin complex, the intramembrane ErbB2 ligand, induces specific phosphorylation of ErbB2 and enhances expression of p27(kip), but does not activate mitogenactivated kinase or protein kinaseB/Akt pathways, Oncogene 21 (2002) 7524–7532. D. Harari, Y. Yarden, Molecular mechanisms underlying ErbB2/ HER2 action in breast cancer, Oncogene 19 (2000) 6102–6114. A.E. Lenferink, D. Busse, W.M. Flanagan, F.M. Yakes, C.L. Arteaga, ErbB2/neu kinase modulates cellular p27(Kip1) and cyclin D1 through multiple signaling pathways, Cancer Res. 61 (2001) 6583–6591.

67

[97] R.J. Lee, C. Albanese, M. Fu, M. D’Amico, B. Lin, G. Watanabe, G.K. Haines 3rd, P.M. Siegel, M.C. Hung, Y. Yarden, J.M. Horowitz, W.J. Muller, R.G. Pestell, Cyclin D1 is required for transformation by activated Neu and is induced through an E2F-dependent signaling pathway, Mol. Cell. Biol. 20 (2000) 672–683. [98] G.M. Wulf, A. Ryo, G.G. Wulf, S.W. Lee, T. Niu, V. Petkova, K.P. Lu, Pin1 is overexpressed in breast cancer and cooperates with Ras signaling in increasing the transcriptional activity of c-Jun towards cyclin D1, EMBO J. 20 (2001) 3459–3472. [99] J.A. Diehl, M. Cheng, M.F. Roussel, C.J. Sherr, Glycogen synthase kinase-3beta regulates cyclin D1 proteolysis and subcellular localization, Genes Dev. 12 (1998) 3499–3511. [100] R.S. Muraoka, A.E. Lenferink, B. Law, E. Hamilton, D.M. Brantley, L.R. Roebuck, C.L. Arteaga, ErbB2/Neu-induced, cyclin D1dependent transformation is accelerated in p27-haploinsufficient mammary epithelial cells but impaired in p27-null cells, Mol. Cell. Biol. 22 (2002) 2204–2219. [101] R.M. Neve, H. Sutterluty, N. Pullen, H.A. Lane, J.M. Daly, W. Krek, N.E. Hynes, Effects of oncogenic ErbB2 on G1 cell cycle regulators in breast tumour cells, Oncogene 19 (2000) 1647–1656. [102] M. Fiorini, C. Ballaro, G. Sala, G. Falcone, S. Alema, O. Segatto, Expression of RALT, a feedback inhibitor of ErbB receptors, is subjected to an integrated transcriptional and post-translational control, Oncogene 21 (2002) 6530–6539. [103] L. Fiorentino, C. Pertica, M. Fiorini, C. Talora, M. Crescenzi, L. Castellani, S. Alema, P. Benedetti, O. Segatto, Inhibition of ErbB-2 mitogenic and transforming activity by RALT, a mitogen-induced signal transducer which binds to the ErbB-2 kinase domain, Mol. Cell. Biol. 20 (2000) 7735–7750. [104] N.G. Azios, F.J. Romero, M.C. Denton, J.K. Doherty, G.M. Clinton, Expression of herstatin, an autoinhibitor of HER-2/neu, inhibits transactivation of HER-3 by HER-2 and blocks EGF activation of the EGF receptor, Oncogene 20 (2001) 5199–5209. [105] H. Lee, R.W. Akita, M.X. Sliwkowski, N.J. Maihle, A naturally occurring secreted human ErbB3 receptor isoform inhibits heregulinstimulated activation of ErbB2, ErbB3, and ErbB4, Cancer Res. 61 (2001) 4467–4473. [106] A. Citri, I. Alroy, S. Lavi, C. Rubin, W. Xu, N. Grammatikakis, C. Patterson, L. Neckers, D.W. Fry, Y. Yarden, Drug-induced ubiquitylation and degradation of ErbB receptor tyrosine kinases: implications for cancer therapy, EMBO. J. 21 (2002) 2407–2417. [107] W. Xu, M. Marcu, X. Yuan, E. Mimnaugh, C. Patterson, L. Neckers, Chaperone-dependent E3 ubiquitin ligase CHIP mediates a degradative pathway for c-ErbB2/Neu, Proc. Natl. Acad. Sci. USA 99 (2002) 12847–12852. [108] A. Rosenthal, P.B. Lindquist, T.S. Bringman, D.V. Goeddel, R. Derynck, Expression in rat fibroblasts of a human transforming growth factor-alpha cDNA results in transformation, Cell 46 (1986) 301–309. [109] D.S. Salomon, R. Brandt, F. Ciardiello, N. Normanno, Epidermal growth factor-related peptides and their receptors in human malignancies, Crit. Rev. Oncol. Hematol. 19 (1995) 183–232. [110] M. Alimandi, A. Romano, M.C. Curia, R. Muraro, P. Fedi, S.A. Aaronson, P.P. Di Fiore, M.H. Kraus, Cooperative signaling of ErbB3 and ErbB-2 in neoplastic transformation of human mammary carcinoma cells, Oncogene 15 (1995) 1813–1821. [111] F. Lonardo, E. Di Marco, C.R. King, J.H. Pierce, O. Segatto, S.A. Di Aaronson, P.P. Fiore, The normal erbB-2 product is an atypical receptor-like tyrosine kinase with constitutive activity in the absence of ligand, New Biol. 2 (1995) 992–1003. [112] Y. Yarden, Agonistic antibodies stimulate the kinase encoded by the neu protooncogene in living cells but the oncogenic mutant is constitutively active, Proc. Natl. Acad. Sci. USA 87 (1990) 2569– 2573. [113] D.J. Slamon, G.M. Clark, S.G. Wong, W.J. Levin, A. Ullrich, W.L. McGuire, Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene, Science 235 (1987) 177–182.

68


[114] D.J. Slamon, W. Godolphin, L.A. Jones, J.A. Holt, S.G. Wong, D.E. Keith, W.J. Levin, S.G. Stuart, J. Udove, A. Ullrich, M.F. Press, Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer, Science 244 (1989) 707–712. [115] D.C. Allred, G.M. Clark, R. Molina, A.K. Tandon, S.J. Schnitt, K.W. Gilchrist, C.K. Osborne, D.C. Tormey, W.L. McGuire, Overexpression of HER-2/Neu and its relationship with other prognostic factors change during the progression of in situ to invasive breast cancer, Hum. Pathol. 23 (1992) 974–979. [116] W.J. Gullick, R. Srinivasan, The type 1 growth factor receptor family: new ligands and receptors and their role in breast cancer, Breast Cancer Res. Treat. 52 (1998) 43–53. [117] N.E. Hynes, D.F. Stern, The biology of erbB-2/neu/HER-2 and its role in cancer, Biochem. Biophys. Acta 1198 (1994) 165–184. [118] L.N. Klapper, M.H. Kirschbaum, M. Sela, Y. Yarden, Biochemical and clinical implications of the ErbB/HER signaling network of growth factor receptors, Adv. Cancer Res. 77 (2000) 25–79. [119] Z. Suo, B. Risberg, M.G. Kalsson, K. Willman, A. Tierens, E. Skovlund, J.M. Nesland, EGFR family expression in breast carcinomas. c-erbB-2 and c-erbB-4 receptors have different effects on survival, J. Pathol. 196 (2002) 17–25. [120] R.J. Gilbertson, R.H. Perry, P.J. Kelly, A.D. Pearson, J. Lunec, Prognostic significance of HER2 and HER4 coexpression in childhood medulloblastoma, Cancer Res. 57 (1997) 3272–3280. [121] B. Bodey, B. Bodey Jr., A.M. Groger, J.V. Luck, S.E. Siegel, C.R. Taylor, H.E. Kaiser, Clinical and prognostic significance of the expression of the c-erbB-2 and c-erbB-3 oncoproteins in primary and metastatic malignant melanomas and breast carcinomas, Anticancer Res. 17 (1997) 1319–1330. [122] N.R. Lemoine, D.M. Barnes, D.P. Hollywood, C.M. Hughes, P. Smith, E. Dublin, S.A. Prigent, W.J. Gullick, H.C. Hurst, Expression of the ERBB3 gene product in breast cancer, Br. J. Cancer 66 (1992) 1116–1121. [123] C.A. Maurer, H. Friess, B. Kretschmann, A. Zimmermann, A. Stauffer, H.U. Baer, M. Kore, M.W. Buchler, Increased expression of erbB3 in colorectal cancer is associated with concomitant increase in the level of erbB2, Hum. Pathol. 29 (1998) 771–777. [124] W. Xia, Y.K. Lau, H.Z. Zhang, F.Y. Xiao, D.A. Johnston, A.R. Liu, L. Li, R.L. Katz, M.C. Hung, Combination of EGFR, HER-2/neu, and HER-3 is a stronger predictor for the outcome of oral squamous cell carcinoma than any individual family members, Clin. Cancer Res. 5 (1999) 4164–4174.

[125] E.S. Yi, D. Harclerode, M. Gondo, M. Stephenson, R.W. Brown, M. Younes, P.T. Cagle, High c-erbB-3 protein expression is associated with shorter survival in advanced non-small cell lung carcinomas, Mod. Pathol. 10 (1997) 142–148. [126] B.J. Simpson, J. Weatherill, E.P. Miller, A.M. Lessells, S.P. Langdon, W.R. Miller, c-erbB-3 protein expression in ovarian tumours, Br. J. Cancer 71 (1995) 758–762. [127] Y. Ito, T. Takeda, M. Sakon, M. Tsujimoto, S. Higashiyama, K. Noda, E. Miyoshi, M. Monden, N. Matsuura, Expression and clinical significance of erb-B receptor family in hepatocellular carcinoma, Br. J. Cancer 84 (2001) 1377–1383. [128] N.H. Chow, S.H. Chan, T.S. Tzai, C.L. Ho, H.S. Liu, Expression profiles of ErbB family receptors and prognosis in primary transitional cell carcinoma of the urinary bladder, Clin. Cancer Res. 7 (2001) 1957–1962. [129] R. Derynck, D.V. Goeddel, A. Ullrich, J.U. Gutterman, R.D. Williams, T.S. Bringman, W.H. Berger, Synthesis of messenger RNAs for transforming growth factors alpha and beta and the epidermal-growth-factor-receptor by human tumors, Cancer Res. 47 (1987) 707–712. [130] A. deFazio, Y.E. Chiew, R.L. Sini, P.W. Janes, R.L. Sutherland, Expression of c-erbB receptors, heregulin and oestrogen receptor in human breast cell lines, Int. J. Cancer 87 (2000) 487–498. [131] H. Friess, P. Berberat, M. Schilling, J. Kunz, M. Kore, M.W. Buchler, Pancreatic cancer: the potential clinical relevance of alterations in growth factors and their receptors, J. Mol. Med. 74 (1996) 35– 42. [132] G. Carpenter, Employment of the epidermal growth factor receptor in growth factor-independent signaling pathways, J. Cell Biol. 146 (1999) 697–702. [133] R. Schweitzer, R. Howes, R. Smith, B.-Z. Shilo, M. Freeman, Inhibition of Drosophila EGF receptor activation by the secreted Argos protein, Nature 376 (1995) 699–702. [134] P. Nagy, G. Vereb, Z. Sebestyen, G. Horvath, S.J. Lockett, S. Damjanovich, J.W. Park, T.M. Jovin, J. Szollosi, Lipid rafts and the local density of ErbB proteins influence the biological role of homo- and heteroassociations of ErbB2, J. Cell Sci. 115 (2002) 4251–4262. [135] E. Tzahar, J.D. Moyer, H. Waterman, E.G. Barbacci, J. Bao, G. Levkowitz, M. Shelly, S. Strano, R. Pinkas-Kramarski, J.H. Pierce, G.C. Andrews, Y. Yarden, Pathogenic poxviruses reveal viral strategies to exploit the ErbB signaling network, EMBO J. 17 (1998) 5948– 5963.

ErbB-4: mechanism of action and biology Graham Carpenter Department of Biochemistry, Vanderbilt University School of Medicine, Nashville, TN 37232-0146, USA

Abstract The most recently described member of the ErbB receptor tyrosine kinase family is ErbB-4. In general, the structure of this receptor and its mechanism of action is similar to that described for ErbB-1. However, significantly less is known about ErbB-4 and there are several novel aspects to its structure, mechanism of action, and biology. This includes the spectrum of ligands that activate ErbB-4, the presence of functionally distinct isoforms, a proteolytic processing pathway to the nucleus, and the capacity to induce a spectrum of cellular responses such as mitogenesis, differentiation, growth inhibition, and survival.

Introduction: receptor identification, structure, isoforms, and gene localization Employing a strategy of homology cloning, ErbB-4 was cloned from a human mammary carcinoma cell line and its cDNA sequence determined [1]. That data showed ErbB-4 to be related in sequence to other ErbB receptors and to be organized in a similar fashion (Fig. 1). A single transmembrane domain separates equal sized ecto- and cytoplasmic domains. Within the ectodomain is a cleaved signal sequence and two cysteine-rich regions (domains II and IV), typical of ErbB receptors; nearly all of the 50 ectodomain cysteine residues of ErbB-1 are conserved in ErbB-4. By analogy with ErbB-1, it seems likely that domain III mediates growth factor binding. Between domain IV and the transmembrane domain, ErbB-4 has a comparatively longer stalk region, which may make it uniquely sensitive, within this receptor family, to ectodomain cleavage (see Receptor Trafficking, below). The cytoplasmic domain contains a juxtamembrane region, a tyrosine kinase domain, and a carboxyterminal domain—all typical of ErbB receptors. ErbB-4 tyrosine autophosphorylation sites have not been mapped, but within other receptors in this family these sites are located in the carboxy-terminal domain. The tyrosine autophosphorylation residues in ErbB-1 that have been identified are all conserved in ErbB-4. While the ErbB-3 kinase domain has mutations that significantly attenuate its kinase activity, these changes are not present in the ErbB-4 kinase domain. ErbB-4 is an active kinase in the absence of a coreceptor, such as ErbB-2, which is required for activation of ErbB-3 as a signaling component.

An interesting feature of the ErbB-4 sequence is a carboxy-terminal sequence TVV, which constitutes a PDZ domain recognition motif. ErbB-2 is reported to contain an internal motif that recognizes PDZ domains. The function of this ErbB-4 motif in receptor proteolytic processing and association with PDZ domain-containing proteins is described below. The human gene for ErbB-4 has localized to chromosome 2 in the q33.3–34 region [2]. Hence, each member of the ErbB receptor family is located on a different chromosome. An interesting feature of the ErbB-4 receptor is the presence of isoforms, generated by mRNA splicing, that suggest functional differences. One isoform contains an altered sequence within the ErbB-4 receptor stalk region [3]. The originally cloned ErbB-4 isoform is designated Jm-a and the altered isoform is designated Jm-b. The different sequences of Jm-a and Jm-b in the stalk region are depicted in Fig. 1. The changes do not affect ligand binding or tyrosine kinase activity, but do influence the sensitivity of the ErbB-4 ectodomain to shedding from the cell surface. A second isoform contains sequence changes within the ErbB-4 cytoplasmic domain and influences a receptor docking site for phosphatidylinositol-3 kinase (PI-3 kinase) [4]. The originally described ErbB-4 sequence [1] has a binding site for this signal transducer; however, this site is deleted in an isoform designated CYT-2. The presence of the PI-3 kinase site is referred to as the CYT-1 isoform. These changes are depicted in Fig. 1. The occurrence of these ecto- and cytoplasmic domain changes suggests that potentially four isoforms of ErbB-4 may exist. However, it has not been formally shown that an ErbB-4 receptor exists that contains the Jm-b ectodomain

70


Fig. 1. Schematic illustration of ErbB-4 primary structure featuring domains and isoform changes. From the amino-terminus structural features include a signal sequence (SS), domains I, II, III, and IV in the ectodomain, the stalk region, which contains alterations in sequence for the Jm-a and Jm-b isoforms, the transmembrane domain (TM), tyrosine kinase domain, carboxy-terminal domain containing autophosphorylation sites (pYn) plus sequence changes attributable to the CYT-1 and CYT-2 isoforms, and the PDZ domain recognition motif (TVV) at the carboxy-terminus.

sequence together with the CYT-2 cytoplasmic domain sequence. Additional information on these isoforms has been reviewed elsewhere [5] and is described below.

Receptor ligands Ligands, which bind to ErbB-4 with high affinity and specificity and which provoke receptor activation and signaling, are divided into two groups, i.e., the neuregulins, also termed heregulins, and certain members of the epidermal growth factor (EGF) family of ErbB-1 ligands (Fig. 2). For consistency, the term neuregulin is used exclusively in this review. There are four neuregulin genes, denoted 1, 2, 3, and 4, and the product of each is capable of recognizing ErbB-4 in a biologically productive manner. Neuregulins 1 and 2 also exist in a number of splicing isoforms, but it is not clear how each isoform associates with ErbB-4. Evidence of the capacity of each neuregulin gene product to activate ErbB-4 has been published for neuregulin-1 [6–8], neuregulin-2 [9,10], neuregulin-3 [11], and neuregulin-4 [12]. Among these, neuregulins 1 and 2 also recognize ErbB-3 as a high affinity binding site, but neuregulins 3 and 4 are reported to interact only with ErbB-4. None of the neuregulins recognize ErbB-1 or ErbB-2 directly and with high affinity. A direct comparison of the capacity of various neuregulins to activate ErbB-4 confirmed that neuregulins 3 and 4 activate ErbB-4, but not ErbB-3 [13]. This study also showed that neuregulin-2a failed to activate ErbB-4, while neuregulin-2b was a potent activator. It does appear, however, that neuregulins 3 and 4 have a lower potency for ErbB-4 activation compared to neuregulins 1b and 2b. There are seven different gene products that are known to act as high affinity ligands for ErbB-1 and a subset of

these are also agonists for ErbB-4. These include betacellulin [14], heparin binding-EGF (HB-EGF) [15], and epiregulin [16–18]. Other ErbB-1 agonists (EGF, transforming growth factor-a, and amphiregulin) have been tested for ErbB-4 interaction with none detected. The newest ErbB-1 agonist, epigen, has not yet been tested for ErbB-4 interaction. There are three additional cloned growth factors that are reported to activate ErbB-4 autophosphorylation. These are designated Don-1 [19], tomoregulin [20], and NTAK [21]. Their potencies seem relatively low and they remain less well characterized as ErbB-4 agonists. Don-1 and NTAK seem to be related to the neuregulin family, while tomoregulin is closer to the EGF family of ligands. Finally, decorin, an extracellular proteoglycan, is reported to associate with ErbB-4 and provoke its activation [22]. The existence of multiple ligands for ErbB-4 raises the issue of whether these ligands provoke different biological

Fig. 2. Growth factor-dependent ErbB-4 activation. Shown are the ligands that are recognized by ErbB-4 and that produce, through homodimerization or heterodimerization with ErbB-2, an activated receptor complex.

ErbB-4: Mechanism of Action and Biology

responses or the utilization of different signal transduction pathways. In regard to the latter point, one study has reported overlapping but distinct patterns of ErbB-4 autophosphorylation following the addition of different agonists [23]. Since relatively little is known about downstream signaling pathways, it is unclear if the observed differences translate to changes in signaling specificity. Also, comparative studies indicate that biological responses, such as cell survival, are not promoted equally by the various ErbB-4 ligands—at least in the cell assays employed. Last, the presence or absence of ErbB-2 can influence these outcomes significantly and has not been surveyed in a rigorous manner for all ErbB-4 agonists.

Receptor activation and signaling Following ligand binding, the ErbB-4 receptor is activated by a process common to other ErbB receptors, i.e., dimerization and autophosphorylation (Fig. 2). Initial studies showed that, unlike ErbB-3, ligand binding provoked autophosphorylation of ErbB-4 in a cellular environment devoid of other ErbB receptors [1,6–12]. In this regard ErbB-4 is analogous to ErbB-1. However, numerous studies have reported that heterodimerization of ErbB-4 with ErbB2 forms a higher affinity binding site, enhances the level of autophosphorylation, and can significantly modify the biological response to ErbB-4 ligands [13,17,18,24–32, 147,148]. In contrast, two studies have reported that ErbB2 does not modify the activation of ErbB-4 by growth factors [33,34]. The mechanism of ErbB-4 homodimerization or heterodimerization has not been studied at the structural level. It has been reported that the isolated ErbB-4 ectodomain homodimerizes in the presence of its ligand [31]. This report also demonstrated the ligand-induced heterodimerization of ErbB-4 and ErbB-2 isolated ectodomains. Another study has described the capacity of isolated ErbB-4 transmembrane domain to dimerize [35]. Activation of a receptor tyrosine kinase initiates biochemical signals that initiate growth responses. The initial step in these signaling events is receptor association with cellular proteins, which in some cases results in the tyrosine phosphorylation of these proteins. To date a few of these receptor proximal signaling molecules have been identified, i.e., Shc, the p85 subunit of PI-3 kinase, GrbB2, GrbB7 JAK, and STAT [4,15,23,27,28,36–40,145,146]. ErbB-4 interaction with Shc and GrbB2 would predict activation of Ras and mitogenic signals, while the p85 subunit of PI-3 kinase would lead to elevated levels of phosphatidylinositol-3,4,5-trisphosphate and the activation of cell survival pathways. Activation of the JAK/STAT pathway would lead to changes in gene expression. The role of GrbB7 in downstream signal pathways and cellular responses is not clear, though it does associate with several activated receptor tyro-

71

sine kinases. To date no signaling molecule or pathway has been identified as novel to ErbB-4 activation.

Receptor trafficking While ErbB-1 and most growth factor receptor tyrosine kinases are rapidly internalized through clathrin-coated pits following ligand binding, ErbB-4 is internalized very slowly after addition of its ligand [41]. That some ligand: ErbB-4 complexes are, in fact, sorted to lysosomes is indicated by the appearance of low molecular weight degradation products of the ligand [42]. ErbB-4 internalization is sufficiently slow that mechanisms other than the clathrin-coated pit need to be considered. The failure of activated ErbB-4 to rapidly enter the endocytic pathway, considered to be a desensitization mechanism, led to the investigation of alternative desensitization mechanisms. These studies demonstrated that the ErbB-4 ectodomain was shed as a fragment of 120 kDa from the cell surface [43] and that this shedding can be stimulated by 120-tetradecanoylphorbol-13-acetate (TPA) [44] or by neuregulin [45] (Fig. 3). The shedding activity was identified as a metalloprotease on the basis of inhibitor sensitivity [43] and subsequently shown to be attributable to TACE (tumor necrosis factor-alpha converting enzyme) [46], a member of the transmembrane ADAM metalloprotease family also designated ADAM17. Whether TACE actually executes this cleavage event is not clear, but it is required for the ectodomain shedding. Cell surface shedding activity releases the ErbB-4 ectodomain fragment into the media and leaves an 80-kDa fragment, representing the transmembrane and cytoplasmic domains, associated with the cell [44]. The potential function of the ectodomain fragment is unclear, but the 80-kDa fragment is an active tyrosine kinase at least in vitro [43]. As previously noted, ectodomain cleavage is stimulatable by TPA or neuregulin, but it should be mentioned that in the absence of these agents there is a basal level of cleavage in most all cells [43]. Control of this basal cleavage is not understood mechanistically. However, the TPA and heregulin mechanisms seem distinct in the following ways [45]. TPA-dependent cleavage is blocked by an inhibitor of protein kinase C, but heregulin-mediate cleavage is not. Also, neuregulin cleavage is associated with liganddependent translocation to a detergent-insoluble fraction (see below) and receptor internalization, but TPA cleavage is not (Fig. 3). It seems plausible to suggest that TPA may mediate activation of TACE and thereby provoke cleavage of a number of cell surface substrates, while neuregulin mediates the colocalization of ErbB-4 and TACE in a detergent-insoluble fraction or internalized compartment and thereby stimulates cleavage. Further examination has shown that the 80-kDa ErbB-4 fragment is processed by a second membrane-localized protease activity, g-secretase, which typically cleaves a

72


Fig. 3. Proteolytic cleavage pathway for ErbB-4. Proteolytic processing is initiated by TPA or cognate growth factor and involves different topological mechanisms, but the same proteases. Ectodomain cleavage involves TACE, while intramembrane proteolysis is affected by g-secretase (PS-1). Growth factor-dependent cleavage seems to involve endocytosis, while TPA-dependent cleavage does not. Processing by either route produces the cytosolic ErbB-4 fragment (s80), which is found in the nucleus.

transmembrane protein within the transmembrane domain [47,48] (Fig. 3). This activity liberates the ErbB-4 cytoplasmic domain from the plasma membrane and into the cytosol. Subsequently this fragment translocates to the nucleus, but its function in the nucleus is not known. The means by which the 80-kDa fragment is recognized by g-secretase is unclear, but may require the PDZ domain recognition sequence located its carboxy-terminus [49]. This recognition sequence has been shown to facilitate a constitutive interaction of ErbB-4 with the multi-PDZ domain containing protein PSD95 [50–52], which may enhance clustering ErbB-4 molecules [51,52] and/or facilitate cleavage by g-secretase [49]. That this proteolytic processing of ErbB-4 also constitutes a signaling pathway is indicated by the fact that g-secretase inhibition prevents the capacity of an ErbB-4 agonist to provoke cell death under serum-free conditions [47,49]. Also, the nuclear localization of ErbB-4 has been noted, using immunohistochemistry, in certain tissues [53–55]. Two studies have described the association of ErbB-4 with detergent-resistant membrane microdomains and the influence of the cognate growth factor on this localization. In cardiac myocytes the majority of ErbB-4 is present in caveolin-enriched microdomains, most likely caveolae, and association of ErbB-4 and caveolin-3 was detected by coprecipitation [56]. Addition of neuregulin to myocytes promoted the rapid exit of some ErbB-4 from this specialized microdomain to the general (detergent-soluble) plasma membrane faction. Interestingly, ErbB-2 was also localized to the caveolin microdomain, but did not change location following the addition of neuregulin.

In a mammary carcinoma cell line, the addition of neuregulin promoted the migration of ErbB-4 [45] and ErbB-2 [57], but not ErbB-3, into a detergent-resistant membrane microdomain. Within the context of this microdomain both ErbB-4 and ErbB-2 were hypertyrosine phosphorylated. While the nature of this detergent-resistant fraction is unclear, it most likely resembles a lipid raft and may be analogous to microdomains found in cells that do not express caveolin and hence do not exhibit caveolae. The significance of these microdomain structures for ErbB-4 trafficking or signaling remains to be defined. Other reports have indicated that ErbB-4 may associate with other cell surface molecules that could influence its trafficking or signaling properties. These include MUC1, a large heavily glycosylated transmembrane molecule that associates with all four ErbB receptors [58], and CD44 [59], a transmembrane proteoglycan that has a role in cell adhesion to the extracellular matrix. The above data suggest that ErbB-4 may be found in association with several other transmembrane proteins depending on cell type and the presence or absence of cognate growth factors. In many cases the quantitative aspects of these associations as well as their biological significance is unclear. Recently, a cytosolic protein termed Nrdp1 (neuregulin receptor degradation protein-1) was isolated on the basis of its interaction with ErbB-3 [60]. When the protein was overexpressed, the cellular levels of ErbB3 and ErbB-4, but not ErbB-1 or -2, were reduced. A subsequent study has shown that Nrdp1 is a ubiquitin ligase [61]. Interestingly, the 80-kDa fragment produced from


ErbB-4 by metalloprotease activity is ubiquitinated and proteosome inhibitors elevate the level of this fragment in cells [43].

Growth responses in experimental systems Numerous articles have reported the influence of neuregulin/heregulin on the growth of cell lines that endogenously express ErbB-4. However, in nearly all of these cell lines ErbB-3 is also expressed and hence it is difficult to discern whether the cell response is mediated by ErbB-3 or ErbB-4 or both. For this reason these articles are not reviewed herein. Investigators have employed a few cell systems to experimentally evaluate growth responses mediated directly by ErbB-4 and in some cases by its coexpression with ErbB-2. 3T3 cell lines that do not normally express ErbB receptors have been used extensively as recipient cells to evaluate growth effects mediated by transfected ErbB receptors. In particular, this cell system is used to evaluate mitogenic responses and transforming activity. Expression of ErbB-4 in these cells has established that growth factor activation of ErbB-4 in the absence of other ErbB receptors provokes a significant increase in cell proliferation [15,26,62] and that either the CYT-1 or the CYT-2 isoforms of ErbB-4 can mediate this activity [63]. Hence, this response seems independent of ErbB-4 association with ErbB-2 or its capacity to activate (CYT-1) or not activate (CYT-2) PI-3 kinase. It is reported that while neuregulin activation of ErbB-4 increases cell proliferation, activation of ErbB-4 by HBEGF does not [15]. Also, this study demonstrated that activation of ErbB-4 by either HB-EGF or neuregulin stimulated chemotaxis. This response was dependent on PI3 kinase activation, as it was provoked by the CYT-1 ErbB4 isoform, but not the CYT-2 isoform [63]. When 3T3 cells are starved for serum, cell death occurs slowly. Expression of the CYT-1 ErbB-4 isoform, but not the CYT-2 isoform, together with neuregulin, was able to prevent cell death in a cell survival response [63]. Hence, these studies demonstrate that ErbB-4 signaling for chemotaxis and cell survival requires activation of PI-3 kinase. Various experimental parameters of the transformation process are measurable in 3T3 cells. The most common assay with ErbB receptors is the focus forming assay—a measure of low density growth capacity. When ErbB-4 was expressed by itself, the data show no [26] or moderate [32,64] focus-forming activity in the presence of exogenous neuregulin. This cellular response to ErbB-4 activation was increased when ErbB-2 was coexpressed [26,32]. Similar results were obtained when colony formation in soft agar, a measure of anchorage-independent growth, was employed [65]. Expression of ErbB-4 by itself did not mediate colony formation in the presence of neuregulin; however, coexpression of ErbB-2 and ErbB-4 did provoke colony forming activity that was dependent on the presence of neuregulin.

73

This is consistent with another report in which a constitutively active mutant of ErbB-4, dimerized by mutagenesis to Cys in the ectodomain, was unable to form colonies in soft agar, while a similar ErbB-2 mutant did form colonies [66]. Some investigators have evaluated ErbB receptor growth stimulating capacity by transfection of the receptors into the hematopoietic cell lines 32D (myeloid) or BaF3 (lymphobastoid). These cells do not exogenously express any ErbB family members and require interleukin-3 (IL-3) for survival and proliferation. In this system one can measure cell survival or proliferation in the absence of IL-3. These studies [14,17,18,24,27,28,67] agree that by itself ErbB-4 has a weak capacity to support cell survival in the presence of its cognate ligand. However, when coexpressed with ErbB-2, most studies show that ErbB-4 more effectively mediates cell survival in a manner that is dependent on the presence of an ErbB-4 ligand. The picture that emerges is that ErbB-4 can mediate a proliferative signal, but its potential to do so is significantly enhanced by the presence of ErbB-2. There is evidence based on other systems that ErbB-4 can provoke cell differentiation responses. In neuronal PC12 cells the expression and activation of ErbB-4 increases neurite extension and induces the expression of GAP-43, a neuronal differentiation marker [68]. In this cell system ErbB-4 is added by transfection and the cells endogenously express the other three ErbB receptors. However, neuregulin only induces differentiation in the cells that express ErbB4. Neuregulin activation of ErbB-4 also protects these cells from apoptotic stimuli by a mechanism that requires activation of PI-3 kinase and cell survival pathways [69,70]. Studies of breast cancer cell lines have shown that ErbB-4 is necessary to mediate antiproliferative and differentiation responses provoked by neuregulin [71]. In these cells, suppression of ErbB-2 expression did not alter the ErbB-4provoked response, indicating that homodimers of ErbB-4 were sufficient for these nonmitogenic responses.

Roles in normal and tumor tissues A survey of ErbB receptor expression in a large number of adult and fetal human tissues showed that ErbB-4 is ubiquitously expressed [72]. Expression is highest in brain and heart, but significant levels are present in the epithelia of skin, gastrointestinal, urinary, reproductive, and respiratory tracts, along with skeletal muscle, circulatory, endocrine, and nervous systems. Also, a variety of human cancers express ErbB-4, although squamous carcinomas seem relatively devoid of this receptor. The authors concluded with the general impression that ErbB-4 expression in normal and tumor tissue is skewed toward the differentiated compartments. If so, this view is consistent with the observation that endogenous ErbB-4 expression in cell lines is relatively uncommon. In the following sections, the expression and

74


function, where addressed, are reviewed on a site-by-site basis.

Mammary tissue Three studies have examined ErbB-4 expression in normal mammary tissue by western blotting [73] or immunohistochemistry [72,74]. The studies argue that ErbB-4 is highest during pregnancy and occurs primarily in the ductal epithelium, especially at the terminal ducts or end buds. Expression at lower levels is detected in nulliparous animals and during lactation and involution. Using polymerase chain reaction analysis, expression of mRNA for both the CYT-1 and CYT-2 isoforms of ErbB-4 have been identified in normal mammary tissue [75]. To assess the function of ErbB-4 in normal mammary development, a dominant-negative construct of ErbB-4 was expressed in the breast as a transgene under control of the MMTV promoter [40]. Mammary gland development was affected at mid-lactation when lubuloalveoli became condensed and deficient in lactation products—acid whey protein and a-lactalbumin. This result points to a critical role of ErbB-4 in the late differentiation of mammary gland function and are consistent with cell culture data showing a critical role for ErbB-4 in mammary cell differentiation [71]. Several groups have evaluated the expression of ErbB-4 in mammary carcinoma, particularly in view of the known high expression of ErbB-1 and ErbB-2 in this cancer [54,76–83]. These studies agree that ErbB-4 is not frequently overexpressed in breast carcinoma, as are ErbB-1 and ErbB-2, but is found at moderate to low levels. While ErbB-1 and ErbB-2 expression is correlated with tumors that are estrogen receptor negative and have a poor prognosis, the opposite is true for ErbB-4. Its expression correlates with the presence of estrogen receptors, a more differentiated tumor grade, and a more favorable prognosis. Interestingly, a significant percentage of breast tumors that express ErbB4 demonstrate nuclear localization [54], perhaps as a result of ErbB-4 proteolysis and trafficking (see above). Also, both the CYT-1 and CYT-2 mRNA isoforms of ErbB-4 are found in mammary tumor specimens, as in normal mammary tissue [75], ErbB-4 is frequently expressed in Paget’s disease of the breast, which involves migration of neoplastic cells into the nipple [149]. It is suggested that neuregulin activation of ErbB-3 or ErbB-4 may mediate this migratory activity.

Other tumors ErbB-4 expression has been noted in several other tumors in addition to mammary carcinoma [53,79,84–90]. These include carcinomas of the colon, prostate, lung, ovary, pancreas, endometrium, bronchus, cervix, stomach, and thyroid. Also, some astrocytomas and soft tissue sarcomas are

reported to express ErbB-4. Expression of all known ErbB4 isoforms (Jm-a, Jm-b, CYT-1, and CYT-2) have been tested in ovarian tumor specimens [90]. The Jm-a, CYT-1, and CYT-2 isoforms are expressed in nearly all samples, while the Jm-b isoform is not detected at significant levels. Perhaps the most interesting and important tumors for ErbB-4 expression are found in pediatric brain tumors, i.e., medulloblastoma [91–93] and ependymoma [94]. In medulloblastomas, ErbB-4 is expressed in approximately 66% of the samples and is coexpressed with ErbB-2 in about 50% of the cases. These tumors also frequently express neuregulin. Patients with coexpression of ErbB-4 and ErbB-2 had the poorest prognosis compared to patients that expressed predominantly ErbB-4 or ErbB-2. Coexpression of ErbB-4, ErbB-2, and neuregulin had the greatest propensity to describe metastasis of the tumor. Also, Jm-a and Jm-b isoforms of ErbB-4 were detected in 53% and 28% of the tumor specimens. In ependymoma, ErbB-4 expression is quite high, about 75% of the cases [94]. ErbB-2 expression is approximately 30%, while the expression of ErbB-1 or ErbB-3 is relatively low. Coexpression of ErbB-2 and ErbB-4 occurs in about 75% or more of these tumors and is correlated with a high proliferative index. Survival analysis indicates that patients who coexpress ErbB-4 and ErbB-2 had the poorest survival outcome. As with medulloblastomas, the Jm-a ErbB-4 isoform was frequently present while the Jm-b isoform was rarely detectable.

Heart development The initial cloning studies of ErbB-4 also made clear that expression of this receptor was very high in the heart, skeletal muscle, and brain. Targeted disruption of the ErbB-4 gene in mice produces, in nullizygous animals, an embryonic lethal phenotype at approximately E10.5 [95]. In the embryonic heart, ErbB-4 is highly expressed in both the atrial and ventricular myocardium (muscle tissue), but is not detectable in the endocardium or epithelial lining. In the heart of ErbB-4 -/- embryos, muscle differentiation into trabeculae fails and this severely reduces blood flow. These embryos also demonstrate impaired development in the central nervous system, which is discussed in the following section. Animals heterozygous for ErbB-4 are apparently normal. The cardiac phenotype of ErbB-4-/-embryos is shared with that exhibited by embryos with targeted disruption of the ErbB-2 [96] or neuregulin-1 [97] genes. This indicates that during embryonic heart trabeculation, neuregulin must activate the ErbB-4/ErbB-2 heterodimer and that ErbB-4 homodimers are insufficient for this development sequelae. Cardiac myocytes isolated from neonatal or adult animals express ErbB-4 and ErbB-2, but ErbB-3 is not detectable [98]. Hence, neuregulin biologic effects on these cells can be attributed to ErbB-4-dependent signaling. The stimula-


tory influence of neuregulin on the proliferation, hypertrophy, differentiation, and survival of these cells has been described [97,98,150].

75

didate gene in humans [121]. This study also examined the behavioral phenotypes of mice heterozygous for neuregulin 1 or ErbB-4. The behavioral results are reported to overlap with that of mouse models for schizophrenia.

Nervous system Miscellaneous systems ErbB-4 expression is widespread in various parts of the brain and nervous system [95,99–103], including the olfactory bulb [103–105] and retina [106]. This is reflected in the phenotype of knockout mice [95]. While the ErbB-4 -/mice die at embryonic day 10.5 due to abnormal development of the heart, axonal guidance is also impaired. In contrast to the cardiac defect, which is shared by mice nullizygous for ErbB-2 or neuregulin, misinervation of axons in the hindbrain of ErbB-4 mutant mice is not observed in ErbB-2 or neuregulin mutants [96,97]. Other studies of ErbB-4 -/- embryos [107] or the expression of a dominant-negative ErbB-4 construct in primary cultures of neuronal cells [108] show additional ErbB-4 requirements for cell migration in the developing brain. The phenotypes observed in these studies include defects in pathfinding by cranial neural crest cells [107] and in the migration of cerebellar granule cells along radial glial fibers [108]. Cerebellar granule cells express significant levels of ErbB-4 and ErbB-2, but do not express ErbB-3 at detectable levels [109]. Hence, neuregulin influences on these cells are attributable to ErbB-4-dependent signaling. In granule cells neuregulin is reported to alter the subunit composition of the GABA and NMDA neurotransmitter receptors, which may alter the conductance properties of these receptors at synapses [109,110]. Also, ErbB-4 mediates increased expression of nitric oxide synthetase in these cells [111]. Several authors have noted expression of the ErbB-4 receptor at neuronal and neuromuscular synapses [50–52, 112,113], where it is postulated to have a role in gene expression mediated by these junctions. However, the presence of ErbB-3 at these sites complicates this interpretation in terms of the exact role of ErbB-4. Several groups have studied the influence of neuregulin on the proliferation, survival, and 2differentiation of cells in the oligodendrite lineage. The results are generally consistent in that neuregulin promotes proliferation and survival of oligodendrite precursors and decreases their differentiation into mature oligodendrites. In fact, neuregulin tends to enhance differentiation into other lineages. However, the data are difficult to interpret in terms of ErbB-4 as in some instances only ErbB-2 and ErbB-4 are expressed in the progenitors [114,115], while in other instances ErbB-3 is also present [116–118]. This could reflect the differing experimental conditions and reagents used. Other studies have reported that ErbB-4 may be involved in neuronal plasticity [119] and that ErbB-4 expression is increased in neurons and microglia/macrophages at the site of closed head injury [120]. Last, a recent study of susceptibility to schizophrenia has identified neuregulin 1 as a can-

Understanding of the role of ErbB-4 in other tissues is only beginning, but given its widespread distribution in embryonic and adult tissues, functions in additional systems will not be surprising. These investigations include the role of ErbB-4 in hypothalamus and reproductive behavior [122], palatogenesis and its disorders [123], tooth development [124], chondrocyte biology [125], and pancreatic islet development [126,127]. Finally, two reports identify ErbB-4 on the outer surface of the blastocyst where it may have a role in implantation [151,152]. Concluding remarks The author regrets that space limitation requires the omission of many studies indirectly related to the focus of this review and apologizes for the failure to cite any publication directly related to ErbB-4 function and biology. The interested reader is referred to other reviews for additional information and references. Many of these can be found in this issue of this journal. Other review articles and their focus are as follows: ErbB-4 [128], ErbB receptors [129–132], ErbB-4 isoforms [5], ErbB-4 in neuronal [133–136], or cardiac development [136,137], dimerization of ErbB receptors [138,139], neuregulin apoptosis [140], and ErbB receptors in mammary physiology and cancer [141–144]. Acknowledgments The author appreciates the efforts of Sue Carpenter in manuscript preparation and Lori Bennett in preparation of figures. Support of National Cancer Institute grant CA97456 is appreciated. References [1] G.D. Plowman, J.-M. Culouscou, G.S. Whitney, J.M. Green, G.W. Carlton, F. Foy, M.G. Neubauer, M. Shoyab, Ligand-specific activation of HER4/p180erbB4, a fourth member of the epidermal growth factor receptor family, Proc. Natl. Acad. Sci. USA 90 (1992) 1746–1750. [2] D.B. Zimonjic, M. Alimandi, T. Miki, N.C. Popescu, M.H. Kraus, Localization of the human HER4/erbB-4 gene to chromosome 2, Oncogene 10 (1995) 1235–1237. [3] K. Elenius, G. Corfas, S. Paul, C.J. Choi, C. Rio, G.D. Plowman, M. Klagsbrun, A novel juxtamembrane domain isoform of HER4/ ErbB4. Isoform-specific tissue distribution and differential processing in response to phorbol ester, J. Biol. Chem. 272 (1997) 26761– 26768.

76


[4] K. Elenius, C.J. Choi, S. Paul, E. Santiestevan, E. Nishi, M. Klagsbrun, Characterization of a naturally occurring ErbB-4 isoform that does not bind or activate phosphatidyl inositol 3-kinase, Oncogene 18 (1999) 2607–2615. [5] T.T. Junttila, M. Sundvall, J.A. Määttä, K. Elenius, ErbB4 and its isoforms. Selective regulation of growth factor responses by naturally occurring receptor variants, Trends Cardiovasc. Med. 10 (2000) 304–308. [6] J.-M. Culouscou, G.D. Plowman, G.W. Carlton, J.M. Green, M. Shoyab, Characterization of a breast cancer cell differentiation factor that specifically activates the HER4/p180erbB4 receptor, J. Biol. Chem. 268 (1993) 18407–18410. [7] G.D. Plowman, J.M. Green, J.-M. Gulouscou, G.W. Carlton, V.M. Rothwell, S. Buckley, Heregulin induces tyrosine phosphorylation of HER4/p180erbB4, Nature 366 (1993) 473–475. [8] E. Tzahar, G. Levkowitz, D. Karunagaran, L. Yi, E. Peles, S. Lavi, D. Chang, N. Liu, A. Yayon, D. Wen, Y. Yarden, ErbB-3 and ErbB4 function as the respective low and high affinity receptors of all neu differentiation factor/heregulin isoforms, J. Biol. Chem. 269 (1994) 25226–25233. [9] H. Chang, D.J. Riese II, W. Gilbert, D.F. Stern, U.J. McMahan, Ligands for ErbB-family receptors encoded by a neuregulin-like gene, Nature 387 (1997) 509–512. [10] K.L. Carraway III, J.L. Weber, M.J. Unger, J. Ledesma, N. Yu, M. Gassmann, C. Lai, Neuregulin-2, a new ligand of ErbB3/ErbB4receptor tyrosine kinases, Nature 387 (1997) 512–516. [11] D. Zhang, M.X. Sliwkowski, M. Mark, G. Frantz, R. Akita, Y. Sun, K. Hillan, C. Crowley, J. Brush, P.J. Godowski, Neuregulin-3 (NRG3): a novel neural tissue-enriched protein that binds and activates ErbB4, Proc. Natl. Acad. Sci. USA 94 (1997) 9562–9667. [12] D. Harari, E. Tzahar, J. Romano, M. Shelly, J.H. Pierce, G.C. Andrews, Y. Yarden, Neuregulin-4: a novel growth factor that acts through the ErbB-4 receptor tyrosine kinase, Oncogene 18 (1999) 2681–2689. [13] S.S. Hobbs, S.L. Coffing, A.T.D. Le, E.M. Cameron, E.E. Williams, M. Andrew, E.N. Blommel, R.P. Hammer, H. Chang, D.J. Riese II, Neuregulin isoforms exhibit distinct patterns of ErbB family receptor activation. Oncogene 21 (2002) 8442–8452. [14] D.J. Riese II, Y. Bermingham, T.M. van Raaij, S. Buckley, G.D. Plowman, D.F. Stern, Betacellulin activates the epidermal growth factor receptor and erbB-4, and induces cellular response patterns distinct from those stimulated by epidermal growth factor or neuregulin-b, Oncogene 12 (1996) 345–353. [15] K. Elenius, S. Paul, G. Allison, J. Sun, M. Klagsbrun, Activation of HER4 by heparin-binding EGF-like growth factor stimulates chemotaxis but not proliferation, EMBO J. 16 (1997) 1268–1278. [16] T. Komurasaki, H. Toyoda, D. Uchida, S. Morimoto, Epiregulin binds to epidermal growth factor receptor and ErbB-4 and induces tyrosine phosphorylation of epidermal growth factor receptor, ErbB2, ErbB-3, and ErbB-4, Oncogene 15 (1997) 2841–2848. [17] M. Shelly, R. Pinkas-Kramarski, B.C. Guarino, H. Waterman, L.-M. Wang, L. Lyass, M. Alimandi, A. Kuo, S.S. Bacus, J.H. Pierce, G.C. Andrews, Y. Yarden, Epiregulin is a potent pan-ErbB ligand that preferentially activates heterodimeric receptor complexes, J. Biol. Chem. 273 (1998) 10496–10505. [18] D.J. Riese II, T. Komurasaki, B.D. Plowman, D.F. Stern, Activation of ErbB4 by the bifunctional epidermal growth factor family hormone epiregulin is regulated by ErbB2, J. Biol. Chem. 273 (1998) 11288–11294. [19] S.J. Busfield, D.A. Michnick, T.W. Chickering, T.L. Revett, J. Ma, E.A. Woolf, C.A. Comrack, B.J. Dussault, J. Woolf, A.J. Goodearl, D.P. Gearing, Characterization of a neuregulin-related gene, Don-1, that is highly expressed in restricted regions of the cerebellum and hippocampus, Mol. Cell. Biol. 17 (1997) 4007–4014. [20] T. Uchida, K. Wada, T. Akamatsu, M. Yonexawa, H. Noguchi, A. Mizoguchi, M. Kasuga, C. Sakamoto, A novel epidermal growth factor-like molecule containing two follistatin modules stimulates

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

tyrosine phosphorylation of erbB-4 in MKN28 gastric cancer cells, Biochem. Biophys. Res. Commun. 266 (1999) 593–602. S. Higashiyama, M. Horikawa, K. Yamada, N. Ichino, N. Nakano, T. Nakagawa, J. Miyagawa, N. Matsushita, T. Nagatsu, N. Taniguchi, H. Ishiguro, A novel brain-derived member of the epidermal growth factor family that interacts with ErbB3 and ErbB4, J. Biochem. 122 (1997) 675–680. M. Santra, I. Eichstetter, R.V. Iozzo, An anti-oncogenic role for decorin. Down-regulation of ErbB2 leads to growth suppression and cytodifferentiations of mammary carcinoma cells, J. Biol. Chem. 275 (2000) 35153–35161. C. Sweeney, C. Lai, D.J. Riese II, A.J. Diamonti, L.C. Cantley, K.L. Carraway III, Ligand discrimination in signaling through an ErbB4 receptor homodimer, J. Biol. Chem. 275 (2000) 19803–19807. D.J. Riese II, T.M. van Raaij, G.D. Plowman, G.C. Andrews, D.F. Stern, The cellular response to neuregulins is governed by complex interactions of the erbB receptor family, Mol. Cell. Biol. 15 (1995) 5770–5776. R.R. Beerli, D. Graus-Porta, K. Wood-Cook, X. Chen, Y. Yarden, N.E. Hynes, Neu differentiation factor activation of ErbB-3 and ErbB-4 is cell specific and displays a differential requirement for ErbB-2, Mol. Cell. Biol. 15 (1995) 6496–6505. K. Zhang, J. Su, N. Liu, D. Wen, D. Chang, A. Thomason, S.K. Yoshinaga, Transformation of NIH 3T3 cells by HER3 or HER4 receptors requires the presence of HER1 or HER2, J. Biol. Chem. 271 (1996) 3884–3890. L.-M. Wang, A. Kuo, M. Alimandi, M.C. Veri, C.-C. Lee, V. Kapoor, N. Ellmore, X.-H. Chen, J.H. Pierce, ErbB2 expression increases the spectrum and potency of ligand-mediated signal transduction through ErbB4, Proc. Natl. Acad. Sci. USA 95 (1998) 6809–6814. R. Pinkas-Kramarski, M. Shelly, B.C. Guarino, L.M. Wang, L. Lyass, I. Alroy, M. Alamandi, A. Kuo, J.D. Moyer, S. Lavi, M. Eisenstein, B.J. Ratzkin, R. Seger, S.S. Bacus, J.H. Pierce, G.C. Andrews, Y. Yarden, ErbB tyrosine kinases and the two neuregulin families constitute a ligand-receptor network, Mol. Cell. Biol. 18 (1998) 6090–6101. V.D. Fitzpatrick, P.I. Pisacane, R.L. Vandlen, M.X. Sliwkowski, Formation of a high affinity heregulin binding site using the soluble extracellular domains of ErbB2 with ErbB3 or ErbB4, FEBS Lett. 431 (1998) 102–106. J.T. Jones, R.W. Akita, M.X. Sliwkowski, Binding specificities and affinities of egf domains for ErbB receptors, FEBS Lett. 447 (1999) 227–231. K.M. Ferguson, P.J. Darling, M.J. Mohan, T.L. Macatee, M.A. Lemmon, Extracellular domains drive homo- but not heterodimerization of erbB receptors, EMBO J. 19 (2000) 4632–4643. F.U. Weiß, C. Wallasch, M. Campiglio, W. Issing, A. Ullrich, Distinct characteristics of heregulin signals mediated by HER3 and HER4, J. Cell. Physiol. 173 (1997) 187–195. Y.J. Ma, D.F. Hill, K.E. Creswick, M.E. Costa, A. Cornea, M.N. Lioubin, G.D. Plowman, S.R. Ojeda, Neuregulins signaling via a glial erbB-2-erbB-4 receptor complex contribute to the neuroendocrine control of mammalian sexual development, J. Neurosci. 19 (1999) 9913–9927. K. Feroz, E. Williams, D.J. Riese II, ErbB2 and ErbB3 do not quantitatively modulate ligand-induced ErbB4 tyrosine kinase phosphorylation, Cell. Signal. 14 (2002) 793–798. J.M. Mendrola, M.B. Berger, M.C. King, M.A. Lemmon, The single transmembrane domains of ErbB receptors sel-associate in cell membranes, J. Biol. Chem. 277 (2002) 4704–4712. J.-M. Culouscou, G.W. Carlton, A. Aruffo, HER4 receptor activation and phosphorylation of Shc proteins by recombinant heregulin-Fc fusion proteins, J. Biol. Chem. 270 (1995) 12857–12863. M.A. Olayioye, D. Graus-Porta, R.R. Beerli, J. Rohrer, B. Gay, N.E. Hynes, ErbB-1 and ErbB-2 acquire distinct signaling properties dependent upon their dimerization partner, Mol. Cell. Biol. 18 (1998) 5042–5051.

ErbB-4: Mechanism of Action and Biology [38] M.A. Olayioye, I. Beuvink, K. Horsh, J.M. Daly, N.E. Hynes, ErbB receptor-induced activation of Stat transcription factors is mediated by Src tyrosine kinases, J. Biol. Chem. 274 (1999) 17209–1718. [39] L. Puricelli, C.J. Proiettii, L. Labriola, M. Salatino, M.E. Balaña, J.A. Ghiso, R. Lupu, O.P. Pignataro, E.H. Charreau, E.B. de Kier Joffé, P.V. Elizalde, Heregulin inhibits proliferation via ERKs and phosphatidyl-inositol 3-kinase activation by regulates urokinase plaminogen activator independently f these pathways in metastatic mammary tumor cells, Int. J. Cancer 100 (2002) 642–653. [40] F.E. Jones, T. Welte, X.-Y. Fu, D.F. Stern, ErbB4 signaling in the mammary gland is required for lobuloalveolar development and Stat5 activation during lactation, J. Cell Biol. 147 (1999) 77–87. [41] J. Baulida, M.H. Kraus, M. Alimandi, P.P. Di Fiore, G. Carpenter, All erbB receptors other that the epidermal growth factor receptor are endocytosis impaired, J. Biol. Chem. 271 (1996) 5251–5257. [42] J. Baulida, G. Carpenter, Heregulin degradation in the absence of rapid receptor-mediated internalization, Exp. Cell Res. 232 (1997) 167–172. [43] M. Vecchi, G. Carpenter, Constitutive proteolysis of the ErbB-4 receptor tyrosine kinase by a unique, sequential mechanism, J. Cell Biol. 139 (1997) 995–1003. [44] M. Vecchi, J. Baulida, G. Carpenter, Selective cleavage of the heregulin receptor ErbB-4 by protein kinase C activation, J. Biol. Chem. 271 (1996) 18989–18995. [45] W. Zhou, G. Carpenter, Heregulin-dependent trafficking and cleavage or ErbB-4, J. Biol. Chem. 275 (2000) 34737–34743. [46] C. Rio, J.D. Buxbaum, J.J. Peschon, G. Corfas, Tumor necrosis factor-a-converting enzyme is required for cleavage of erbB4/HER4, J. Biol. Chem. 275 (2000) 10379–10387. [47] C.-Y. Ni, M.P. Murphy, T.E. Golde, G. Carpenter, g-Secretase cleavage and nuclear localization of ErbB-4 receptor tyrosine kinase, Science 294 (2001) 2179–2181. [48] H.-J. Lee, K.-M. Jung, Y.Z. Huang, L.B. Bennett, J.S. Lee, L. Mei, T.-W. Kim, Presenilin-dependent g-secretase-like intramembrane cleavage of ErbB4. J. Biol. Chem. 277 (2002) 6318–6323. [49] C.-Y. Ni, H. Yuan, G. Carpenter, Role of the ErbB-4 carboxyterminus in g-secretase cleavage. J. Biol. Chem. 278 (2003) 4561–4565. [50] R.A.G. Garcia, K. Vasudevan, A. Buonanno, The neuregulin receptor ErbB-4 interacts with PDZ-containing proteins at neuronal synapses, Proc. Natl. Acad. Sci. USA 97 (2000) 2596–2601. [51] Y.Z. Huang, S. Won, D.W. Ali, Q. Wang, M. Tanowitz, Q.S. Du, K.A. Pelkey, D.J. Yang, W.C. Xiong, M.W. Salter, L. Mei, Regulation of neuregulin signaling by PSD-95 interacting with ErbB4 at CNS synapses, Neuron 26 (2000) 443–455. [52] Y.Z. Huang, Q. Wang, S. Won, Z.G. Luo, W.C. Xiong, L. Mei, Compartmentalized NRG signaling and PDZ domain-containing proteins in synapse structure and function, Int. J. Dev. Neurosci. 20 (2002) 173–185. [53] R. Srinivasan, E. Benton, F. McCormick, H. Thomas, W.J. Gullick, Expression of the c-erbB-3/HER-3 and c-erbB-4/HER-4 growth factor receptors and their ligands, neuregulin-1a, neuregulin-1b, and betacellulin, in normal endometrium and endometrial cancer, Clin. Cancer Res. 5 (1999) 2877–2883. [54] R. Srinivasan, C.E. Gillett, D.M. Barnes, W.J. Gullick, Nuclear expression of the c-erbB-4/HER-4 growth factor receptor in invasive breast cancers, Cancer Res. 60 (2000) 1483–1487. [55] M. Zhang, D. Ding, R. Salvi, Expression of heregulin and ErbB/Her receptors in adult chinchilla cochlear and vestibular sensory epithelium, Hearing Res. 169 (2002) 56–68. [56] Y.-Y. Zhao, O. Feron, C. Dessy, X. Han, M.A. Marchionni, R.A. Kelly, Neuregulin signaling in the heart. Dynamic targeting of erbB4 to caveolar microdomains in cardiac myocytes, Circ. Res. 84 (1999) 1380–1387. [57] Zhou, W., Carpenter, G. Heregulin-dependent translocation and hyperphosphorylation of ErbB-2, Oncogene 20 (2001) 3918–3920. [58] J.A. Schroeder, M.C. Thompson, M.M. Gardner, S.J. Gendler, Transgenic MUC1 interacts with epidermal growth factor receptor and

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

[73]

[74]

[75]

[76]

77

correlates with mitogen-activated protein kinase activation in the mouse mammary gland, J. Biol. Chem. 276 (2001) 13057–13064. W.-H. Yu, J.F. Woessner Jr, J.D. McNeish, I. Stamenkovic, CD44 anchors the assembly of matrilysin/MMP-7 with heparin-binding epidermal growth factor precursor and ErbB4 and regulates female reproductive organ remodeling, Genes Dev. 16 (2002) 307–323. A.J. Diamonti, P.M. Guy, C. Ivanof, K. Wong, C. Sweeney, K.L. Carraway III, An RBCC protein implicated in maintenance of steady-state neuregulin receptor levels, Proc. Natl. Acad. Sci. USA 99 (2002) 2866–2871. X.-B. Qiu, A.L. Goldberg, Nrdp1/FLRF is a ubiquitin ligase promoting ubiquitination and degradation of the epidermal growth factor receptor family member, ErbB3, Proc. Natl. Acad. Sci. USA 99 (2002) 14843–14848. F. Xu, Y. Yu, X.-F. Le, C. Boyer, G.B. Mills, R.C. Bast Jr, The outcome of heregulin-induced activation of ovarian cancer cells depends on the relative levels of HER-2 and HER-3 expression, Clin. Cancer Res. 5 (1999) 3653–3660. V. Kainulainen, M. Sundvall, J.A. Mäatä, E. Santiestevan, M. Klagsbrun, K. Elenius, A natural ErbB4 isoform that does not activate phosphoinositide 3-kinase mediates proliferation but not survival or chemotaxis, J. Biol. Chem. 275 (2000) 8641–8649. B.D. Cohen, J.M. Green, L. Foy, H.P. Fell, HER4-mediated biological and biochemical properties in NIH 3T3 cells. Evidence for HER1-HER4 heterodimers, J. Biol. Chem. 271 (1996) 4813– 4818. B.D. Cohen, P.A. Kiener, J.M. Green, L. Foy, H.P. Fell, K. Zhang, The relationship between human epidermal growth-like factor receptor expression and cellular transformation in NIH3T3 cells, J. Biol. Chem. 271 (1996) 30897–30903. D.J. Penington, I. Bryant, D.J. Riese II, Constitutively active ErbB4 and ErbB2 mutants exhibit distinct biological activities, Cell Growth Diff. 13 (2002) 247–256. C.K. Tang, J.P. Goldstein, F. Czubayko, M. Alimandi, L.-M. Wang, J.H. Pierce, M.E. Lippman, ErbB-4 ribozymes abolish neuregulininduced mitogenesis, Cancer Res. 58 (1998) 3415–3422. A. Valovsky, Z. Lupowitz, S. Erlich, R. Pinkas-Kramarski, ErbB-4 activation promotes neurite outgrowth in PC12 cells, J. Neurochem. 74 (2000) 979–987. S. Erlich, Y. Goldshmit, Z. Lupowitz, R. Pinkas-Kramarski, ErbB-4 activation inhibits apoptosis in PC12 cells, Neuroscience 107 (2001) 353–362. Y. Goldshmit, S. Erlich, R. Pinkas-Kramarski, Neuregulin rescues PC12-ErbB4 cells from cell death induced by H2O2, J. Biol. Chem. 276 (2001) 46379–46385. C.I. Sartor, H. Zhou, E. Kozlowska, K. Guttridge, E. Kawata, L. Caskey, J. Harrelson, N.S. Hynes, S. Ethier, B. Calvo, H.S. Earp III, HER4 mediates ligand-dependent antiproliferative and differentiation responses in human breast cancer cells, Mol. Cell. Biol. 21 (2001) 4265–4275. R. Srinivasan, R. Poulsom, H.C. Hurst, W.J. Gullick, Expression of the c-erbB-4/HER4 protein and mRNA in normal human fetal and adult tissues and in a survey of nine solid tumour types, J. Pathol. 185 (1998) 236–245. J.A. Schroeder, D.C. Lee, Dynamic expression and activation of ErbB receptors in the developing mouse mammary gland, Cell Growth Diff. 9 (1998) 451–464. K.M. Darcy, D. Zangani, A.L. Wohlhueter, R.-Y. Huang, M.M. Vaughan, J.A. Russell, M.M. Ip, Changes in ErbB2 (her-2/neu), ErbB3, and ErbB4 during growth, differentiation, and apoptosis of normal rat mammary epithelial cells, J. Histochem. Cytochem. 48 (2000) 63–80. C. Sawyer, I. Hiles, M. Page, M. Crompton, C. Dean, Two erbB-4 transcripts are expressed in normal breast and in most breast cancer, Oncogene 17 (1998) 919–924. S.S. Bacus, C.R. Zelnick, D.M. Chin, Y. Yarden, D.B. Kaminsky, J. Bennington, D. Wen, J.N. Marcus, D.L. Page, Medullary carcinoma

78

[77]

[78]

[79]

[80]

[81]

[82]

[83]

[84]

[85]

[86]

[87]

[88]

[89]

[90]

[91]

[92]

[93]

The EGF Receptor Family is associated with expression of intercellular adhesion molecule-1, Am. J. Pathol. 145 (1994) 1337–1348. J.M. Knowlden, J.M.W. Gee, L.T. Seery, L. Farrow, W.J. Gullick, I.O. Ellis, R.W. Blamey, J.F.R. Robertson, R.I. Nicholson, C-erbB3 and c-erbB4 expression is a feature of the endocrine responsive phenotype in clinical breast cancer, Oncogene 17 (1998) 1949–1957. U. Vogt, K. Bielawski, C.M. Schlotter, U. Bosse, B. Falkiewicz, A.J. Podhajska, Amplification of erbB-4 oncogene occurs less frequently than that of erbB-2 in primary human breast cancer, Gene 223 (1998) 375–380. R. Bei, L. Masuelli, E. Moriconi, V. Visco, A. Moretti, M.H. Kraus, R. Muraro, Immune responses to all ErbB family receptors detectable in serum of cancer patients, Oncogene 18 (1999) 1267–1275. S.S. Bacus, Y. Yarden, M. Oren, D.M. Chin, L. Lyass, C.R. Zelnick, A. Kazarov, W. Toyofuku, J. Gray-Bablin, R.R. Beerli, N.E. Hynes, M. Nikiforov, R. Haffner, A. Gudkov, K. Keyomarsi, Neu differentiation factor (Heregulin) activates a p53-dependent pathway in cancer cells, Oncogene 12 (1996) 2535–2547. Z. Suo, B. Risberg, M.G. Karlsson, K. Willman, A. Tierens, E. Skovlund, J.M. Nesland, EGFR family expression in breast carcinomas. c-erbB-2 an c-erbB-4 receptors have different effects on survival, J. Pathol. 196 (2002) 17–25. Z. Suo, H.S. Berner, B. Risberg, M.G. Karlsson, J.M. Nesland, Estrogen receptor-a and C-ERBB-4 expression in breast carcinomas, Virchows Arch. 439 (2001) 62–69. S.S. Ba·cus, D. Chin, Y. Yarden, C.R. Zelnick, D.F. Stern, Type 1 receptor tyrosine kinases are differentially phosphorylated in mammary carcinoma and differentially associated with steroid receptors, Am J. Pathol. 148 (1996) 549–558. O. Merimsky, J. Issakov, J. Bickels, Y. Kollender, G. Flusser, V. Soyfer, I. Schwartz, M. Inbar, I. Meller, ErbB-4 expression in limb soft-tissue sarcoma: correlation with the results of neoadjuvant chemotherapy, Eur. J. Cancer 38 (2002) 1335–1342. J.C. Lyne, M.F. Melhem, G.G. Finley, D. Wen, N. Liu, D.H. Deng, R. Salup, Tissue expression of neu differentiation factor/heregulin and its receptor complex in prostate cancer and its biologic effects on prostate cancer cells in vitro, Cancer J. Sci. Am. 3 (1997) 21–30. D.R.F. Haugen, L.A. Akslen, J.E. Varhaug, J.R. Lillehaug, Expression of c-erbB-3 and c-erbB-4 proteins in papillary thyroid carcinomas, Cancer Res. 56 (1996) 1184–1188. O. Merimsky, A. Staroselsky, M. Inbar, Y. Schwartz, N. Wigler, A. Mann, S. Marmor, J. Greif, Correlation between c-erbB-4 receptor expression and response to gemcitabine-cisplatin chemotherapy in non-small-cell lung cancer, Ann. Oncol. 12 (2001) 1127–1131. H.U. Graber, H. Friess, B. Kaufmann, D. Willi, A. Zimmermann, M. Korc, W. Büchler, ErbB-4 mRNA expression is decreased in nonmetastatic pancreatic cancer, Int. J. Cancer (Pred. Oncol.) 84 (1999) 24–27. C. Furger, R.J. Fiddes, D.I. Quinn, R.J. Bova, R.J. Daly, R.L. Sutherland, Granulosa cell tumors express erbB4 and are sensitive to the cytotoxic action of heregulin-b2/PE40, Cancer Res. 58 (1998) 1773–1778. L.M.R. Gilmour, K.G. Macleod, A. McCaig, W.J. Gullick, J.F. Smyth, S.P. Langdon, Expression of erbB-4/HER-4 growth factor receptor isoforms in ovarian cancer, Cancer Res. 61 (2001) 2169–2176. R.J. Gilbertson, R.H. Perry, P.J. Kelly, A.D.J. Pearson, J. Lunec, Prognostic significance of HER2 and HER4 coexpression in childhood medulloblastoma, Cancer Res. 57 (1997) 3272–3280. R.J. Gilbertson, S.C. Clifford, W. MacMeekin, C. Wright, R.H. Perry, P. Kelly, A.D.J. Pearson, J. Lunec, Expression of the ErbBneuregulin signaling network during human cerebellar development: implications for the biology of medulloblastoma, Cancer Res. 58 (1998) 3932–3941. R. Gilbertson, R. Hernan, T. Pietsch, L. Pinto, P. Scotting, R. Allibone, D. Ellison, R. Perry, A. Pearson, H. Lunec, Novel ERBB4

[94]

[95]

[96]

[97] [98]

[99]

[100]

[101]

[102]

[103]

[104]

[105]

[106]

[107]

[108]

[109]

[110]

[111]

juxtamembrane splice variants are frequently expressed in childhood medulloblastoma, Genes Chrom. Cancer 31 (2001) 288–294. R.J. Gilbertson, L. Bentley, R. Hernan, T.T. Junttila, A.J. Frank, H. Haapasalo, M. Connelly, C. Wetmore, T. Curran, K. Elenius, D.W. Ellison, ERBB receptor signaling promotes ependymoma cell proliferation and represents a potential novel therapeutic target for this disease, Clin. Cancer Res. 8 (2002) 3054–3064. M. Gassmann, F. Casagranda, D. Orioli, H. Simon, C. Lai, R. Klein, G. Lemke, Aberrant neural and cardiac development in mice lacking the ErbB4 neuregulin receptor, Nature 378 (1995) 390–394. K.-F. Lee, H. Simon, H. Chen, B. Bates, M.-C. Hung, C. Hauser, Requirement for neuregulin receptor erbB2 in neural and cardiac development, Nature 378 (1995) 394–398. D. Meyer, C. Birchmeier, Multiple essential functions of neuregulin in development, Nature 378 (1995) 386–390. Y.-Y. Zhao, D.R. Sawyer, R.R. Baliga, D.J. Opel, X. Han, M.A. Marchionni, R.A. Kelly, Neuregulins promote survival and growth of cardiac myocytes. Persistence of ErbB2 and ErbB4 expression in neonatal and adult ventricular myocytes, J. Biol. Chem. 273 (1998) 10261–10269. K.M. Gerecke, J.M. Wyss, I. Karavanova, A. Buonanno, S.L. Carroll, ErbB transmembrane tyrosine kinase receptors are differentially expressed throughout the adult rat central nervous system, J. Comp. Neurol. 433 (2001) 86–100. H. Steiner, M. Blum, S.T. Kitai, P. Fedi, Differential expression of ErbB3 and ErbB4 neuregulin receptors in dopamine neurons and forebrain areas of the adult rat, Exp. Neurol. 159 (1999) 494– 503. D. Meyer, T. Yamaai, A. Garratt, E. Reithmacher-Sonnenberg, D. Kane, L.E. Theill, C. Birchmeier, Isoform-specific expression and function of neuregulin, Development 124 (1997) 2575–2586. R. Pinkas-Kramarski, R. Eilam, I. Alroy, G. Levkowitz, P. Lonai, Y. Yarden, Differential expression of NDF/neuregulin receptors ErbB3 and ErbB-4 and involvement in inhibition of neuronal differentiation, Oncogene 15 (1997) 2803–2815. T. Lindholm, S. Cullheim, M. Deckner, T. Carlstedt, M. Risling, Expression of neuregulin and ErbB3 and ErbB4 after a traumatic lesion in the ventral funiculus of the spinal cord and in the intact primary olfactory system, Exp. Brain Res. 142 (2002) 81–90. I. Perroteau, M. Oberto, A. Ieraci, P. Bovolin, A. Fasolo, ErbB-3 and ErbB-4 expression in the mouse olfactory system, Ann. NY Acad. Sci. 855 (1998) 255–259. M. Oberto, I. Soncin, P. Bovolin, S. Voyron, M. De Bortoli, C. Dati, A. Fasolo, I. Perroteau, ErbB-4 and neuregulin expression in the adult mouse olfactory bulb after peripheral denervation, Eur. J. Neurosci. 14 (2001) 513–521. O. Bermingham-McDonogh, K.L. McCabe, T.A. Reh, Effects of GGF/neuregulins on neuronal survival and neurite outgrowth correlate with erbB2/neu expression in developing rat retina, Development 122 (1996) 1427–1438. J.P. Golding, P. Trainor, R. Krumlauf, M. Gassmann, Defects in pathfinding by cranial neural crest cells in mice lacking the neuregulin receptor ErbB4, Nat. Cell Biol. 2 (2000) 103–109. C. Rio, H.I. Rieff, P. Qi, G. Corfas, Neuregulin and erbB receptors play a critical role in neuronal migration, Neuron 19 (1997) 39– 50. M. Ozaki, M. Sasner, R. Yano, H.S. Lu, A. Buonanno, Neuregulinb induces expression of an NMDA-receptor subunit, Nature 390 (1997) 691–694. H.I. Rieff, L.T. Raetzman, D.W. Sapp, H.H. Yeh, R.E. Siegel, G. Corfas, Neuregulin induces GABAA receptor subunit expression and neurite outgrowth in cerebellar granule cells, J. Neurosci. 19 (1999) 10757–10766. R. Krainock, S. Murphy, Heregulin upregulates the expression of nitric oxide synthase (NOS)-1 in rat cerebellar granule neurons via the ErbB4 receptor, J. Neurochem. 76 (2001) 312–315.

ErbB-4: Mechanism of Action and Biology [112] S. Zhu, C. Lai, S. Thomas, S.J. Burden, Neuregulin receptors, erbB3 and erbB4, are localized at neuromuscular synapses, EMBO J. 14 (1995) 5842–5848. [113] J.C. Trinidad, G.D. Fischbach, J.B. Cohen, The agrin/MuSK signaling pathway is spatially segregated from the neuregulin/ErbB receptor signaling pathway at the neuromuscular junction, J. Neurosci. 20 (2000) 8762–8770. [114] T. Vartanian, A. Goodearl, A. Viehöver, G. Fischbach, Axonal neuregulin signals cells of the oligoendrocyte lineage through activation of HER4 and Schwann cells through HER2 and HER3, J. Cell Biol. 137 (1997) 211–220. [115] V. Calaora, B. Rogister, K. Bismuth, K. Murray, H. Brandt, P. Leprince, M. Marchionni, M. Cubois-Calcq, Neuregulin signaling regulates neural precursor growth and the generation of oligodendrocytes in vitro, J. Neurosci. 21 (2001) 4740–4751. [116] P.D. Canoll, J.M. Musacchio, R. Hardy, R. Reynolds, M.A. Marchionni, J.L. Salzer, GGF/Neuregulin is a neuronal signal that promotes the proliferation and survival and inhibits the differentiation of oligodendrocyte progenitors, Neuron 17 (1996) 229–243. [117] G.D. Deadwyler, S. Pouly, J.P. Antel, G.H. Devries, Neuregulins and ErbB receptor expression in adult human oligodendrocytes, Glia 32 (2000) 304–312. [118] A.I. Flores, B.S. Mallon, T. Matsui, W. Ogawa, A. Rosenzweig, T. Okamoto, W.B. Macklin, Akt-mediated survival of oligodendrocytes induced by neuregulins, J. Neurosci. 20 (2000) 7622–7630. [119] R. Eilam, R. Pinkas-Kramarski, B.J. Ratzkin, M. Seal, Y. Yarden, Activity-dependent regulation of Neu differentiation factor/ neuregulin expression in rat brain, Proc. Natl. Acad. Sci. USA 95 (1998) 1888–1893. [120] S. Erlich, E. Shohami, R. Pinkas-Kramarski, Closed head injury induces up-regulation of ErbB-4 receptor at the site of injury, Mol. Cell. Neurosci. 16 (2000) 597–608. [121] H. Stefansson, E. Sigurdsson, V. Steinthorsdottir, S. Bjornsdottir, T. Sigmundsson, S. Ghosh, J. Brynjolfsson, S. Cunnarsdottir, O. Ivarsson, T.T. Chou, O. Hjaltason, B. Birgisdottir, H. Jonsson, V.G. Gudnadottir, E. Gudmundsdottir, B. Bjornsson, B. Ingvarsson, A. Inason, S. Sigfusson, H. Hardardottir, R.P. Harvey, D. Lai, M. Zhou, D. Brunner, V. Mutel, A. Gonzalo, G. Lemke, J. Sainz, G. Johannesson, T. Andresson, D. Gudbjartsson, A. Manolescu, M.L. Figge, M.E. Gurney, A. Kong, J.R. Gulcher, H. Petursson, K. Stefansson, Nueregulin 1 and susceptibility to schizophrenia, Am. J. Hum. Genet. 71 (2002) 877–892. [122] J. Hou, B. Li, Z. Yuang, N. Fager, M.Y.J. Ma, Functional integrity of ErbB-4/2 tyrosine kinase receptor complex in the hypothalamus is required for maintaining normal reproduction in young adult female rats, Endocrinology 143 (2002) 1901–1912. [123] K.Y. Wang, F.H.F. Chang, C.P. Chiang, K.C. Chen, M.Y.P. Kuo, Temporal and spatial expression of erbB4 in ectodermal and mesenchymal cells during primary palatogenesis in noncleft and cleft strains of mice, J. Oral Pathol. Med. 27 (1998) 141– 146. [124] K. Fried, M. Risling, H. Tidcombe, M. Gassman, C. Lillesaar, Expression of ErbB3, ErbB4, and neuregulin-1 mRNA during tooth development. Dev. Dynam. 224 (2002) 356–360. [125] K. Nawachi, M. Inoue, S. Kubota, T. Nishida, G. Yosimichi, T. Nakanishi, M. Kanyama, T. Kuboki, H. Yatani, T. Yamaai, M. Takigawa, Tyrosine kinase-type receptor ErbB4 in chondrocytes: interaction with connective tissue growth factor and distribution in cartilage, FEBS Lett. 528 (2002) 109–113. [126] M.R. Kritzik, T. Krahl, A. Good, D. Gu, C. Lai, H. Fox, N. Sarvetnick, Expression of ErbB receptors during pancreatic islet development and regrowth, J. Endocrinol. 165 (2000) 67–77. [127] M.-A. Huotari, P.J. Miettinen, J. Palgi, T. Koivisto, J. Ustinov, D. Harari, Y. Yarden, T. Otonkoski, ErbB signaling regulates lineage determination of developing pancreatic islet cells in embryonic organ culture, Endocrinology 143 (2002) 4437–4446.

79

[128] W. Zhou, G. Carpenter, ErbB-4: a receptor tyrosine kinase, Inflamm. Res. 51 (2002) 91–101. [129] Y. Yarden, M.X. Sliwkowski, Untangling the ErbB signalling network, Nat. Rev. 2 (2001) 127–137. [130] C. Sweeney, K.L. Carraway III, Ligand discrimination by ErbB receptors: differential signaling through differential phosphorylation site usage, Oncogene 19 (2000) 5568–5573. [131] D.J. Riese II, D.F. Stern, Specificity within the EGF family/ErbB receptor family signaling network, BioEssays 20 (1998) 41–48. [132] I. Alroy, Y. Yarden, The ErbB signaling network in embryogenesis and oncogenesis: signal diversification through combinatorial ligand-receptor interactions, FEBS Lett. 410 (1997) 83–86. [133] M. Gassmann, G. Lemke, Neuregulins and neuregulin receptors in neural development, Curr. Opin. Neurobiol. 7 (1997) 87–92. [134] S. Murphy, R. Krainock, M. Tham, Neuregulin signaling via ErbB receptor assemblies in the nervous system, Mol. Neurobiol. 25 (2002) 67–77. [135] A. Buonanno, G.C. Fischbach, Neuregulin and ErbB receptor signaling pathways in the nervous system, Curr. Opin. Neurobiol. 11 (2001) 287–296. [136] S. Burden, Y. Yarden, Neuregulins and their receptors: a versatile signaling module in organogenesis and oncogenesis, Neuron 18 (1997) 847–855. [137] M.A. Marchionni, Neu track on neuregulin, Nature 378 (1995) 334–335. [138] K.L. Carraway III, L.C. Cantley, A neu acquaintance for ErbB3 and ErbB4: a role for receptor heterodimerization in growth signaling, Cell 78 (1994) 5–8. [139] M.A. Olayioye, R.M. Neve, H.A. Lane, N.E. Hynes, The ErbB signaling network: receptor heterodimerization in development and cancer, EMBO J. 13 (2000) 3159–3167. [140] X.-F. Le, C.R. Varela, R.C. Bast Jr, Heregulin-induced apoptosis, Apoptosis 7 (2002) 483–491. [141] K.L. Carraway, C.A. Carothers Carraway, K.L. Carraway III, Roles of ErbB-3 and ErbB-4 in physiology and pathology of the mammary gland, J. Mamm. Gland Biol. Neopl. 2 (1997) 187–198. [142] D.F. Stern, Tyrosine kinase signalling in breast cancer ErbB family receptor tyrosine kinases, Breast Cancer Res. 2 (2000) 176–183. [143] H.S. Earp, T.L. Dawson, X. Li, H. Yu, Heterodimerization and functional interaction between EGF receptor family members: a new signaling paradigm with implications for breast cancer research, Breast Cancer Res. Treat. 35 (1995) 115–132. [144] Y. Yarden, The EGFR family and its ligands in human cancer: signalling mechanisms and therapeutic opportunities, Eur. J. Cancer 37 (2001) S3–S8. [145] L. Sepp-Lorenzino, I. Eberhard, Z. Ma, C. Cho, H. Serve, F. Liu, N. Rosen, R. Lupu, Signal transduction pathways induced by heregulin in MDA-MB-453 breast cancer cells, Oncogene 12 (1996) 1679–1687. [146] R.J. Fiddes, D.H. Campbell, P.W. Janes, S.P. Sivertsen, H. Sasaki, C. Wallasch, R.J. Daly, Analysis of Grb7 recruitment by heregulinactivated erbB receptors reveals a novel target selectivity for erbB3, J. Biol. Chem. 273 (1998) 7717–7724. [147] D. Karunagaran, E. Tzahar, R.R. Beerli, X. Chen, D. Graus-Porta, B.J. Ratzkin, R. Seger, N.E. Hynes, Y. Yarden, ErbB-2 is a common auxiliary subunit of NDF and EGF receptors: implications for breast cancer, EMBO J. 15 (1996) 254–264. [148] D. Graus-Porta, R.R. Beerli, J.M. Daly, N.E. Hynes, ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling, EMBO J. 16 (1997) 1647–1655. [149] V.R.J. Schelfhout, E.D. Coene, B. Belaney, S. Thys, D.L. Page, C.R. DePotter, Pathogenesis of Paget’s Disease: epidermal heregulin-a, motility factor, and the HER receptor family, J. Natl. Cancer Inst. 92 (2000) 622–628. [150] R.R. Baliga, D.R. Pimental, Y.-Y. Zhao, W.W. Simmons, M.A. Marchionni, D.B. Sawyer, R.A. Kelly, NRG-1 induced cardiomy-

80


ocyte hypertrophy. Role of PI-3-kinase, p70S6K, and MEK-MAPKRSK, Am. J. Physiol. 277 (1999) H2026–H2037. [151] X. Wang, H. Wang, H. Matsumoto, S.K. Roy, S.K. Das, B.C. Paria, Dual source and target of heparin-binding EGF-like growth factor during the onset of implantation in the hamster, Development 129 (2002) 4125–4134.

[152] K. Chobotova, I. Spyropoulou, J. Carver, S. Manek, J.K. Heath, W.J. Gullick, D.H. Barlow, I.L. Sargent, H.J. Mardon, Heparin-binding epidermal growth factor and its receptor ErbB4 mediate implantation of the human blastocyst, Mech. Develop. 119 (2002) 137– 144

Trafficking of the ErbB receptors and its influence on signaling H. Steven Wiley Biological Sciences Division, Pacific Northwest National Laboratory, Richland, WA 99353, USA

Abstract Although members of the ErbB receptor family are found predominantly at the cell surface, these receptors undergo constant cycling between the plasma membrane and the endosomal compartment. In the absence of an activating ligand, these receptors are slowly internalized (t1/2 ~ 30 min) but are quickly recycled. The constitutive degradation rate of the epidermal growth factor (EGF) receptor (EGFR) is slower than other ErbB family members and only the EGFR appears to alter its trafficking pattern in response to ligand binding. This altered pattern is characterized by accelerated internalization and enhanced lysosomal targeting. Ligand-regulated trafficking of the EGFR is mediated by a series of motifs distributed through the cytoplasmic domain of the receptor that are exposed by a combination of activation-mediated conformation changes and the binding of proteins such as Grb2. As a consequence of induced internalization, most EGFR signaling occurs within endosomes whereas signaling by the other members of the ErbB family appear to be generated predominantly from the cell surface. Overexpression of ErbB family members can disrupt normal receptor trafficking by driving heterodimerization of receptors with disparate trafficking patterns. Because different ErbB receptor substrates are localized in different cellular compartments, disrupted trafficking could be an important factor in the altered signaling patterns observed as a consequence of receptor overexpression.

Introduction Intracellular trafficking of the epidermal growth factor (EGF) receptor (EGFR) has been intensely studied for a quarter century and is generally considered the model by which the behaviors of other tyrosine kinase receptors are evaluated. Even in the earliest studies, it was observed that addition of EGF induced a rapid internalization of receptors and their eventually delivery to lysosomes [1,2]. This linear sequence of bind–internalize–degrade has been investigated at the morphological, biochemical, and molecular levels and we even have sophisticated computer models to recapitulate the process [3]. What is frequently overlooked, however, is the physiological significance of regulated EGFR trafficking. This issue has recently become more pertinent because of the observation that other members of the ErbB receptor family (ErbB2, ErbB3, and ErbB4) do not display ligand-induced internalization [4–6]. Perhaps the EGFR is the anomalous family member, or perhaps the function of the different receptors is dependent on their individual trafficking behaviors. It is becoming clear that all members of the ErbB receptor family do indeed undergo internalization and trafficking, but at rates distinct from those displayed by the EGFR. It seems likely that the different trafficking patterns of these receptors have functional significance. Understanding how trafficking impacts cell signaling could provide important insights into the evolu-

tionary pressures that produced such divergent receptor behavior.

Constitutive trafficking of ErbB receptors It is productive to look at protein trafficking in the context of the normal flow of membrane and lipids through cells. All membrane proteins display a pattern of intracellular trafficking. All receptors, including the members of the ErbB family, are synthesized and routed to specific cellular locations in the absence of any activating ligands. The distribution can be with respect to apical and basolateral disposition, as in the case of polarized epithelial cells, or a particular distribution between surface and endosomal compartments [7,8]. Receptor distribution is dynamic, just as membrane flow in cells is dynamic and is dictated by the presence or absence of specific trafficking motifs, typically present in the cytoplasmic domain of receptors [9]. Even receptors that lack any targeting sequences will display the ability to traffick to the cell surface and be internalized. The “default trafficking” of ErbB receptors is most obviously displayed by EGFR that lack their entire cytoplasmic domain, except for a 2-amino acid sequence necessary to keep the receptor anchored in the membrane. Despite the total absence of cytoplasmic sequences, these receptors are internalized at a finite, albeit slow rate, but are rapidly

82


Fig. 1. Trafficking of ErbB receptor family. Only the epidermal growth factor receptor (EGFR) and ErbB2 proteins are shown for clarity, but the behavior of ErbB3 and ErbB4 are similar to that displayed by ErbB2. Activated EGFR and EGFR:ErbB2 heterodimers are internalized through a coated pit pathway, but other members of the ErbB family are likely internalized by a smooth pit pathway. The numbers next to the arrows represent the approximate mean time of the specific process. The time constants for heterodimerization and formation of multivesicular bodies are unknown. The mean time for lysosomal degradation is a combination of the time necessary for both multivesicular body formation and lysosomal fusion.

recycled back to the cell surface [10]. These “tailless” receptors also display a very slow turnover rate, suggesting that the normal constitutive turnover rate of EGFR is dependent on specific sequence motifs [11]. Tailless receptors are particularly informative because they provide a basis on which to determine whether a particular receptor trafficking pattern shows evidence of specific, or regulated, behavior. Note that although the internalization of these receptors is slow in comparison to activate, wild-type EGFR, it is still substantial on the timescale of many experiments. The typical value of 1–2% internalized per minute translates to a mean time of internalization of approximately 30–60 min [12]. The overwhelming majority of EGFR trafficking studies have been done using probes that follow activated EGFR. Usually, radiolabeled or fluorescently labeled EGF is pro-

vided to cells and its distribution is followed over time [13,14]. This approach has skewed the interpretation of ligand-modulated trafficking behavior because empty receptors are invisible. More information can be gained from studies using EGFR fused to green fluorescent protein or bound to labeled, nonantagonistic antibodies [14,15]. These studies have revealed that although EGFR are primarily localized at the cell surface, they constantly shuttle and recycle through the cell. Constitutive distribution and trafficking of the EGFR can be remarkably cell-type specific. In fibroblasts, EGFR are predominantly at the cell surface whereas some epithelial cells display a large pool of intracellular receptors that shuttle rapidly between endosomes and the cell surface [7]. The physiological significance of differences in empty receptor distribution is unknown.

Trafficking of the ErbB Receptors and Its Influence on Signaling

Studies using radiolabeled anti-EGFR Fab antibody fragments have indicated that although empty EGFR in fibroblasts are internalized relatively slowly (t1/2 ~ 20–30 min), they are quickly recycled back to the cell surface [10] (Fig. 1). This recycling displays biphasic kinetics with a predominant fast component (t1/2 ~ 5 min) and a more prolonged phase of ~20 min. The constitutive flux of EGFR through cells is similar to rates reported for nonreceptor molecules such as the class II major histocompatibility complex (t1/2 ~ 30 min for internalization and t1/2 of 2.5 min for recycling [16]). In addition, the half-life for internalization of “internalization defective” low density lipoprotein (LDL), transferrin, and asialoglycoprotein receptors all range between 12 and 35 min [17–20]. Once internalized, these receptors all rapidly recycle back to the cell surface, again indicating that the default pathway for internalized membrane proteins is rapid recycling. If constitutive internalization and recycling rates constants are 0.02 and 0.1–0.2 min-1, respectively, this would localize 80–90% of the cycling proteins to the cell surface, consistent with the observed cellular distribution of EGFR in the absence of ligand as well as the distribution observed for other internalization defective receptors [17,19–21]. In the absence of ligand, EGFRs undergo metabolic turnover with a half-life of approximately 10–14 h in both fibroblasts and epithelial cells [7,22] and 20–48 h in transformed cells [23,24]. Thus, an average receptor will cycle through the endocytic pathway dozens of times during its life span, with a low probability of being degraded. This disparity between the relatively high constitutive flux of EGFR through the endocytic pathway and the low probability of lysosomal degradation has significant consequences. It means that a small increase in the fraction of receptors shunted to the lysosomal pathway can have a large effect on receptor degradation rates. Thus, an increase in receptor degradation rates does not necessarily require an increase in receptor internalization. Rather, it could be due to an increase in the fraction of shuttling receptors targeted to the lysosomal compartment [25]. Much less is known regarding the constitutive turnover of the other members of the ErbB family, but their behavior appears similar to that described for the EGFR. The constitutive internalization rate of ErbB2 appears to be the same as that of the EGFR [6,24], indicating that in the absence of ligands, both receptors are still internalized as a consequence of normal membrane flow through the cell. The situation is probably similar in the case of ErbB3 and ErbB4. In the absence of ligand, the distribution of both EGFR and ErbB2 in cells appears to be very similar (Fig. 2).

Ligand-regulated trafficking An interesting aspect of the ErbB family of receptors is that only the EGFR appears to undergo ligand-induced

83

internalization (Fig. 3). The rate of EGFR endocytosis is increased 5–10-fold following ligand activation, whereas the internalization rate of ErbB2, ErbB3, and ErbB4 appear to be similar in both the empty and the occupied states [26,27]. The mechanistic reason for the lack of induced internalization in the ErbB2-4 receptors appears to be the absence of specific internalization sequences in their cytoplasmic domains [28,29]. Functionally, this means that only the EGFR displays acute, ligand-induced receptor degradation. Perhaps compensating for the lack of induced degradation is a more rapid rate of constitutive turnover of the other receptor family members. For example, it has been noted that the half-life of ErbB2 in CHO cells is only 3.5 h, versus 10 h for the EGFR [30]. A similar turnover of ErbB2 has been observed in 3T3 cells [31]. Turnover rates of ErbB3 have been reported to be in the range of 2 h [27], whereas ErbB4 is around 5 h [6]. Rapid constitutive turnover of growth factor receptors is not unusual. The metabolic halflife of the empty platelet-derived growth factor (PDGF) receptor has been reported to be 80% homologous HER2 (erbB2) tyrosine kinase are variably higher (reviewed by Fry in this issue and see Arteaga [29]), supporting their overall EGFR specificity. In general, the IC50 of these compounds against the EGFR was generated by using the purified EGFR as a substrate in an in vitro kinase reaction. Upon binding to the ATP site, some of these inhibitors have been shown to induce the formation of inactive EGFR homodimers both in vitro and in intact cells [30,31]. Receptor homodimers induced by AG-1478, a quinazoline similar to ZD1839 and OSI-774 [30], can be competed with 100-fold higher concentrations of ATP, suggesting that these inhibitors bind in the EGFR kinase domain with higher affinity than ATP itself. Because of the high intracellular concentration of ATP, higher concentrations of these inhibitors are required to block EGFR phosphorylation in intact cells (in vivo) than to inhibit the purified EGFR kinase in vitro. The homology between EGFR and HER2 has been exploited for the generation of bifunctional inhibitors like CI-1033, EKB-569, and GW2016 (Table 1). Chemical modification of some of these structures has led to the generation of irreversible inhibitors that bind covalently to specific cysteines in the ATP-binding pocket of the EGFR [32], like CI-1033 and EKB-569 (reviewed by Fry in this issue). This covalent binding might provide the irreversible inhibitors with a longer in situ half-life at the receptor’s

ErbB-targeted Therapeutic Approaches in Human Cancer

ATP-binding site. Whether this property results in increased efficacy (and possible enhanced toxicity) for this approach compared to reversible EGFR kinase inhibitors requires further investigation. Because minor structural modifications of each individual inhibitor could alter their ATP sitebinding properties, at this time there seems to be a consensus that, for considerations of safety and therapeutic efficacy, these agents should not be considered interchangeable. Inhibition of the EGFR with small-molecule tyrosine kinase inhibitors can lead to cell cycle delay and apoptosis [33–35]. Antisense p27 oligonucleotides abrogate cell cycle arrest in A431 cells treated with the EGFR inhibitor AG1478 [33]. This result implies that upregulation of p27Kip1, inhibition of Cdk2, and restoration of Rb function facilitate the antimitogenic effect of these inhibitors. Administration of ZD1839 to nude mice bearing established A431 xenografts results in tumor elimination. Upon cessation of drug treatment, some but not all tumors recur [36], suggesting that these drugs can be both cytostatic and tumoricidal against EGFR-dependent cancers. ZD1839 induces apoptosis of mammary epithelial cells. This proapoptotic effect is mediated by inhibition of phosphorylation of the proapoptotic protein BAD in Ser112 [37]. Treatment with the reversible inhibitors ZD1839 and AG-1478 does not to reduce EGF binding sites or EGFR protein content in A431 and MDA-468 cells ([33] and Arteaga, unpublished data). Treatment with PKI-166 results in a dose-dependent increase in EGFR protein levels in A431 cells and no change in LAPC4 human prostate cancer cells [38]. Interestingly, treatment of N87 gastric cancer cells with the irreversible EGFR/HER2 inhibitor CI-1033 alquilates Cys805 in HER2, disrupting the association of HER2 with chaperone complexes and leading to ubiquitination and degradation of HER2 by proteasomal proteinases and inhibition of growth [39]. Submicromolar concentrations of ZD1839 block HER2 phosphorylation and growth of HER2-overexpressing breast cancer cells that also express low levels of EGFR [34,40,41]. Interestingly, this effect is seen at concentrations of the small molecule that do not recognize chimeric HER2 receptors directly [34], suggesting that it is the result of blockade of EGFR-mediated transactivation of HER2. The EGFRspecific kinase inhibitor AG-1478 suppresses mammary tumor formation in MMTV/TGFa ¥ MMTV/neu bigenic mice and, similar to ZD1839, inhibits the phosphorylation of both EGFR and neu, the rat homolog of human HER2, in tumor lysates [42]. AG-1517 (PD153035), another antiEGFR quinazoline similar in structure to ZD1839 and OSI-774 [43], induces the formation of inactive EGFR/ HER2 heterodimers in SKBR-3 cells and prevents the response to exogenous heregulin [30]. This result suggests the possibility that some EGFR tyrosine kinase inhibitors may (1) recognize HER2 directly, and (2) stabilize HER2 in an inactive complex thus impeding its partnering with HER3 and HER4.

129

Patient selection and timing of EGFR-targeted therapies As discussed by N. Hynes, in this issue, several human tumors including cancers of the upper aerodigestive tract (non-small cell lung, head and neck, esophagus, gastric), colon, pancreas, breast, ovary, bladder, kidney, and gliomas display EGFR RNA and/or protein overexpression. This occurs with or without gene amplification of the EGFR locus and is often associated with overexpression of receptor ligands like TGFa or amphiregulin [44–47]. In general, these studies utilized a plethora of methods to detect EGFR in situ and defined “overexpression” loosely without an adequate quantitation of receptor levels in tumor sites. More important, a systematic analysis of the levels of EGFR that are activated/phosphorylated in situ, which should reflect the level of receptor utilization by tumors and possibly predict those tumors that will respond to anti-EGFR therapies, is sorely missing. A phase I trial with indium-labeled EGFR mouse mAb 225 revealed selective localization of the EGFR antibody in 11 of 11 squamous cancers of the lung that had not been prescreened for EGFR levels [48]. In patients with head and neck cancers, administration of the EGFR antibody RG 83852 results in antibody localization in tumor tissue and skin but not in stroma [49]. These studies suggest that differential expression of EGFR in tumor versus nontumor host tissues can provide an exploitable therapeutic window in cancers with high frequency of EGFR overexpression without the prior determination of individual tumor EGFR levels. More recently, studies with the EGFR antibody C225 and the EGFR tyrosine kinase inhibitor ZD1839 (Iressa) have demonstrated responses in human tumor xenografts and cell lines expressing a wide range of EGFR levels from very low to very high [34,40,50–52]. These data support the critical need of assays that will determine a threshold level of total and/or activated (phosphorylated) tumor EGFR predictive of benefit from an EGFR-targeted therapy. Until such predictive assay(s) is/are available, patients with “EGFR-negative” advanced cancers as defined by current laboratory methods are not being excluded from enrollment into clinical trials of EGFR inhibitors. The ideal timing of EGFR-targeted therapies during the natural history of cancers is not known. Currently, most clinical trials with EGFR inhibitors are being done in patients with advanced metastatic cancers. The EGFR has been reported to be overexpressed in hyperplastic and preneoplastic epithelial lesions [44,53–55], suggesting that EGFR overexpression occurs years before the onset of invasive cancer and metastatatic disease. Preclinical studies in transgenic mice overexpressing the EGFR ligand TGFa and neu (rat/mouse homolog of HER2) in the mammary gland indicate that EGFR kinase inhibitors are more effective in preventing mammary hyperplasias than in inhibiting established carcinomas [42]. Pharmacological inhibition of

130


the EGFR with the small molecule EKI-785 in APCMin mice markedly reduces intestinal polyp number but not polyp size [56], implying that EGFR function is required for the initiation of intestinal tumors but that it might be dispensable for polyp expansion and progression. Nonetheless, the overall tolerability of EGFR inhibitors (see below) provides an opportunity for testing them against preneoplastic lesions and/or in subjects at high risk, an approach not defensible with cytotoxic chemotherapy. Interestingly, treatment with ZD1839 results in significant inhibition of proliferation and increase in the apoptotic index in human xenografts consisting of ductal carcinoma in situ (DCIS) and adjacent normal breast epithelium established in athymic nude mice [55]. Based on these data, elucidating the role of EGFR signaling inhibitors on preventing the progression of preneoplastic and/or preinvasive lesions is clearly indicated. This will require investigation in appropriately selected populations at increased risk for the subsequent development of cancer.

Clinical studies with EGFR inhibitors The following several issues must be taken into account when designing trials with anti-EGFR therapies: (1) These agents are less toxic and better tolerated than conventional cancer chemotherapy; (2) their optimal biological dose may not match their maximally tolerated dose and, in many cases, it has not been established; (3) the tumor types that will derive the most benefit from these agents are unknown; (4) the EGFR expression level and/or other molecular determinants predictive of a therapeutic benefit are unknown as well; and (5) finally, preclinical models suggest that they are supra-additive when added to conventional chemotherapy or hormonal agents and, in some cases, may reverse acquired resistance to these drugs. A critical issue in clinical trials with EGFR inhibitors has been the definition of their optimal biological dose and schedule at which complete and sustained receptor saturation and/or inhibition are achieved. This premise is based on preclinical studies that suggested that complete receptor occupancy was required for maximal inhibition of function [8,57]. This approach is radically different from dosefinding approaches for conventional nontargeted chemotherapeutic agents, where dose selection has been based on determining dose-limiting toxicities. The optimal biological dose may be chosen by extrapolating from preclinical models, i.e., establishing a parallelism between doses resulting in steady-state concentrations in plasma that are equivalent to those required to inhibit tumor cell growth ex vivo. This approach has been explored with both anti-HER2 [58] and anti-EGFR therapies [59]. However, possible differences in EGFR function and turnover, EGFR and/or receptor ligand levels, intracellular ATP concentrations, drug-protein binding in situ, and so on, between tumor and nontumor tissues could lead to choosing a suboptimal dose

and/or schedule. This has been exemplified in studies with the mAb C225. In the initial phase I studies, a difference in dose was found between the optimal biological dose projected from preclinical mouse models and the higher C225 dose required to achieve saturation of drug clearance in humans, a finding probably related to the fact that the monoclonal antibody binds to human EGFR but not to mouse receptors [59]. Additional factors to consider include a wide variation in interpatient pharmacokinetic parameters that has been observed both with both monoclonal antibodies and smallmolecule tyrosine kinase inhibitors [60]. Therefore, the best way to identify the biologically effective dose for these compounds might be by analyzing appropriate pharmacodynamic endpoints. A recent report suggests that this approach might have been fruitful. Treatment with ZD1839 at several dose levels ≥150 mg/day over 28 days resulted in maximal inhibition of EGFR and mitogen-activated protein kinase (MAPK) in keratinocytes from the suprabasal dermis as measured by immunohistochemistry using phospho-specific antibodies [61]. Because of these data, two well-tolerated doses (250 and 500 mg/day) were chosen for phase II and III efficacy studies. Interestingly, in advanced non-small cell lung cancer (NSCLC) an objective response rate of 18% was found at both dose levels of ZD1839 [62], suggesting that, as predicted by the pharmacodynamic studies, both doses were adequate to inhibit the EGFR tyrosine kinase in vivo. In addition, the proportion of patients exhibiting disease stabilization and prompt improvement in symptoms was identical at both doses of ZD1839 [62]. At the time of the writing of this report, ZD1839 has been approved in Japan for the treatment of advanced NSCLC. Similar single-agent clinical activity in patients with advanced NSCLC has been preliminarily reported with OSI-774 (Roman PerezSoler, Albert Einstein College of Medicine, personal communication). At therapeutic doses, these agents can induce reversible skin rash, diarrhea, nausea, and elevation of liver enzymes. In general, these effects are mild and well tolerated [60,63,64]. Clinical activity of C225 in NSCLC has not been reported yet, but considering its ability to block receptor function, it may not be different to that seen with ZD1839 and OSI-774. The dose of C225 selected for efficacy studies in humans was derived from those that induced an activity in human plasma that blocks the binding of EGF to receptors immobilized on a Biacore chip [59]. C225 has been shown to revert resistance to chemotherapy in preclinical models [65]. This observation led to studies in chemotherapy-resistant patients with colorectal and head and neck cancers. In both trials, the EGFR antibody was able to reverse the clinical resistance to chemotherapy in approximately 20% of patients [52,66]. Preliminary data indicate that C225 has single-agent activity in metastatic colorectal cancer (Jose Baselga, University of Barcelona, personal communication). ABX-EGF has shown clinical activity in metastatic renal cancer [101]. The toxicity profile of EGFR antibodies is

ErbB-targeted Therapeutic Approaches in Human Cancer

similar to that of the small-molecule tyrosine kinase inhibitors plus occasional allergic reactions and fever [74–76,81]. Interestingly, diarrhea is exceedingly rare with treatment with the C225 antibody [59].

Individual strengths of EGFR antibodies and small-molecule tyrosine kinase inhibitors Although the initial interaction of receptor antibodies and small-molecule kinase inhibitors occurs with different portions of the EGFR molecule, the net result is the inhibition of receptor function. At present it is difficult to predict which of these two well-tolerated approaches will be more effective. This problem might not be easy to solve, given the strengths of both types of interventions. It has been speculated that one advantage of small molecules over anti-EGFR antibodies might be the potential to inhibit the tyrosine kinase activity associated with the mutant EGFR vIII, frequently amplified in glioblastomas [68]. This receptor lacks residues 6–276 in the ectodomain and exhibits constitutive tyrosine kinase activity [69]. It has also been reported in breast, ovarian, and NSCLCs [70]. At this time, however, this possible advantage of small molecules, as it applies to EGFR vIII, cannot be substantiated by experimental data. In fact, ZD1839 fails to inhibit NR6M tumors that overexpress EGFR vIII [71]. On the other hand, C225 binds to ectodomain of EGFR vIII in glioma cells (Michael Needle, Imclone, personal communication). The mAb 806, raised against EGFR vIII, inhibits the growth of EGFR vIIIexpressing U87 MG xenografts while sparing normal tissues expressing the wild-type receptor [72]. Small molecules are administered orally, which makes them highly convenient for chronic therapy. Their potential ability to cross-react with HER2 is an attractive feature of these compounds if we consider the ability of EGFR and HER2 to cooperate as heterodimers [73] as well as epidemiological data indicating that cooverexpression of members of the erbB (HER) network confers a poor clinical outcome compared to tumors with only high levels of EGFR [74–76]. Three recent reports show that low concentrations of ZD1839 blocked basal phosphorylation of HER2 and growth of HER2-overexpressing human breast tumor cells [34,40,41]. These results suggest that some small molecules may block EGFR-mediated transphosphorylation of HER2 and/or recognize the HER2 kinase directly. The small-molecule AG-1478 but not C225 inhibits fibronectin/ integrin-mediated transactivation of the EGFR [77,78], suggesting that small molecules might be better in blocking lateral inputs from heterologous signaling networks. Similarly, ZD1839 blocks insulin-like growth factor (IGF)mediated phosphorylation of mitogen-activated protein kinase (MAPK) and the proapoptotic protein BAD, leading to mammary cell apoptosis [37]. Some EGFR antibodies may work, at least in part, by immune effector-mediated destruction of tumor cells

131

Table 2 Individual strengths of anti-EGFR agentsa Humanized EGFR antibodies

Small molecule ATP-competitive tyrosine kinase inhibitors

Prolonged half-life and stability in vivo Downregulation of cell surface receptors Possible ADCC None or minimal gastrointestinal toxicity

Convenient oral delivery Inhibition of EGFR-HER2 cross-talk Some inhibit HER2 directly No anaphylactic or allergic reactions Blockade of lateral intracellular signals

a EGFR, epidermal growth factor receptor; ADCC, antibody-dependent cell-mediated cytotoxicity.

[23–26]. This immune mechanism of action might be unique to some anti-EGFR IgGs and potentially provide an advantage over small-molecule tyrosine kinase inhibitors. Because of their stability and prolonged half-life, antibodies may require infrequent administration. Therefore, these complementary cellular mechanisms of action (summarized in Table 2), plus that both types of drugs bind to different noncompeting sites of the EGFR, suggest the possibility of a synergistic anticancer effect of EGFR antibodies and small-molecule tyrosine kinase inhibitors when used in combination.

Tumor cell resistance to EGFR inhibitors Cellular or molecular mechanisms that will explain de novo or acquired resistance to EGFR inhibitors in patients have not been reported yet. However, the known mechanisms of receptor signaling have provided some leads. For example, it is well established that overexpression of HER2 potentiates EGFR signaling [79] and contributes to EGFRmediated transformation and tumor progression [80]. Cancers that cooverexpress both EGFR and HER2 fare worse than those that overexpress either receptor [74,75,81]. In some experimental systems, inactivation of HER2 is required to block EGFR-mediated transformation [82,83]. Overexpression of HER2 counteracts the ability of EGFR kinase inhibitors to block EGFR activity [84]. Conversely, high levels of activated EGFR abrogate the efficacy of the HER2 antibody trastuzumab against HER2 gene-amplified MKN7 human gastric cancer cells [85], and this resistance is reversed by the EGFR inhibitor PKI-166 [86]. In addition, the EGFR antibody C225 synergizes with HER2 antibodies against HER2-overexpressing ovarian cancer cells [87]. Finally, ZD1839 inhibits HER2 phosphorylation per se in intact cells [34,40] and potentiates the antitumor effect of trastuzumab against BT-474 breast cancer xenografts with high HER2 gene amplification [34]. Taken together, these results lead to the hypothesis that overexpression of HER2 is a preferential mechanism of de novo or acquired

132


resistance to EGFR inhibitors and that combinations of EGFR and HER2 inhibitors will be synergistic against tumors with EGFR-positive tumors that also express high HER2 levels. This hypothesis is currently being tested by phase II studies of trastuzumab in combination with either ZD1839 or OSI-774. Several studies have shown that both EGFR antibodies and small-molecule tyrosine kinase inhibitors reduce vascular endothelial growth factor (VEGF) protein levels and microvessel density in tumors that regress upon EGFR blockade [19,88,89]. A431 tumor cells with acquired resistance to C225 exhibit increased expression and secretion of VEGF. Forced expression of VEGF in C225-sensitive A431 cells renders them resistant to the EGFR antibody in vivo [90]. These data imply that (1) subversion of EGFRdependent tumor neoangiogenesis is required for the antitumor effect of EGFR inhibitors, and (2) enhanced angiogenesis can endow tumors with resistance to EGFR blockade. In addition, these results provide a strong rationale for combinations of anti-EGFR agents with inhibitors of angiogenesis. Elucidation of the preferential molecular mechanisms of escape from anti-EGFR therapies will define new rational targets against which drugs are either available or to be developed. Drugs against these targets can be combined with EGFR inhibitors to prevent de novo or acquired resistance and enhance therapeutic efficacy. For example, overexpression of the IGF-I receptor has been recently reported to abrogate the antitumor effect of EGFR tyrosine kinase inhibitors against human cancer cells [91]. In these studies, simultaneous blockade of IGF-I receptor signaling restored tumor cell sensitivity to the EGFR inhibitors, providing a rationale for combined antireceptor therapies. If tumor specific and well tolerated, combinations of antisignaling agents, like those suggested immediately above, should become a robust alternative to current cytotoxic chemotherapy. Acknowledgments Supported by NIH grant R01 CA80195 and VanderbiltIngram Cancer Center Support Grant CA68485. References [1] J.R. Woodburn, The epidermal growth factor receptor and its inhibition in cancer therapy, Pharmacol. Ther. 82 (1999) 241–250. [2] D.W. Threadgill, A.A. Dlugosz, L.A. Hansen, T. Tennenbaum, U. Lichti, D. Yee, C. LaMantia, T. Mourton, K. Herrup, R.C. Harris, et al., Targeted disruption of mouse EGF receptor: effect of genetic background on mutant phenotype, Science 269 (1995) 230–234. [3] P.J. Miettinen, J.E. Berger, J. Meneses, Y. Phung, R.A. Pedersen, Z. Werb, R. Derynck, Epithelial immaturity and multiorgan failure in mice lacking epidermal growth factor receptor, Nature 376 (1995) 337–341. [4] M. Sibilia, J.P. Steinbach, L. Stingl, A. Aguzzi, E.F. Wagner, A strain-independent postnatal neurodegeneration in mice lacking the EGF receptor, EMBO J. 17 (1998) 719–731.

[5] G.B. Mann, K.J. Fowler, A. Gabriel, E.C. Nice, R.L. Williams, A.R. Dunn, Mice with a null mutation of the TGF alpha gene have abnormal skin architecture, wavy hair, and curly whiskers and often develop corneal inflammation, Cell 73 (1993) 249–261. [6] N.C. Luetteke, T.H. Qiu, S.E. Fenton, K.L. Troyer, R.F. Riedel, A. Chang, D.C. Lee, Targeted inactivation of the EGF and amphiregulin genes reveals distinct roles for EGF receptor ligands in mouse mammary gland development, Development 126 (1999) 2739– 2750. [7] N. Redemann, B. Holzmann, T. von Ruden, E.F. Wagner, J. Schlessinger, A. Ullrich, Anti-oncogenic activity of signallingdefective epidermal growth factor receptor mutants, Mol. Cell. Biol. 12 (1992) 491–498. [8] J.D. Sato, T. Kawamoto, A.D. Le, J. Mendelsohn, J. Polikoff, G.H. Sato, Biological effects in vitro of monoclonal antibodies to human epidermal growth factor receptors, Mol. Biol. Med. 1 (1983) 511– 529. [9] Z. Fan, H. Masui, I. Altas, J. Mendelsohn, Blockade of epidermal growth factor receptor function by bivalent and monovalent fragments of 225 anti-epidermal growth factor receptor monoclonal antibodies, Cancer Res. 53 (1993) 4322–4328. [10] Z. Fan, Y. Lu, X. Wu, J. Mendelsohn, Antibody-induced epidermal growth factor receptor dimerization mediates inhibition of autocrine proliferation of A431 squamous carcinoma cells, J. Biol. Chem. 269 (1994) 27595–27602. [11] Z. Fan, J. Mendelsohn, H. Masui, R. Kumar, Regulation of epidermal growth factor receptor in NIH3T3/HER14 cells by antireceptor monoclonal antibodies, J. Biol. Chem. 268 (1993) 21073–21079. [12] M.J. Van de Vijver, R. Kumar, J. Mendelsohn, Ligand-induced activation of A431 cell epidermal growth factor receptors occurs primarily by an autocrine pathway that acts upon receptors on the surface rather than intracellularly, J. Biol. Chem. 266 (1991) 7503– 7508. [13] H.S. Wiley, M.F. Woolf, L.K. Opresko, P.M. Burke, B. Will, J.R. Morgan, D.A. Lauffenburger, Removal of the membrane-anchoring domain of epidermal growth factor leads to intracrine signaling and disruption of mammary epithelial cell organization, J. Cell Biol. 143 (1998) 1317–1328. [14] L.N. Klapper, H. Waterman, M. Sela, Y. Yarden, Tumor-inhibitory antibodies to HER-2/ErbB-2 may act by recruiting c-Cb1 and enhancing ubiquitination of HER-2, Cancer Res. 60 (2000) 3384– 3388. [15] H. Masui, T. Kawamoto, J.D. Sato, B. Wolf, G. Sato, J. Mendelsohn, Growth inhibition of human tumor cells in athymic mice by antiepidermal growth factor receptor monoclonal antibodies, Cancer Res. 44 (1984) 1002–1007. [16] X. Wu, Z. Fan, H. Masui, N. Rosen, J. Mendelsohn, Apoptosis induced by an anti-epidermal growth factor receptor monoclonal antibody in a human colorectal carcinoma cell line and its delay by insulin, J. Clin. Invest. 95 (1995) 1897–1905. [17] S.M. Huang, J.M. Bock, P.M. Harari, Epidermal growth factor receptor blockade with C225 modulates proliferation, apoptosis, and radiosensitivity in squamous cell carcinomas of the head and neck, Cancer Res. 59 (1999) 1935–1940. [18] N.I. Goldstein, M. Prewett, K. Zuklys, P. Rockwell, J. Mendelsohn, Biological efficacy of a chimeric antibody to the epidermal growth factor receptor in a human tumor xenograft model, Clin. Cancer Res. 1 (1995) 1311–1318. [19] D. Kedar, C.H. Baker, J.J. Killion, C.P. Dinney, I.J. Fidler, Blockade of the epidermal growth factor receptor signaling inhibits angiogenesis leading to regression of human renal cell carcinoma growing orthotopically in nude mice, Clin. Cancer Res. 8 (2002) 3592– 3600. [20] D. Peng, Z. Fan, Y. Lu, T. DeBlasio, H. Scher, J. Mendelsohn, Anti-epidermal growth factor receptor monoclonal antibody 225 upregulates p27KIP1 and induces G1 arrest in prostatic cancer cell line DU145, Cancer Res. 56 (1996) 3666–3669.

ErbB-targeted Therapeutic Approaches in Human Cancer [21] Z. Fan, B.Y. Shang, Y. Lu, J.L. Chou, J. Mendelsohn, Reciprocal changes in p27(Kip1) and p21(Cip1) in growth inhibition mediated by blockade or overstimulation of epidermal growth factor receptors, Clin. Cancer Res. 3 (1997) 1943–1948. [22] J. Albanell, J. Codony-Servat, F. Rojo, J.M. Del Campo, S. Sauleda, J. Anido, G. Raspall, J. Giralt, J. Rosello, R.I. Nicholson, J. Mendelsohn, J. Baselga, Activated extracellular signal-regulated kinases: association with epidermal growth factor receptor/transforming growth factor alpha expression in head and neck squamous carcinoma and inhibition by anti-epidermal growth factor receptor treatments, Cancer Res. 61 (2001) 6500–6510. [23] U. Rodeck, M. Herlyn, D. Herlyn, C. Molthoff, B. Atkinson, M. Varello, Z. Steplewski, H. Koprowski, Tumor growth modulation by a monoclonal antibody to the epidermal growth factor receptor: immunologically mediated and effector cell-independent effects, Cancer Res. 47 (1987) 3692–3696. [24] J.H. Sampson, L.E. Crotty, S. Lee, G.E. Archer, D.M. Ashley, C.J. Wikstrand, L.P. Hale, C. Small, G. Dranoff, A.H. Friedman, H.S. Friedman, D.D. Bigner, Unarmed, tumor-specific monoclonal antibody effectively treats brain tumors, Proc. Natl. Acad. Sci. USA 97 (2000) 7503–7508. [25] H. Modjtahedi, S. Eccles, J. Sandle, G. Box, J. Titley, C. Dean, Differentiation or immune destruction: two pathways for therapy of squamous cell carcinomas with antibodies to the epidermal growth factor receptor, Cancer Res. 54 (1994) 1695–1701. [26] H. Bier, T. Hoffmann, I. Haas, A. van Lierop, Anti-(epidermal growth factor) receptor monoclonal antibodies for the induction of antibody-dependent cell-mediated cytotoxicity against squamous cell carcinoma lines of the head and neck, Cancer Immunol. Immunother. 46 (1998) 167–173. [27] M. Naramura, S.D. Gillies, J. Mendelsohn, R.A. Reisfeld, B.M. Mueller, Therapeutic potential of chimeric and murine anti(epidermal growth factor receptor) antibodies in a metastasis model for human melanoma, Cancer Immunol. Immunother. 37 (1993) 343–349. [28] N. Prenzel, O.M. Fischer, S. Streit, S. Hart, A. Ullrich, The epidermal growth factor receptor family as a central element for cellular signal transduction and diversification, Endocr. Relat. Cancer 8 (2001) 11–31. [29] C.L. Arteaga, The epidermal growth factor receptor: from mutant oncogene in nonhuman cancers to therapeutic target in human neoplasia, J. Clin. Oncol. 19 (2001) 32S–40S. [30] C.L. Arteaga, T.T. Ramsey, L.K. Shawver, C.A. Guyer, Unliganded epidermal growth factor receptor dimerization induced by direct interaction of quinazolines with the ATP binding site, J. Biol. Chem. 272 (1997) 23247–23254. [31] R.B. Lichtner, A. Menrad, A. Sommer, U. Klar, M.R. Schneider, Signaling-inactive epidermal growth factor receptor/ligand complexes in intact carcinoma cells by quinazoline tyrosine kinase inhibitors, Cancer Res. 61 (2001) 5790–5795. [32] D.W. Fry, Site-directed irreversible inhibitors of the erbB family of receptor tyrosine kinases as novel chemotherapeutic agents for cancer, Anticancer Drug Des. 15 (2000) 3–16. [33] D. Busse, R.S. Doughty, T.T. Ramsey, W.E. Russell, J.O. Price, W.M. Flanagan, L.K. Shawver, C.L. Arteaga, Reversible G(1) arrest induced by inhibition of the epidermal growth factor receptor tyrosine kinase requires up-regulation of p27(KIP1) independent of MAPK activity, J. Biol. Chem. 275 (2000) 6987–6995. [34] S.L. Moulder, F.M. Yakes, S.K. Muthuswamy, R. Bianco, J.F. Simpson, C.L. Arteaga, Epidermal growth factor receptor (HER1) tyrosine kinase inhibitor (Iressa) inhibits HER2/neu (erbB2)-overexpressing breast cancer cells in vitro and in vivo, Cancer Res. 61 (2001) 8887–8895. [35] J.D. Moyer, E.G. Barbacci, K.K. Iwata, L. Arnold, B. Boman, A. Cunningham, C. DiOrio, J. Doty, M.J. Morin, M.P. Moyer, M. Neveu, V.A. Pollack, L.R. Pustilnik, M.M. Reynolds, D. Sloan, A. Theleman, P. Miller, Induction of apoptosis and cell cycle arrest by

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

133

CP-358,774, an inhibitor of epidermal growth factor receptor tyrosine kinase, Cancer Res. 57 (1997) 4838–4848. A.E. Wakeling, S.P. Guy, J.R. Woodburn, S.E. Ashton, B.J. Curry, A.J. Barker, K.H. Gibson, ZD1839 (Iressa): an orally active inhibitor of epidermal growth factor signaling with potential for cancer therapy, Cancer Res. 62 (2002) 5749–5754. A.P. Gilmore, A.J. Valentijn, P. Wang, A.M. Ranger, N. Bundred, M.J. O’Hare, A. Wakeling, S.J. Korsmeyer, C.H. Streuli, Activation of BAD by therapeutic inhibition of epidermal growth factor receptor and transactivation by insulin-like growth factor receptor, J. Biol. Chem. 277 (2002) 27643–27650. I.K. Mellinghoff, C. Tran, C.L. Sawyers, Growth inhibitory effects of the dual ErbB1/ErbB2 tyrosine kinase inhibitor PKI-166 on human prostate cancer xenografts, Cancer Res. 62 (2002) 5254– 5259. A. Citri, I. Alroy, S. Lavi, C. Rubin, W. Xu, N. Grammatikakis, C. Patterson, L. Neckers, D.W. Fry, Y. Yarden, Drug-induced ubiquitylation and degradation of ErbB receptor tyrosine kinases: implications for cancer therapy, EMBO J. 21 (2002) 2407–2417. M.M. Moasser, A. Basso, S.D. Averbuch, N. Rosen, The tyrosine kinase inhibitor ZD1839 (“Iressa”) inhibits HER2-driven signaling and suppresses the growth of HER2-overexpressing tumor cells, Cancer Res. 61 (2001) 7184–7188. N.G. Anderson, T. Ahmad, K. Chan, R. Dobson, N.J. Bundred, ZD1839 (Iressa), a novel epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor, potently inhibits the growth of EGFRpositive cancer cell lines with or without erbB2 overexpression, Int. I. Cancer 94 (2001) 774–782. A.E. Lenferink, J.F. Simpson, L.K. Shawver, R.J. Coffey, J.T. Forbes, C.L. Arteaga, Blockade of the epidermal growth factor receptor tyrosine kinase suppresses tumorigenesis in MMTV/Neu + MMTV/TGF-alpha bigenic mice, Proc. Natl. Acad. Sci. USA 97 (2000) 9609–9614. D.W. Fry, A.J. Kraker, A. McMichael, L.A. Ambroso, J.M. Nelson, W.R. Leopold, R.W. Connors, A.J. Bridges, A specific inhibitor of the epidermal growth factor receptor tyrosine kinase, Science 265 (1994) 1093–1095. D.S. Salomon, R. Brandt, F. Ciardiello, N. Normanno, Epidermal growth factor-related peptides and their receptors in human malignancies, Crit. Rev. Oncol. Hematol. 19 (1995) 183–232. V. Rusch, J. Baselga, C. Cordon-Cardo, J. Orazem, M. Zaman, S. Hoda, J. McIntosh, J. Kurie, E. Dmitrovsky, Differential expression of the epidermal growth factor receptor and its ligands in primary non-small cell lung cancers and adjacent benign lung, Cancer Res. 53 (1993) 2379–2385. J.R. Grandis, M.F. Melhem, W.E. Gooding, R. Day, V.A. Holst, M.M. Wagener, S.D. Drenning, D.J. Tweardy, Levels of TGF-alpha and EGFR protein in head and neck squamous cell carcinoma and patient survival, J. Natl. Cancer Inst. 90 (1998) 824–832. G. Fontanini, M. De Laurentiis, S. Vignati, S. Chine, M. Lucchi, V. Silvestri, A. Mussi, S. De Placido, G. Tortora, A.R. Bianco, W. Gullick, C.A. Angeletti, G. Bevilacqua, F. Ciardiello, Evaluation of epidermal growth factor-related growth factors and receptors and of neoangiogenesis in completely resected stage I-IIIA non-small-cell lung cancer: amphiregulin and microvessel count are independent prognostic indicators of survival, Clin. Cancer Res. 4 (1998) 241– 249. C.R. Divgi, S. Welt, M. Kris, F.X. Real, S.D. Yeh, R. Gralla, B. Merchant, S. Schweighart, M. Unger, S.M. Larson, et al., Phase I and imaging trial of indium 111-labeled anti-epidermal growth factor receptor monoclonal antibody 225 in patients with squamous cell lung carcinoma, J. Natl. Cancer Inst. 83 (1991) 97–104. R. Perez-Soler, N.J. Donato, D.M. Shin, M.G. Rosenblum, H.Z. Zhang, C. Tornos, H. Brewer, J.C. Chan, J.S. Lee, W.K. Hong, et al., Tumor epidermal growth factor receptor studies in patients with non-small-cell lung cancer or head and neck cancer treated with monoclonal antibody RG 83852, J. Clin. Oncol. 12 (1994) 730–739.

134


[50] F.M. Sirotnak, M.F. Zakowski, V.A. Miller, H.I. Scher, M.G. Kris, Efficacy of cytotoxic agents against human tumor xenografts is markedly enhanced by coadministration of ZD1839 (Iressa), an inhibitor of EGFR tyrosine kinase, Clin. Cancer Res. 6 (2000) 4885–4892. [51] F. Ciardiello, R. Caputo, R. Bianco, V. Damiano, G. Pomatico, S. De Placido, A.R. Bianco, G. Tortora, Antitumor effect and potentiation of cytotoxic drugs activity in human cancer cells by ZD-1839 (Iressa), an epidermal growth factor receptor-selective tyrosine kinase inhibitor, Clin. Cancer Res. 6 (2000) 2053–2063. [52] L. Saltz, M. Rubin, H. Hochster, N.S. Tchekmeydian, H. Waksal, M. Needle, A. LoBuglio, Cetuximab (IMC-225) plus irinotecan (CPT11) is active in CPT-11-refractory colorectal cancer that expresses epidermal growth factor receptors, Proc. Am. Soc. Clin. Oncol. 20 (2001) 3a. [53] V. Rusch, D. Klimstra, I. Linkov, E. Dmitrovsky, Aberrant expression of p53 or the epidermal growth factor receptor is frequent in early bronchial neoplasia and coexpression precedes squamous cell carcinoma development, Cancer Res. 55 (1995) 1365–1372. [54] J.M. Kurie, H.J. Shin, J.S. Lee, R.C. Morice, J.Y. Ro, S.M. Lippman, W.N. Hittelman, R. Yu, J.J. Lee, W.K. Hong, Increased epidermal growth factor receptor expression in metaplastic bronchial epithelium, Clin. Cancer Res. 2 (1996) 1787–1793. [55] K.C. Chan, W.F. Knox, J.M. Gee, J. Morris, R.I. Nicholson, C.S. Potten, N.J. Bundred, Effect of epidermal growth factor receptor tyrosine kinase inhibition on epithelial proliferation in normal and premalignant breast, Cancer Res. 62 (2002) 122–128. [56] R.B. Roberts, L. Min, M.K. Washington, S.J. Olsen, S.H. Settle, R.J. Coffey, D.W. Threadgill, Importance of epidermal growth factor receptor signaling in establishment of adenomas and maintenance of carcinomas during intestinal tumorigenesis, Proc. Natl. Acad. Sci. USA 99 (2002) 1521–1526. [57] T. Kawamoto, J. Mendelsohn, A. Le, G.H. Sato, C.S. Lazar, G.N. Gill, Relation of epidermal growth factor receptor concentration to growth of human epidermoid carcinoma A431 cells, J. Biol. Chem. 259 (1984) 7761–7766. [58] J. Baselga, D. Tripathy, J. Mendelsohn, S. Baughman, C.C. Benz, L. Dantis, N.T. Sklarin, A.D. Seidman, C.A. Hudis, J. Moore, P.P. Rosen, T. Twaddell, I.C. Henderson, L. Norton, Phase II study of weekly intravenous recombinant humanized anti-p185HER2 monoclonal antibody in patients with HER2/neu-overexpressing metastatic breast cancer, J. Clin. Oncol. 14 (1996) 737–7344. [59] J. Baselga, D. Pfister, M.R. Cooper, R. Cohen, B. Burtness, M. Bos, G. D’Andrea, A. Seidman, L. Norton, K. Gunnett, J. Falcey, V. Anderson, H. Waksal, J. Mendelsohn, Phase I studies of antiepidermal growth factor receptor chimeric antibody C225 alone and in combination with cisplatin, J. Clin. Oncol. 18 (2000) 904–914. [60] J. Baselga, D. Rischin, M. Ranson, H. Calvert, E. Raymond, D.G. Kieback, S.B. Kaye, L. Gianni, A. Harris, T. Bjork, S.D. Averbuch, A. Feyereislova, H. Swaisland, F. Rojo, J. Albanell, Phase I safety, pharmacokinetic and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types, J. Clin. Oncol. 20 (2002) 4292–4302. [61] J. Albanell, F. Rojo, S. Averbuch, A. Feyereislova, J.M. Mascaro, R. Herbst, P. LoRusso, D. Rischin, S. Sauleda, J. Gee, R.I. Nicholson, J. Baselga, Pharmacodynamic studies of the epidermal growth factor receptor inhibitor ZD1839 in skin from cancer patients: histopathologic and molecular consequences of receptor inhibition, J. Clin. Oncol. 20 (2002) 110–124. [62] M. Fukuoka, S. Yano, G. Giaccone, T. Tamura, K. Nakagawa, J.-Y. Douillard, Y. Nishiwaki, J. Vansteenkiste, S. Kudoh, D. Rischin, R. Eek, T. Horai, K. Noda, I. Takata, E. Smit, S. Averbuch, A. MacLeod, A. Feyereislova, R.-P. Dong, J. Baselga, A multi-institutional randomized phase III trial of gefitinib for previously treated patients with advanced non-small-cell lung cancer (THE IDEAL 1 Trial), J. Clin. Oncol. 21 (2003) (in press).

[63] M. Hidalgo, L.L. Siu, J. Nemunaitis, J. Rizzo, L.A. Hammond, C. Takimoto, S.G. Eckhardt, A. Tolcher, C.D. Britten, L. Denis, K. Ferrante, D.D. Von Hoff, S. Silberman, E.K. Rowinsky, Phase I and pharmacologic study of OSI-774, an epidermal growth factor receptor tyrosine kinase inhibitor, in patients with advanced solid malignancies, J. Clin. Oncol. 19 (2001) 3267–3279. [64] M. Ranson, L.A. Hammond, D. Ferry, M. Kris, A. Tullo, P.I. Murray, V. Miller, S. Averbuch, J. Ochs, C. Morris, A. Feyereislova, H. Swaisland, E.K. Rowinsky, ZD1839, a selective oral epidermal growth factor receptor-tyrosine kinase inhibitor, is well tolerated and active in patients with solid, malignant tumors: results of a phase I trial, J. Clin. Oncol. 20 (2002) 2240–2250. [65] M.C. Prewett, A.T. Hooper, R. Bassi, L.M. Ellis, H.W. Waksal, D.J. Hicklin, Enhanced antitumor activity of anti-epidermal growth factor receptor monoclonal antibody IMC-C225 in combination with Irinotecan (CPT-11) against human colorectal tumor xenografts, Clin. Cancer Res. 8 (2002) 994–1003. [66] J.M. Baselga, J. Trigo, J. Bourhis, J. Tortochaux, H. Cortes-Funes, R. Hitt, P. Gascon, M. Muesser, A. Harstrick, A. Eckardt, Cetuximab (C225) plus cisplatin/carboplatin is active in patients with recurrent/metastatic squamous cell carcinoma of the head and neck (SCCHN) progressing on a same dose and schedule platinum-based regimen, 2002, pp. 226a. [67] H. Bier, T. Hoffmann, U. Hauser, M. Wink, M. Ochler, A. Kovar, M. Muser, R. Knecht, Clinical trial with escalating doses of the antiepidermal growth factor receptor humanized monoclonal antibody EMD 72 000 in patients with advanced squamous cell carcinoma of the larynx and hypopharynx, Cancer Chemother. Pharmacol. 47 (2001) 519–524. [68] G. Sauter, T. Maeda, F.M. Waldman, R.L. Davis, B.G. Feuerstein, Patterns of epidermal growth factor receptor amplification in malignant gliomas, Am. J. Pathol. 148 (1996) 1047–1053. [69] A.J. Wong, J.M. Ruppert, S.H. Bigner, C.H. Grzeschik, P.A. Humphrey, D.S. Bigner, B. Vogelstein, Structural alterations of the epidermal growth factor receptor gene in human gliomas, Proc. Natl. Acad. Sci. USA 89 (1992) 2965–2969. [70] D.K. Moscatello, M. Holgado-Madruga, A.K. Godwin, G. Ramirez, G. Gunn, P.W. Zoltick, J.A. Biegel, R.L. Hayes, A.J. Wong, Frequent expression of a mutant epidermal growth factor receptor in multiple human tumors, Cancer Res. 55 (1995) 5536–5539. [71] A.B. Heimberger, C.A. Learn, G.E. Archer, R.E. McLendon, T.A. Chewning, F.L. Tuck, J.B. Pracyk, A.H. Friedman, H.S. Friedman, D.D. Bigner, J.H. Sampson, Brain tumors in mice are susceptible to blockade of epidermal growth factor receptor (EGFR) with the oral, specific, EGFR-tyrosine kinase inhibitor ZD1839 (Iressa), Clin. Cancer Res. 8 (2002) 3496–3502. [72] R.B. Luwor, T.G. Johns, C. Murone, H.J. Huang, W.K. Cavenee, G. Ritter, L.J. Old, A.W. Burgess, A.M. Scott, Monoclonal antibody 806 inhibits the growth of tumor xenografts expressing either the de2-7 or amplified epidermal growth factor receptor (EGFR) but not wild-type EGFR, Cancer Res. 61 (2001) 5355–5361. [73] D. Graus-Porta, R.R. Beerli, J.M. Daly, N.E. Hynes, ErbB-2, the preferred heterodimerization partner of all ErbB receptors, is a mediator of lateral signaling, EMBO J. 16 (1997) 1647–1655. [74] M. Tateishi, T. Ishida, S. Kohdono, M. Hamatake, Y. Fukuyama, K. Sugimachi, Prognostic influence of the co-expression of epidermal growth factor receptor and c-erbB-2 protein in human lung adenocarcinoma, Surg. Oncol. 3 (1994) 109–113. [75] W. Xia, Y.K. Lau, H.Z. Zhang, F.Y. Xiao, D.A. Johnston, A.R. Liu, L. Li, R.L. Katz, M.C. Hung, Combination of EGFR, HER-2/neu, and HER-3 is a stronger predictor for the outcome of oral squamous cell carcinoma than any individual family members, Clin. Cancer Res. 5 (1999) 4164–4174. [76] A.L. Harris, S. Nicholson, J.R. Sainsbury, J. Farndon, C. Wright, Epidermal growth factor receptors in breast cancer: association with early relapse and death, poor response to hormones and interactions with neu, J. Steroid Biochem. 34 (1989) 123–131.

ErbB-targeted Therapeutic Approaches in Human Cancer [77] L. Moro, M. Venturino, C. Bozzo, L. Silengo, F. Altruda, L. Beguinot, G. Tarone, P. Defilippi, Integrins induce activation of EGF receptor: role in MAP kinase induction and adhesion-dependent cell survival, EMBO J. 17 (1998) 6622–6632. [78] D. Liu, J. Aguirre Ghiso, Y. Estrada, L. Ossowski, EGFR is a transducer of the urokinase receptor initiated signal that is required for in vivo growth of a human carcinoma, Cancer Cell. 1 (2002) 445–457. [79] D. Karunagaran, E. Tzahar, R.R. Beerli, X. Chen, D. Graus-Porta, B.J. Ratzkin, R. Seger, N.E. Hynes, Y. Yarden, ErbB-2 is a common auxiliary subunit of NDF and EGF receptors: implications for breast cancer, EMBO J. 15 (1996) 254–264. [80] W.J. Muller, C.L. Arteaga, S.K. Muthuswamy, P.M. Siegel, M.A. Webster, R.D. Cardiff, K.S. Meise, F. Li, S.A. Halter, R.J. Coffey, Synergistic interaction of the Neu proto-oncogene product and transforming growth factor alpha in the mammary epithelium of transgenic mice, Mol. Cell Biol. 16 (1996) 5726–5736. [81] R.I. Nicholson, R.A. McClelland, P. Finlay, C.L. Eaton, W.J. Gullick, A.R. Dixon, J.F. Robertson, I.O. Ellis, R.W. Blamey, Relationship between EGF-R, c-erbB-2 protein expression and Ki67 immunostaining in breast cancer and hormone sensitivity, Eur. J. Cancer 7 (1993) 1018–1023. [82] X. Qian, W.C. Dougall, M.E. Hellman, M.I. Greene, Kinasedeficient neu proteins suppress epidermal growth factor receptor function and abolish cell transformation, Oncogene 9 (1994) 1507–1514. [83] D. Graus-Porta, R.R. Beerli, N.E. Hynes, Single-chain antibodymediated intracellular retention of ErbB-2 impairs Neu differentiation factor and epidermal growth factor signaling, Mol. Cell Biol. 15 (1995) 1182–1191. [84] J.G. Christensen, R.E. Schreck, E. Chan, X. Wang, C. Yang, L. Liu, J. Cui, L. Sun, J.M. Cherrington, D.B. Mendel, High levels of HER-2 expression alter the ability of epidermal growth factor receptor (EGFR) family tyrosine kinase inhibitors to inhibit EGFR phosphorylation in vivo, Clin. Cancer Res. 7 (2001) 4230–4238. [85] H.A. Lane, I. Beuvink, A.B. Motoyama, J.M. Daly, R.M. Neve, N.E. Hynes, ErbB2 potentiates breast tumor proliferation through modulation of p27(Kip1)-Cdk2 complex formation: receptor overexpression does not determine growth dependency, Mol. Cell Biol. 20 (2000) 3210–3223. [86] A.B. Motoyama, N.E. Hynes, H.A. Lane, The efficacy of ErbB receptor targeted anti-cancer therapeutics is influenced by the availability of EGF-related peptides, Cancer Res. 62 (2002) 3151–3158. [87] D. Ye, J. Mendelsohn, Z. Fan, Augmentation of a humanized antiHER2 mAb 4D5 induced growth inhibition by a human-mouse chimeric anti-EGF receptor mAb C225, Oncogene 18 (1999) 731– 738. [88] A.M. Petit, J. Rak, M.C. Hung, P. Rockwell, N. Goldstein, B. Fendly, R.S. Kerbel, Neutralizing antibodies against epidermal growth factor and ErbB-2/neu receptor tyrosine kinases down-regulate vascular endothelial growth factor production by tumor cells in vitro and in vivo: angiogenic implications for signal transduction therapy of solid tumors, Am. J. Pathol. 151 (1997) 1523–1530. [89] F. Ciardiello, R. Caputo, R. Bianco, V. Damiano, G. Fontanini, S. Cuccato, S. De Placido, A.R. Bianco, G. Tortora, Inhibition of

[90]

[91]

[92]

[93]

[94]

[95]

[96]

[97]

[98]

[99]

[100]

135

growth factor production and angiogenesis in human cancer cells by ZD1839 (Iressa), a selective epidermal growth factor receptor tyrosine kinase inhibitor, Clin. Cancer Res. 7 (2001) 1459–1465. A. Viloria-Petit, T. Crombet, S. Jothy, D. Hicklin, P. Bohlen, J.M. Schlaeppi, J. Rak, R.S. Kerbel, Acquired resistance to the antitumor effect of epidermal growth factor receptor-blocking antibodies in vivo: a role for altered tumor angiogenesis, Cancer Res. 61 (2001) 5090–5101. A. Chakravarti, J.S. Loeffler, N.J. Dyson, Insulin-like growth factor receptor I mediates resistance to anti-epidermal growth factor receptor therapy in primary human glioblastoma cells through continued activation of phosphoinositide 3-kinase signaling, Cancer Res. 62 (2002) 200–207. D.H. Lynch, X.D. Yang, Therapeutic potential of ABX-EGF: a fully human anti-epidermal growth factor receptor monoclonal antibody for cancer treatment, Semin. Oncol. 29 (2002) 47–50. R.T. Curnow, Clinical experience with CD64-directed immunotherapy. An overview, Cancer Immunol. Immunother. 45 (1997) 210–215. P.K. Wallace, J.L. Romet-Lemonne, M. Chokri, L.H. Kasper, M.W. Fanger, C.E. Fadul, Production of macrophage-activated killer cells for targeting of glioblastoma cells with bispecific antibody to FcgammaRI and the epidermal growth factor receptor, Cancer Immunol. Immunother. 49 (2000) 493–503. D.W. Rusnak, K. Lackey, K. Affleck, E.R. Wood, K.J. Alligood, N. Rhodes, B.R. Keith, D.M. Murray, W.B. Knight, R.J. Mullin, T.M. Gilmer, The effects of the novel, reversible epidermal growth factor receptor/ErbB-2 tyrosine kinase inhibitor, GW2016, on the growth of human normal and tumor-derived cell lines in vitro and in vivo, Mol. Cancer Ther. 1 (2001) 85–94. J.B. Smaill, G.W. Rewcastle, J.A. Loo, K.D. Greis, O.H. Chan, E.L. Reyner, E. Lipka, H.D. Showalter, P.W. Vincent, W.L. Elliott, W.A. Denny, Tyrosine kinase inhibitors. 17. Irreversible inhibitors of the epidermal growth factor receptor: 4-(phenylamino)quinazoline-and 4-(phenylamino)pyrido, J. Med. Chem. 43 (2000) 1380–1397. G.S. Rao, S. Murray, S.P. Ethier, Radiosensitization of human breast cancer cells by a novel ErbB family receptor tyrosine kinase inhibitor, Int. J. Radiat. Oncol. Biol. Phys. 48 (2000) 1519–1528. C.J. Torrance, P.E. Jackson, E. Montgomery, K.W. Kinzler, B. Vogelstein, A. Wissner, M. Nunes, P. Frost, C.M. Discafani, Combinatorial chemoprevention of intestinal neoplasia, Nat. Med. 6 (2000) 1024–1028. C.J. Bruns, C.C. Solorzano, M.T. Harbison, S. Ozawa, R. Tsan, D. Fan, J. Abbruzzese, P. Traxler, E. Buchdunger, R. Radinsky, I.J. Fidler, Blockade of the epidermal growth factor receptor signaling by a novel tyrosine kinase inhibitor leads to apoptosis of endothelial cells and therapy of human pancreatic carcinoma, Cancer Res. 60 (2000) 2926–2935. G. Schwartz, J.P. Dutcher, N.J. Vogelzang, J. Gollob, J. Thompson, R.M. Bukowski, R.A. Figlin, M. Lohner, L. Roskos, C.C. Hwang, K. Foon, G. Schwab, E. Rowinsky, Phase 2 clinical trial evaluating the safety and effectiveness of ABX-EGF in renal cell cancer (RCC), Proc. Am. Soc. Clin. Oncol. 21 (2002) 24a.

Mechanism of action of erbB tyrosine kinase inhibitors David W. Fry Pfizer Global Research and Development, 2800 Plymouth Road, Ann Arbor, MI 48105, USA

Abstract Over the last decade, drug discovery efforts have generated a myriad of compounds that inhibit the activity of the erbB family of tyrosine kinases with potencies and selectivity that have surpassed original expectations. These characteristics, along with improved pharmaceutical properties, have enabled inhibitors from this class of agents to finally realize their therapeutic potential, and indeed, some are currently producing significant clinical responses. Interestingly, those properties that are essential for a clinically active inhibitor of the erbB family are most readily attained with compounds that bind at the ATP site, and the most successful compounds have shown a distinct convergence to certain common chemical features. The reasons for this trend are beginning to be realized through the generation of an increasing array of crystalline structures for protein kinases as well as advances in molecular modeling. This has allowed a more complete understanding of the precise physical interactions that occur between erbB tyrosine kinase inhibitors and their target(s), which, in turn, has begun to shed light on the mechanism by which these molecules attain their remarkable affinity and specificity.

Introduction Inhibition of the erbB receptor kinases has been proposed as a rational approach to cancer chemotherapy for nearly 15 years. Initially, drug discovery programs focused on erbB1, and while these efforts produced a profusion of diverse chemical structures, most of these inhibitors suffered from a lack of potency and specificity, which perhaps compromised attempts to demonstrate antitumor activity based solely on the target [1–5]. The field was greatly enhanced, however, with the identification of 4-anilinoquinazolines, which clearly set a new benchmark for defining potency and specificity for inhibitors of epidermal growth factor (EGF) receptor tyrosine kinase [6–10]. Subsequent to these initial reports, more detailed structural activity relationships were developed that further underscored the potency and selectivity that could be attained with this structural class of molecules [8,9,11,12]. These studies reported IC50 values as low as 6 pM against the EGF receptor tyrosine kinase with unprecedented specificity relative to other receptor and intracellular tyrosine kinases, as well as potent, specific inhibition of EGFmediated processes in viable cells. Eventually, a number of alternative chemical classes were also shown to be equally efficient structural templates including other bicyclic pharmacophores such as pyridopyrimidines [7,13,14], quinoline3-carbonitriles [15], pyrrolopyrimidines [16], and pyrazolopyrimidines [17], as well as tricyclic molecules

such as imidazoloquinazolines, pyrroloquinazolines, and pyrazoloquinazolines [18,19]. Very early discovery approaches to protein kinase inhibition tended to deemphasize ATP-competitive inhibitors for fear that they would lack selectivity due to high sequence homology in the ATP-binding domain of kinases as well as other ATP-utilizing enzymes. In addition, since the Km for ATP of most kinases is usually in the micromolar range, the binding site was thought perhaps not to be the most viable avenue to potent inhibitors. As mentioned above, these notions have been dispelled by dozens of inhibitors that bind with nanomolar or even picomolar affinities in the ATP pockets of these kinases with excellent specificity. Although many different chemical templates have been explored over the years, the more successful compounds possess certain common structural elements that are responsible for the tight-binding affinity to their target(s). Many of these compounds are in various stages of clinical development [20,21] and this review will highlight the erbB receptor interactions of five of the more advanced agents that are in either phase II or phase III clinical trials (Fig. 1). The recently disclosed structure of erbB-1 cocrystallized with Tarceva [22], as well as some of the earlier predictions for inhibitor-enzyme interactions using homology models of erbB1 [15,19,23–25], have shed light on how these inhibitors can attain their remarkable affinity and selectivity. Although many structural similarities exist between the molecules in Fig. 1 that likely provide common elements for interaction with the

138


Fig. 1. Chemical structures for five erbB receptor tyrosine kinase inhibitors that are currently in clinical trials.

erbB receptors, varied substitution patterns also create a diversity that distinguishes each compound by offering additional mechanisms of contact with its target, resulting in different selectivities across the erbB family. Thus, the agents in Fig. 1, although similar in many ways, are actually quite distinct from each other and, as a group, represent a fairly comprehensive arsenal against this kinase family. Information regarding the tertiary structure of protein kinases has increased enormously over the past decade, with over 60 crystal structures having been solved for various tyrosine and serine/threonine kinases [24]. Most have been cocrystalized with ATP or an ATP-competitive inhibitor, which has also helped to improve homology models for those kinases where crystals are not yet available. To date, the only member of the erbB family of tyrosine kinases to be crystallized is erbB1 with or without the erbB1 kinase inhibitor Tarceva bound in the ATP pocket [22]. It should be emphasized that many of the generalizations described below are based on the Tarceva cocrystal model, and although most of the extrapolations to other inhibitors are likely accurate, the true binding modes could deviate depending on the structural modifications of each individual inhibitor. Nevertheless, this review attempts, in a broad sense, to indicate those potential inhibitor-enzyme interactions that could contribute to potency and selectivity for these five inhibitors. A generic 4-anilinoquinazoline has been used for illustrative purposes (Fig. 5).

Common interactions with erbB-1 As determined by crystallography [22], the erbB-1 kinase domain consists of two major lobes separated by a narrow cleft that defines the ATP-binding pocket and the site where ATP-competitive inhibitors interact with the enzyme. The N1-C8 edge of the quinazolines is directed into the groove with the substituents at the C6 and C7 position directed out toward solvent and the aniline pointed into a hydrophobic pocket (Fig. 2). The affinity and specificity for these inhibitors is provided in part by a series of known and proposed hydrogen bonding and hydrophobic interactions. Again, from the cocrystal structure, the amide NH of Met769 points toward N1 of the quinazoline, forming a stable hydrogen bond (Fig. 3). The other ring nitrogen, N3, was originally proposed, in a previously described molecular model, to hydrogen bond to the side-chain hydroxy of Thr766 [19]; however, the erbB-1 crystal structure [22] indicated that the distance between these two atoms was too great (4.1 Å) for a direct interaction. Instead, a water molecule was detected, which bridges the gap between N3 and Thr766. The relative contribution that these ligand/protein hydrogen bonds make to the affinity of the inhibitor can be estimated from previously reported structural activity relationship data on quinazoline inhibitors [19]. Replacement of N1 with carbon resulted in a greater than 3700-fold decrease in inhibitor potency, indicating that this is one of the more important

Mechanism of Action of erbB Tyrosine Kinase Inhibitors

139

Fig. 2. Molecular surface model of erbB-1 showing the ATP-binding pocket and potential binding orientation of a generic quinazoline based on the erbB-1/Tarceva cocrystal.

interactions between quinazoline inhibitors and erbB-1. Replacement of N3 with carbon decreased potency by about 200-fold, indicating a significant but far smaller contribution to the affinity. This is compatible with an expected weaker binding energy when molecular interactions are mediated through a water bridge. The success of the quinoline3-carbonitrile series [15,26], from which EKB-569 was derived (Fig. 1), actually relies on displacement of the bridging water molecule. In this series, N3 has been replaced with a carbon where the resulting C3 position bears a cyano group (Fig. 1) that projects its nitrogen toward Thr766. The reduction in distance between the two molecules allows a direct hydrogen bond thus eliminating the need for a water bridge. Other protein interactions with the quinazoline template are proposed for the C2 and C8 positions. The hydrogen atoms from these ring carbons are directed toward the carbonyl oxygens of Gln767 and Met769, respectively with distances of 3.1 Å and 3.2 Å, thus making these positions likely contact points [22]. Because C2 is positioned between two nitrogens the C-H may be acidic enough to form a hydrogen bond with Gln767. This is not likely to be the case for C8; however, other attractive forces such as van der Waals could play a role at this site. The importance of these locations is again underscored by prior structural activity rela-

tionship work where six examples of substituting C8 with N decreased potency from 1 to 3 orders of magnitude [13]. Additionally, substitution of the hydrogens at C2 or C8 with methyl or methoxy destroyed all inhibition [12]. The dramatic decreases in binding affinity resulting from these modifications are most likely due in part to loss of the proposed hydrogen bonding opportunity; however, it is also possible that unfavorable repulsive electrostatic interactions with the 8-aza compounds, or steric hindrance upon alkyl substitution with the peptide backbone, might also play a role. The 4-aniline, including the 3¢ and 4¢ substitutents, clearly plays an important role in the affinity and selectivity for these inhibitors. The phenyl ring is tilted out of plane from the quinazoline (42° for Tarceva), allowing the substituent at the 3¢ position, which is often a halogen (Cl or Br) or, in the case of Tarceva, an acetylene moiety, to fit more precisely into a well-defined hydrophobic pocket (Fig. 2). This area of the receptor is formed by the side chains of residues Thr766, Cys751, Leu764, Met742, Thr830, Phe832, and part of Lys721. The pocket appears to be utilized heavily by all of the inhibitors in Fig. 1, but not by ATP, which might explain in part the much greater affinity for these inhibitors than the enzyme’s natural substrate. The acetylene moiety of Tarceva was predicted to be less than 4 Å from Thr766, Lys721, and

140


Fig. 3. Molecular model of erbB-1 showing key amino acids that are known to interact or potentially interact with ATP-competitive inhibitors.

Leu764 [22], implying favorable hydrophobic interactions. The 4-(3¢-bromo-aniline) of previously described inhibitors has also been proposed to possibly form favorable interactions with Cys751 and Met742 [19], and since this particular cysteine occurs only in erbB-1, the interaction may also provide an explanation for the selectivity that Iressa and Tarceva exhibit toward this particular kinase (Table 1).

Broadening selectivity to include erbB-2 Although Iressa and Tarceva were designed to selectively inhibit erbB-1, it is clear that solid tumors frequently express multiple members of the erbB family that can act in concert

to produce a more transformed phenotype and a less favorable clinical prognosis in patients with breast, ovarian, and other tumors [27]. Consequently, a more recent approach to exploiting this receptor family as a target in cancer chemotherapy has been to broaden the selectively to include inhibition of multiple members of the erbB receptor family tyrosine kinases. Thus far, this has been accomplished through two different methods. The first approach has been to modify the 4-anilino substituents from the relatively small 3¢-ethynylaniline (Tarceva) or 3¢-chloroaniline-4¢-fluoro (Iressa) to much larger groups. Surprisingly, a wide range of bulky groups off the 4¢ position of the aniline are not only tolerated by erbB-1 but confer inhibitory activity against other members of the erbB family, especially erbB2 [28–

Mechanism of Action of erbB Tyrosine Kinase Inhibitors Table 1 Inhibition of erbB family tyrosine kinases by agents in advanced clinical trialsa Inhibitor

ErbB-1

ErbB-2

ErbB-4

Reference

Iressa Tarceva GW2016 CI-1033 EKB-569

27 2.9 11 0.8 38

3700 1050 9 17 1255

Not reported Not reported 367 7 Not reported

[60] [29] [61] [32] [40]

a

Numbers are IC50 values in nanomolar units.

30]. GW2016 contains 3¢-chloro-4¢-[(3-fluorobenzyl)oxy] aniline, which retains excellent potency against erbB1 but at the same time has equal potency against erbB2 (Table 1). This substitution also confers some inhibitory activity against erbB4, although with 30-fold less potency. The mechanism by which this modification allows dual specificity for erbB1 and erbB2 is not entirely understood yet. The fact that erbB1 can tolerate such dramatic changes in the molecule might indicate that the binding mode for GW2016 is different from the erbB1/Tarceva structure. However, if one assumes the basic orientation of the quinazoline is similar, then the hydrophobic pocket in erbB1 must accommodate the increased bulkiness, possibly through conformational changes in the molecule and/or the protein. Either way, the data indicate that the benzyloxy moiety must interact with certain structural components of erbB2 in a way that is evidently not possible with the unsubstituted aniline. One such 4¢-benzyloxy derivative has been modeled in another kinase, p38 [30]. The benzyloxyaniline was indeed accommodated in the back of the hydrophobic pocket and from this model analogies were proposed between this kinase structure and that of erbB2 through sequence homology that exists between the two kinases. A more accurate explanation for the selectivity pattern of this chemical series, however, will have to await either a crystal structure of erbB2 or at least homology models based on the current erbB-1 structure.

Irreversible inhibitors Another approach to expanding inhibitory activity to other members of the erbB family has exploited a specific cysteine in the ATP-binding domain. Cys773 is positioned at a fortuitous location in the ATP pocket of erbB1 that, in conjunction with the specific binding mode for quinazolines and other chemical templates, allows the substituent at the 6 position to approach the sulfhydryl of this amino acid to within less than 3 Å as proposed by molecular modeling [23]. Substitution of a mild alkylating moiety at the 6 position such as the acrylamide in CI-1033 (Fig. 1) brings the electrophilic-carbon atom into close proximity with the nucleophilic thiol atom of Cys773, which facilitates the rapid

141

formation of an addition product [23]. Although the concept and practice of making irreversible inhibitors has existed for decades, the property that distinguishes these inhibitors from previous ATP site-directed irreversible inhibitors is their selectivity. One of the concerns with this approach has always been a potential for nonselective interaction with proteins other than the proposed target. The intrinsic reactivity of CI-1033, however, is low enough such that it will not interact nonspecifically with random nucleophiles. As a result it reacts only when bound into the ATP pocket after the reactive end of the acrylamide has been brought into the immediate vicinity of the nucleophilic Cys773, thus maximizing the probability of a reaction. This selectivity is borne out by the fact that CI-1033 has little or no activity against a panel of other protein kinases [31]. This feat has been accomplished with several different chemical templates including quinazolines, pyridopyrimidines, and quinolinecarbonitriles [23,32–35] and with a variety of alkylating sidechains at the C6 position including acrylamide, butynamide, and crotylamide [34]. EKB-569, a member of the quinolinecarbonitrile series, also exploits this mechanism through C6 substitution of dimethylamino-2-butenamide (Fig. 1). The rationale for irreversible inhibition was originally based on the potential to provide additional therapeutic benefit through prolonged target suppression and the ability to achieve adequate drug exposure with shorter plasma halflives and lower Cmax values [36]. However, another potential advantage is the ability for irreversible agents to inhibit other members of the erbB family. An examination of the sequences for erbB1, 2, 3, and 4 [37] shows that cysteines are also present in erbB-2 and erbB-4 that are analogous to Cys773 in erbB-1 (Fig. 4). This provides the same opportunity for irreversible inhibition of these kinases as has been shown for erbB-1 and, indeed, CI-1033 has been shown to have equivalent activity against erbB1, 2, and 4 [31] (Table 1). ErbB-3 contains a serine at that position (Fig. 4), but as has been reported previously [38], this receptor has little or no kinase activity and participates in signaling through this receptor family via heterodimerization and transphosphorylation with the other erbB members [27]. Reported IC50 values for EKB-569 against erbB1 and erbB2 show that this irreversible inhibitor is not as potent as CI-1033 against erbB2 [39] (Table 1). However, as discussed above, this compound differs from the quinazoline template in that N3 has been replaced by carbon with the deficiency being compensated by a cyano group at C3. As described in earlier studies of irreversible inhibitors, small changes in the position of the alkylating side chain can have profound effects on the reaction efficiency with the cysteine(s) [23]. Thus, it is conceivable that the quinolin-3-carbonitriles may deviate from quinazolines in binding orientation within erbB2 in a way that makes irreversible alkylation less favorable. The potential interactions of quinazoline inhibitors with erbB-1 that have been discussed above have been summarized in Fig. 5.

142


Fig. 4. Partial amino acid sequences of erbB-1, 2, 3, and 4 surrounding the cysteine targeted by irreversible inhibitors. The amino acid numbering does not include the leader sequence.

Other mechanisms Although the structural design of the inhibitors in Table 1 was originally directed exclusively toward inhibition of the catalytic activity of erbB receptors, recent data indicate that secondary mechanisms may also contribute to inactivation of this receptor family and its associated signal transduction pathways. Thus, another effect of kinase inhibition that may have therapeutic consequences is inhibitormediated degradation of the receptors. Two major pathways have been described for degradation of erbB receptors. The first is ligand-induced endocytosis and degradation of erbB1 through clathrin-coated pits [40–42] and antibody- or oncogenic mutation-induced degradation of erbB2, both occurring at least in part through c-Cbl ubiquitin ligase [43,44]. Since this mechanism of degradation appeared to require kinase activity and tyrosine phosphorylation of the Cterminal tails of the receptors [43,45], erbB kinase inhibitors would most likely inhibit this pathway. However, there is another mechanism of receptor downregulation that occurs in response to stress, perturbations in structural integrity, or suppression of kinase activity that involves shuffling of chaperone complexes associated with the receptor with subsequent endocytosis, ubiquitylation, and degradation [46,47]. This process, which is mediated by proteasomal proteinases, involves disruption of a complex containing Hsp90 and recruitment of Hsp70. The observation that

kinase dead mutants of erbB1 are preferentially degraded through this pathway in response to geldanamycin [46] led to the possibility that ATP-competitive inhibitors of the kinase domain may also promote this pathway. Recent studies have shown that inhibitors of the catalytic activity of erbB2 do indeed promote downregulation and degradation of erbB2 through the pathway described for geldanamycin and analogs [47,48], but in a specific manner, by binding selectively to the ATP-binding domain. This effect appears to be more robust for irreversible inhibitors. As an example, when N87 gastric carcinoma cells were incubated with CI1033, there was an enhancement of erbB-2 ubiquitylation with subsequent degradation of erbB2 resulting in a decrease in receptor expression levels [46]. Similar experiments with a reversible congener of CI-1033, in which the double bond of the acrylamide was reduced, had a much lesser effect. This could be interpreted to mean that the alkylation process, and thus permanent structural modification of erbB-2, is the major recognition feature for degradation. The contribution that degradation of erbB2 makes to the therapeutic effects of CI-1033 is not fully known; however, the concept is viable since downregulation of erbB2 has been proposed as part of the mechanism for other therapeutic agents such as Herceptin [43,49,50] and geldanamycin [51,52]. Another mechanism by which erbB kinase inhibitors may impede signal transduction via this receptor family is an interference with the dynamics of the signaling process.

Fig. 5. Generic quinazoline illustrating the main positions on the molecule that interact with erbB-1.

Mechanism of Action of erbB Tyrosine Kinase Inhibitors

Homo- and heterodimerization between members of the erbB family clearly play an important role in signal transduction through this receptor family [27,53]. Certain evidence indicates that small molecular weight erbB tyrosine kinase inhibitors may hinder signaling by disruption or alteration of this natural dimerization process. This notion was encouraged by the unexpected finding that Iressa, which enzymatically appears to be an erbB1-specific tyrosine kinase inhibitor, was quite active against erbB2-overexpressing tumor cells [54–56]. Furthermore, another erbB-1selective quinazoline, AG-1478, delayed mammary tumor formation and increased survival time of bigenic mice that were infected with MMTV/Neu + MMTV/transforming growth factor-a [57]. Although these data are not entirely understood, an earlier report had shown that small molecular weight quinazoline inhibitors, AG-1478 and AG-1517, induced inactive homodimers of EGFr and erbB-2 as well as inactive heterodimers of EGFr/erbB2 [58]. It was proposed that this effect may decrease the population of active receptors on a tumor cell and thus contribute to the ability of these compounds to reduce signaling through this receptor family. Finally, more recent evidence is beginning to emerge indicating that Iressa can induce the formation of inactive-unphosphorylated erbB1/erbB2 and erbB1/erbB3 heterodimers, which sequester erbB2 and prevent formation of active phosphorylated erbB2/erbB3 dimers [59].

Acknowledgments

[9]

[10]

[11]

[12]

[13]

[14]

[15]

I would like to give special thanks to James Dunbar for generating the molecular modeling images and to Peter L. Toogood for critical reading of the manuscript.

References [1] D.W. Fry, Protein tyrosine kinases as therapeutic targets in cancer chemotherapy and recent advances in the development of new inhibitors, Exp. Opin. Invest. Drugs 3 (1994) 577–595. [2] A. Levitzki, A. Gazit, Tyrosine kinase inhibition: an approach to drug development, Science 267 (1995) 1782–1788. [3] T.R. Burke, Protein-tyrosine kinase inhibitors, Drugs Fut. 17 (1992) 119–131. [4] A. Levitzki, Tyrphostins: tyrosine kinase blockers as novel antiproliferative agents and dissectors of signal transduction, FASEB J. 6 (1992) 3275–3282. [5] D.S. Lawrence, J.K. Niu, Protein kinase inhibitors: the tyrosinespecific protein kinases, Pharmacol. Ther. 77 (1998) 81–114. [6] W.H.J. Ward, P.N. Cook, A.M. Slater, H. Davies, G.A. Holdgate, L.R. Green, Epidermal growth factor receptor tyrosine kinase. Investigation of catalytic mechanism, structure-based searching and discovery of a potent inhibitor, Biochem. Pharmacol. 48 (1994) 659–666. [7] A.M. Thompson, A.J. Bridges, D.W. Fry, A.J. Kraker, W.A. Denny, Tyrosine kinase inhibitors. 7. 7-amino-4-(phenylamino)- and 7amino-4-[(phenylmethyl)amino]pyrido[4,3-d]pyrimidines: a new class of inhibitors of the tyrosine kinase activity of the epidermal growth factor receptor, J. Med. Chem. 38 (1995) 3780–3788. [8] G.W. Rewcastle, W.A. Denny, A.J. Bridges, H.R. Zhou, D.R. Cody, A. McMichael, D.W. Fry, Tyrosine kinase inhibitors. 5. Synthesis and

[16]

[17]

[18]

[19]

[20]

[21]

143

structure-activity relationships for 4-[(phenylmethyl)amino]- and 4-(phenylamino)quinazolines as potent adenosine 5¢-triphosphate binding site inhibitors of the tyrosine kinase domain of the epidermal growth factor receptor, J. Med. Chem. 38 (1995) 3482–3487. K.H. Gibson, W. Grundy, A.A. Godfrey, J.R. Woodburn, S.E. Ashton, B.J. Curry, L. Scarlett, A.J. Barker, D.S. Brown, Epidermal growth factor receptor tyrosine kinase: structure-activity relationships and antitumour activity of novel quinazolines, Bioorg. Med. Chem. Lett. 7 (1997) 2723–2728. D.W. Fry, A.J. Kraker, A. McMichael, L.A. Ambroso, J.M. Nelson, W.R. Leopold, R.W. Connors, A.J. Bridges, Specific Inhibitor of the epidermal growth factor receptor tyrosine kinase, Science 265 (1994) 1093–1095. W.A. Denny, G.W. Rewcastle, A.J. Bridges, D.W. Fry, A.J. Kraker, Structure-activity relationships for 4-anilinoquinazolines as potent inhibitors at the ATP binding site of the epidermal growth factor receptor in vitro, Clin. Exper. Pharmacol. Physiol. 23 (1996) 424–427. A.J. Bridges, H. Zhou, D.R. Cody, G.W. Rewcastle, A. McMichael, H.D.H. Showalter, D.W. Fry, A.J. Kraker, W.A. Denny, Tyrosine kinase inhibitors. 8. An unusually steep structure-activity relationship for analogues of 4-(3-bromoanilino)-6,7-dimethoxyquinazoline (PD 153035), a potent inhibitor of the epdiermal growth factor receptor, J. Med. Chem. 39 (1996) 267–276. G.W. Rewcastle, B.D. Palmer, A.M. Thompson, A.J. Bridges, D.R. Cody, H. Zhou, D.W. Fry, A. McMichael, W.A. Denny, Tyrosine kinase inhibitors. 10. Isomeric 4-[(3-bromophenyl)amino]pyrido[d]pyrimidines are potent ATP binding site inhibitors of the tyrosine kinase function of the epidermal growth factor receptor, J. Med. Chem. 39 (1996) 1823–1835. G.W. Rewcastle, A.J. Bridges, D.W. Fry, J.R. Rubin, W.A. Denny, Tyrosine kinase inhibitors. 12. Synthesis and structure-activity relationships for 6-substituted 4-(phenylamino)pyrimido-[5,4-d]pyrimidines designed as inhibitors of the epidermal growth factor receptor, J. Med. Chem. 40 (1997) 1820–1826. A. Wissner, D.M. Berger, D.H. Boschelli, M.B. Floyd, L.M. Greenberger, B.C. Gruber, B.D. Johnson, N. Mamuya, R. Nilakantan, M.F. Reich, R. Shen, H.R. Tsou, E. Upeslacis, Y.F. Wang, B.Q. Wu, F. Ye, N. Zhang, 4-Anilino-6,7-dialkoxyquinolime-3-carbonitrile inhibitors of epidermal g th factor receptor kinase and their bioisosteric relationship to the 4-anilino-6,7-dialkoxyquinazoline inhibitors, J. Med. Chem. 43 (2000) 3244–3256. P.M. Traxler, P. Furet, H. Mett, E. Buchdunger, T. Meyer, N. Lydon, 4-(Phenylamino)pyrrolopyrimidines: potent and selective, ATP site directed inhibitors of the EGF-receptor protein tyrosine kinase, J. Med. Chem. 39 (1996) 2285–2292. P. Traxler, G. Bold, J. Frei, M. Lang, N. Lydon, H. Mett, E. Buchdunger, T. Meyer, M. Mueller, P. Furet, Use of a pharmacophore model for the design of EGF-R tyrosine kinase inhibitors: 4-(Phenylamino)pyrazolo[3,4-d]pyrimidines, J. Med. Chem. 40 (1997) 3601–3616. G.W. Rewcastle, B.D. Palmer, A.J. Bridges, H.D.H. Showalter, S. Li, J. Nelson, A. Mcmichael, A.J. Kraker, D.W. Fry, W.A. Denny, Tyrosine kinase inhibitors. 9. Synthesis and evaluation of fused tricyclic quinazoline analogues as ATP site inhibitors of the tyrosine kinase activity of the epidermal growth factor receptor, J. Med. Chem. 39 (1996) 918–928. B.D. Palmer, S. Trumpp-Kallmeyer, D.W. Fry, J.M. Nelson, H.D.H. Showalter, W.A. Denny, Tyrosine kinase inhibitors. 11. Soluble analogues of pyrrolo- and pyrazoloquinazolines as epidermal growth factor receptor inhibitors: synthesis, biological evaluation, and modeling of the mode of binding, J. Med. Chem. 40 (1997) 1519–1529. G.S. Cockerill, K.E. Lackey, Small molecule inhibitors of the class 1 receptor tyrosine kinase family, Curr. Top. Med. Chem. 2 (2002) 1001–1010. A.E. Wakeling, Epidermal growth factor receptor tyrosine kinase inhibitors, Curr. Opin. Pharmacol. 2 (2002) 382–387.

144


[22] J. Stamos, M.X. Sliwkowski, C. Eigenbrot, Structure of the EGF receptor kinase domain alone and in complex with a 4-anilinoquinazoline inhibitor, J. Biol. Chem. 277 (2002) 46265–46272. [23] D.W. Fry, A.J. Bridges, W.A. Denny, A. Doherty, K.D. Greis, J.L. Hicks, K.E. Hook, P.R. Keller, W.R. Leopold, J.A. Loo, D.J. McNamara, J.M. Nelson, V. Sherwood, J.B. Smaill, S. TrumppKallmeyer, E.M. Dobrusin, Specific irreversible inactivation of the epidermal growth factor receptor and erbB2, by a new class of tyrosine kinase inhibitor, Proc. Natl. Acad. Sci. USA 95 (1998) 12022–12027. [24] L.M. Toledo, N.B. Lydon, D. Elbaum, The structure-based design of ATP-site directed protein kinase inhibitors, Curr. Med. Chem. 6 (1999) 775–805. [25] G. McMahon, L. Sun, C. Liang, C. Tang, Protein kinase inhibitors: structural determinants for target specificity, Curr. Opin. Drug Dis. Dev. 1 (1998) 131–146. [26] D.H. Boschelli, 4-Anilino-3-quinolinecarbonitriles: an emerging class of kinase inhibitors, Curr. Top. Med. Chem. 2 (2002) 1051–1063. [27] M.A. Olayioye, R.M. Neve, H.A. Lane, N.E. Hynes, The ErbB signaling network: receptor heterodimerization in development and cancer, EMBO J. 19 (2000) 3159–3167. [28] D.W. Rusnak, K. Affleck, S.G. Cockerill, C. Stubberfield, R. Harris, M. Page, K.J. Smith, S.B. Guntrip, M.C. Carter, R.J. Shaw, A. Jowett, J. Stables, P. Topley, E.R. Wood, P.S. Brignola, S.H. Kadwell, B.R. Reep, R.J. Mullin, K.J. Alliggood, B.R. Keith, R.M. Crosby, D.M. Murray, W.B. Knight, T.M. Glimer, K. Lackey, The characterization of novel, dual erbB-2/EGFR, tyrosine kinase inhibitors: potential therapy for cancer, Cancer Res. 61 (2001) 7196–7203. [29] P.S. Brignola, K. Lackey, S.H. Kadwell, C. Hoffman, E. Horne, H.L. Carter, J.D. Stuart, K. Blackburn, M.B. Moyer, K.J. Alligood, W.B. Knight, E.R. Wood, Comparison of the biochemical and kinetic properties of the type 1 receptor tyrosine kinase intracellular domains. Demonstration of differential sensitivity to kinase inhibitors, J. Biol. Chem. 277 (2002) 1576–1585. [30] S. Cockerill, C. Stubberfield, J. Stables, M. Carter, S. Guntrip, K. Smith, S. McKeown, R. Shaw, P. Topley, L. Thomsen, K. Affleck, A. Jowett, D. Hayes, M. Willson, P. Woollard, D. Spalding, Indazolylamino quinazolines and pyridopyrimidines as inhibitors of the EGFr and C-erbB-2, Bioorg. Med. Chem. Lett. 11 (2001) 1401–1405. [31] L.F. Allen, P.F. Lenehan, I.A. Eiseman, W.L. Elliott, D.W. Fry, Potential benefits of the irreversible pan-erbB inhibitor, CI-1033, in the treatment of breast cancer, Semin. Oncol. 29 (2002) 11–21. [32] J.B. Smaill, B.D. Palmer, G.W. Rewcastle, W.A. Denny, D.J. McNamara, E.M. Dobrusin, A.J. Bridges, H.R. Zhou, H.D.H. Showalter, R.T. Winters, W.R. Leopold, D.W. Fry, J.M. Nelson, V. Slintak, W.L. Elliot, B.J. Roberts, P.W. Vincent, S.J. Patmore, Tyrosine kinase inhibitors. 15. 4-(Phenylamino)quinazoline and 4-(phenylamino)pyrido[d] pyrimidine acrylamides as irreversible inhibitors of the ATP binding site of the epidermal growth factor receptor, J. Med. Chem. 42 (1999) 1803–1815. [33] J.B. Smaill, G.W. Rewcastle, A.J. Bridges, H.R. Zhou, H.D.H. Showalter, D.W. Fry, J.M. Nelson, V. Sherwood, W.L. Elliott, P.W. Vincent, D.E. DeJohn, J.A. Loo, K.D. Greis, O.H. Chan, E.L. Reyner, E. Lipka, W.A. Denny, Tyrosine kinase inhibitors. 17. Irreversible inhibitors of the epidermal growth factor receptor: 4-(phenylamino)quinazoline- and 4-(phenylamino)pyrido[3,2-d]-pyrimidine6-acrylamides bearing additional solubilizing functions (vol 43, pg 1380, 2000), J. Med. Chem. 43 (2000) 1380–1397. [34] H.R. Tsou, N. Mamuya, B.D. Johnson, M.F. Reich, B.C. Gruber, F. Ye, R. Nilakantan, R. Shen, C. Discafani, R. DeBlanc, R. Davis, F.E. Koehn, L.M. Greenberger, Y.F. Wang, A. Wissner, 6-Substituted-4-(3bromophenylamino) quinazolines as putative irreversible inhibitors of the epidermal growth factor receptor (EGFR) and human epidermal growth factor receptor (HER-2) tyrosine kinases with enhanced antitumor activity, J. Med. Chem. 44 (2001) 2719–2734. [35] C.M. Discafani, M.L. Carroll, M.B. Floyd Jr., I.J. Hollander, Z. Husain, B.D. Johnson, D. Kitchen, M.K. May, M.S. Malo, A.A. Minnick Jr, R. Nilakantan, R. Shen, Y.F. Wang, A. Wissner, L.M.

[36]

[37] [38]

[39]

[40] [41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

Greenberger, Irreversible inhibition of epidermal growth factor receptor tyrosine kinase with in vivo activity by N-[4-[(3-bromophenyl) amino]-6-quinazolinyl]-2-butynamide (CL-387,785), Biochem. Pharmacol. 57 (1999) 917–925. D.W. Fry, Site-directed irreversible inhibitors of the erbB family of receptor tyrosine kinases as novel chemotherapeutic agents for cancer, Anticancer Drug Des. 15 (2000) 3–16. R.A. Stein, J.V. Staros, Evolutionary analysis of the ErbB receptor and ligand families, J. Mol. Evol. 50 (2000) 397–412. P.M. Guy, J.V. Platko, L.C. Cantley, R.A. Cerione, K.L. Carraway, Insect cell-expressed p180(erbB3) possesses an impaired tyrosine kinase activity, Proc. Natl. Acad. Sci. USA 91 (1994) 8132–8136. C.J. Torrance, P.E. Jackson, E. Montgomery, K.W. Kinzler, B. Vogelstein, A. Wissner, M. Nunes, P. Frost, C.M. Discafani, Combinatorial chemoprevention of intestinal neoplasia, Nature Med. 6 (2000) 1024–1028. G. Carpenter, The EGF receptor: a nexus for trafficking and signaling, BioEssays 22 (2000) 697–707. H. Waterman, Y. Yarden, Molecular mechanisms underlying endocytosis and sorting of ErbB receptor tyrosine kinases, FEBS Lett. 490 (2001) 142–152. A. Sorkin, T. McKinsey, W. Shih, T. Kirchhausen, G. Carpenter, Stoichiometric interaction of the epidermal growth factor receptor with the clathrin-associated protein complex AP-2, J. Biol. Chem. 270 (1995) 619–625. L.N. Klapper, H. Waterman, M. Sela, Y. Yarden, Tumor-inhibitory antibodies to HER-2/erbB-2 may act by recruiting c-cbl and enhancing ubiquitination of HER-2, Cancer Res. 60 (2000) 3384–3388. G. Levkowitz, S. Oved, L.N. Klapper, D. Harari, S. Lavi, M. Sela, Y. Yarden, c-Cbl is a suppressor of the neu oncogene, J. Biol. Chem. 275 (2000) 35532–35539. G. Levkowitz, H. Waterman, S.A. Ettenberg, M. Katz, A.Y. Tsygankov, I. Alroy, S. Lavi, K. Iwai, Y. Reiss, A. Ciechanover, S. Lipkowitz, Y. Yarden, Ubiquitin ligase activity and tyrosine phosphorylation underlie suppression of growth factor signaling by c-Cbl/Sli1, Mol. Cell. 4 (1999) 1029–1040. A. Citri, I. Alroy, S. Lavi, C. Rubin, W. Xu, N. Grammatkakis, C. Patterson, L. Neckers, D.W. Fry, Y. Yarden, Drug-induced ubiquitylation and degradation of ErbB receptor tyrosine kinase: simplications for cancer therapy, EMBO J. 21 (2002) 2407–2417. W. Xu, E. Mimnaugh, M.F. Rosser, C. Nicchitta, M. Marcu, Y. Yarden, L. Neckers, Sensitivity of mature Erbb2 to geldanamycin is conferred by its kinase domain and is mediated by the chaperone protein Hsp90, J. Biol. Chem. 276 (2001) 3702–3708. O. Tikhomirov, G. Carpenter, Geldanamycin induces ErbB-2 degradation by proteolytic fragmentation, J. Biol. Chem. 275 (2000) 26625–26631. M. Cuello, S.A. Ettenberg, A.S. Clark, M.M. Keane, R.H. Posner, M.M. Nau, P.A. Dennis, S. Lipkowitz, Down-regulation of the erbB2 receptor by trastuzumab (herceptin) enhances tumor necrosis factorrelated apoptosis-inducing ligand-mediated apoptosis in breast and ovarian cancer cell lines that overexpress erbB-2, Cancer Res. 61 (2001) 4892–4900. M.X. Sliwkowski, J.A. Lofgren, G.D. Lewis, T.E. Hotaling, B.M. Fendly, J.A. Fox, Nonclinical studies addressing the mechanism of action of trastuzumab (Herceptin), Semin. Oncol. 26 (1999) 60–70. R.C. Schnur, M.L. Corman, R.J. Gallaschun, B.A. Cooper, M.F. Dee, J.L. Doty, M.L. Muzzi, J.D. Moyer, C.I. Diorio, E.G. Barbacci, P.E. Miller, A.T. Obrien, M.J. Morin, B.A. Foster, V.A. Pollack, D.M. Savage, D.E. Sloan, L.R. Pustilnik, M.P. Moyer, Inhibition of the oncogene product p185(erbB-2) in vitro and in vivo by geldanamycin and dihydrogeldanamycin derivatives, J. Med. Chem. 38 (1995) 3806–3812. F.F. Zheng, S.D. Kuduk, G. Chiosis, P.N. Munster, L. Sepp Lorenzino, S.J. Danishefsky, N. Rosen, Identification of a geldanamycin dimer that induces the selective degradation of HER-family tyrosine kinases, Cancer Res. 60 (2000) 2090–2094.

Mechanism of Action of erbB Tyrosine Kinase Inhibitors [53] Y. Yarden, M.X. Sliwkowski, Untangling the erbB signalling network, Nature Rev. 2 (2001) 127–136. [54] M.M. Moasser, A. Basso, S.D. Averbuch, N. Rosen, The tyrosine kinase inhibitor ZD1839 (“Iressa”) inhibits HER2-driven signaling and suppresses the growth of HER2-overexpressing tumor cells, Cancer Res. 61 (2001) 7184–8188. [55] F. Rojo, J. Albanell, J. Anido, I. Rodriguez-Viltro, M. Guzman, S. Averbuch, J. Baselga, Dose dependent pharmacodynamic effects of ZD1839 (“Iressa”) correlate with tumor growth inhibition in BT-474 breast cancer xenografts, Proc. Am. Assoc. Cancer Res. 43 (2002) Abstract 3893. [56] S.L. Moulder, F.M. Yakes, S.K. Muthuswamy, R. Bianco, J.F. Simpson, C.L. Arteaga, Epidermal growth factor receptor (HER1) tyrosine kinase inhibitor ZD1839 (Iressa) inhibits HER2/neu (erbB2)overexpressing breast cancer cells in vitro and in vivo, Cancer Res. 61 (2001) 8887–8895. [57] A.E.G. Lenferink, J.F. Simpson, L.K. Shawver, R.J. Coffey, J.T. Forbes, C.L. Arteaga, Blockade of the epidermal growth facctor receptor tyrosine kinase suppresses tumorigenesis in MMTV/Neu + MMTV/TGF-alpha bigenic mice, Proc. Natl. Acad. Sci USA 97 (2000) 9609–9614.

145

[58] C.L. Arteaga, T.T. Ramsey, L.K. Shawver, C.A. Guyer, Unliganded epidermal growth factor receptor dimerization induced by direct interaction of quinazolines with the ATP binding site, J. Biol. Chem. 272 (1997) 23247–23254. [59] J. Anido, J. Albanell, F. Rojo, J. Arribas, S. Averbuch, J. Baselga, ZD1839 (“Iressa”), a specific epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor, induces the formation of inactive EGFR/HER2 and EGFR/HER3 heterodimers and prevents heregulin signaling in HER2 overexpressing breast cancer cells, Proc. Am. Assoc. Cancer Res. 43 (2002) Abstract 3881. [60] A.E. Wakeling, S.P. Guy, J.R. Woodburn, S.E. Ashton, B.J. Curry, A.J. Barker, K.H. Gibson, ZD1839 (Iressa)an orally active inhibitor of epidermal growth factor signaling with potential for cancer therapy, Cancer Res. 62 (2002) 5749–5754. [61] D.W. Rusnak, K. Lackey, D. Affleck, E.R. Wood, K.J. Alligood, N. Rhodes, B.R. Keith, D.M. Murray, W.B. Knight, R.J. Mullin, T.M. Gilmer, The effects of the novel, reversible epidermal growth factor receptor/erbB-2 tyrosine kinase inhibitor, GW2016, on the growth of human normal and tumor-derived cell lines in vitro and in vivo, Mol. Cancer. Ther. 1 (2001) 85–94.

Signaling by the Drosophila epidermal growth factor receptor pathway during development Ben-Zion Shilo Department of Molecular Genetics, Weizmann Institute of Science, Rehovot 76100, Israel

Abstract In 1997 we wrote a review entitled “A thousand and one roles for the Drosophila epidermal growth factor (EGF) receptor (DER/EGFR).” We are not there yet in terms of the number of developmental roles assigned to this receptor in Drosophila. Nevertheless, DER has certainly emerged as one of the key players in development, since it is used repeatedly to direct cell fate choices, cell division, cell survival, and migration. A battery of activating ligands and an inhibitory ligand achieves this versatility. For the ligands that are produced as membrane-bound precursors, trafficking and processing are the key regulatory steps, determining the eventual temporal and spatial pattern of receptor activation. In most cases DER is activated at a short range, in the cells adjacent to the ones producing the active ligand. This activation dictates a binary choice. In some instances DER is also activated over a longer range, and multiple cell fate choices may be induced, according to its level of activation. A battery of negative feedback loops assures the limited range of DER induction. The distinct responses to DER activation in the different tissues depend upon combinatorial interactions with other signaling pathways and tissue-specific factors, at the level of target-gene regulation.

Introduction The Drosophila epidermal growth factor (EGF) receptor (DER/EGFR) is a single member of the EGFR/ErbB family in the fly genome. The DER protein is similar to the mammalian family members in overall structure. At the extracellular region it has the typical four domains, including two cysteine-rich domains, required for ligand binding. Similar to the C. elegans receptor Let-23, the juxta-membrane cysteine-rich domain (domain IV) is duplicated in DER [1]. The signal peptide and extreme N-terminus is represented in alternative splice forms, encoding two different protein isoforms. However, it is not clear whether there is a distinct role to each of the forms [2]. Expression of DER per se is not a critical regulatory step, as the receptor is broadly expressed during development [3]. The ligands activating the receptor and the signals it transduces represent one of the key channels of communication between cells during development. This review will begin with a survey of the diverse roles carried out by DER during fly development. We will then discuss the versatility of ligand structure and regulation as key factors in providing diverse modes of receptor activation. Emphasis will be placed on the central regulatory events in ligand activation, namely trafficking and processing. The capacity to activate DER in a restricted spatial domain depends largely on a set

of negative feedback loops that will be discussed. Finally, interpretation of the signals of receptor activation in a tissuespecific context will be explored.

Multiple roles during development The multitude of roles played by DER during development have complicated the identification of its developmental roles by simple analysis of loss of function phenotypes during embryonic or postembryonic stages [4]. Instead, the use of dominant negative receptor constructs, temperature-sensitive or hypomorphic mutations in the receptor, or mutations in distinct ligands identified discrete roles. The list of 30 odd distinct roles for DER is presented in Table 1. These roles encompass the induction of cell fates in a multitude of contexts, triggering cell proliferation, and even guidance of cell migration. For the most part, the responses to DER activation are manifested by the induction of sets of target genes. In addition to the diversity in the final outcome, tight and distinct regulation of the duration of activation, as well as the strength and range of activation, is essential to assure the correct developmental response. We will outline below some of the regulatory features providing this diversity.

148


Table 1 Roles of DER during development DER function Embryogenesis Patterning the neuroectoderm during gastrulation Patterning the ventral ectoderm Specification of muscle precursors Specification of tracheal invagination and branching Cell recruitment to chordotonal organs Specification of oenocytes Segmentation Patterning the dorsal midline Formation and proliferation of malpigian tubules Germ band retraction Tip cell invagination in the stomatogastric nervous system Muscle attachment to tendons Viability of midline cells Imaginal discs Determination of wing vs. leg disc Determination of notum vs. wing pouch field, and wing dorsal compartment Determination of eye vs. antennal disc Wing disc Determination of veins Inter vein identity Eye disc Spacing R8 photoreceptors Proliferation anterior to the furrow and viability Induction of the second mitotic wave Neuronal, cone, and pigment cell determination Brain Neuronal differentiation in lamina Leg Induction of bract cells Proximal-distal patterning Spermatogenesis Restriction of stem cell renewal capacity in somatic cyst cells Encapsulation of germ cells by somatic support cells Oogenesis Encapsulation in germ cells by follicle cells in germarium Determination of posterior follicle cells Migration of border cells Determination of dorsal follicle cells Patterning the dorsal appendages Abdomen Dorsoventral patterning of the adult abdomen

Stages

Ligands

References

5, 6 10, 11 11 11, 12 11 11 12, 13 8 12–15 12, 13 13 15 14–16

Spitz Spitz, Vein, Argos Spitz, Vein, Argos Spitz Spitz, Argos Spitz, Argos Spitz, Argos Spitz, Argos Spitz ? Spitz Vein Spitz, Argos

[62–64] [6,19,46] [16,65] [66,67] [5,68,69] [70,71] [72–74] [75] [76,77] [78] [79] [16] [78,80–82]

E11 2nd instar 2nd instar

Spitz Vein

[83] [84–87] [88]

3rd instar Pupa

Spitz, Keren?

[37,89–91] [89]

3rd instar 3rd instar 3rd instar 3rd instar and pupa

Keren? Spitz Spitz, Argos

[92] [92–94] [95] [96,97]

Pupa

Spitz, Argos

[98]

Pupa Pupa

Spitz, Argos Vein

[99] [17,18]

Adult

Spitz ?

[100] [44]

Adult

? Gurken Gurken (in collaboration with PVF1/PVR) Gurken Gurken, Spitz, Argos

[44] [9,11] [58] [10,101] [20,40,102]

?

[103]

Adult

DER, Drosophila epidermal growth factor (EGF) receptor.

Five DER ligands provide versatile modes of DER activation The presence of four activating ligands and one inhibitory ligand allows versatile combinations of DER activation. Three of the ligands, Spitz, Keren, and Gurken, are produced as transmembrane precursors. The primary activating ligand is Spitz, a transforming growth factor-a (TGFa) homologue that is responsible for DER activation in most tissues [5]. As described below, the active, secreted form of Spitz is produced by tightly regulated cleavage of the membrane-bound precursor [6]. A ligand structurally related to Spitz has recently been identified and termed Keren [7,8]. In general, this ligand is regulated in a similar manner to

Spitz, and may complement its activity in certain tissues. In contrast to Spitz, the membrane-bound precursor of Keren can undergo unregulated low-level cleavage that may be utilized in several tissues [7]. Gurken, a third TGF-a homologue, is restricted to the activation of DER in the follicle cells of the ovary [9–11]. Gurken is tightly regulated at several levels. First, it is transcribed only in the germline cells of the ovary. Second, regulatory sequences on the gurken transcript restrict the localization of the RNA to the vicinity of the oocyte nucleus [10]. Finally, the Gurken protein is concentrated accordingly [12]. Tight localization of gurken transcript and protein play an instructive role in induction of dorsal follicle cells fates [13].

Signaling by the Drosophila Epidermal Growth Factor Receptor Pathway During Development

Vein, a secreted ligand, possesses an inherently weaker activation capacity, and is used in tissues where low activation levels are required [14,15]. In some tissues Vein functions as the main ligand. It induces the muscle attachment cell fate, following its accumulation at the receiving cell [16]. In the leg, Vein induces several distal cell fates [17,18]. Vein is also utilized as a positive feedback reinforcement to the initial activation of the receptor by other ligands [19,20]. Finally, Argos functions as a secreted ligand that binds the receptor but inhibits activation by competing with the activating ligands [21–23]. It is induced in response to DER signaling, and plays a major role in restricting the activation range of the activating ligands [24]. The structure of DER ligands in schematized in Fig. 1. Networks of ligands generate the final pattern of DER activation. The most frequently used network involves a primary activation of the pathway by Spitz. Since argos and, in some contexts, also vein are transcriptional targets of DER, a circuitry of ligands is created. In the case of Argos, being an inhibitory ligand, this assures the spatial restriction of DER activation by Spitz. In contrast, the induction of Vein, which is less potent in activating DER, promotes lower levels of DER induction in cells that are more distant from the source of the signal. Ligand processing as a key regulatory step Three of the five DER ligands, Spitz, Keren, and Gurken, are produced as a precursor molecule with a transmembrane domain. Processing of these molecules to produce a secreted ligand was shown to be a key regulatory step in DER activation. This paradigm was first established for Spitz and subsequently applied to the other two ligands. Spitz is produced as an inactive membrane precursor and is ubiquitously expressed. Even when expressed at high levels, the precursor form is inactive [6]. The spatial and temporal pattern of Spitz-induced DER activation is thus dependent upon regulated processing of Spitz. The Spitz precursor is normally retained in the endoplasmic reticulum. In addition, the protein has a high turnover rate. In this state, it is prevented from reaching cellular compartments where cleavage by nonspecific proteases may take place [25,26]. Indeed, Spitz constructs where the retention is compromised exhibit a basal activity [7]. A clue to the mechanisms regulating Spitz processing emerged from the identification of mutations in two genes that give rise to phenotypes similar to spitz, namely Star and rhomboid [27]. The first regulated step in Spitz processing is the trafficking of the protein from the endoplasmic reticulum to the Golgi compartment. This step is carried out by Star, a novel type II transmembrane protein [28], that serves as a cargo receptor and associates with Spitz [25,26]. Once reaching the Golgi, Spitz encounters Rhomboid, a seven-transmembrane domain protein [29]. Rhomboid is essential for Spitz cleavage. Furthermore, different observations suggest that Rhomboid is the protease cleaving Spitz

149

[30]. The catalytic domain of Rhomboid resides within its conserved transmembrane domains, giving rise to regulated intramembrane proteolysis. This mode of cleavage was identified in other systems such as Presenillins and cholesterol metabolism [31]. Furthermore, it is conserved even in bacteria, where Rhomboid homologues are involved in releasing the peptide responsible for “quorum sensing” [32,33]. The cleavage of Spitz by Rhomboid appears to take place within the Golgi, rather than on the plasma membrane. Trafficking and processing of Spitz is schematized in Fig. 2. While Spitz and Star are broadly expressed, the expression of Rhomboid is extremely dynamic [29,34]. Interestingly, it is synonymous with the pattern of DER-induced MAPK activation (as followed by the double phosphorylated MAPK-dpERK) [35]. It thus appears that the expression of Rhomboid is the limiting step in DER activation. This was indeed demonstrated experimentally, by showing that ectopic expression of Rhomboid in diverse tissues and contexts is sufficient to give rise to high levels of DER activation [36,37]. What regulates the dynamic expression of Rhomboid? Promoter dissection has revealed an extremely complex organization of enhancer elements, which specifically regulate the modular expression of rhomboid gene [38,39]. It is interesting that in some tissues DER activation induces rhomboid expression. In tissues where multiple cycles of DER activation are required, the induction of rhomboid expression in the responding cells leads to additional rounds of ligand processing [20,40]. Elucidation of the mechanisms regulating Spitz processing led to insights regarding the other two transmembrane ligands. It was shown that Gurken can undergo cleavage that is Rhomboid and Star dependent in cell culture [8,41]. In the ovary, Gurken is found predominantly in the cleaved form and is endocytosed by the follicle cells receiving the signal [42]. A member of the Rhomboid family (Brho, Rho2, or Stet) that is expressed in the oocyte may carry out Gurken processing [8,43,44]. The mechanism regulating Gurken trafficking within the oocyte, before and after cleavage, is still not clear. Genetic interactions show that Star may be required [41]. In addition, Cornichon, which is another potential cargo receptor, is also required for Gurken signaling [11]. The capacity of Rho2/Brho to cleave DER ligands within the ER in culture may imply that Gurken is processed in the ER, and its trafficking from the ER is regulated post cleavage. Finally, Keren is a ligand that is most similar in structure to Spitz, and its processing is regulated in an analogous manner by Star and Rho [7,8]. Since the retention of Keren in the ER is less stringent, some cleavage by Rhomboid was observed in culture even in the absence of Star. In flies this feature is reflected by the observation that, in contrast to Spitz, overexpression of the Keren precursor can lead to hyperactivation of DER [7]. Which aspects of this elaborate retention, trafficking, and cleavage mechanism may also be utilized for the ligands of

Fig. 1. Activating and inhibitory Drosophilia epidermal growth factor (EGF) receptor (DER) ligands. Five ligands are interacting with DER. Spitz, Keren, and Gurken are produced as transmembrane precursors and are cleaved (arrows) to generate the active secreted ligand. Vein is produced as a secreted protein and also has an Ig domain. Argos is produced a secreted protein, and its EGF domain (red) mediates binding to DER and inhibits binding of other ligands, as well as receptor dimerization. Fig. 2. Intracellular trafficking and cleavage of Spitz. Processing of transmembrane Drosophilia epidermal growth factor (EGF) receptor (DER) ligands is tightly regulated and has been studied in detail for Spitz. (1) Spitz precursor is normally retained in the endoplasmic reticulum (ER). (2) Star is also localized predominantly to the ER. It can associate with Spitz and facilitate its translocation to the Golgi. (3) Rhomboid is localized to the Golgi. It catalyzes the cleavage of Spitz that has been transported to the Golgi by Star. (4) Following cleavage, the extracellular domain of Spitz is secreted outside the cell.


151

152


the mammalian EGF receptor family? Homologues of Star protein were found only in Anopheles gambiae and Bombyx mori [26], while other trafficking molecules like Cornichon are highly conserved. Mammalian Rhomboid homologues have been identified [45], and some were even capable of cleaving Spitz [30]. However, it remains to be seen if they also cleave precursors of the vertebrate ligands.

Feedback loops The general theme emerging from examination of the diverse array of DER functions, is that this pathway provides a relatively short-range signaling module between cells. However, it is not confined only to the cells immediately contacting the ligand source. DER signaling that takes place several cells away from the source is the exception. The two cases that stand out include the patterning of the embryonic ventral ectoderm by Spitz emanating from the midline [6,36,46], and the induction of leg segments by expression of Vein in the distal tip of the pupal leg disc [17,18]. In both these instances the ligand functions as a morphogen, inducing more than one cell fate depending upon the level of the ligand. The further the cells are from the ligand source, the lower the level of DER activation. Graded activation of DER expressed by the follicle cells is also achieved by the graded distribution of Gurken in the egg. However, in most other cases, activation of DER provides a binary switch, either in the cells adjacent to the ligand source or within the cells producing the processed ligand. The capacity to induce a response that is spatially restricted within a population of identical cells entails several feedback circuits that are inherent to the pathway. Three negative feedback circuits, which play a cardinal role in this restriction, have been identified. Argos expression is induced in the cells receiving high levels of DER activation [24]. The protein is secreted and reaches several cell rows away from the site of production. Argos maintains a steadystate level of signaling such that the cells receiving maximal levels of Spitz maintain DER activation, in spite of the production of Argos, while in the cells further away from the source, Argos attenuates activation by Spitz [35]. The assumption is that the distribution profiles of Spitz and Argos are different, such that Spitz levels are higher closer

to the source while Argos levels are higher further away. Visualization of the actual distribution profile of the two ligands and dissection of the factors affecting them awaits further experiments. The other two feedback circuits work in a cell autonomous manner and are less universally used than Argos. Kekkon and Sprouty are induced only in some of the tissues in which DER is activated. In those cases they display a fairly broad expression in cells receiving both high and intermediate levels of DER activation [47–49]. The assumption is that their broad expression leads to a uniform reduction in the response to signaling. Thus, cells in which high levels of DER activation took place retain activation, while cells exposed to lower activation levels shut off signaling. In mutants for sprouty, and in some contexts also for kekkon, broadening of the response to DER activation ensues. Kekkon is a transmembrane protein that binds the DER extracellular domain and attenuates receptor dimerization [50]. Sprouty is an intracellular protein that may interfere with DER signaling at several levels [51]. While Argos and Kekkon associate with DER itself, Sprouty interacts with signaling elements that are shared by other receptor tyrosine kinases including the fibroblast growth factor (FGF) receptors. Indeed, Sprouty was initially shown to be induced by other receptor tyrosine kinases, such as Breathless, and attenuate their activity [52]. The positive and negative DER feedback loops are schematized in Fig. 3. Inhibitors that are present in the tissue where DER is activated but are not transcriptional targets of the pathway were also identified. They include Yan, an ETS-domain protein that competes with Pointed, a transcription factor that is triggered by DER [53]. Expression of DCbl in the follicle cells of the ovary was shown to be critical for attenuating DER activation. In the absence of DCbl, endocytosis and degradation of DER is compromised and elevated signaling ensues [54].

Is the DER pathway linear? In mammalian systems, the activated EGF receptor serves as a docking site for several distinct signaling modules, leading to multiple outputs. This raises the ques-

Fig. 3. Positive and negative DER feedback loops. (A) High levels of Drosophilia epidermal growth factor (EGF) receptor (DER) activation by Spitz trigger through MAPK a series of positive and negative feedback loops. The positive feedbacks entails induction of Vein expression, which facilitates moderate levels of DER activation in adjacent cells. It also includes, in some cases, the induction of Rhomboid expression, which facilitates processing of Spitz in these cells. Negative feedback responses encompass the induction of expression of Argos, which is secreted to attenuate DER activation in more distant cells. It also entails the induction of Kekkon expression to reduce the levels of free DER, and the expression of Sprouty, which compromises signaling downstream to the activated receptor. While induction of Argos is universal to all DER-responding tissues, induction of the other responses occurs only in the context of some tissues. (B) The feedback responses have different thresholds of induction and consequently distinct domains of expression. Together, they ensure that the final pattern of activated DER (as monitored by activated MAPK-dpERK) will be spatially restricted. The cells expressing Rhomboid provide the source of Spitz. Induction of Vein and Argos takes place only in cells receiving high levels of DER activation, but their biological effect is also exerted on neighboring cells since they are secreted proteins. Kekkon and Sprouty have lower thresholds of induction and are expressed also in cells located further away from the ligand source. They attenuate DER signaling in a cell-autonomous manner.


tion whether the DER signaling pathway is bifurcating or linear. The issue is especially pertinent when considering the quantitative aspects of DER signaling. Bifurcating signaling may provide a mechanism for generating tighter thresholds in response to small differences in the level of activating ligand. The main intracellular signaling pathway activated by DER is the Ras/MAPK pathway. The obligatory use of this pathway for most aspects of DER signaling is implied by the fact that mutations in the intracellular components of the pathway give rise to phenotypes similar to loss of the receptor or ligand [55]. More specifically, absence of Ras in the wing imaginal disc had no effect of PI3 kinase signaling [56]. Conversely, activated Ras mimics DER gain of function phenotypes [57]. Similarities in phenotypes or genetic interactions were not detected with other intracellular signaling pathways, suggesting that as a rule, activation of DER triggers a linear intracellular cascade. One exception to the activation of the canonical Ras pathway by DER appears to take place in the border cells of the ovary, where a combination of signals from DER and the platelet-derived growth factor (PDGF)/vascular endothelial growth factor (VEGF) receptor (PVR) guide the migration of these cells towards the oocyte. In most cases of DER signaling, transcriptional activation is the final output, averaging the cumulative levels of DER activation around the circumference of the cell. In cases of cell migration, the site of receptor activation on the cell surface is critical, and the final response is likely to be local rather than transcriptional. In the migrating border cells, high and uniform activation of DER by ubiquitous expression of an activating ligand stalled migration, while activated Raf had no effect on migration [58]. This suggests that in the context of the border cells, DER triggers a different, yet unknown signaling pathway that is important for migration.

Tissue-specific responses to DER activation The wide spectrum of tissues in which DER activation provides a developmental switch raises the question of the basis for tissue specificity. In most of these cases, the actual activation of the DER pathway provides a binary switch, and the output of the switch, in terms of the battery of target genes, depends on the tissue context and on other signaling pathways. The issue has been addressed in detail in three tissues where the regulatory sequence of a target gene have been dissected in detail, shedding light on the interplay of DER signaling with tissue-specific factors and other signaling pathways [59–61]. The emerging theme is that all these inputs are integrated at the level of enhancer sequences. Only in cases where all necessary factors are bound to the DNA will transcription of the target gene ensue. For example, in the case of the DPax2 gene, expression in the induced cone cells will take

153

place if three requirements have been simultaneously fulfilled. They include DER pathway activation (as represented by the expression of the transcription factor Pointed), triggering of the Notch pathway (culminating in activation of Su(H)), and the competence of the cells to become photoreceptor cells as indicated by the expression of the Lozenge transcription factor [59]. This combinatorial arrangement allows the integration of tissue-specific factors that will induce a given gene only in the desired context. Furthermore, in terms of interactions between different signaling pathways, it provides a modular and highly versatile system. Integration at the promoter level allows use of a different set of pathways, as well as alteration of their interrelationship (synergistic or antagonistic) for any given tissue. In cases where DER activation is not a binary switch but an instructive signal dictating different cell fates, the mechanisms responsible for converting small changes at the level of activation to tight thresholds of gene expression remain to be identified.

Future directions Many of the open issues in DER signaling involve the intersection between cell biology and developmental patterning. Understanding in detail how the ligands are transported within the producing cell to allow regulated cleavage, and how they are distributed in the extracellular milieu, will be critical. One may expect the convergence of players dedicated to the DER pathway with components of the cellular machinery that are shared with other pathways. A detailed understanding of the spatial regulation of DER activation is still missing. Rhomboid expression triggers the processing of Spitz. Following secretion, what is the distribution of Spitz and Argos outside the producing cells, and why is short-range signaling observed in some tissues and long-range signaling in others? In some tissues such as the chordotonal organs, the cells producing secreted Spitz are refractive to DER activation, while in other cases, such as the future wing veins, the cells producing secreted ligand(s) are the ones undergoing DER activation. The mechanistic basis for the refractivity to DER activation in the producing cells remains to be explored. In view of the conservation of the pathway, it will be interesting to examine if some of the mammalian ligands are regulated in a similar manner to Spitz in terms of retention, trafficking, and cleavage by Rhomboid. The DER pathway emerges as a universal module for inducing spatially restricted responses. Many of the key regulatory features of the pathway are tuned to this purpose, including most notably inducible inhibitory molecules and restricted diffusion of the activating ligands. Will some of these aspects also be found in mammalian organisms, or has the pathway been adapted to more systemic responses in vertebrates?

154


Acknowledgments I thank all members of the lab for lively discussions and insightful comments on the manuscript. B.S. is an incumbent of the Hilda and Cecil Lewis chair in Molecular Genetics.

References [1] E. Livneh, L. Glazer, D. Segal, J. Schlessinger, B.Z. Shilo, The Drosophila EGF receptor gene homolog: conservation of both hormone binding and kinase domains, Cell 40 (1985) 599–607. [2] E.D. Schejter, D. Segal, L. Glazer, B.Z. Shilo, Alternative 5¢ exons and tissue-specific expression of the Drosophila EGF receptor homolog transcripts, Cell 46 (1986) 1091–1101. [3] N.B. Zak, R.J. Wides, E.D. Schejter, E. Raz, B.Z. Shilo, Localization of the DER/flb protein in embryos: implications on the faint little ball lethal phenotype, Development 109 (1990) 865–874. [4] R.J. Clifford, T. Schupbach, Coordinately and differentially mutable activities of torpedo, the Drosophila melanogaster homolog of the vertebrate EGF receptor gene, Genetics 123 (1989) 771–787. [5] B.J. Rutledge, K. Zhang, E. Bier, Y.N. Jan, N. Perrimon, The Drosophila spitz gene encodes a putative EGF-like growth factor involved in dorsal-ventral axis formation and neurogenesis, Genes Dev. 6 (1992) 1503–1517. [6] R. Schweitzer, M. Shaharabany, R. Seger, B.Z. Shilo, Secreted Spitz triggers the DER signaling pathway and is a limiting component in embryonic ventral ectoderm determination, Genes Dev. 9 (1995b) 1518–1529. [7] A. Reich, B.Z. Shilo, Keren, a new ligand of the Drosophila epidermal growth factor receptor, undergoes two modes of cleavage, EMBO J. 21 (2002) 4287–4296. [8] S. Urban, J.R. Lee, M. Freeman, A family of Rhomboid intramembrane proteases activates all Drosophila membrane-tethered EGF ligands, EMBO J. 21 (2002a) 4277–4286. [9] A. Gonzalez-Reyes, H. Elliott, D. St Johnston, Polarization of both major body axes in Drosophila by gurken-torpedo signalling, Nature 375 (1995) 654–658. [10] F.S. Neuman-Silberberg, T. Schupbach, The Drosophila dorsoventral patterning gene gurken produces a dorsally localized RNA and encodes a TGF alpha-like protein, Cell 75 (1993) 165–174. [11] S. Roth, F.S. Neuman-Silberberg, G. Barcelo, T. Schupbach, cornichon and the EGF receptor signaling process are necessary for both anterior-posterior and dorsal-ventral pattern formation in Drosophila, Cell 81 (1995) 967–978. [12] F.S. Neuman-Silberberg, T. Schupbach, The Drosophila TGFalpha-like protein Gurken: expression and cellular localization during Drosophila oogenesis, Mech. Dev. 59 (1996) 105–113. [13] F.S. Neuman-Silberberg, T. Schupbach, Dorsoventral axis formation in Drosophila depends on the correct dosage of the gene gurken, Development 120 (1994) 2457–2463. [14] B. Schnepp, T. Donaldson, G. Grumbling, S. Ostrowski, R. Schweitzer, B.Z. Shilo, A. Simcox, EGF domain swap converts a Drosophila EGF receptor activator into an inhibitor, Genes Dev. 12 (1998) 908–913. [15] B. Schnepp, G. Grumbling, T. Donaldson, A. Simcox, Vein is a novel component in the Drosophila epidermal growth factor receptor pathway with similarity to the neuregulins, Genes Dev. 10 (1996) 2302–2313. [16] T. Yarnitzky, L. Min, T. Volk, The Drosophila neuregulin homolog Vein mediates inductive interactions between myotubes and their epidermal attachment cells, Genes Dev. 11 (1997) 2691–2700. [17] G. Campbell, Distalization of the Drosophila leg by graded EGFreceptor activity, Nature 418 (2002) 781–785.

[18] M.I. Galindo, S.A. Bishop, S. Greig, J.P. Couso, Leg patterning driven by proximal-distal interactions and EGFR signaling, Science 297 (2002) 256–259. [19] M. Golembo, T. Yarnitzky, T. Volk, B.Z. Shilo, Vein expression is induced by the EGF receptor pathway to provide a positive feedback loop in patterning the Drosophila embryonic ventral ectoderm, Genes Dev. 13 (1999) 158–162. [20] J.D. Wasserman, M. Freeman, An autoregulatory cascade of EGF receptor signaling patterns the Drosophila egg, Cell 95 (1998) 355–364. [21] M. Freeman, C. Klambt, C.S. Goodman, G.M. Rubin, The argos gene encodes a diffusible factor that regulates cell fate decisions in the Drosophila eye, Cell 69 (1992) 963–975. [22] M.H. Jin, K. Sawamoto, M. Ito, H. Okano, The interaction between the Drosophila secreted protein argos and the epidermal growth factor receptor inhibits dimerization of the receptor and binding of secreted spitz to the receptor, Mol. Cell. Biol. 20 (2000) 2098– 2107. [23] R. Schweitzer, R. Howes, R. Smith, B.Z. Shilo, M. Freeman, Inhibition of Drosophila EGF receptor activation by the secreted protein Argos, Nature 376 (1995a) 699–702. [24] M. Golembo, R. Schweitzer, M. Freeman, B.Z. Shilo, Argos transcription is induced by the Drosophila EGF receptor pathway to form an inhibitory feedback loop, Development 122 (1996b) 223–230. [25] J.R. Lee, S. Urban, C.F. Garvey, M. Freeman, Regulated intracellular ligand transport and proteolysis control EGF signal activation in Drosophila, Cell 107 (2001) 161–171. [26] R. Tsruya, A. Schlesinger, A. Reich, L. Gabay, A. Sapir, B.Z. Shilo, Intracellular trafficking by Star regulates cleavage of the Drosophila EGF receptor ligand Spitz, Genes Dev. 16 (2002) 222–234. [27] U. Mayer, C. Nusslein-Volhard, A group of genes required for pattern formation in the ventral ectoderm of the Drosophila embryo, Genes Dev. 2 (1988) 1496–1511. [28] A.L. Kolodkin, A.T. Pickup, D.M. Lin, C.S. Goodman, U. Banerjee, Characterization of Star and its interactions with sevenless and EGF receptor during photoreceptor cell development in Drosophila, Development 120 (1994) 1731–1745. [29] E. Bier, L.Y. Jan, Y.N. Jan, rhomboid, a gene required for dorsoventral axis establishment and peripheral nervous system development in Drosophila melanogaster, Genes Dev. 4 (1990) 190–203. [30] S. Urban, J.R. Lee, M. Freeman, Drosophila rhomboid-1 defines a family of putative intramembrane serine proteases, Cell 107 (2001) 173–182. [31] M.S. Brown, J. Ye, R.B. Rawson, J.L. Goldstein, S. Urban, J.R. Lee, M. Freeman, Regulated intramembrane proteolysis: a control mechanism conserved from bacteria to humans, Cell 100 (2000) 391–398. [32] M. Gallio, G. Sturgill, P. Rather, P. Kylsten, A conserved mechanism for extracellular signaling in eukaryotes and prokaryotes, Proc. Natl. Acad. Sci. USA 99 (2002) 12208–12213. [33] S. Urban, D. Schlieper, M. Freeman, Conservation of intramembrane proteolytic activity and substrate specificity in prokaryotic and eukaryotic rhomboids, Curr. Biol. 12 (2002b) 1507–1512. [34] M.A. Sturtevant, M. Roark, J.W. O’Neill, B. Biehs, N. Colley, E. Bier, The Drosophila rhomboid protein is concentrated in patches at the apical cell surface, Dev. Biol. 174 (1996) 298–309. [35] L. Gabay, R. Seger, B.Z. Shilo, In situ activation pattern of Drosophila EGF receptor pathway during development, Science 277 (1997) 1103–1106. [36] M. Golembo, E. Raz, B.Z. Shilo, The Drosophila embryonic midline is the site of Spitz processing, and induces activation of the EGF receptor in the ventral ectoderm, Development 122 (1996a) 3363–3370. [37] M.A. Sturtevant, M. Roark, E. Bier, The Drosophila rhomboid gene mediates the localized formation of wing veins and interacts genetically with components of the EGF-R signaling pathway, Genes Dev. 7 (1993) 961–973.

Signaling by the Drosophila Epidermal Growth Factor Receptor Pathway During Development [38] Y.T. Ip, R.E. Park, D. Kosman, E. Bier, M. Levine, The dorsal gradient morphogen regulates stripes of rhomboid expression in the presumptive neuroectoderm of the Drosophila embryo, Genes Dev. 6 (1992) 1728–1739. [39] E. Zelzer, B.Z. Shilo, Interaction between the bHLH-PAS protein Trachealess and the POU-domain protein Drifter, specifies tracheal cell fates, Mech. Dev. 91 (2000) 163–173. [40] A. Sapir, R. Schweitzer, B.Z. Shilo, Sequential activation of the EGF receptor pathway during Drosophila oogenesis establishes the dorsoventral axis, Development 125 (1998) 191–200. [41] C. Ghiglione, E.A. Bach, Y. Paraiso 3rd, K.L. Carraway, S. Noselli, N. Perrimon, Mechanism of activation of the Drosophila EGF receptor by the TGFalpha ligand Gurken during oogenesis, Development 129 (2002) 175–186. [42] F. Peri, C. Bokel, S. Roth, Local Gurken signaling and dynamic MAPK activation during Drosophila oogenesis, Mech. Dev. 81 (1999) 75–88. [43] A. Guichard, M. Roark, M. Ronshaugen, E. Bier, Brother of rhomboid, a rhomboid-related gene expressed during early Drosophila oogenesis, promotes EGF-R/MAPK signaling, Dev. Biol. 226 (2000) 255–266. [44] C. Schulz, C.G. Wood, D.L. Jones, S.I. Tazuke, M.T. Fuller, Signaling from germ cells mediated by the rhomboid homolog stet organizes encapsulation by somatic support cells, Development 129 (2002) 4523–4534. [45] J.C. Pascall, K.D. Brown, Characterization of a mammalian cDNA encoding a protein with high sequence similarity to the Drosophila regulatory protein Rhomboid, FEBS Lett. 429 (1998) 337–340. [46] E. Raz, B.Z. Shilo, Establishment of ventral cell fates in the Drosophila embryonic ectoderm requires DER, the EGF receptor homolog, Genes Dev. 7 (1993) 1937–1948. [47] S. Kramer, M. Okabe, N. Hacohen, M.A. Krasnow, Y. Hiromi, Sprouty: a common antagonist of FGF and EGF signaling pathways in Drosophila, Development 126 (1999) 2515–2525. [48] M. Musacchio, N. Perrimon, The Drosophila kekkon genes: novel members of both the leucine-rich repeat and immunoglobulin superfamilies expressed in the CNS, Dev. Biol. 178 (1996) 63–76. [49] A. Reich, A. Sapir, B. Shilo, Sprouty is a general inhibitor of receptor tyrosine kinase signaling, Development 126 (1999) 4139–4147. [50] C. Ghiglione 3rd, K.L. Carraway, L.T. Amundadottir, R.E. Boswell, N. Perrimon, J.B. Duffy, The transmembrane molecule kekkon 1 acts in a feedback loop to negatively regulate the activity of the Drosophila EGF receptor during oogenesis, Cell 96 (1999) 847–856. [51] T. Casci, J. Vinos, M. Freeman, Sprouty, an intracellular inhibitor of Ras signaling, Cell 96 (1999) 655–665. [52] N. Hacohen, S. Kramer, D. Sutherland, Y. Hiromi, M.A. Krasnow, Sprouty encodes a novel antagonist of FGF signaling that patterns apical branching of the Drosophila airways, Cell 92 (1998) 253– 263. [53] L. Gabay, H. Scholz, M. Golembo, A. Klaes, B.Z. Shilo, C. Klambt, EGF receptor signaling induces pointed P1 transcription and inactivates Yan protein in the Drosophila embryonic ventral ectoderm, Development 122 (1996) 3355–3362. [54] L.M. Pai, G. Barcelo, T. Schupbach, D-cbl, a negative regulator of the Egfr pathway, is required for dorsoventral patterning in Drosophila oogenesis, Cell 103 (2000) 51–61. [55] F.J. Diaz-Benjumea, E. Hafen, The sevenless signalling cassette mediates Drosophila EGF receptor function during epidermal development, Development 120 (1994) 569–578. [56] D.A. Prober, B.A. Edgar, Interactions between Ras1, dMyc, and dPI3K signaling in the developing Drosophila wing, Genes Dev. 16 (2002) 2286–2299. [57] A.H. Brand, N. Perrimon, Raf acts downstream of the EGF receptor to determine dorsoventral polarity during Drosophila oogenesis, Genes Dev. 8 (1994) 629–639. [58] P. Duchek, P. Rorth, Guidance of cell migration by EGF receptor signaling during Drosophila oogenesis, Science 291 (2001) 131–133.

155

[59] G.V. Flores, H. Duan, H. Yan, R. Nagaraj, W. Fu, Y. Zou, M. Noll, U. Banerjee, S. Urban, J.R. Lee, M. Freeman, Combinatorial signaling in the specification of unique cell fates, Cell 103 (2000) 75–85. [60] M.S. Halfon, A. Carmena, S. Gisselbrecht, C.M. Sackerson, F. Jimenez, M.K. Baylies, A.M. Michelson, Ras pathway specificity is determined by the integration of multiple signal-activated and tissuerestricted transcription factors, Cell 103 (2000) 63–74. [61] C. Xu, R.C. Kauffmann, J. Zhang, S. Kladny, R.W. Carthew, Overlapping activators and repressors delimit transcriptional response to receptor tyrosine kinase signals in the Drosophila eye, Cell 103 (2000) 87–97. [62] J.B. Skeath, The Drosophila EGF receptor controls the formation and specification of neuroblasts along the dorsal-ventral axis of the Drosophila embryo, Development 125 (1998) 3301–3312. [63] T. von Ohlen, C.Q. Doe, Convergence of dorsal, dpp, and egfr signaling pathways subdivides the Drosophila neuroectoderm into three dorsal-ventral columns, Dev. Biol. 224 (2000) 362–372. [64] Y. Yagi, T. Suzuki, S. Hayashi, Interaction between Drosophila EGF receptor and vnd determines three dorsoventral domains of the neuroectoderm, Development 125 (1998) 3625–3633. [65] E. Buff, A. Carmena, S. Gisselbrecht, F. Jimenez, A.M. Michelson, Signalling by the Drosophila epidermal growth factor receptor is required for the specification and diversification of embryonic muscle progenitors, Development 125 (1998) 2075–2086. [66] M. Llimargas, J. Casanova, EGF signalling regulates cell invagination as well as cell migration during formation of tracheal system in Drosophila, Dev. Genes Evol. 209 (1999) 174–179. [67] P. Wappner, L. Gabay, B.Z. Shilo, Interactions between the EGF receptor and DPP pathways establish distinct cell fates in the tracheal placodes, Development 124 (1997) 4707–4716. [68] P. Lage, Y.N. Jan, A.P. Jarman, Requirement for EGF receptor signalling in neural recruitment during formation of Drosophila chordotonal sense organ clusters, Curr. Biol. 7 (1997) 166–175. [69] M. Okabe, H. Okano, Two-step induction of chordotonal organ precursors in Drosophila embryogenesis, Development 124 (1997) 1045–1053. [70] P.R. Elstob, V. Brodu, A.P. Gould, spalt-dependent switching between two cell fates that are induced by the Drosophila EGF receptor, Development 128 (2001) 723–732. [71] T.E. Rusten, R. Cantera, J. Urban, G. Technau, F.C. Kafatos, R. Barrio, Spalt modifies EGFR-mediated induction of chordotonal precursors in the embryonic PNS of Drosophila promoting the development of oenocytes, Development 128 (2001) 711–722. [72] L. Dubois, M. Lecourtois, C. Alexandre, E. Hirst, J.P. Vincent, Regulated endocytic routing modulates wingless signaling in Drosophila embryos, Cell 105 (2001) 613–624. [73] L. O’Keefe, S.T. Dougan, L. Gabay, E. Raz, B.Z. Shilo, S. DiNardo, Spitz and Wingless, emanating from distinct borders, cooperate to establish cell fate across the Engrailed domain in the Drosophila epidermis, Development 124 (1997) 4837–4845. [74] F. Payre, A. Vincent, S. Carreno, ovo/svb integrates Wingless and DER pathways to control epidermis differentiation, Nature 400 (1999) 271–275. [75] K. Dumstrei, C. Nassif, C. Abboud, A. Aryai, V. Hartenstein, EGFR signaling is required for the differentiation and maintenance of neural progenitors along the dorsal midline of the Drosophila embryonic head, Development 125 (1998) 3417–3426. [76] B. Kerber, S. Fellert, M. Hoch, Seven-up, the Drosophila homolog of the COUP-TF orphan receptors, controls cell proliferation in the insect kidney, Genes Dev. 12 (1998) 1781–1786. [77] V. Sudarsan, S. Pasalodos-Sanchez, S. Wan, A. Gampel, H. Skaer, A genetic hierarchy establishes mitogenic signalling and mitotic competence in the renal tubules of Drosophila, Development 129 (2002) 935–944. [78] E. Raz, B.Z. Shilo, Dissection of the faint little ball (flb) phenotype: determination of the development of the Drosophila central nervous

156

[79]

[80]

[81]

[82]

[83]

[84]

[85]

[86]

[87]

[88]

[89]

[90]

The EGF Receptor Family system by early interactions in the ectoderm, Development 114 (1992) 113–123. M. Gonzalez-Gaitan, H. Jackle, Tip cell-derived RTK signaling initiates cell movements in the Drosophila stomatogastric nervous system anlage, EMBO Rep. 1 (2000) 366–371. A. Bergmann, M. Tugentman, B.Z. Shilo, H. Steller, Regulation of cell number by MAPK-dependent control of apoptosis: a mechanism for trophic survival signaling, Dev. Cell 2 (2002) 159–170. C. Klambt, J.R. Jacobs, C.S. Goodman, The midline of the Drosophila central nervous system: a model for the genetic analysis of cell fate, cell migration, and growth cone guidance, Cell 64 (1991) 801–815. C. Stemerdink, J.R. Jacobs, Argos and Spitz group genes function to regulate midline glial cell number in Drosophila embryos, Development 124 (1997) 3787–3796. K. Kubota, S. Goto, K. Eto, S. Hayashi, EGF receptor attenuates Dpp signaling and helps to distinguish the wing and leg cell fates in Drosophila, Development 127 (2000) 3769–3776. A. Baonza, F. Roch, E. Martin-Blanco, DER signaling restricts the boundaries of the wing field during Drosophila development, Proc. Natl. Acad. Sci. USA 97 (2000) 7331–7335. S.H. Wang, A. Simcox, G. Campbell, Dual role for Drosophila epidermal growth factor receptor signaling in early wing disc development, Genes Dev. 14 (2000) 2271–2276. M. Zecca, G. Struhl, Control of growth and patterning of the Drosophila wing imaginal disc by EGFR-mediated signaling, Development 129 (2002a) 1369–1376. M. Zecca, G. Struhl, Subdivision of the Drosophila wing imaginal disc by EGFR-mediated signaling, Development 129 (2002b) 1357–1368. J.P. Kumar, K. Moses, EGF receptor and Notch signaling act upstream of Eyeless/Pax6 to control eye specification, Cell 104 (2001) 687–697. E. Martin-Blanco, F. Roch, E. Noll, A. Baonza, J.B. Duffy, N. Perrimon, A temporal switch in DER signaling controls the specification and differentiation of veins and interveins in the Drosophila wing, Development 126 (1999) 5739–5747. R. Nagaraj, A.T. Pickup, R. Howes, K. Moses, M. Freeman, U. Banerjee, Role of the EGF receptor pathway in growth and patterning of the Drosophila wing through the regulation of vestigial, Development 126 (1999) 975–985.

[91] F. Roch, G. Jimenez, J. Casanova, EGFR signalling inhibits Capicuadependent repression during specification of Drosophila wing veins, Development 129 (2002) 993–1002. [92] A. Baonza, T. Casci, M. Freeman, A primary role for the epidermal growth factor receptor in ommatidial spacing in the Drosophila eye, Curr. Biol. 11 (2001) 396–404. [93] M. Dominguez, J.D. Wasserman, M. Freeman, Multiple functions of the EGF receptor in Drosophila eye development, Curr. Biol. 8 (1998) 1039–1048. [94] T. Xu, G.M. Rubin, Analysis of genetic mosaics in developing and adult Drosophila tissues, Development 117 (1993) 1223–1237. [95] N.E. Baker, S.Y. Yu, The EGF receptor defines domains of cell cycle progression and survival to regulate cell number in the developing Drosophila eye, Cell 104 (2001) 699–708. [96] M. Freeman, Reiterative use of the EGF receptor triggers differentiation of all cell types in the Drosophila eye, Cell 87 (1996) 651–660. [97] M. Tio, K. Moses, The Drosophila TGF alpha homolog Spitz acts in photoreceptor recruitment in the developing retina, Development 124 (1997) 343–351. [98] Z. Huang, B.Z. Shilo, S. Kunes, A retinal axon fascicle uses spitz, an EGF receptor ligand, to construct a synaptic cartridge in the brain of Drosophila, Cell 95 (1998) 693–703. [99] D. del Alamo, J. Terriente, F.J. Diaz-Benjumea, Spitz/EGFr signalling via the Ras/MAPK pathway mediates the induction of bract cells in Drosophila legs, Development 129 (2002) 1975–1982. [100] A.A. Kiger, H. White-Cooper, M.T. Fuller, Somatic support cells restrict germline stem cell self-renewal and promote differentiation, Nature 407 (2000) 750–754. [101] J.V. Price, R.J. Clifford, T. Schupbach, The maternal ventralizing locus torpedo is allelic to faint little ball, an embryonic lethal, and encodes the Drosophila EGF receptor homolog, Cell 56 (1989) 1085–1092. [102] A.M. Queenan, A. Ghabrial, T. Schupbach, Ectopic activation of torpedo/Egfr, a Drosophila receptor tyrosine kinase, dorsalizes both the eggshell and the embryo, Development 124 (1997) 3871– 3880. [103] A. Kopp, R.K. Blackman, I. Duncan, Wingless, decapentaplegic and EGF receptor signaling pathways interact to specify dorso-ventral pattern in the adult abdomen of Drosophila, Development 126 (1999) 3495–3507.

The epidermal growth factor system in Caenorhabditis elegans Nadeem Moghal and Paul W. Sternberg HHMI and Division of Biology, Caltech, Pasadena, CA 91125, USA

Abstract The single known epidermal growth factor-like growth factor and single epidermal growth factor receptor in Caenorhabditis elegans mediate two types of processes, each via a distinct signal transduction pathway. Several instances of cell fate specification during organogenesis require the RAS-MAP kinase pathway, as well as multiple nuclear factors. By contrast, appropriate myoepithelial contractions during ovulation involve IP3-mediated signal transduction. Positive modulators of the RAS pathway include KSR, SUR-8, phosphatase PP2A, and a zinc cation diffusion facilitator. Negative regulators of the RAS pathway include homologs of CBL, GAP-1, ACK, and MAP kinase phosphatase, while negative regulators of the IP3 pathway are enzymes that modify IP3. In addition to its stimulation of RAS activity, the GRB2 homolog SEM-5 acts negatively on both signaling pathways, as does the Ack-related kinase ARK-1.

Introduction Our understanding of epidermal growth factor receptor (EGFR) signaling in Caenorhabditis elegans has been worked out chiefly by the identification of genetic loci controlling development, the ordering of these genes into pathways, and the cloning of these genes. The multiple roles of EGF signaling in C. elegans were elucidated by analyzing mutations that had preferential effects on various functions of the EGFR homolog LET-23 [1,2]. Analysis of potential phosphotyrosine sites in the cytoplasmic tail of LET-23 indicated distinct sites are involved in stimulation of two signal transduction pathways and inhibition of one of the pathways [103]. As detailed below, we have achieved a good understanding of core EGF signaling, and the ways in which signaling can be regulated.

Development Vulva development Of the processes regulated by EGFR signaling in C. elegans, vulva development has provided the greatest opportunity for understanding the complexities of signaling by this pathway in vivo. First, the vulva is not essential for the development of viable, self-fertilizing hermaphrodites, and second, it has been possible to isolate many viable, non-null mutations in genes that affect development of this tissue. During vulva development, the somatic gonad initially stimulates EGFR signaling in a group of precursor cells to

specify vulval versus nonvulval fates. Later, the somatic gonad stimulates the EGFR a second time to specify subfates within vulval tissue, while the vulval tissue reciprocally signals back to the somatic gonad via EGF to promote uterine fates. Some of the striking features to emerge from these studies indicate that although a core signaling pathway drives EGFR signaling, this pathway is subject to multiple redundantly acting positive and negative modulatory signals. In addition, virtually every step of signaling from receptor localization to communication to RNA Pol II is regulated in a precise manner. Induction (vulval versus nonvulval fates) Core signaling in the EGFR pathway At the beginning of the third larval stage of development (L3), six Pn.p cells (P3.p–P8.p) located along the ventral side of the hermaphrodite express LET-23, an EGFR family member [2,3]. These Pn.p cells are competent to respond to LIN-3 [4], an EGF-like molecule, and are termed the vulval precursor cells (VPCs). At this time, lin-3 expression is detectable in the anchor cell [4], a cell located in the somatic gonad, just above P6.p (Fig. 1A). At the beginning of the L3 larval stage, LET-23 expression increases on the surface of P6.p, and it decreases in the other VPCs [3]. Mutations in lin-3 and let-23 indicate that these genes are absolutely required for the induction of vulval fates, as severe mutations in these genes result in no vulval tissue [1,5,6] (Fig. 2). Ablation of the anchor cell also impairs vulval induction [7], as well as ablation of the gonadal precursor cells [8], supporting the molecular/genetic model that LIN-3 is the

158


ligand for LET-23. P6.p is stimulated by LIN-3 to undergo a characteristic pattern of three cell divisions, resulting in eight progeny cells, known as a primary fate [9,10]. During this time, P5.p and P7.p, which flank P6.p, undergo a different pattern of cell division, resulting in a distinct set of progeny cells, known as the secondary fate [9,10]. The secondary fate is thought to result from the lateral activation of LIN-12/NOTCH receptors [10–13] and low levels of activation of LET-23 on P5.p and P7.p [14]. In wild-type animals, P5.p–P7.p always adopt these vulval fates, while the remaining VPCs fuse with the underlying hyp7 epidermal syncytium (Fig. 1B). Animals in which fewer than three VPCs adopt vulval fates are termed vulvaless (Fig. 1C), and animals in which more than three VPCs adopt vulval fates are termed multivulva (Fig. 1D). Proper development is dependent on the polarized basolateral localization of LET-23 on P6.p, which places LET23 on the surface closest to the anchor cell. Mutations in the genes encoding the PDZ domain-containing proteins LIN2, LIN-7, and LIN-10 result in a uniform membrane distribution of LET-23 and a vulvaless phenotype [3,5,15]. LIN-7 and LIN-10 only have PDZ domains [3,16], whereas LIN-2 also has an SH3 and a guanylate kinase domain, making it a member of the MAGUK family [17]. LIN-7 binds to the C-terminus of LET-23, and LIN-2 binds to both LIN-7 and LIN-10, suggesting that proper localization of LET-23 is dependent on quaternary complex formation [3,15]. The vulvaless phenotypes of lin-2, lin-7, and lin-10 mutants are rescued by overexpression of wild-type LET-23, suggesting that the polarized localization of LET-23 is necessary to achieve proper receptor activation under physiologic conditions [3]. The cloning of a number of other genes whose individual mutation causes a vulvaless phenotype has led to the identification of many of the conserved EGFR pathway components. These include SEM-5 (GRB2) [18], LET-341 (SOS) [19], LET-60 (RAS) [20,21], LIN-45 (RAF) [22], MEK-2 (MEK) [23,24], and SUR-1/MPK-1 (MAP kinase) [25,26]. let-60 expression is detected in the VPCs at the time of vulval induction, and increases in these cells during the inductive process [27]. Mosaic analyses and tissue-specific promoters driving transgenes suggest that LET-23, LET-60, and SUR-1/MPK-1 function in the VPCs, consistent with these proteins acting in a linear pathway in the same cell [13,28–30]. Positive modulators of EGFR signaling. Genetic screens in sensitized backgrounds have led to the discovery of a number of genes in which individual mutation does not affect vulval development, but in conjunction with other mutations, reduces vulval induction. These genes thus define modulators of EGFR signaling and are distinct from the core components. The most fruitful of these sensitized screens involved the isolation of suppressors of the multivulva phenotype caused by a gain-of-function mutation in let-60. Single mutations in ksr-1 [31,32], sur-8/soc-2 [33,34], and

sur-6 [35] do not cause abnormal vulval development. However, they strongly reduce the penetrance of the let60(gf) multivulva phenotype, and they cause strong vulvaless phenotypes in the presence of weak mutations in core genes such as sur-1/mpk-1 and lin-45. ksr-1 encodes a RAFrelated serine/threonine kinase without the CR1 and CR2 domains. KSR-1 likely functions upstream of the LIN-1 Etsdomain transcription factor and it may interact directly with SUR-1/MPK-1 through an FXFP docking site for MAP kinase [36]. Transgenes harboring mutations affecting the ATP-binding site in KSR-1 or the catalytic aspartate have the same rescuing activity as wild-type DNA, indicating kinase activity is not crucial for KSR-1 function [37]. SUR8 possesses 18 tandem leucine-rich repeats, similar to yeast adenylate cyclases. Mutation of sur-8 does not suppress the multivulva phenotype of a gain-of-function lin-45 transgene, suggesting SUR-8 functions upstream of RAF activation. SUR-8 can interact with the LET-60 effector domain in a region different from that where LIN-45 interacts, suggesting a ternary complex may exist between SUR-8, RAS, and RAF [34]. SUR-6 is most similar to a regulatory B subunit of protein phosphatase 2A (PP2A). Mutation of sur-6 does not suppress the multivulva phenotype conferred by a gain-of-function lin-45 transgene, suggesting it also acts upstream of RAF activation. A sur-6 mutation causes a strong synthetic vulvaless phenotype with a sur-8 mutation, but not with a ksr-1 mutation, suggesting that SUR-6 functions with KSR-1 [35]. RNAi against the A and C PP2A catalytic core components also inhibits the let-60(gf) multivulva phenotype, indicating that SUR-6 positively regulates PP2A, and that the PP2A complex promotes RAS-dependent vulva development [35]. CDF-1 is a cation diffusion facilitator most similar to rat ZnT-1 that lowers the concentration of cytosolic zinc cations [38]. Mutation of cdf1 also suppresses the multivulva phenotype of let-60(gf); however, unlike the other modulators, a mutation in cdf-1 results in a weak vulvaless phenotype by itself [38]. Mutation of cdf-1 does not suppress the multivulva phenotype of a lin-1 mutation, suggesting that zinc cations act upstream of MAP kinase to inhibit its activation. Consistent with this model, directly increasing the zinc cation concentration inhibits vulval induction in C. elegans, and MAP kinase activation by insulin in Xenopus oocytes [38]. Thus, CDF1 likely acts as a positive regulator of EGFR signaling by lowering the concentration of zinc cations. Reverse genetic approaches led to the identification of a second ksr-like gene, ksr-2 [39], and one SH2-domain containing protein tyrosine phosphatase, ptp-2, which is most closely related to SHP-2/corkscrew [40]. KSR-2 lacks conserved residues in the kinase domain, and similar to KSR-1 is not likely to act as a protein kinase. As with ksr-1, mutation of ksr-2 by itself does not affect vulval induction. However, a ksr-2; ksr-1 double mutant displays a synthetic vulvaless phenotype that suppresses activated LET-60, but not activated SUR-1/MPK-1. Thus, KSR-1 and KSR-2 are likely to function redundantly upstream of MAP

The Epidermal Growth Factor System in Caenorhabditis Elegans

159

Fig. 1. C. elegans hermaphrodites displaying different amounts of vulval induction. Photographs were taken with Nomarski optics. Animals are on their side. Arrows point to cell nuclei and nucleoli. Scale bars: 20 mm. (A) Wild-type animal at the beginning of the L3 larval stage when vulval induction begins. Vulva precursor cells (VPCs) have divided once. (B and C). Animals are at the mid-L4 larval stage, at the end of vulval induction. (B) Wild-type animal. (C) let-23(sy1) loss-of-function vulvaless mutant that has less than wild-type vulval induction. (D) lin-3(syIs1) gain-of-function multivulva mutant that has more than wild-type vulval induction. lin-3(syIs1gf) animals contain an integrated transgenic array of multiple copies of the lin-3 gene [14].

kinase activation. Deletion of ptp-2 does not affect vulval development by itself, but it suppresses the multivulva phenotypes caused by gain-of-function mutations in let-23 and let-60, and a loss-of-function mutation in lin-15 (see next section). In addition, a ptp-2 mutation cooperates with a weak mutation in sem-5 to cause a synthetic vulvaless phenotype, indicating it also is a positive modulator of EGFR signaling. Single mutations in either mig-2 [41], a RAC/RHO-like GTPase, or unc-73 [42], a TRIO-like RAC exchange factor, result in a weakly penetrant vulvaless phenotype [43]. Double mutants in mig-2 and ced-10 [44], which encodes a RAC-like protein, have a more penetrant vulvaless phenotype, suggesting RAC is involved in EGFRdependent RAS signaling [43]. Negative regulators of EGFR signaling. Forward and reverse genetics have led to the identification of a number of molecules that negatively regulate LET-23 signaling. The identification of mutations that suppress loss-of-function mutations in the let-23 pathway led to the cloning of unc101 [45], sli-1 [46,47], gap-1 [48], and sur-5 [49]. UNC-101 is a functional homolog of the AP47 medium chain of the trans-Golgi clathrin-associated AP-1 complex. UNC-101 acts redundantly with a second AP47 homolog, apm-1 [50]. SLI-1 belongs to the CBL family of adaptors/E3 ubiquitin ligases. Mutation of sli-1 suppresses a strong mutation in let-23, but not let-60, suggesting it acts upstream of RAS [51]. Although strong inhibitory activity by SLI-1 requires the PTB and RING finger domains, SLI-1 displays some inhibitory activity in the absence of the RING finger, indicating it may antagonize EGFR signaling independent of E3 ubiquitin ligase activity [51]. A C-terminal tyrosine in the sequence NSSRYKETP in LET-23 is necessary for inhibition by sli-1 [51]. This sequence is similar to a ZAP-70

binding site for the c-CBL PTB domain [52,53], supporting a model of direct binding of SLI-1 to LET-23. Mutation of gap-1 suppresses loss-of-function mutations in let-60 and upstream components, but not lin-45 [106]. Sequence analysis indicates that GAP-1 is most similar to the GAP-1 and GAP-1m RAS GTPase activating proteins in Drosophila and vertebrates, respectively, suggesting that it is a GAP for LET-60. sur-5 was cloned as a suppressor of a dominantnegative let-60 allele. In contrast to the previously described negative regulators, a sur-5 mutation does not suppress any of the standard loss-of-function mutations in the core components, including a weak allele of let-23. Moreover, it only suppresses a subset of dominant negative let-60 alleles. SUR-5 has some homology to acetyl-coenzyme A synthetase, but it is unclear how it regulates EGFR signaling. ark-1 was cloned by the isolation of a mutation that caused a synthetic multivulva phenotype in the presence of a sli-1 mutation [54]. ARK-1 is most similar to the ACK tyrosine kinase, and yeast two hybrid studies indicate that a C-terminal proline-rich domain in ARK-1 can bind to SEM5. Furthermore, ARK-1 inhibition of RAS signaling is dependent on GRB-2, suggesting that GRB-2 can both positively and negatively regulate signaling by the EGFR (see ovulation). Mutation of ark-1 suppresses a strong mutation in let-23, but not let-60, suggesting that ARK-1 may inhibit EGFR signaling upstream of RAS activation. A reverse genetics approach resulted in the generation of a mutation in lip-1, a gene predicted to encode a MAP kinase phosphatase [55]. Consistent with its sequence homology, mutation of lip-1 suppresses the vulvaless phenotypes of loss-of-function mutations in the let-23 pathway. Targets of MAP kinase and regulation by homeobox genes. A number of proteins that are involved in transcriptional

160


Fig. 2. Signaling by the LET-23 (EGF) receptor tyrosine kinase. Core signaling molecules whose individual mutation leads to strong phenotypes are indicated in blue. Positive regulators whose mutation only leads to strong phenotypes in sensitized backgrounds are indicated in green. Negative regulators are indicated in red. EGF, epidermal growth factor.

regulation have been identified that likely transduce the signal of MAP kinase, and regulate the responsiveness of the VPCs to EGFR signaling. Two potential direct targets for SUR-1/MPK-1 include LIN-1 [56] and LIN-31 [57], an Etsdomain and winged helix domain transcription factor, respectively. A loss-of-function mutation in lin-1 results in a spontaneous multivulva phenotype [5], and a gain-offunction mutation results in a weak vulvaless phenotype [58]. LIN-31 appears to regulate the choice of fates by the VPCs. In lin-31 mutants, VPCs randomly adopt either vulval or nonvulval fates [57]. LIN-1 and LIN-31 can form a complex, and both can be phosphorylated by MAP kinase [58,59]. Phosphorylation of LIN-31 by MAP kinase disrupts the complex and facilitates stimulation of vulval fates by activated LIN-31 [59]. In contrast, forced dimerization of LIN-1 with LIN-31 inhibits vulval induction, suggesting that MAP kinase induces vulval fates by disrupting LIN-1/LIN31 complexes [59].

Homeobox genes, which are expressed in distinct regions along the anteroposterior axis of the animal, play an important role in regulating the sensitivity of the VPCs to EGFR signaling. The lin-39 Hox gene is expressed in all the VPCs [60,61]. Mutation of lin-39 impairs the ability of the LET-23 pathway to trigger vulval fates [61,62]. Conversely, the mab-5 Hox gene is strongly expressed in P8.p–P11.p [60,63], which normally do not respond to LIN-3 signaling. MAB-5 inhibits the ability of constitutively active LET-23 to promote ectopic vulval fates in P8.p, and ectopic expression of MAB-5 in P3.p–P6.p inhibits the normal adoption of vulval fates by P5.p–P7.p [62]. Interactions with other signaling pathways. Proper specification of vulval fates is dependent on communication between three different signaling pathways. High levels of EGFR signaling promote primary vulval fates [14], and high levels of LIN-12/NOTCH promote secondary vulval fates


[11,64]. Simultaneous activation of the two pathways antagonizes each other [65]. LIN-12 signaling induces the expression of the LIP-1 MAP kinase phosphatase, providing one explanation for how the NOTCH pathway can inhibit the EGFR pathway [55]. In contrast, WNT signaling cooperates with EGFR signaling to promote vulval fates. However, relative to EGFR signaling, WNT signaling plays a different role in promoting vulval fates. Whereas strong mutations in the core components of the let-23 pathway abolish vulval induction, a null mutation in the bar-1 (b-catenin) gene results in only a partially penetrant vulvaless phenotype [66]. Hyperactivation of WNT signaling through a mutation in pry-1 [67,68], an axin-like inhibitor of WNT signaling, results in the ectopic adoption of vulval fates by the VPCs, and can bypass certain mutations in the let-23 pathway [69]. One mechanism of cooperation between the two pathways likely involves joint regulation of the lin-39 Hox gene [61,66]. Regulation by chromatin remodeling and histone acetylation. A group of redundantly acting genes termed the “synmuv” genes that negatively regulates vulval fates implicates chromatin structure as being an important mechanism of regulating EGFR signaling. These genes can be subdivided into A and B classes, and any combination of an A class mutation with a B class mutation results in a synthetic multivulva phenotype [70]. lin-15A [71,72], lin-15B [71,72], and lin-36 [73] encode novel proteins. However, LIN-35 is related to the retinoblastoma family of proteins [74], and lin-53 encodes the Rb-binding protein, RbAp48 [74]. Consistent with the genetic identification of lin-53 as RbAp48, which can bind the HDAC-1 histone deacetylase, components of the NuRD nucleosome remodeling and histone deacetylase complex also have been implicated in this pathway [75]. Some of these components include HDAC-1 [74] and the Mi-2 chromatin remodeling protein homolog, LET-418 [76]. In addition to the NuRD complex, DPL-1, which is related to the DP family of transcription factors [77], and efl-1/2, which are E2F-like genes, also have been shown to have synmuv properties [77]. Genetic [72–74,77] and cell ablation [72] experiments indicate that the synmuv phenotype is strongly dependent on LET-23 and its downstream signaling components, but not LIN-3, suggesting that the synmuv genes promote ligand-independent activity of the EGFR pathway. Expression, transgenic, and mosaic analyses suggest that both the vulval precursor cells and the surrounding epidermal syncytium might be a source for an inhibitory signal by these genes [73,74,77–79]. Regulation by the mediator complex and other nuclear proteins. The endpoint of EGFR signaling ultimately involves communication with RNA Pol II. Studies have shown that proteins found in mediator complexes, which affect communication between sequence-specific transcription factors and RNA Pol II, provide both positive and negative regulation of EGFR signaling. A loss-of-function mutation in the

161

sur-2 mediator component results in a moderately strong vulvaless phenotype [80], reminiscent of mutations in the let-23 core pathway. Mutations in the lin-25 gene [81], which encodes a novel protein, also results in a vulvaless phenotype similar to that of mutations in sur-2. Since LIN25 protein levels are reduced in the absence of SUR-2 [82], both proteins might function together through a mediator complex to promote EGFR signaling. Conversely, the DPY22/SOP-1 mediator component acts as an inhibitor of EGFR signaling [83], and might be important in preventing inappropriate EGFR signaling in the VPCs that do not adopt vulval fates. eor-1 and eor-2 encode a PLZF-related protein and a novel nuclear protein, respectively [84]. Single mutations in these genes do not affect vulval development. However, an eor-2 mutation causes a synthetic vulvaless phenotype in the presence of a sur-8 mutation, and mutation of either eor-1 or eor-2 reduces the penetrance of the multivulva phenotype caused by a loss-of-function mutation in lin-1. Thus, the EOR proteins likely act downstream or parallel to a very distal step in EGFR signaling, and provide another layer of complexity. Specification of the vulval F fate (VulF) The eight progeny of P6.p are located at the apex of the vulva and can be separated into two distinct classes. The four outermost cells adopt the E fate (VulE), while the four innermost cells adopt the F fate (VulF). These fates also can be molecularly visualized by transcriptional induction of a zinc metalloprotease gene, zmp-1, in the VulE cells, but not the VulF cells [85,86]. The LIN-3-producing anchor cell is only required up to the first cell division of P6.p in order to execute inductive signaling by the EGFR [7]. If the anchor cell is ablated after this point, three rounds of cell division still occur. However, if the anchor cell is ablated between the two-cell to early four-cell stage, VulE and VulF fates are abnormal [84]. Normal fates can be restored in this background by constitutive activation of LET-23, but not by a loss-of-function mutation in the lin-15 gene [85]. Furthermore, expression of dominant-negative LET-60 after the two-cell stage also impairs specification of the VulE and VulF fates, without affecting the vulval versus nonvulval cell fate decision of the VPCs [85]. Thus, it appears that EGFR signaling is used a second time after the initial vulval induction to specify subfates within vulval tissue. However, signaling in these vulval cells is different from that in the VPCs. In contrast to vulval induction, the RAS-dependent signaling pathway in the vulval cells requires LIN-1, but not LIN-31, and is not inhibited by the synmuv genes [85]. Reverse signaling from the vulva to the somatic gonad After vulval induction has occurred and vulval subfates have been specified, four of the 12 daughters of the ventral uterine pi cells are induced to become uv1 cells, which join the uterine and vulval epithelia by contacting the vulF vulval

162


cells [87,88]. These presumptive uv1 cells lie above the VulE and VulF vulval cells. lin-3 is expressed during the L4 larval stage in the VulF cells, and LET-23 is expressed in the uv1 cells [89]. Ablation of P6.p or the progeny that give rise to the four VulF cells prevent the uv1 fate from being specified [89]. Similarly, a mutation in the egl-38 pax gene that eliminates lin-3 expression in VulF, but not the anchor cell, also prevents uv1 specification [89]. Thus, it is likely that after being stimulated by EGF from the somatic gonad, the vulva produces EGF to reciprocally specify a uterine fate. Rare animals harboring a mutation in lin-45 or expressing dominant-negative LET-60 that do not have defects in vulval development display defects in uv1 specification [89]. These data suggest that the EGFR acts in the presumptive uv1 cells through activation of RAS and RAF. A gain-of-function mutation in lin-1 inhibits uv1 fates, suggesting that LIN-1 acts as an inhibitor of EGFR signaling in these cells, similar to in the VPCs [89]. However, animals in which the vulvaless phenotype caused by a let-23 mutation is strongly suppressed by a mutation in gap-1 still display a strong defect in uv1 specification [89]. Thus, GAP1 appears to be a more critical inhibitor of EGFR signaling in the VPCs than in the presumptive uv1 cells. Viability Strong mutations in the core EGFR pathway genes let23, sem-5, let-341, let-60, lin-45, mek-2, and sur-1/mpk-1 result in lethality during the early L1 larval stage [1,5,18–20,22–24,26,30,48,90]. In contrast, mutations in the modulators ksr-1, ksr-2, sur-6, and sur-8, or the mediator/nuclear proteins sur-2, lin-25, eor-1, and eor-2, do not cause highly penetrant L1 lethality [32,39,80,84,91,92]. However, in different double mutant backgrounds with other weakly penetrant mutations, mutation of either the modulators or the nuclear proteins causes a much higher incidence of larval lethality. Lethality resulting from loss of LET-23 function can be rescued by a gain-of-function mutation in let-60, and lethality due to the presence of a homozygous dominant-negative mutation in let-60 is suppressed by a mutation in lin-1 [48,93]. These data suggest that a linear signaling cascade, similar to the one in the VPCs, also operates to promote larval growth. Mosaic analysis indicates that lethality due to loss of LET-60 results from failure to specify the duct cell fate in the excretory system, suggesting that the duct cell precursor is the site of action for this cascade [29]. However, mosaic data for let-23 and studies with egl-15, an FGF receptor-like gene, suggest a more complicated model. Mosaic experiments suggest that LET-23 is required in at least two different lineages to promote viability [28]. Furthermore, a strong mutation in egl-15 also results in L1 lethality that is rescued by a gain-of-function mutation in let-60 [94]. Thus, it may be that the combined action of two different receptor tyrosine kinases is necessary to provide sufficient RAS activity for larval growth. Further experiments are required to reconcile these observations.

P12 fate There are 12 postembryonic ventral cord precursor cells (P1–P12) [9]. During the early phase of the L1 larval stage, before P11 and P12 have entered the ventral cord, both cells have the ability to adopt the P12 fate [8]. If one cell is ablated, the remaining cell adopts the P12 fate. Similar to vulval induction, specification of the P12 fate is dependent on the action of both EGF and WNT signaling. Mutations in either lin-44 (WNT) or lin-17 (FRIZZLED) cause the normally specified P12 fate to be replaced by a second P11 fate [95,96]. Similarly single mutations in let-23, sem-5, and let-60 also result in two P11 fates [96]. Conversely, overexpression of LIN-3 or a lin-15 loss-of-function mutation results in two P12 fates [96]. Thus, EGF and WNT signaling promotes the P12 fate, and similar to in the VPCs, EGFR signaling in the ventral cord precursors is antagonized by the synmuv pathway. Also similar to vulval induction, P12 specification requires the integration of EGF and WNT signaling at the level of a Hox gene. Mutation of the Hox gene egl-5 results in two P11 fates, and this phenotype is epistatic to the two P12 fates induced by either overexpression of LIN-3 or a lin-15 mutation [96]. Furthermore, overexpression of EGL-5 rescues the phenotype of two P11 fates observed in animals harboring mutations in either the wnt or egfr pathway, suggesting that egl-5 is a common target for these signaling pathways [96]. Mutations in either sur-2 or lin-25 cause a low incidence of two P11 fates that is enhanced in the presence of a let-23 mutation [91]. Mutations in either eor-1 or eor-2 result in a more penetrant phenotype of two P11 fates that is not enhanced in a eor-1; eor-2 double mutant [84]. Thus, the EOR proteins likely function together to promote the P12 fate. Mutation of an eor gene with either a loss-of-function mutation in sur-8 or a gain-offunction mutation in lin-1 results in a high frequency of two P11 fates, indicating that this EGFR pathway also is modulated by SUR-8 and inhibited by LIN-1 [84]. Spicule development The spicules are part of the male-specific copulatory apparatus. Each of the two spicules comprises two sensory neurons, glial cells and structural cells that secrete the hardened cuticle that gives the spicules their rigidity [97,104]. A motor neuron is associated with each spicule, and the spicules attach to protractor and retractor muscles. The spicule sensory neurons are involved in vulval location, spicule insertion, and sperm transfer [8,98,105]. The spicules are inserted into the vulva during copulation and anchor the male during sperm transfer. The spicules develop from a single precursor cell, called B, present at hatching that divides only in the male [9,104]. The anterior daughter of B (B.a) generates eight progeny that are grouped into four pairs of cells. The fate of each cell in a pair is dependent on whether it occupies an anterior or posterior position. Just anterior and dorsal to these cells are cells derived from the


male-specific blast cells F and U. The F and U progeny are necessary for the specification of anterior fates in the B.a progeny, suggesting that the F and U lineages supply an inductive signal [99]. These inductions require EGFR signaling [100]. Similar to vulval induction, LIN-3, LET-23, SEM-5, LET-60, and LIN-45 are required for anterior fate specification, and the lin-15 gene acts as an inhibitor of this pathway. lin-3 is expressed in F and U cells (B.J. Hwang and P.W. Sternberg, unpublished observations), suggesting that EGF is the inductive signal from the F and U lineages, and that the EGFR-RAS pathway is activated in the B.a progeny. Each anterior fate specified by EGF signaling leads to different outcomes that are not correlated with the extent of proliferation or the cell types produced. Thus, the EGFR pathway acts to choose alternative developmental pathways rather than induce a specific cell biological change. In contrast to vulva and P12 development, the uninduced cells do not fuse with an epidermal syncytium, but instead adopt an alternative cellular fate. Specific target genes have not been studied.

163

factors, and methods of detecting signal transduction rather than just the end product, namely differentiated vulval cells. Analysis of target genes is also crucial to elucidate how EGFR signaling is integrated with LIN-12 (Notch) and WNT signaling pathways. Some of these targets might be among the microarray upregulated genes [102]. While the ovulation and cell fate specification pathways are clearly separable, determinants of specificity among the RASdependent processes remain to be discovered, as well as the mechanism by which HOX genes regulate responsiveness to EGF signaling [61,62,96]. Acknowledgments P.W.S. is an Investigator with the Howard Hughes Medical Institute. N.M. is a Fellow of the California Breast Cancer Research Program and was previously a fellow of the Leukemia and Lymphoma Society. References

Behavior Ovulation Some mutations in let-23 and lin-3 result in a sterile phenotype [1,5]. This phenotype results from a failure in ovulation, a rhythmic behavior in which individual oocytes are engulfed by the spermatheca so that they can be fertilized by sperm. Ovulation appears to be a RAS-independent process [46]. The signaling pathway regulating ovulation, identified by extragenic suppressors of lin-3 (and let-23) sterility, involves IP3 (inositol 1,4,5-triphophate) [93]. One type of suppressor is a loss-of-function mutation in IP3 kinase; the second class comprises a gain-of-function mutation in the IP3 receptor, suggesting that LET-23 increases IP3 levels to promote ovulation [93]. Activation of both the RAS and IP3 pathways bypasses the need for LET-23 for viability, vulval development, and ovulation. Deletion of an IP3-5-phosphatase results in double ovulations and suppresses the sterile phenotype of mutations in let-23 and lin3, further supporting the hypothesis that LET-23 regulates this behavior by controlling IP3 levels [101]. Proximal effectors of LET-23 in this pathway, such as the phospholipase, are not known. However, ARK-1 and SEM-5 inhibit the IP3 pathway [54]. Since ARK-1 binds to SEM-5, it is possible that GRB-2-mediated recruitment of ARK-1 to the EGFR inhibits the activation of both RAS and IP3 signaling cascades.

Prospects Future progress is likely to come from identification of target genes to enable the analysis of the many nuclear

[1] R. Aroian, P. Sternberg, Multiple functions of let-23, a Caenorhabditis elegans receptor tyrosine kinase gene required for vulval induction, Genetics 128 (1991) 251–267. [2] R.V. Aroian, M. Koga, J.E. Mendel, Y. Ohshima, P.W. Sternberg, The let-23 gene necessary for Caenorhabditis elegans vulval-induction encodes a tyrosine kinase of the EGF receptor subfamily, Nature 348 (1990) 693–699. [3] J. Simske, S. Kaech, S. Harp, S. Kim, LET-23 receptor localization by the cell junction protein LIN-7 during C. elegans vulval induction, Cell 85 (1996) 195–204. [4] R. Hill, P. Sternberg, The gene lin-3 encodes an inductive signal for vulval development in C. elegans, Nature 358 (1992) 470–476. [5] E. Ferguson, H. Horvitz, Identification and characterization of 22 genes that affect the vulval cell lineages of the nematode C. elegans, Genetics 110 (1985) 17–72. [6] J. Liu, P. Tzou, R. Hill, P. Sternberg, Structural requirements for the tissue-specific and tissue-general functions of the Caenorhabditis elegans epidermal growth factor LIN-3, Genetics 153 (1999) 1257–1269. [7] J. Kimble, Alterations in cell lineage following laser ablation of cells in the somatic gonad of C. elegans, Dev. Biol. 87 (1981) 286–300. [8] J. Sulston, J. White, Regulation and cell autonomy during postembryonic development of C. elegans, Dev. Biol. 78 (1980) 577–597. [9] J. Sulston, H. Horvitz, Post-embryonic cell lineages of the nematode, Caenorhabditis elegans, Dev. Biol. 56 (1977) 110–156. [10] P. Sternberg, H. Horvitz, Pattern formation during vulval development in C. elegans, Cell 44 (1986) 761–772. [11] I. Greenwald, P. Sternberg, H. Horvitz, The lin-12 locus specifies cell fates in C. elegans, Cell 34 (1983) 435–444. [12] J. Yochem, K. Weston, I. Greenwald, The Caenorhabditis elegans lin12 gene encodes a transmembrane protein with overall similarity to Drosophila Notch, Nature 335 (1988) 547–550. [13] J. Simske, S. Kim, Sequential signaling during Caenorhabditis elegans vulval induction, Nature 375 (1995) 142–146. [14] W. Katz, R. Hill, T. Clandinin, P. Sternberg, Different levels of the C. elegans growth factor LIN-3 promote distinct vulval precursor fates, Cell 82 (1995) 297–307. [15] S. Kaech, C. Whitfield, S. Kim, The LIN-2/LIN-7/LIN-10 complex mediates basolateral membrane localization of the C. elegans EGF receptor LET-23 in vulval epithelial cells, Cell 94 (1998) 761–771.

164


[16] C. Whitfield, C. Benard, T. Barnes, S. Hekimi, S. Kim, Basolateral localization of the Caenorhabditis elegans epidermal growth factor receptor in epithelial cells by the PDZ protein LIN-10, Mol. Biol. Cell 10 (1999) 2087–2100. [17] R. Hoskins, A. Hajnal, S. Harp, S. Kim, The C. elegans vulval induction gene lin-2 encodes a member of the MAGUK family of cell junction proteins, Development 122 (1996) 97–111. [18] S. Clark, M. Stern, H. Horvitz, C. elegans cell signalling gene sem5 encodes a protein with SH2 and SH3 domains, Nature 356 (1992a) 340–344. [19] C. Chang, N. Hopper, P. Sternberg, Caenorhabditis elegans SOS-1 is necessary for multiple Ras-mediated developmental signals, EMBO J. 19 (2000) 3283–3294. [20] G. Beitel, S. Clark, H. Horvitz, Caenorhabditis elegans ras gene let-60 acts as a switch in the pathway of vulval induction, Nature 348 (1990) 503–509. [21] M. Han, P. Sternberg, let-60, a gene that specifies cell fates during C. elegans vulval induction, encodes a ras protein, Cell 63 (1990) 921–931. [22] M. Han, A. Golden, Y. Han, P. Sternberg, C. elegans lin-45 raf gene participates in let-60 RAS-stimulated vulval differentiation, Nature 363 (1993) 133–140. [23] K. Kornfeld, K.-L. Guan, H. Horvitz, The Caenorhabditis elegans gene mek-2 is required for vulval induction and encodes a protein similar to the protein kinase MEK, Genes Dev. 9 (1995a) 756–768. [24] Y. Wu, M. Han, K.-L. Guan, MEK-2, a Caenorhabditis elegans MAP kinase kinase, functions in Ras-mediated vulval induction and other developmental events, Genes Dev. 9 (1995) 742–755. [25] M. Lackner, K. Kornfeld, L. Miller, H. Horvitz, S. Kim, A MAP kinase homolog, mpk-1, is involved in ras-mediated induction of vulval cell fates in Caenorhabditis elegans, Genes Dev. 8 (1994) 160–173. [26] Y. Wu, M. Han, Suppression of activated Let-60 Ras protein defines a role of Caenorhabditis elegans Sur-1 MAP kinase in vulval differentiation, Genes Dev. 8 (1994) 147–159. [27] J. Dent, M. Han, Post-embryonic expression pattern of C. elegans let-60 ras reporter constructs, Mech. Dev. 72 (1998) 179–182. [28] M. Koga, Y. Ohshima, Mosaic analysis of the let-23 gene function in vulval induction of Caenorhabditis elegans, Development 121 (1995) 2655–2666. [29] J. Yochem, M. Sundaram, M. Han, Ras is required for a limited number of cell fates and not for general proliferation in Caenorhabditis elegans, Mol. Cell. Biol. 17 (1997) 2716–2722. [30] M. Lackner, S. Kim, Genetic analysis of the Caenorhabditis elegans MAP kinase gene mpk-1, Genetics 150 (1998) 103–117. [31] K. Kornfeld, D. Hom, H. Horvitz, The ksr-1 gene encodes a novel protein kinase involved in Ras-mediated signaling in C. elegans, Cell 83 (1995b) 903–913. [32] M. Sundaram, M. Han, The C. elegans ksr-1 gene encodes a novel Raf-related kinase involved in Ras-mediated signal transduction, Cell 83 (1995) 889–901. [33] L. Selfors, J. Schutzman, C. Borland, M. Stern, soc-2 encodes a leucine-rich repeat protein implicated in fibroblast growth factor receptor signaling, Proc. Natl. Acad. Sci. USA 95 (1998) 6903–6908. [34] D. Sieburth, Q. Sun, M. Han, SUR-8, a conserved Ras-binding protein with leucine-rich repeats, positively regulates Ras-mediated signaling in C. elegans, Cell 94 (1998) 119–130. [35] D. Sieburth, M. Sundaram, R. Howard, M. Han, A PP2A regulatory subunit positively regulates Ras-mediated signaling during Caenorhabditis elegans vulval induction, Genes Dev. 13 (1999) 2562–2569. [36] D. Jacobs, D. Glossip, H. Xing, A.J. Muslin, K. Kornfeld, Multiple docking sites on substrate proteins form a modular system that mediates recognition by ERK MAP kinase, Genes Dev 13 (1999) 163–175. [37] S. Stewart, M. Sundaram, Y. Zhang, J. Lee, M. Han, K. Guan, Kinase suppressor of Ras forms a multiprotein signaling complex and modulates MEK localization, Mol. Cell. Biol. 19 (1999) 5523–5534.

[38] J.J. Bruinsma, T. Jirakulaporn, A.J. Muslin, K. Kornfeld, Zinc ions and cation diffusion facilitator proteins regulate Ras-mediated signaling, Dev. Cell 2 (2002) 567–578. [39] M. Ohmachi, C.E. Rocheleau, D. Church, E. Lambie, T. Schedl, M.V. Sundaram, C. elegans ksr-1 and ksr-2 have both unique and redundant functions and are required for MPK-1 ERK phosphorylation, Curr. Biol. 12 (2002) 427–433. [40] M. Gutch, A. Flint, J. Keller, N. Tonks, M. Hengartner, The Caenorhabditis elegans SH2 domain-containing protein tyrosine phosphatase PTP-2 participates in signal transduction during oogenesis and vulval development, Genes Dev. 12 (1998) 571–585. [41] I.D. Zipkin, R.M. Kindt, C.J. Kenyon, Role of a new Rho family member in cell migration and axon guidance in C. elegans, Cell 90 (1997) 883–894. [42] R. Steven, T.J. Kubiseski, H. Zheng, S. Kulkarni, J. Mancillas, A. Ruiz Morales, C.W. Hogue, T. Pawson, J. Culotti, UNC-73 activates the Rac GTPase and is required for cell and growth cone migrations in C. elegans, Cell 92 (1998) 785–795. [43] R.S. Kishore, M.V. Sundaram, ced-10 Rac and mig-2 function redundantly and act with unc-73 trio to control the orientation of vulval cell divisions and migrations in Caenorhabditis elegans, Dev. Biol. 241 (2002) 339–348. [44] P.W. Reddien, H.R. Horvitz, CED-2/CrkII and CED-10/Rac control phagocytosis and cell migration in Caenorhabditis elegans, Nat. Cell. Biol. 2 (2000) 131–136. [45] J. Lee, G. Jongeward, P. Sternberg, unc-101, a gene required for many aspects of Caenorhabditis elegans development and behavior, encodes a clathrin-associated protein, Genes Dev. 8 (1994) 60–73. [46] G. Jongeward, T. Clandinin, P. Sternberg, sli-1, a negative regulator of let-23-mediated signaling in C. elegans, Genetics 139 (1995) 1553–1566. [47] C. Yoon, J. Lee, G. Jongeward, P. Sternberg, Similarity of sli-1, a regulator of vulval development in C. elegans, to the mammalian proto-oncogene c-cbl, Science 269 (1995) 1102–1105. [48] M. Han, R. Aroian, P. Sternberg, The let-60 locus controls the switch between vulval and nonvulval cell fates in Caenorhabditis elegans, Genetics 126 (1990) 899–913. [49] T. Gu, S. Orita, M. Han, Caenorhabditis elegans SUR-5, a novel but conserved protein, negatively regulates LRT-60 ras activity during vulval induction, Mol. Cell. Biol. 18 (1998) 4556–4564. [50] J. Shim, P.W. Sternberg, J. Lee, Distinct and redundant functions of mu1 medium chains of the AP-1 clathrin-associated protein complex in the nematode Caenorhabditis elegans, Mol. Biol. Cell. 11 (2000) 2743–2756. [51] C.H. Yoon, C. Chang, N.A. Hopper, G.M. Lesa, P.W. Sternberg, Requirements of multiple domains of SLI-1, a Caenorhabditis elegans homologue of c-Cbl, and an inhibitory tyrosine in LET-23 in regulating vulval differentiation, Mol. Biol. Cell 11 (2000) 4019–4031. [52] M.L. Lupher Jr., Z. Songyang, S.E. Shoelson, L.C. Cantley, H. Band, The Cbl phosphotyrosine-binding domain selects a D(N/D)XpY motif and binds to the Tyr292 negative regulatory phosphorylation site of ZAP-70, J. Biol. Chem. 272 (1997) 33140–33144. [53] W. Meng, S. Sawasdikosol, S.J. Burakoff, M.J. Eck, Structure of the amino-terminal domain of Cbl complexed to its binding site on ZAP70 kinase, Nature 398 (1999) 84–90. [54] N. Hopper, J. Lee, P. Sternberg, ARK-1 inhibits EGFR signaling in C. elegans, Mol. Cell 6 (2000) 65–75. [55] T. Berset, E.F. Hoier, G. Battu, S. Canevascini, A. Hajnal, Notch inhibition of RAS signaling through MAP kinase phosphatase LIP1 during C. elegans vulval development, Science 291 (2001) 1055–1058. [56] G. Beitel, S. Tuck, I. Greenwald, H. Horvitz, The Caenorhabditis elegans gene lin-1 encodes an ETS-domain protein and defines a branch of the vulval induction pathway, Genes Dev. 9 (1995) 3149–3162. [57] L. Miller, M. Gallegos, B. Morisseau, S. Kim, lin-31, a Caenorhabditis elegans HNF-3/fork head transcription factor homolog, speci-


[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65] [66]

[67]

[68]

[69]

[70]

[71]

[72]

[73]

[74]

[75]

[76]

[77]

[78]

fies three alternative cell fates in vulval development, Genes Dev. 7 (1993) 933–947. D. Jacobs, G. Beitel, S. Clark, H. Horvitz, K. Kornfeld, Gain-offunction mutations in the Caenorhabditis elegans lin-1 ETS gene identify a C-terminal regulatory domain phosphorylated by ERK MAP kinase, Genetics 149 (1998) 1809–1822. P. Tan, M. Lackner, S. Kim, MAP kinase signaling specificity mediated by the LIN-1 Ets/LIN-31 WH transcription factor complex during C. elegans vulval induction, Cell 93 (1998) 569–580. B. Wang, M. Muller-Immergluck, J. Austin, N. Robinson, A. Chisholm, C. Kenyon, A homeotic gene-cluster patterns the anteroposterior body axis of C. elegans, Cell 74 (1993) 29–42. J. Maloof, C. Kenyon, The Hox gene lin-39 is required during C. elegans vulval induction to select the outcome of Ras signaling, Development 125 (1998) 181–190. T. Clandinin, W. Katz, P. Sternberg, Caenorhabditis elegans HOMC genes regulate the response of vulval precursor cells to inductive signal, Dev. Biol. 182 (1997) 150–161. S. Salser, C. Loer, C. Kenyon, Multiple HOM-C gene interactions specify cell fates in the nematode central nervous system, Genes Dev. 7 (1993) 1714–1724. P. Sternberg, H. Horvitz, The combined action of two intercellular signaling pathways specifies three cell fates during vulval induction in C. elegans, Cell 58 (1989) 679–693. M. Wang, P. Sternberg, Competence and commitment of Caenorhabditis elegans vulval precursor cells, Dev. Biol. 212 (1999) 12–24. D. Eisenmann, J. Maloof, J. Simske, C. Kenyon, S. Kim, The B-catenin homolog BAR-1 and LET-60 Ras coordinately regulate the Hox gene lin-39 during Caenorhabditis elegans vulval development, Development 125 (1998) 3667–3680. J. Maloof, J. Whangbo, J. Harris, G. Jongeward, C. Kenyon, A Wnt signaling pathway controls Hox gene expression and neuroblast migration in C. elegans, Development 126 (1999) 37–49. H.C. Korswagen, D.Y. Coudreuse, M.C. Betist, S. Van De Water, D. Zivkovic, H.C. Clevers, The Axin-like protein PRY-1 is a negative regulator of a canonical Wnt pathway in C. elegans, Genes Dev. 16 (2002) 1291–1302. J.E. Gleason, H.C. Korswagen, D.M. Eisenmann, Activation of Wnt signaling bypasses the requirement for RTK/Ras signaling during C. elegans vulval induction, Genes Dev. 16 (2002) 1281–1290. E. Ferguson, H. Horvitz, The multivulva phenotype of certain Caenorhabditis elegans mutants results from defects in two functionally redundant pathways, Genetics 123 (1989) 109–121. S. Clark, W. Lu, H. Horvitz, The Caenorhabditis elegans locus lin15, a negative regulator of a tyrosine kinase signaling pathway, encodes two different proteins, Genetics 137 (1994) 987–997. L. Huang, P. Tzou, P. Sternberg, The lin-15 locus encodes two negative regulators of Caenorhabditis elegans vulval development, Mol. Biol. Cell. 5 (1994) 395–411. J. Thomas, H. Horvitz, The C. elegans gene lin-36 acts cell autonomously in the lin-35 Rb pathway, Development 126 (1999) 3449–3459. X. Lu, H. Horvitz, lin-35 and lin-53, Two genes that antagonize a C. elegans Ras pathway, encode proteins similar to Rb and its binding protein RbAp48, Cell 95 (1998) 981–991. F. Solari, J. Ahringer, NURD-complex genes antagonise Ras-induced vulval development in Caenorhabditis elegans, Curr. Biol. 10 (2000) 223–226. T. von Zelewsky, F. Palladino, K. Brunschwig, H. Tobler, A. Hajnal, F. Muller, The C. elegans Mi-2 chromatin-remodelling proteins function in vulval cell fate determination, Development 127 (2000) 5277–5284. C.J. Ceol, H.R. Horvitz, dpl-1 DP and efl-1 E2F act with lin-35 Rb to antagonize Ras signaling in C. elegans vulval development, Mol. Cell 7 (2001) 461–473. R. Herman, E. Hedgecock, Limitation of the size of the vulval primordium of Caenorhabditis elegans by lin-15 expression in surrounding hypodermis, Nature 348 (1990) 169–171.

165

[79] R. Herman, Mosaic analysis, in: H.F. Epstein, D.C. Shakes (Eds.), Methods in Cell Biology. Vol. 48: Caenorhabditis elegans: Modern Biological Analysis of an Organism, Academic Press, Inc., San Diego, 1995, pp. 123–146. [80] N. Singh, M. Han, sur-2, a novel gene, functions late in the let-60 ras-mediated signaling pathway during Caenorhabditis elegans vulval induction, Genes Dev. 9 (1995) 2251–2265. [81] S. Tuck, I. Greenwald, lin-25, a gene required for vulval induction in Caenorhabditis elegans, Genes Dev. 9 (1995) 341–357. [82] L. Nilsson, X. Li, T. Tiensuu, R. Auty, I. Greenwald, S. Tuck, Caenorhabditis elegans lin-25: cellular focus, protein expression and requirement for sur-2 during induction of vulval fates, Development 125 (1998) 4809–4819. [83] N. Moghal, P.W. Sternberg, A component of the transcriptional mediator complex inhibits RAS-dependent vulval fate specification in C. elegans, Development 130 (2003) 57–69. [84] R.M. Howard, M.V. Sundaram, C. elegans EOR-1/PLZF and EOR2 positively regulate Ras and Wnt signaling and function redundantly with LIN-25 and the SUR-2 Mediator component, Genes Dev. 16 (2002) 1815–1827. [85] M. Wang, P.W. Sternberg, Patterning of the C. elegans 1 degrees vulval lineage by RAS and Wnt pathways, Development 127 (2000) 5047–5058. [86] T. Inoue, D.R. Sherwood, G. Aspöck, J.A. Butler, B.P. Gupta, M. Kirouac, M. Wang, P.-Y., Lee, J.M. Kramer, I. Hope, T. Bürglin, P.W. Sternberg, Gene expression markers for C. elegans vulval cells, Mech. Dev. 2 (2003) 235–241. [87] A. Newman, P. Sternberg, Coordinated morphogenesis of epithelia during development of the Caenorhabditis elegans uterine-vulva connection, Proc. Natl. Acad. Sci. USA 93 (1996) 9329–9333. [88] A. Newman, J. White, P. Sternberg, Morphogenesis of the C. elegans hermaphrodite uterus, Development 122 (1996) 3617–3626. [89] C. Chang, A. Newman, P. Sternberg, Reciprocal EGF signaling back to the uterus from the induced C. elegans vulva coordinates morphogenesis of epithelia, Curr. Biol. 9 (1999) 237–246. [90] S.G. Clark, M.J. Stern, H.R. Horvitz, Genes involved in two Caenorhabditis elegans cell-signaling pathways, Cold Spring Harb. Symp. Quant. Biol. 57 (1992b) 363–373. [91] L. Nilsson, T. Tiensuu, S. Tuck, Caenorhabditis elegans lin-25: a study of its role in multiple cell fate specification events involving Ras and the identification and characterization of evolutionarily conserved domains, Genetics 156 (2000) 1083–1096. [92] C.E. Rocheleau, R.M. Howard, A.P. Goldman, M.L. Volk, L.J. Girard, M.V. Sundaram, A lin-45 raf enhancer screen identifies eor1, eor-2 and unusual alleles of Ras pathway genes in Caenorhabditis elegans, Genetics 161 (2002) 121–131. [93] T. Clandinin, J. DeModena, P. Sternberg, Inositol trisphospate mediates a Ras-independent response to LET-23 receptor tyrosine kinase activation in C. elegans, Cell 92 (1998) 523–533. [94] D.L. DeVore, H.R. Horvitz, M.J. Stern, An FGF receptor signaling pathway is required for the normal cell migrations of the sex myoblasts in C. elegans hermaphrodites, Cell 83 (1995) 611– 620. [95] M. Herman, H. Horvitz, The Caenorhabditis elegans gene lin-44 controls the polarity of asymmetric cell divisions, Development 120 (1994) 1035–1047. [96] L. Jiang, P. Sternberg, Interactions of EGF, Wnt and Hom-C genes specify the P12 neuroectoblast fate in C. elegans, Development 125 (1998) 2337–2347. [97] L. Jiang, P.W. Sternberg, Socket cells mediate spicule morphogenesis in Caenorhabditis elegans males, Dev. Biol. 211 (1999) 88– 99. [98] L.R. Garcia, P. Mehta, P.W. Sternberg, Regulation of distinct muscle behaviors controls the C. elegans male’s copulatory spicules during mating, Cell 107 (2001) 777–788. [99] H.M. Chamberlin, P.W. Sternberg, Multiple cell interactions are required for fate specification during male spicule development in Caenorhabditis elegans, Development 118 (1993) 297–324.

166


[100] H.M. Chamberlin, P.W. Sternberg, The lin-3/let-23 signaling pathway mediates induction during male spicule development in Caenorhabditis elegans, Development 120 (1994) 2713–2721. [101] Y. Bui, P.W. Sternberg, C. elegans inositol 5-phosphatase homologue negatively regulates inositol 1,4,5-triphosphate signaling in ovulation, Mol. Biol. Cell 13 (2002) 1641–1651. [102] B. Romagnolo, M. Jiang, M. Kiraly, C. Breton, R. Begley, J. Wang, J. Lund, S.K. Kim, Downstream targets of let-60 Ras in Caenorhabditis elegans, Dev. Biol. 247 (2002) 127–136. [103] C.H. Yoon, C. Chang, N.A. Hopper, G.M. Lesa, P.W. Sternberg, Requirements of multiple domains of SLT-1, a Caenorhabditis

elegans homologue of c-Cbl, and an inhibitory tyrosine in LET-23 in regulating vulval differentiation, Mol. Biol. Cell. 11 (2000) 4019–4031. [104] J.E. Sulston, D.G. Albertson, J.N. Thomson, The Caenorhabditis elegans male: postembryonic development of nongonadal structures, Dev. Biol. 78 (1980) 542–576. [105] K.S. Liu, P.W. Sternberg, Sensory regulation of male mating behavior in Caenorhabditis elegans, Neuron 14 (1995) 79–89. [106] A. Hajnal, C. Whitfield, S. Kim, Inhibition of Caenorhabditis elegans vulval induction by gap-1 and by let-23 receptor tyrosine kinase, Genes & Devel. 11 (1997) 2715–2728.

The EGF Receptor Family: Biologic Mechanisms and Role in Cancer

Hematopoietic Receptor Family

IL-1 Receptor Family

Molecular Mechanisms in Gastrointestinal Cancer

Drug Resistance in Cancer: Mechanisms and Models

Molecular Mechanisms of Cancer

Biologic Markers in Immunotoxicology

IL-6 Ligand and Receptor Family

Tumor Angiogenesis: Basic Mechanisms and Cancer Therapy

The Role of Religion in Marriage and Family Counseling (The Family Therapy and Counseling Series)

Copper Amine Oxidases: Structures, Catalytic Mechanisms and Role in Pathophysiology

The Role of Religion in Marriage and Family Counseling (The Family Therapy and Counseling Series)

Biologic Therapy of Leukemia

The AH Receptor in Biology and Toxicology

Our Family Has Cancer, Too!

Perspectives in Receptor Research

Cancer. The Role of Genes, Lifestyle and Environment

Cancer metastatis, molecular and cellular mechanisms and clinical intervention

Semaphorins: Receptor and Intracellular Signaling Mechanisms (Advances in Experimental Medicine and Biology)

The Delta Receptor

Bioinformatics in Cancer and Cancer Therapy

The Delta Receptor

The Nuclear Receptor FactsBook

Mechanisms and Concepts in Toxicology

ACTH Receptor

b Receptor

Techniques and Mechanisms in Electrochemistry

Progress in Brain Research Volume 43 Somatosensory and Visceral Receptor Mechanisms

C5a Receptor

The EGF Receptor Family: Biologic Mechanisms and Role in Cancer

Hematopoietic Receptor Family

IL-1 Receptor Family

Molecular Mechanisms in Gastrointestinal Cancer

Drug Resistance in Cancer: Mechanisms and Models

Molecular Mechanisms of Cancer

Biologic Markers in Immunotoxicology

IL-6 Ligand and Receptor Family

Tumor Angiogenesis: Basic Mechanisms and Cancer Therapy

The Role of Religion in Marriage and Family Counseling (The Family Therapy and Counseling Series)

Copper Amine Oxidases: Structures, Catalytic Mechanisms and Role in Pathophysiology

The Role of Religion in Marriage and Family Counseling (The Family Therapy and Counseling Series)

Biologic Therapy of Leukemia

The AH Receptor in Biology and Toxicology

Our Family Has Cancer, Too!

Perspectives in Receptor Research

Cancer. The Role of Genes, Lifestyle and Environment

Cancer metastatis, molecular and cellular mechanisms and clinical intervention

Semaphorins: Receptor and Intracellular Signaling Mechanisms (Advances in Experimental Medicine and Biology)

The Delta Receptor

Bioinformatics in Cancer and Cancer Therapy

The Delta Receptor

The Nuclear Receptor FactsBook

Mechanisms and Concepts in Toxicology

ACTH Receptor

b Receptor

Techniques and Mechanisms in Electrochemistry

Progress in Brain Research Volume 43 Somatosensory and Visceral Receptor Mechanisms

C5a Receptor

Recommend Documents