ã

Oncogene (2001) 20, 1594 ± 1600 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc

Cell communication networks: epidermal growth factor receptor transactivation as the paradigm for interreceptor signal transmission Andreas Gschwind1, Esther Zwick1, Norbert Prenzel1, Michael Leserer1 and Axel Ullrich*,1 1

Max-Planck-Institut fuÈr Biochemie, Martinsried, Germany

Communication between di€erent cellular signaling systems has emerged as a common principle that enables cells to integrate a multitude of signals from its environment. Transactivation of the epidermal growth factor receptor (EGFR) represents the paradigm for cross-talk between G protein-coupled receptors (GPCRs) and receptor tyrosine kinases (RTKs). The recent identi®cation of Zn2+-dependent metalloproteinases and transmembrane growth factor precursors as critical elements in GPCR-induced EGFR transactivation pathways has de®ned new components of a cellular communication network of rapidly increasing complexity. Further elucidation of the molecular details of the EGFR transactivation mechanism will provide new understanding of its relevance for normal physiological processes and their pathophysiological deviations. Oncogene (2001) 20, 1594 ± 1600. Keywords: EGFR; GPCR; transactivation; cross-talk; ectodomain shedding; ADAMs Introduction Cross-communication between diverse signaling networks is a well established concept in the cellular stimulus-response conversion system that allows the translation of complex environmental conditions into appropriate reactions and adaptations. The EGFR which belongs to a family of four closely related receptor tyrosine kinases has been recognized as a convergence point for diverse signal transduction pathways (reviewed in Hackel et al., 1999; Moghal and Sternberg, 1999). Depending on the activating ligand EGFR family members are able to form various homo- or heterodimers with di€erent biological signaling capacities (reviewed in Olayioye et al., 2000). While signaling through the EGFR family kinases is essential for the development of fruit ¯ies, nematodes and mice (Moghal and Sternberg, 1999), their critical importance for the human organism is underscored by their frequent involvement in cancer development due to overexpression, activating muta-

*Correspondence: A Ullrich, Max-Planck-Institut fuÈr Biochemie, Department of Molecular Biology, Am Klopferspitz 18A, 82152 Martinsried, Germany

tions or autocrine stimulation by diverse EGF-like ligands. The six known ligands that act as direct agonists for the EGFR, EGF, heparin-binding EGF-like growth factor (HB-EGF), transforming growth factor alpha (TGFa), amphiregulin, betacellulin and epiregulin are synthesized as transmembrane precursors and must therefore be proteolytically cleaved by metalloproteases to release the mature growth factor (reviewed in Massague and Pandiella, 1993). The agonist occupied EGFR undergoes autophosphorylation and activates the Ras/mitogen activated protein (MAP) kinase pathway through initial tyrosine phosphorylation of the adaptor protein SHC, subsequent formation of a SHC-Grb2-Sos complex and Ras-mediated induction of Raf function. This prototypical signaling pathway couples EGFR stimulation to gene transcription and mitogenesis. In addition to its cognate ligands, the EGFR is activated by stimuli that do not directly interact with the EGFR ectodomain including GPCR ligands, other RTK agonists, cytokines, chemokines and cell adhesion elements. Furthermore the EGFR signal may also be released by non-physiological in¯uences such as UV and gamma radiation, osmotic shock, membrane depolarization, heavy metal ions and radical-generating agents such as hydrogenperoxide (King et al., 1989; reviewed in Carpenter, 1999; Zwick et al., 1999a; Prenzel et al., 1999a,b; Leserer et al., 2000). Due to recent advances in the elucidation of the EGFR transactivation mechanism we now begin to understand the biological and pathophysiological signi®cance of these complex signaling networks.

Cross-talk between GPCRs and the EGFR Occurrence of EGFR transactivation in diverse cell types Following the early observation by Faure et al. (1994) that agonist stimulation of COS7 cells transiently expressing Gq-, or Gi-coupled receptors results in MAPK activation, in 1996, Daub et al. (1996) described a critical role of the EGFR in GPCRinduced mitogenesis of rat ®broblasts. They demonstrated that the EGFR and its relative HER2 were rapidly tyrosine phosphorylated after stimulation of Rat1 cells with the GPCR agonists endothelin-1 (ET1), lysophosphatidic acid (LPA) or thrombin. This

EGFR transactivation A Gschwind et al

transactivation of a receptor tyrosine kinase coupled GPCR-ligand engagement to extracellular regulated kinase (ERK)-activation, induction of fos gene expression and DNA synthesis which were abrogated either by the selective EGFR inhibitor tyrphostin AG1478 or by expression of a dominant-negative EGFR mutant. Further investigations revealed that the GPCREGFR cross-talk mechanism was installed in a variety of other cell types such as human keratinocytes, primary mouse astrocytes, PC12 cells and vascular smooth muscle cells (Daub et al., 1997; Zwick et al., 1997; Eguchi et al., 1998) and established it as a widely relevant pathway towards the activation of the MAP kinase signal. Subsequent work provided evidence for widespread use of EGFR signal transactivation by diverse GPCRs and the capacity of di€erent G-proteins to generate the necessary connections (Table 1). Interestingly, LPA-induced transactivation of the EGFR in COS-7 cells was attenuated by pertussis toxin (PTX), whereas thrombin-stimulated EGFR tyrosine phosphorylation and downstream signaling was not a€ected (Daub et al., 1997). Furthermore, agonist stimulation of ectopically expressed Gq-coupled bombesin (BombR) or Gi-coupled M2 muscarinic acetylcholine receptor (M2R) triggered EGFR transactivation followed by tyrosine phosphorylation of SHC and formation of SHC-Grb2 complexes. These results demonstrated that EGFR transactivation occurs via both PTX-insensitive and -sensitive pathways and that EGFR inhibition strongly impairs MAP kinase activation by Gq- and Gi-coupled receptors in COS-7 cells. More recent studies showed that EGFR tyrosine phosphorylation induced by the GPCR agonist substance P is PTX-sensitive in U-373 MG cells (Castagliuolo et al., 2000) and that Ga13 subunits mediate LPA-induced actin polymerization and actin stress ®ber formation in Swiss 3T3 cells and mouse ®broblasts via EGFR transactivation (Gohla et al., 1998, 1999). In summary, Gi-, Gq- as well as G13-

coupled receptors have been reported to transactivate the EGFR after agonist stimulation in diverse cell systems, whereas up to now there is no data available concerning an analogous function of Gs-coupled receptors.

1595

Characteristics of the EGFR transactivation signal Several studies indicate that the EGFR transactivation mechanism is subject to di€erent cell type-characteristic regulatory in¯uences. In neuron-like PC12 cells, vascular smooth muscle cells and intestinal epithelial cells intracellular Ca2+ concentration has been demonstrated to be a critical parameter in Gq-coupled receptor mediated EGFR transactivation upon bradykinin, angiotensin II (AngII), ET-1 or carbachol stimulation (Zwick et al., 1997; Solto€, 1998; Eguchi et al., 1998; Murasawa et al., 1998; Iwasaki et al., 1999). Moreover, in ovarian cancer cells it was shown that through activation of Gq-coupled heptahelical receptors the chemokine interleukin-8 (IL-8) induces EGFR activation in a Ca2+-dependent manner (Venkatakrishnan et al., 2000) and a recent study provides evidence that the pathogenic dermatonecrotic toxin produced by Pasteurella multocida (rPMT) exerts its mitogenic e€ect on HEK 293 cells through a Gqdependent pathway involving EGFR-dependent ERK activation (Seo et al., 2000). Activation of the Ser/Thr protein kinase C (PKC) was shown to be required for Gq-coupled receptors to induce EGFR transactivation in HEK 293 and PC12 cells (Tsai et al., 1997; Solto€, 1998) as well as in response to gonadotropin-releasing hormone in a variety of cell lines such as gonadotropic aT3-1 cells (Grosse et al., 2000). In rat liver epithelial cells, however, Li et al. (1998) proposed an AngII-stimulated EGFR-dependent signaling pathway to Ras only when PKC activity was inhibited. In contrast, others have reported that bradykinin-stimulated ERK activation

Table 1 Cross-talk between G protein-coupled receptors and the epidermal growth factor receptor (adapted from Zwick et al., 1999a,b) GPCR ligand Endothelin-1, LPA, Thrombin Bradykinin Bombesin, Carbachol, LPA Angiotensin II Thrombin, LPA Thrombin Carbachol

G proteins involved ? Gq Gq, Gi Gq ? ? Gq

Carbachol

Gq

LPA LPA LPA

G13 ? ?

Bombesin Substance P Interleukin-8

? Gi Gq

Cell type or tissue

Cellular response

Reference

Rat-1 ERK activation, FOS transcription Daub et al., 1996 PC-12 ERK activation Zwick et al., 1997 COS-7 ERK activation Daub et al., 1997 vascular smooth muscle ERK activation Eguchi et al., 1998 HaCaT ERK activation Daub et al., 1997 primary astrocytes ERK activation Daub et al., 1997 HEK 293 Modulation of Kv1.2 ion Tsai et al., 1997 channel activity T84 ERK activation, inhibition Keely et al., 1998 of Cl7 secretion Swiss 3T3 stress fiber formation Gohla et al., 1998 HeLa ERK activation Cunnick et al., 1998 NIH3T3 MKK1/2 activation, Cunnick et al., 1998 DNA synthesis PC3 EGFR tyrosine phosphorylation Prenzel et al., 1999 U-373 MG ERK activation, DNA synthesis Castagliuolo et al., 2000 SK-OV-3 ERK activation, morphology Venkatakrishnan et al., 2000 changes

Abbreviations: extracellular regulated kinase (ERK), lysophosphatidic acid (LPA), mitogen-activated protein kinase kinase 1 and 2 (MKK1/2) Oncogene

EGFR transactivation A Gschwind et al

1596

requires simultaneous signaling via both PKC and EGFR in COS-7 cells that are transiently expressing the bradykinin B2 receptor (Adomeit et al., 1999). In this study Adomeit et al. (1999) report that inhibition of phosphoinositide 3-kinase g (PI3Kg) failed to diminish bradykinin-induced ERK activation, whereas in the same cell system Daub et al. (1997) demonstrated that the PI3K inhibitors wortmannin and LY294002 signi®cantly reduced ERK activity stimulated through M2R, BombR or in response to LPA or EGF. The latter result was con®rmed in a more recent study where accumulation of PI3K lipid products was directly measured after LPA stimulated EGFR transactivation in COS-7 cells (La€argue et al., 1999). Interestingly, in vascular smooth muscle cells AngIIinduced p70S6K activation is mediated via both ERK and PI3K/Akt cascades that bifurcate at the point of EGFR-dependent Ras activation (Eguchi et al., 1999). In agreement with these results Iwasaki et al. (1999) reported that ET-1-mediated vascular growth requires ERK and p70S6K signaling through transactivation of the EGFR. Besides the function of PKC in GPCR-mediated EGFR transactivation Matsubara and coworkers reported Ca2+/calmodulin-dependent receptor activation in Ang II-stimulated cardiac ®broblasts (Murasawa et al., 1998). Similarly in PC12 cells Zwick et al. (1999a,b) demonstrated the involvement of a Ca2+calmodulin-dependent kinase II (CaMK II) activity in K+-but not bradykinin-induced EGFR signal transactivation (reviewed in Prenzel et al., 2000). The role of another Ca2+-dependent kinase, PYK2, in the transmission of mitogenic signals is somewhat controversial. While Solto€ (1998) and Keely et al. (2000) suggested a role of this tyrosine kinase in Gq-mediated EGFR tyrosine phosphorylation in PC12 (Solto€, 1998) and intestinal epithelial cells (Keely et al., 2000) respectively, Zwick et al. (1999a,b) reported Ca2+-dependent, but PYK2-independent EGFR transactivation in response to bradykinin in PC12 cells. Furthermore, tyrosine phosphorylated Src is often found in association with the EGFR (Luttrell et al., 1999) or with PYK2 (Solto€, 1998; Keely et al., 2000) upon stimulation of Gq-coupled receptors and has therefore been proposed to function as a mediator of EGFR transactivation. Since other reports have demonstrated Src-independent EGFR transactivation, but Src-dependent SHC tyrosine phosphorylation and ERK activation (Daub et al., 1997; Adomeit et al., 1999; Slack, 2000) it seems likely that Src is recruited by the transactivated EGFR and thereby contributes to activation of the Ras signaling pathway. Role of endocytosis in signaling of the GPCR-transactivated EGFR In HEK293 and COS-7 cells stimulation of the Gicoupled b2 adrenergic receptor (b2AR) results in its association with activated Src and b-arrestin (Luttrell et al., 1999). The formation of this complex is thought to be an essential step in Src-dependent activation of

Oncogene

the Ras signaling pathway. Recently, however, it was suggested that mitogenic signaling through the b2AR and a2AAR requires transactivation of the EGFR which for ecient ERK activation needs to be linked to the endocytotic machinery (Maudsley et al., 2000; Pierce et al., 2000). Hence, clathrin-mediated endocytosis has been attributed a critical role in GPCRmediated ERK activation (Figure 1). Interestingly, receptor internalization appears to be di€erentially regulated since upon AR stimulation and EGFR transactivation the b 2 subtype is internalized together with the EGFR while the a2AAR remains on the surface of transfected COS-7 cells (Pierce et al., 2000). Remarkably, concanavalin A (ConA)-treatment of vascular smooth muscle cells blocked AT1R internalization, AngII-induced EGFR transactivation and MAP kinase activation demonstrating the necessity of an intact cytoskeleton for receptor downregulation and signal transmission (Tang et al., 2000). Ligand-independent pathways of GPCR-induced EGFR transactivation While the overall role of Ca2+, PKC and Src as mediators of EGFR transactivation is established, little is known about the detailed mechanism whereby agonist-occupied GPCRs cause tyrosine phosphorylation of the EGFR. In time-course experiments GPCRinduced EGFR tyrosine phosphorylation was detectable after 2 min of stimulation (Daub et al., 1996; Iwasaki et al., 1999). This rapid response and the fact that soluble EGF-like growth factors could not be detected in the culture medium (Daub et al., 1996; Tsai et al., 1997; Eguchi et al., 1998) suggested that EGFR transactivation is exclusively mediated through intracellular mechanisms and did not involve the interaction of the EGFR with a ligand. This notion gained further support by the observation that transient overexpression of Gb1g2 subunits was sucient to trigger tyrosine phosphorylation of SHC suggesting that a physical interaction with Gbg subunits played a role in EGFR transactivation (Luttrell et al., 1997). Metalloprotease-mediated shedding of EGF-like ligands as a key step in GPCR-induced EGFR transactivation Critical reexamination of the `ligand-independent' EGFR transactivation hypothesis resulted in an unexpected explanation of the ®ndings by Daub et al. (1996) and led to the establishment of a `Triple Membrane Passing Signal' (TMPS) mechanism that involves a metalloprotease activity and the processing of transmembrane EGF-like growth factor precursors. Prenzel et al. (1999) ®rst showed that a chimeric RTK consisting of the EGFR ectodomain and the transmembrane and intracellular portion of the PDGFR was transactivated upon treatment of Rat1 ®broblasts with GPCR ligands while endogenous PDGFR was not. Hence, GPCR-induced transactivation of the arti®cial RTK did not involve an intracellular pathway and was dependent on the extracellular ligand-binding

EGFR transactivation A Gschwind et al

domain of the EGFR. In addition, using the diphtheria toxin mutant CRM197 that speci®cally blocks HBEGF function or the metalloprotease inhibitor batimastat (BB94), LPA-, carbachol- or tetra-decanoylphorbol-13-acetate (TPA)-induced transactivation of the EGFR and tyrosine phosphorylation of SHC was completely abrogated in COS-7 and HEK 293 cells. Flow cytometric analysis directly con®rmed cell surface ectodomain shedding of proHB-EGF upon treatment

with GPCR agonists or TPA. Together, these experimental data yield the novel triple membrane-passing signal (TMPS) mechanism of EGFR transactivation (Prenzel et al., 2000), (Figure 2) which provides new insights towards the understanding of interreceptor communication and helps to interpret previously reported physiological and pathophysiological phenomena (reviewed in Carpenter, 2000; Leserer et al., 2000). Therein, GPCR stimulation induces a metallo-

1597

Figure 1 Role of clathrin-mediated endocytosis in signaling of the GPCR-transactivated EGFR: Stimulation of GPCRs as exempli®ed by the b2AR and the a2AAR leads to complex formation of the activated heptahelical receptor with Src and b-arrestin. Subsequent activation of the Ras/MAP-kinase pathway is based on transactivation of the EGFR that needs to be linked to the endocytotic machinery for ecient downstream signaling

Figure 2 Triple-membrane-passing signal mechanism of the EGFR transactivation: GPCR-induced and metalloprotease-mediated proteolytic cleavage of EGF-like growth factor precursors leads to transactivation of the EGFR. This new mechanistic signaling model for ligand-dependent interreceptor communication encloses three membrane passages and couples receptor activation to the Ras/MAP kinase pathway Oncogene

EGFR transactivation A Gschwind et al

1598

Oncogene

protease activity which cleaves the HB-EGF precursor and allows the released growth factor to bind to the extracellular receptor domain thereby transactivating the EGFR signaling potential. This process also allows the transactivation of EGFRs on neighboring cells but only over short distances and under participation of the heparan sulfate proteoglycan matrix which in retrospect explains the failure of Daub et al. (1996) to detect EGF-like activity in conditioned medium of GPCR-ligand-stimulated Rat1 cell cultures. This novel mechanistic concept is reinforced by recently reported experimental data that show thrombin-induced EGFR transactivation and cell migration of rat and baboon smooth muscle cells to be dependent on HB-EGF shedding by a metalloprotease (Kalmes et al., 2000). Still, the identity of the metalloprotease and the way it is activated upon GPCR stimulation remains to be de®ned. A growing body of evidence, however, suggests the involvement of members of the ADAM (a disintegrin and metalloprotease) family of zincdependent proteases in the processing of EGF-like precursors (Werb and Yan, 1998). Knock out of the gene coding for ADAM17, also called tumor necrosis factor alpha-converting enzyme (TACE), was embryonic lethal in mice and was similar to that obtained after homozygous inactivation of the EGFR gene suggesting a role of TACE in cleavage of EGFR ligands in mouse development (Peschon et al., 1998). Further analysis of embryonic ®broblasts derived from TACE knock-out mice also displayed a de®ciency in release of soluble TGFa compared to wild-type cells, strongly suggesting that TACE is responsible for the cleavage of certain EGF-like precursors (Peschon et al., 1998). One of the unsolved problems in the TMPS pathway is the question of metalloprotease activation through a GPCR-mediated signal. In this regard using a reconstitution system with TACE7/7 ®broblasts Black's group has now reported that the cytoplasmic domain of TACE is dispensable at least for phorbolesterstimulated shedding of TNFa and other substrates (Reddy et al., 2000) although this domain has been observed to be phosphorylated after TPA treatment of cells (Black et al., 1997). Similarly, TPA-induced phosphorylation of the cytoplasmic domain of MDC9 (ADAM9) which was implicated in shedding of proHB-EGF in Vero cells (Izumi et al., 1998) involved protein kinase C delta (PKCd) but was not shown to be related to MDC9 activation. Therefore, much remains to be learned about domain requirements of metalloproteases for cell surface ectodomain cleavage especially in response to physiological stimuli such as growth factors in order to understand the complex regulatory mechanisms governing TMPS-mediated EGFR transactivation and cellular responses such as cell proliferation, migration and survival. Studies by Fan and Derynck (1999) provide evidence for the involvement of MAP kinase pathways in the regulation of shedding events upon growth factor stimulation. Treatment of CHO cells with platelet-derived growth factor (PDGF), ®broblast growth factor (FGF) or

EGF induced cleavage of proTGFa through the ERK MAP kinase pathway whereas basal proTGFa cleavage in the absence of growth factors was dependent on p38 MAP kinase signaling. EGFR transactivation in cancer cells Upregulated EGFR signaling has often been correlated with the formation and progression of human cancers and there is growing evidence for the signi®cance of metalloprotease-mediated release of EGF-like precursors in oncogenesis and progression. As a ®rst example the metalloprotease inhibitor BB94 was shown to inhibit bombesin and TPA-induced transactivation of the EGFR in PC3 human prostate cancer cells and to reduce high constitutive levels of EGFR tyrosine phosphorylation in unstarved cells (Prenzel et al., 1999). Furthermore, Dong et al. (1999) reported that BB94 reduced cell proliferation and cell migration of a human mammary epithelial cell line by interfering with the release of EGFR ligands. BB94 also inhibited proliferation of colon and breast cancer cell lines which were known to depend on autocrine signaling through the EGFR. In addition, EGFR function was reported to be critical for GPCR-stimulated mitogenic signaling in several other cancer cell lines including U-373 MG astroglioma cells (Castagliuolo et al., 2000) and SKOV-3 ovarian cancer cells (Venkatakrishnan et al., 2000). Together, these data strongly suggest an important and broad role for GPCR-EGFR cross-talk in the pathophysiology of cancer which exceeds by far that already predicted by the early discovery of the verbB/EGFR oncogene/protooncogene relationship (Ullrich et al., 1984). In addition to being a member of a promiscuous family of RTKs that forms functionally diverse heterodimers in combination with a variety of directly binding ligands, the EGFR and its relatives are now subject to activation through many members of the very large GPCR family and the di€erential use of di€erent G-proteins by these receptors expands the physiological and pathophysiological potential of this network even further. Transactivation of IGF1R and PDGR by GPCRs Cross-talk between GPCRs and other RTKs besides the EGFR has been reported to contribute to GPCRmediated mitogenic pathways in a variety of cell systems. One of the ®rst such demonstrations involved thrombin-induced IGF1R transactivation in primary rat smooth muscle cells (Linseman et al., 1995; Rao et al., 1995). Recently, Roudabush et al. (2000) showed that ERK1/2 activation after IGF1R stimulation is mediated by the EGFR in COS-7 cells. Interestingly, they proposed an IGF1R-EGFR cross-talk pathway based on metalloprotease-induced shedding of proHBEGF. Another study suggested the release of pertussis toxin-sensitive heterotrimeric Gi proteins from agonistoccupied IGF1R (Hallak et al., 2000) which is critical for ERK activation in certain rat neurons. Together,

EGFR transactivation A Gschwind et al

one could envision a pathway connecting IGF1 signaling with the activation of the EGFR by release of a population of Gbg subunits by agonist occupied IGF1R that by an unknown mechanism activate a metalloprotease resulting in the release of EGF-like ligands from the cell surface. In L cells LPA was shown to induce transactivation of the PDGF-b-R and subsequent signaling steps analogous to EGFR transactivation in COS-7 cells (Herrlich et al., 1998). Furthermore, several studies have demonstrated AngII-stimulated transactivation of the PDGF-b-R in vascular smooth muscle cells (Linseman et al., 1995; Heeneman et al., 2000) and reported the involvement of PTX-sensitive G-proteins in PDGFstimulated ERK activation in airway smooth muscle cells (Conway et al., 1999). In addition, there is evidence that in the same cell system activation of the MAP kinase pathway is dependent on both Src and complex formation of Grb2 with PI3K. Very recently, a new ligand for the PDGF-a-R, PDGF-C, has been identi®ed that is activated by proteolysis (Li et al., 2000). When overexpressed PDGF-C causes hyperproliferation of cardiac ®broblasts in transgenic mice. PDGF-C displays a two-domain structure consisting of N-terminal complement proteins C1r/C1s, Uegf, and BMP-1 like (CUB) domain and a C-terminal PDGF/ VEGF core domain connected by a hinge region. Only the core domain functions as a high-anity ligand for the PDGF-a-R, which is released from the CUB domain by an unknown protease presumably at the cell surface. These ®ndings suggest that PDGF-C could be another growth factor that is released in its active

form the cell surface after limited proteolysis similar to the members of the EGF-family of ligands.

1599

Conclusion Progress in understanding the communication mechanism between GPCRs and RTKs has revolutionized the classical view of signal transmission from the cell surface to the nucleus which so far appeared to be composed of largely parallel linear pathways. In the new scenario the EGFR and HER2 as well as the PDGFR and IGF1R can be transactivated by agonist-occupied GPCRs and the number of RTKs that are downstream signaling elements of haptahelical receptors is likely to increase in the future. Although the mechanistic details of EGFR transactivation are still not fully understood it has become obvious that metalloproteases play an important role in signaling through GPCRs in both physiological and pathophysiological processes. Questions that remain to be answered include the identity of additional elements of the TMPS pathway, the tissueand substrate speci®city of participating metalloproteases and the signi®cance of auto- and transregulatory in¯uences by other signaling circuits. Furthermore, the elucidation of the pathophysiological signi®cance of EGFR signal transactivation can be expected to yield novel intervention targets for the treatment of diseases such as cancer and other hyperproliferative disorders.

References Adomeit A, Graness A, Gross S, Seedorf K, Wetzker R and Liebmann C. (1999). Mol. Cell. Biol., 19, 5289 ± 5297. Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA, Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ and Cerretti DP. (1997). Nature, 385, 729 ± 733. Carpenter G. (1999). J. Cell. Biol., 146, 697 ± 702. Carpenter G. (2000). Science's stke (www.stke.org) January 18, 2000. Castagliuolo I, Valenick L, Liu J and Pothoulakis C. (2000). J. Biol. Chem., 275, 26545 ± 26550. Conway AM, Rahkit S, Pyne S and Pyne NJ. (1999). Biochem. J., 337, 171 ± 177. Cunnick J, Dorsey J, Standley T, Turkson J, Kraker A, Fry D, Jove R and Wu J. (1998). J. Biol. Chem., 273, 14468 ± 14475. Daub H, Wallasch C, Lankenau A, Herrlich A and Ullrich A. (1997). EMBO J., 16, 7032 ± 7044. Daub H, Weiss FU, Wallasch C and Ullrich A. (1996). Nature, 379, 557 ± 560. Dong J, Opresko LK, Dempsey PJ, Lau€enburger DA, Co€ey RJ and Wiley HS. (1999). Proc. Natl. Acad. Sci. USA, 96, 6235 ± 6240. Eguchi S, Iwasaki H, Ueno H, Frank GD, Motley ED, Eguchi K, Marumo F, Hirata Y and Inagami T. (1999). J. Biol. Chem., 274, 36843 ± 36851.

Eguchi S, Numaguchi K, Iwasaki H, Matsumoto T, Yamakawa T, Utsunomiya H, Motley ED, Kawakatsu H, Owada KM, Hirata Y, Marumo F and Inagami TJ. (1998). J. Biol. Chem., 273, 8890 ± 8896. Fan H and Derynck R. (1999). EMBO J., 18, 6962 ± 6972. Faure M, Voyno-Yasenetskaya TA and Bourne HR. (1994). J. Biol. Chem., 269, 7851 ± 7854. Gohla A, Harhammer R and Schultz G. (1998). J. Biol. Chem., 273, 4653 ± 4659. Gohla A, O€ermanns S, Wilkie TM and Schultz G. (1999). J. Biol. Chem., 274, 17901 ± 17907. Grosse R, Roelle S, Herrlich A, Hohn J and Gudermann T. (2000). J. Biol. Chem., 275, 12251 ± 12260. Hackel PO, Zwick E, Prenzel N and Ullrich A. (1999). Curr. Opin. Cell. Biol., 11, 184 ± 189. Hallak H, Seiler AE, Green JS, Ross BN and Rubin R. (2000). J. Biol. Chem., 275, 2255 ± 2258. Heeneman S, Haendeler J, Saito Y, Ishida M and Berk BC. (2000). J. Biol. Chem., 275, 15926 ± 15932. Herrlich A, Daub H, Knebel A, Herrlich P, Ullrich A, Schultz G and Gudermann T. (1998). Proc. Natl. Acad. Sci. USA, 95, 8985 ± 8990. Iwasaki H, Eguchi S, Ueno H, Marumo F and Hirata Y. (1999). Endocrinology, 140, 4659 ± 4668. Izumi Y, Hirata M, Hasuwa H, Iwamoto R, Umata T, Miyado K, Tamai Y, Kurisaki T, Sehara-Fujisawa A, Ohno S and Mekada E. (1998). EMBO J., 17, 7260 ± 7272. Oncogene

EGFR transactivation A Gschwind et al

1600

Oncogene

Kalmes A, Vesti BR, Daum G, Abraham JA and Clowes AW. (2000). Circ. Res., 87, 92 ± 98. Keely S, Uribe J and Barrett K. (1998). J. Biol. Chem., 273, 27111 ± 27117. Keely SJ, Calandrella SO and Barrett KE. (2000). J. Biol. Chem., 275, 12619 ± 12625. King CR, Borello I, Porter L, Comoglio P and Schlessinger J. (1989). Oncogene, 4, 13 ± 18. La€argue M, Raynal P, Yart A, Peres C, Wetzker R, Roche S, Payrastre B and Chap H. (1999). J. Biol. Chem., 274, 32835 ± 32841. Leserer M, Gschwind A and Ullrich A. (2000). IUBMB Life, 49, 405 ± 409. Li X, Ponten A, Aase K, Karlsson L, Abramsson A, Uutela M, Backstrom G, Hellstrom M, Bostrom H, Li H, Soriano P, Betsholtz C, Heldin CH, Alitalo K, Ostman A and Eriksson U. (2000). Nat. Cell. Biol., 2, 302 ± 309. Li X, Lee JW, Graves LM and Earp HS. (1998). EMBO J., 17, 2574 ± 2583. Linseman DA, Benjamin CW and Jones DA. (1995). J. Biol. Chem., 270, 12563 ± 12568. Luttrell LM, Della Rocca GJ, van Biesen T, Luttrell DK and Lefkowitz RJ. (1997). J. Biol. Chem., 272, 4637 ± 4644. Luttrell LM, Ferguson SS, Daaka Y, Miller WE, Maudsley S, Della Rocca GJ, Lin F, Kawakatsu H, Owada K, Luttrell DK, Caron MG and Lefkowitz R. (1999). Science, 283, 655 ± 661. Maudsley S, Pierce KL, Zamah AM, Miller WE, Ahn S, Daaka Y, Lefkowitz RJ and Luttrell LM. (2000). J. Biol. Chem., 275, 9572 ± 9580. Massague J and Pandiella A. (1993). Annu. Rev. Biochem., 62, 515 ± 554. Moghal N and Sternberg PW. (1999). Curr. Opin. Cell. Biol., 11, 190 ± 196. Murasawa S, Mori Y, Nozawa Y, Gotoh N, Shibuya M, Masaki H, Maruyama K, Tsutsumi Y, Moriguchi Y, Shibazaki Y, Tanaka Y, Iwasaka T, Inada M and Matsubara H. (1998). Circ. Res., 82, 1338 ± 1348. Olayioye MA, Neve RM, Lane HA and Hynes NE. (2000). EMBO J., 19, 3159 ± 3167. Peschon JJ, Slack JL, Reddy P, Stocking KL, Sunnarborg SW, Lee DC, Russell WE, Castner BJ, Johnson RS, Fitzner JN, Boyce RW, Nelson N, Kozlosky CJ, Wolfson MF, Rauch CT, Cerretti DP, Paxton RJ, March CJ and Black RA. (1998). Science, 282, 1281 ± 1284.

Pierce KL, Maudsley S, Daaka Y, Luttrell LM and Lefkowitz RJ. (2000). Proc. Natl. Acad. Sci. USA, 97, 1489 ± 1494. Prenzel N, Zwick E, Leserer M and Ullrich A. (2000). Breast Cancer Res., 2, 184 ± 190. Prenzel N, Zwick E, Daub H, Leserer M, Abraham R, Wallasch C and Ullrich A. (1999). Nature, 402, 884 ± 888. Rao GN, Delafontaine P and Runge MS. (1995). J. Biol. Chem., 270, 2781 ± 2785. Reddy P, Slack JL, Davis R, Cerretti DP, Kozlosky CJ, Blanton RA, Shows D, Peschon JJ and Black RA. (2000). J. Biol. Chem., 275, 14608 ± 14614. Roudabush FL, Pierce KL, Maudsley S, Khan KD and Luttrell LM. (2000). J. Biol. Chem., 275, 22583 ± 22589. Seo B, Choy EW, Maudsley S, Miller WE, Wilson BA and Luttrell LM. (2000). J. Biol. Chem., 275, 2239 ± 2245. Slack BE. (2000). Biochem. J., 348, 381 ± 387. Solto€ SP. (1998). J. Biol. Chem., 273, 23110 ± 23117. Tang H, Nishishita T, Fitzgerald T, Landon EJ and Inagami T. (2000). J. Biol. Chem., 275, 13420 ± 13426. Tsai W, Morielli AD and Peralta EG. (1997). EMBO J., 16, 4597 ± 4605. Ullrich A, Coussens L, Hay¯ick JS, Dull TJ, Gray A, Tam AW, Lee J, Yarden Y, Libermann TA, Schlessinger J, Downward J, Mayes ELV, Whittle N, Water®eld MD and Seeburg PH. (1984). Nature, 309, 418 ± 425. Venkatakrishnan G, Salgia R and Groopman JE. (2000). J. Biol. Chem., 275, 6868 ± 6875. Werb Z and Yan Y. (1998). Science, 282, 1279 ± 1280. Zwick E, Daub H, Aoki N, Yamaguchi-Aoki Y, Tinhofer I, Maly K and Ullrich A. (1997). J. Biol. Chem., 272, 24767 ± 24770. Zwick E, Hackel PO, Prenzel N and Ullrich A. (1999a). Trends Pharmac. Sci., 20, 408 ± 412. Zwick E, Wallasch C, Daub H and Ullrich A. (1999b). J. Biol. Chem., 274, 20989 ± 20996.