Oncogene (2001) 20, 1661 ± 1668 ã 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc RGS-containing RhoGEFs: the missing link between transforming G proteins and Rho?

Shigetomo Fukuhara1,2, Hiroki Chikumi1 and J Silvio Gutkind*,1

1Oral and Pharyngeal Cancer Branch, National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, Maryland MD 20892-4340, USA

Oncogene (2001) 20, 1661 ± 1668. aq, and a12 (Wilkie et al., 1992). The aq family of heterotrimeric G protein a subunits is coupled to Keywords: Rho; G proteins; Ras; signal transduction; phospholipase-C b to induce the hydrolysis of G protein-coupled receptors phosphatidylinositol bis phosphate, the consequent rise in the intracellular concentration of Ca2+, and the activation of protein kinase C, whereas the as and ai Introduction families are known to activate and inhibit adenylyl cyclase to increase and decrease the level of cyclic Many mitogens promote cell proliferation by acting AMP, respectively. ai proteins also regulate ion on cell surface receptors that either possess an channels, certain phospholipases and phosphodies- intrinsic protein-tyrosine kinase activity (Yarden et terases (Hamm, 1998). The targets for both members al., 1986) or interact with heterotrimeric GTP- of the a12 family, a12 and a13, were unknown until binding proteins (G proteins). The latter are recently. In addition, eleven-G protein g subunits and collectively known as G protein-coupled receptors ®ve G protein b subunits have been cloned so far and (GPCRs) and comprise the largest group of cell they regulate certain ion channels, phospholipase-C, surface molecules involved in signal transmission. and phosphatidylinositol-3 kinases (Clapham and With more than 1000 members, they are encoded by Neer, 1997). the biggest gene family in the human genome Through these G proteins subunits, GPCRs (Flower, 1999), and the importance and diversity regulate a diverse array of physiological functions of their physiological roles are directly supported by such as neurotransmission, chemoreception, photo- their link to a number of hereditary and acquired reception, metabolism, growth, and di€erentiation. In diseases (Spiegel, 1996; Stadel et al., 1997). Further- addition to these physiological responses, there is more, these receptors and their ligands are the target also emerging evidence of the involvement of of over 50% of all current therapeutic agents aberrant GPCR function in cellular transformation (Flower, 1999). and oncogenesis (see Gutkind, 1998 for a recent GPCRs exhibit a common structural motif consisting review). In this regard, recent work suggests that the of seven membrane-spanning regions (Dohlman et al., Rho family of small GTP-binding proteins plays a 1987), and can be activated by a diverse array of central role in the regulation of cell proliferation by external stimuli, including growth factors, vasoactive GPCRs, and that the persistent activation of Rho- polypeptides, chemoattractants, neurotransmitters, hor- mediated pathways by heterotrimeric G proteins can mones, phospholipids, photons, odorants, and taste lead to cellular transformation. In this review, we ligands. In general, agonist binding provokes rapid will focus on recent e€orts addressing the molecular conformational changes in the transmembrane a mechanisms by which GPCRs and G protein a helices, which result in the exposure of previously subunits activate Rho, with emphasis on their masked G protein binding sites in the intracellular contribution to the understanding on how this signal loops (Altenbach et al., 1996; Bourne, 1997; Wess, transducing system regulates normal and aberrant 1997). This causes the exchange of GDP for GTP cell growth. bound to the G protein a subunit, and GTP-bound a subunits or bg complexes then initiate intracellular signaling responses by acting on a variety of e€ector Transforming G protein coupled receptors molecules. To date, sixteen distinct mammalian G protein a subunits have been identi®ed and divided into Most oncogenes were discovered by virtue of their four families based on their sequence similarity: as, ai, ability to transform cells in culture or to induce tumorigenesis in experimental animals (Weinberg, 1985). They include activated tyrosine kinase growth factor receptors, their ligands, and aberrant versions of *Correspondence: JS Gutkind, molecules thought to participate in their growth- 2Current address: Division of Integrative Cell Biology, Department of Embryogenesis, Institute of Molecular Embryology and Genetics, promoting pathways (Aaronson, 1991). On the other Kumamoto University, Kumamoto 860-0811, Japan hand, the discovery of the mas oncogene provided the RGS-containing RhoGEFs S Fukuhara et al 1662 ®rst evidence that GPCRs can harbor transforming but behaves as a constitutively active Gq-coupled potential (Young et al., 1986). The natural agonist for receptor and, as predicted, is sucient to subvert the mas oncogene product is still unknown, but is likely normal growth control when expressed in cultured to be a serum component. Similarly, serotonin 1C ®broblasts (Arvanitakis et al., 1997). The KSHV ± receptors were shown to be oncogenic if ectopically GPCR can e€ectively activate signaling pathways expressed in NIH3T3 cells and activated by the typical of transforming receptors, and may represent serotonin present in the culture medium (Julius et al., a key molecule in the pathogenesis of Kaposi's 1989). Using a more de®ned experimental system, Sarcoma because of its central role in the promotion certain muscarinic receptor subtype were shown to of VEGF-driven angiogenesis and spindle cell prolif- transform NIH3T3 cells in a strictly agonist-dependent eration (Bais et al., 1998; Sodhi et al., 2000). As manner (Gutkind et al., 1991). Since neurotransmitter described in the accompanying review by Heasley acetylcholine, the natural agonist for these receptors, is (2001), there is also compelling evidence for a role of not present in serum, it was possible to correlate GPCRs in many human neoplasias, including small cell biochemical responses with the biological e€ects lung carcinoma, colon adenomas and carcinomas, elicited by an exogenously added agonist. This gastric hyperplasia and cancer, prostate cancer, and approach established that while Gq-coupled receptors pancreatic carcinoma, which express wild-type GPCRs (m1, m3, and m5), were highly transforming, Gi- that are persistently stimulated by tumor-released coupled receptors (m2 and m4), were without focus- agonists in an autocrine or paracrine fashion. Indeed, forming activity. Together, these ®ndings demonstrated the use of blocking antibodies and receptor antagonists that certain GPCRs could behave as potent agonist- has been shown to be e€ective in a variety of in vitro dependent oncogenes, and raised the possibility that and in vivo tumor models, which lends support to the activating mutations in GPCRs may render them potential causative role of GPCRs in these diseases transforming. Indeed, this was subsequently shown (Langdon et al., 1992; Sethi et al., 1992; Thomas et al., for the a1b G protein-linked receptor to noradrenaline 1992). It also provides the rationale for the develop- (Allen et al., 1991). In this case, speci®c point ment of novel treatment modalities in cancer therapy. mutations in the C-terminal juxtamembrane region of the third intracellular loop of the a1b receptor relieved the requirement of ligand activation for its transform- Transforming G proteins ing ability. Several GPCRs have now been cloned using The potent transforming potential of certain GPCRs standard transfection and tumorigenesis assays. and their likely contribution to human neoplasias Furthermore, the discovery of activated mutants of prompted many laboratories to explore the biological GPCRs in a variety of neoplastic diseases and consequences of expressing constitutively activate oncogenic viruses provided direct evidence of a role mutants of heterotrimeric G protein a subunits. These of these cell surface receptors and their coupled G experiments were possible due to the observation that proteins in cancerous growth. For example, activating all G protein a subunits exhibit a highly conserved mutations in TSH receptors were found in 30% of region, designated G-3 (Bourne et al., 1991), which is thyroid adenomas (Parma et al., 1993), and mutation- involved in binding and hydrolysis of GTP. This region ally activated LH receptors have been identi®ed in a includes a consensus sequence DVGGQ. Replacement form of familial male precocious puberty that results of the glutamine residue (Q) for leucine (L) in the from hyperplastic growth of Leydig cells (Shenker et corresponding sequence of the ras gene (codon 61) al., 1993). Reports are also available indicating that increases the steady-state levels of GTP-bound p21ras active mutants of FSH receptors are expressed in and unmasks its oncogenic potential (Barbacid, 1987). ovarian sex cord tumors and ovarian small cell Accordingly, constitutively activated Ga subunits (QL carcinoma (Kotlar et al., 1997), CCK-B receptors in mutants) were generated and tested for their ability to colorectal cancer (Hellmich et al., 2000), and Ca2+- transform rodent ®broblasts. In these initial studies, sensing receptors in a form of autosomal-dominant activated mutants of members of the Gas,Gai and Gaq hypocalcemia that is concomitant with a variety of family were found to provoke alterations in cell growth neoplasms (Ho€ et al., 1999). As reviewed by (reviewed in Dhanasekaran et al., 1995). However, Rosenkilde et al. (2001), functional GPCRs have been whereas Gas is not transforming, the focus-forming also found in the genome of a number of DNA viruses. ability of Gai and Gaq was found to be several orders Interestingly, some of them have been implicated in of magnitude levels lower than that of activated neoplasia, including the herpes virus saimiri (HVS) GPCRs, which suggested that transforming GPCRs GPCR (also known as ECRF3) (Nicholas et al., 1992), might activate pathways in addition to those regulated and the Kaposi's sarcoma associated herpes virus by these heterotrimeric G proteins. (KSHV) GPCR (also known as KSHV-ORF74) During the course of these studies, a new family of (Arvanitakis et al., 1997). The HVS ± GPCR behaves G proteins a subunits was discovered and termed a12, as a functional chemokine receptor, with ligand and which includes a12 and its highly homologous G Gi-coupling speci®city identical to that of the IL8 protein, a13 (Strathmann and Simon, 1991). These G receptor (Nicholas et al., 1992). The KSHV ± GPCR is proteins appear to be ubiquitously expressed, and they also highly related to the chemokine receptor family, exhibit 67% of aminoacid identity with each other, but

Oncogene RGS-containing RhoGEFs S Fukuhara et al 1663 only 35 ± 44% of aminoacid identity with a subunits of the expression of activated Rho (Fromm et al., 1997). other classes, such as Gi and Gq (Strathmann and In line with these observations, the focus forming Simon, 1991). Interestingly, the ®nding that a12 is ability of G12 was abolished upon treatment with C3, closely related to concertina (cta), a Drosophila gene and restored by the co-expression of a C3-toxin involved in embryogenesis, provided an early indica- insensitive form of RhoA (Fromm et al., 1997). In tion that this G protein class might be involved in addition, recent work indicates that the expression of growth regulation (Parks and Wieschaus, 1991). Loss activated G12 and G13 is sucient to induce an increase of function mutations in the cta gene were identi®ed in the cellular levels of the GTP-bound form of Rho in because of their early embryonic lethality provoked by vivo (Gohla et al., 1998; Kranenburg et al., 1999). the disruption of ventral furrow formation in Droso- Together, these ®ndings indicate that G12 and G13 can phila embryos. This information, and the limited stimulate Rho-dependent pathway, and that this small transforming capacity of other activated G protein a GTPase plays a central role in the transmission of subunits, prompted several laboratories to examine proliferative and transforming signals by the G12 whether this class of G proteins harbors oncogenic family of heterotrimeric G proteins. potential. Indeed, activated Ga12 and Ga13 were found to behave as remarkable potent transforming genes, nearly as potent as other widely studied oncogenes Contribution of the Gq and G12 family of heterotrimeric such as v-raf or the human ras oncogene (Gutkind et G proteins to the activation of Rho by distinct G al., 1998; Jiang et al., 1993; Voyno-Yasenetskaya et al., protein-linked receptors 1994; Xu et al., 1993, 1994). Of interest, the human Ga12 gene was independently cloned by standard Of interest, the ability of Gq-coupled receptors to expression cloning approaches from an Ewing Sarcoma induce cell proliferation and transformation also cDNA library, although whether Ga12 contributes to involves the activation of Rho GTPases (Fromm et this or other neoplastic diseases is still unknown (Chan al., 1997). Thus, both G12 and Gq families of et al., 1993). heterotrimeric G proteins can signal to Rho, and their relative contribution to Rho activation by a variety of GPCRs has recently begun to be thoroughly investi- Biochemical pathways regulated by the G12-class gated. This research was made possible by the of G proteins development of embryonic ®broblast cell lines derived from knocked out animals lacking each class of G The potent biological activity of Ga12 and Ga13 and protein a subunits (see review by O€ermanns, 1999). their limited primary sequence similarity to other G For example, m1-muscarinic and a1-adrenergic recep- proteins led many laboratories to investigate the nature tors failed to induce SRF-dependent gene expression in of the biologically relevant downstream e€ector cells lacking Gq and G11, and this response was molecules for this intriguing G protein family. Initial restored by the expression of wild type Gq (Mao et experiments using blocking antibodies suggested that al., 1998b). Similarly, the stimulation of stress ®ber 2+ G13 regulates the inhibition of voltage-dependent Ca formation by m1-muscarinic and metabotropic gluta- channels in response to bradykinin (Wilk-Blaszczak et mate-1a receptors was not observed in Gq-andG11- al., 1994), and utilizing activated mutants of a12 and de®cient ®broblasts, whereas LPA, bradykinin B2 and a13 evidence was provided to support a role for these serotonin 2C receptors failed to induce stress ®ber novel G proteins regulating a number of signaling formation in G13-de®cient ®broblasts (Gohla et al., + + pathways, including phospholipase A2,Na/H ex- 1999). Using dominant-negative G proteins as a changer, c-Jun N-terminal kinase, and cytoskeletal complementary approach, it has recently been reported components (reviewed in Gutkind et al., 1998). Of that inhibitory G12 constructs can prevent the induc- interest, the latter provided a clue for the role of Rho tion of stress ®bers by thrombin, vasopressin 1 A and acting downstream from G12. Rho small GTPases are endothelin A receptors, whereas dominant negative G13 known to regulate the actin-based cytoskeleton (Ridley proteins prevent such an induction when elicited by and Hall, 1992), and activated G12 and G13 were found LPA, bradykinin B2 and serotonin 2C receptors to mimic the e€ect of activated forms of Rho on stress (Gohla et al., 1999). Taken together, we can conclude ®ber formation and focal adhesion assembly. These that many GPCRs, including those displaying trans- responses were inhibited by C3 exoenzyme, which forming potential, can stimulate Rho-dependent path- ADP-ribosylates and inactivates Rho (Buhl et al., ways by acting either on Gq or on speci®c members of 1995). Furthermore, activated mutants of Ga12 and the G12 family of heterotrimeric G proteins. Ga13 were able to elicit transcriptional responses characteristics of Rho, such as expression from the c- fos Serum Responsive Element (SRE), and these Regulation of Rho small GTPase nuclear events were prevented by the expression of C3 toxin (Fromm et al., 1997). Moreover, Rho In spite of the functional link between GPCRs and Rho, GTPases themselves display oncogenic activity, and the molecular mechanism by which heterotrimeric G the phenotypic appearance of G12- and G13-induced proteins stimulate Rho has just begun to be elucidated. foci in NIH3T3 cells resembles that which is caused by In general, the functional activity of Rho protein is

Oncogene RGS-containing RhoGEFs S Fukuhara et al 1664 tightly regulated by regulatory proteins such as guanine Ga13 through its RGS domain in vitro. Interestingly, nucleotide dissociation inhibitors (RhoGDIs), GTPase- the association of Ga13 with p115 RhoGEF stimulated activating proteins (GAPs) and guanine nucleotide its guanine-nucleotide exchange activity towards Rho exchange factors (GEFs) (Narumiya, 1996; Takai et (Hart et al., 1998), and the RGS domain of p115 al., 1995; Van Aelst and D'Souza-Schorey, 1997). The RhoGEF displayed GAP activity in vitro on both Ga12 GDP-bound, inactive form of Rho is located in the and Ga13 (Kozasa et al., 1998). Furthermore, the two cytosol in a complex with RhoGDI, which stabilizes the closest mammalian homologs of DRhoGEF2, PDZ- GDP-Rho complex, thus inhibiting the exchange of RhoGEF and leukemia-associated RhoGEF (LARG), GDP for GTP (Ueda et al., 1990). GAPs negatively which were originally cloned as KIAA0380 and regulate the activity of Rho proteins by accelerating KIAA0382 cDNAs by a random cloning strategy at their intrinsic GTPase activity, whereas GEFs mediate the Kazusa DNA Institute, respectively, were shown to the activation of Rho proteins by catalyzing the interact physically with Ga12 and Ga13 through their exchange of GDP for GTP on Rho (Boguski and RGS domains in vivo, and a number of experimental McCormick, 1993). Of interest, most GEFs for Rho approaches provided evidence that these Rho-GEFs family G proteins, including dbl, ost, lfc, lbc, vav, ect2, act downstream from G12 and G13 in a novel signaling tim, and net (reviewed in Van Aelst and D'Souza- route leading to Rho activation (Fukuhara et al., 1999, Schorey, 1997; Whitehead et al., 1997) were ®rst 2000). identi®ed because of their ability to transform murine Interestingly, SRE activation in response to throm- ®broblasts in standard focus-formation assays. These bin and LPA was attenuated by the expression of proteins share a common structural motif consisting of dominant negative molecules for RGS-RhoGEFs, a 250-amino acid stretch of sequence similarity with suggesting that these RhoGEFs may mediate some of Dbl, known as DH domain, adjacent to a pleckstrin- the biological responses elicited by these and other G12- homology (PH) domain (Van Aelst and D'Souza- coupled receptors (Fukuhara et al., 1999; Mao et al., Schorey, 1997; Whitehead et al., 1997). The DH domain 1998b). In this regard, p115-RhoGEF, PDZ-RhoGEF is responsible for nucleotide exchange activity towards and LARG are themselves focus-forming when over- GTPases of the Rho family (Hart et al., 1991, 1994). expressed in NIH3T3 cells (Figure 2), and a recent One of these GEFs, the protein product of the vav report indicates that the transformation of these cells proto-oncogene proto-Vav, also exhibits a SH2 domain induced by G2A, an oncogenic GPCR, can be ¯anked by two SH3 domains (Bustelo and Barbacid, suppressed by co-expression of the RGS domain of 1992; Margolis et al., 1992), and tyrosine phosphoryla- Lsc, the murine homolog of p115 RhoGEF (Zohn et tion of proto-Vav by hematopoietic speci®c tyrosine al., 2000). Thus, given the importance of Rho in cell kinases can activate its GEF activity for Rac both in growth regulation, it is possible to speculate that the vitro and in vivo (Crespo et al., 1997; Teramoto et al., activation of RGS-containing RhoGEFs may represent 1997). No other GEF for Rho-like proteins has been a key component of the mitogenic and transforming shown to be regulated by tyrosine phosphorylation, or pathway utilized by many GPCRs. to contain a phosphotyrosine-binding domain such as a SH2 or PTB domains. Thus, a mechanism for their activation was still largely unknown. Activation of RGS-RhoGEFs by Ga12 and Ga13

It is still unclear how G12 and G13 regulate the activity Discovery of RGS domain-containing RhoGEFs of RGS domain-containing RhoGEFs. To address this question, the ability of wild-type and deletion mutants The molecular dissection of the early events involved in of PDZ ± RhoGEF to induce Rho-dependent signaling embryo development in fruit ¯ies provided the ®rst clue has been investigated. Over-expression of PDZ ± of how heterotrimeric G proteins activate Rho. In an RhoGEF in NIH3T3 cells can induce Rho-dependent interesting turn, genetic analysis indicated that a gene expression through SRF, however this response is Drosophila RhoGEF, DRhoGEF2, acts downstream signi®cantly enhanced by the deletion of its N-terminal of the concertina gene (Barrett et al., 1997; Hacker and region or by that of its RGS domain, whereas the Perrimon, 1998), the Drosophila Ga12 homolog in- deletion of PDZ domain does not a€ect its activity volved in gastrulation (Parks and Wieschaus, 1991), (Fukuhara et al., 1999). These ®ndings suggested that thus suggesting a functional relationship between Ga12 the RGS domain of PDZ-RhoGEF might act as a proteins and DRhoGEF2. Two independent lines of negative regulator for its guanine nucleotide exchange research revealed that Ga12 proteins act directly on activity on Rho, and that the binding of G12 and G13 three mammalian DRhoGEF2 homologs, p115Rho- to the RGS domain may relieve its inhibitory e€ect. GEF, PDZ-RhoGEF and LARG (Fukuhara et al., However, the sole association of RGS domain contain- 1999, 2000; Kozasa et al., 1998) (Figure 1). For ing RhoGEFs with Ga12 and Ga13 may not be example, Kozasa et al. (1998) found that the N- sucient for the activation of Rho signaling in vivo, terminal region of p115 RhoGEF, a Rho speci®c GEF, because a Ga13 mutant that lacks a palmitoylation is similar to a conserved region of regulator of G signal cannot induce Rho-dependent signaling path- protein signaling (RGS), and shown that p115 ways, although this Ga mutant is still able to associate RhoGEF can associate physically with Ga12 and with p115 RhoGEF and PDZ ± RhoGEF (Bhattachar-

Oncogene RGS-containing RhoGEFs S Fukuhara et al 1665

Figure 1 The novel family of RGS-domain containing Rho-guanine nucleotide exchange factors (RhoGEFs). Structural features are depicted (see text for details)

Figure 2 Focus formation assay in NIH3T3 cells transfected with RGS-domain containing Rho-GEFs. NIH3T3 cells were transfected with 1 mg of empty pCEFL expression vector (control), pCEFL PDZ-RhoGEF, pCEFL LARG, or pCEFL p115- RhoGEF plasmid DNA, as indicated. Cultures were maintained in Dulbecco's modi®ed Eagle's medium containing 5% of calf serum. Plates were stained 3 weeks after transfection yya and Wedegaertner, 2000). As palmitoylation of mechanisms: tyrosine kinases and/or EGF receptors Ga13 promotes its plasma membrane localization, Ga13 may regulate signaling from G13 to Rho, whereas G12- association with RGS domain-containing RhoGEFs induced Rho activation may not require tyrosine may not only stimulate their exchange activity on Rho, kinases. but may also mediate their proper localization, which The involvement of Btk family of tyrosine kinases, in turn may be essential for Rho activation. Tec and Bmx, in the activation of Rho by G12/13 has Although the recent discovery of RGS domain- been also suggested (Jiang et al., 1998; Mao et al., containing RhoGEFs indicates the existence of a direct 1998a). For example, co-transfection of Tec and Bmx mechanism whereby G proteins of the G12 family synergistically enhance the induction of SRF-depen- activate Rho, several lines of evidence suggest that dent gene expression by G13 (Mao et al., 1998a), and tyrosine kinases may either regulate this signaling Tec was shown to be activated by the constitutive route, or that alternative pathways mediating Rho active mutant of G13 (Mao et al., 1998a). Further- activation by Ga12/13 through tyrosine kinases may also more, a dominant negative mutant of Tec partially exist. For example, a tyrosine kinase inhibitor, inhibits SRF-dependent gene expression by G12 and tyrphostin A25, has been shown to inhibit G13-induced G13 in NIH3T3 cells and by thrombin in Gq/11 neurite retraction and cell rounding, which are Rho- de®cient ®broblasts. On the other hand, a recent dependent functions in PC12 cells whereas this study reported that the active form of G13 enhances inhibitor failed to inhibit these responses when induced the enzymatic activity of Pyk2, and that this non- by G12 (Katoh et al., 1998). Similarly, stress ®ber receptor tyrosine kinase may be also involved in the formation and focal adhesion assembly induced by Rho-dependent activation of SRE induced by G12/13 LPA stimulation and the activated form of G13, but (Shi et al., 2000). However, FAK, a related tyrosine not by that of G12, are inhibited by tyrphostin A25 and kinase of Pyk2, is also activated by the stimulation of AG1478, which are inhibitors of epidermal growth GPCRs and G12 family of heterotrimeric G proteins, factor (EGF) tyrosine kinase receptors (Gohla et al., but this e€ect is inhibited by C3 toxin, suggesting that 1998, 1999). In addition, Kranenburg et al. (1999) FAK, and probably, Pyk2 act downstream from Rho reported that Rho activation by LPA is blocked by the (Needham and Rozengurt, 1998). In addition, several treatment with tyrosine kinase inhibitors such as lines of evidence support that FAK may act as a genistein and tyrphotin A47. These ®ndings suggest negative rather than a positive regulator of Rho that G12 and G13 communicate to Rho by distinct pathways. For example, FAK and Pyk2 de®cient

Oncogene RGS-containing RhoGEFs S Fukuhara et al 1666

Figure 3 Proposed mechanism whereby G protein-coupled receptors stimulate Rho-dependent signaling pathways (see text for details)

®broblasts display an increased number of stress ®bers underlying mechanism whereby G12-andG13-linked and focal adhesion while they are impaired in cell cell surface receptors can stimulate Rho-dependent migration (Ilic et al., 1995, 1996), which may result pathways. On the other hand, whether p115-RhoGEF, from the constitutive activation of Rho and the PDZ-RhoGEF and LARG participate in distinct or inhibition of focal adhesion turnover (Ren et al., overlapping biological functions is still unknown, but 2000). Thus, we can conclude that both receptor and powerful approaches a€orded by the use of mouse non-receptor tyrosine kinases may a€ect the extent of genetics are expected to soon reveal the unique features stimulation of Rho-dependent pathways in response to characterizing each of these RhoGEFs. Similarly, these Ga12/13. However, whether these kinases regulate the RhoGEFs exhibit a number of distinctive structural activation of Rho by these heterotrimeric G proteins motifs, which may play a role in the ®ne-tuning of their or participate in signaling events downstream from function, or may allow their activation by additional Rho is still unclear, and will surely be the focus of cell surface receptors or intracellular molecules. Avail- future investigation. able evidence also suggests that Gaq and its coupled receptors may activate Rho through a distinct, yet to be uncovered biochemical route (Fukuhara et al., 2000) Conclusion that is under intense investigation. We can expect that current e€orts in a number of laboratories will soon The emerging picture of recent studies is that Ga12 and help elucidate fully how GPCRs activate Rho- Ga13 can associate physically with Rho-exchange GTPases, thus providing a unique opportunity to factors containing RGS domains such as p115- begin unraveling the complexity of the signaling RhoGEF, PDZ-RhoGEF and LARG, which in turn networks by which the large family of GPCRs regulate results in the activation of their guanine nucleotide cell proliferation under normal physiological conditions exchange activity towards Rho (Figure 3). Together, and in cancerous growth. these exciting ®ndings have helped elucidate the

References

Aaronson SA. (1991). Science, 254, 1146 ± 1153. AltenbachC,YangK,FarrensDL,FarahbakhshZT, Allen LF, Lefkowitz RJ, Caron MG and Cotecchia S. (1991). Khorana HG and Hubbell WL. (1996). Biochemistry, 35, Proc. Natl. Acad. Sci. USA, 88, 11354 ± 11358. 12470 ± 12478.

Oncogene RGS-containing RhoGEFs S Fukuhara et al 1667 Arvanitakis L, Geras-Raaka E, Varma A, Gershengorn MC Jiang H, Wu D and Simon MI. (1993). FEBS Lett., 330, and Cesarman E. (1997). Nature, 385, 347 ± 350. 319 ± 322. Bais C, Santomasso B, Coso O, Arvanitakis L, Raaka EG, JiangY,MaW,WanY,KozasaT,HattoriSandHuangXY. Gutkind JS, Asch AS, Cesarman E, Gershengorn MC and (1998). Nature, 395, 808 ± 813. Mesri EA. (1998). Nature, 391, 86 ± 89. Julius D, Livelli TJ, Jessell TM and Axel R. (1989). Science, Barbacid M. (1987). Annu. Rev. Biochem., 56, 779 ± 827. 244, 1057 ± 1062. Barrett K, Leptin M and Settleman J. (1997). Cell, 91, 905 ± Katoh H, Aoki J, Yamaguchi Y, Kitano Y, Ichikawa A and 915. Negishi M. (1998). J. Biol. Chem., 273, 28700 ± 28707. Bhattacharyya R and Wedegaertner PB. (2000). J. Biol. Kotlar TJ, Young RH, Albanese C, Crowley Jr. WF, Scully Chem., 275, 14992 ± 14999. RE and Jameson JL. (1997). J. Clin. Endocrinol. Metab., Boguski MS and McCormick F. (1993). Nature, 366, 643 ± 82, 1020 ± 1026. 654. Kozasa T, Jiang X, Hart MJ, Sternweis PM, Singer WD, Bourne HR. (1997). Curr. Opin. Cell. Biol., 9, 134 ± 142. Gilman AG, Bollag G and Sternweis PC. (1998). Science, Bourne HR, Sanders DA and McCormick F. (1991). Nature, 280, 2109 ± 2111. 349, 117 ± 127. Kranenburg O, Poland M, van Horck FP, Drechsel D, Hall Buhl AM, Johnson NL, Dhanasekaran N and Johnson GL. A and Moolenaar WH. (1999). Mol. Biol. Cell., 10, 1851 ± (1995). J. Biol. Chem., 270, 24631 ± 24634. 1857. Bustelo XR and Barbacid M. (1992). Science, 256, 1196 ± Langdon S, Sethi T, Ritchie A, Muir M, Smyth J and 1199. Rozengurt E. (1992). Cancer Res., 52, 4554 ± 4557. ChanAM,FlemingTP,McGovernES,ChedidM,MikiT Mao J, Xie W, Yuan H, Simon MI, Mano H and Wu D. and Aaronson SA. (1993). Mol. Cell. Biol., 13, 762 ± 768. (1998a). EMBO J., 17, 5638 ± 5646. Clapham DE and Neer EJ. (1997). Annu.Rev.Pharmacol. Mao J, Yuan H, Xie W, Simon MI and Wu D. (1998b). J. Toxicol., 37, 167 ± 203. Biol. Chem., 273, 27118 ± 27123. Crespo P, Schuebel KE, Ostrom AA, Gutkind JS and Margolis B, Hu P, Katzav S, Li W, Oliver JM, Ullrich A, Bustelo XR. (1997). Nature, 385, 169 ± 172. Weiss A and Schlessinger J. (1992). Nature, 356, 71 ± 74. Dhanasekaran N, Heasley LE and Johnson GL. (1995). Narumiya S. (1996). J. Biochem. (Tokyo), 120, 215 ± 228. Endocr. Rev., 16, 259 ± 270. Needham LK and Rozengurt E. (1998). J. Biol. Chem., 273, Dohlman HG, Caron MG and Lefkowitz RJ. (1987). 14626 ± 14632. Biochemistry, 26, 2657 ± 2664. Nicholas J, Cameron KR and Honess RW. (1992). Nature, Flower DR. (1999). Biochim. Biophys. Acta., 1422, 207 ± 234. 355, 362 ± 365. Fromm C, Coso OA, Montaner S, Xu N and Gutkind JS. O€ermanns S. (1999). Naunyn Schmiedebergs Arch. Pharma- (1997). Proc. Natl. Acad. Sci. USA, 94, 10098 ± 10103. col., 360, 5±13. Fukuhara S, Chikumi H and Gutkind JS. (2000). FEBS Parks S and Wieschaus E. (1991). Cell, 64, 447 ± 458. Lett., 485, 183 ± 188. Parma J, Duprez L, Van Sande J, Cochaux P, Gervy C, Fukuhara S, Murga C, Zohar M, Igishi T and Gutkind JS. Mockel J, Dumont J and Vassart G. (1993). Nature, 365, (1999). J. Biol. Chem., 274, 5868 ± 5879. 649 ± 651. Gohla A, Harhammer R and Schultz G. (1998). J. Biol. Ren X, Kiosses WB, Sieg DJ, Otey CA, Schlaepfer DD and Chem., 273, 4653 ± 4659. Schwartz MA. (2000). J. Cell. Sci., 113, 3673 ± 3678. Gohla A, O€ermanns S, Wilkie TM and Schultz G. (1999). J. Ridley AJ and Hall A. (1992). Cell, 70, 389 ± 399. Biol. Chem., 274, 17901 ± 17907. Rosenkilde MM, Waldhoer M, LuÈ ttichau HR and Schwartz Gutkind JS. (1998). Oncogene, 17, 1331 ± 1342. TW. (2001). Oncogene (in press). Gutkind JS, Coso OA and Xu N. (1998). G Proteins, Sethi T, Langdon S, Smyth J and Rozengurt E. (1992). Receptors and Disease. Allen M. Spiegel (ed.). Humana Cancer Res., 52, 2737s ± 2742s. Press: Totowa, NJ, pp 101 ± 117. ShenkerA,LaueL,KosugiS,MerendinoJr.JJ,MinegishiT Gutkind JS, Novotny EA, Brann MR and Robbins KC. and Cutler Jr. GB. (1993). Nature, 365, 652 ± 654. (1991). Proc. Natl. Acad. Sci. USA, 88, 4703 ± 4707. Shi CS, Sinnarajah S, Cho H, Kozasa T and Kehrl JH. Hacker U and Perrimon N. (1998). Genes Dev., 12, 274 ± 284. (2000). J. Biol. Chem., 275, 24470 ± 24476. Hamm HE. (1998). J. Biol. Chem., 273, 669 ± 672. Sodhi A, Montaner S, Patel V, Zohar M, Bais C, Mesri EA Hart MJ, Eva A, Evans T, Aaronson SA and Cerione RA. and Gutkind JS. (2000). Cancer Res., 60, 4873 ± 4880. (1991). Nature, 354, 311 ± 314. Spiegel AM. (1996). J. Clin. Endocrinol. Metab., 81, 2434 ± Hart MJ, Eva A, Zangrilli D, Aaronson SA, Evans T, 2442. Cerione RA and Zheng Y. (1994). J. Biol. Chem., 269, 62 ± Stadel JM, Wilson S and Bergsma DJ. (1997). Trends 65. Pharmacol. Sci., 18, 430 ± 437. Hart MJ, Jiang X, Kozasa T, Roscoe W, Singer WD, Gilman Strathmann MP and Simon MI. (1991). Proc. Natl. Acad. AG, Sternweis PC and Bollag G. (1998). Science, 280, Sci. USA, 88, 5582 ± 5586. 2112 ± 2114. Takai Y, Sasaki T, Tanaka K and Nakanishi H. (1995). Heasley LE. (2001). Oncogene (in press). Trends Biochem. Sci., 20, 227 ± 231. Hellmich MR, Rui XL, Hellmich HL, Fleming RY, Evers Teramoto H, Salem P, Robbins KC, Bustelo XR and BM and Townsend CM, Jr. (2000). J. Biol. Chem., 275, Gutkind JS. (1997). J. Biol. Chem., 272, 10751 ± 10755. 32122 ± 32128. Thomas F, Arvelo F, Antoine E, Jacrot M and Poupon MF. Ho€ AO, Cote GJ, Fritsche Jr. HA, Qiu H, Schultz PN and (1992). Cancer Res., 52, 4872 ± 4877. Gagel RF. (1999). Neoplasia, 1, 485 ± 491. Ueda T, Kikuchi A, Ohga N, Yamamoto J and Takai Y. Ilic D, Furuta Y, Kanazawa S, Takeda N, Sobue K, (1990). J. Biol. Chem., 265, 9373 ± 9380. Nakatsuji N, Nomura S, Fujimoto J, Okada M and Van Aelst L and D'Souza-Schorey C. (1997). Genes Dev., 11, Yamamoto T. (1995). Nature, 377, 539 ± 544. 2295 ± 2322. Ilic D, Kanazawa S, Furuta Y, Yamamoto T and Aizawa S. Voyno-Yasenetskaya TA, Pace AM and Bourne HR. (1994). (1996). Exp. Cell. Res., 222, 298 ± 303. Oncogene, 9, 2559 ± 2565.

Oncogene RGS-containing RhoGEFs S Fukuhara et al 1668 Weinberg RA. (1985). Science, 230, 770 ± 776. Xu N, Voyno-Yasenetskaya T and Gutkind JS. (1994). Wess J. (1997). FASEB J., 11, 346 ± 354. Biochem. Biophys. Res. Commun., 201, 603 ± 609. Whitehead IP, Campbell S, Rossman KL and Der CJ. (1997). Yarden Y, Escobedo JA, Kuang WJ, Yang-Feng TL, Daniel Biochim. Biophys. Acta., 1332, F1 ± F23. TO, Tremble PM, Chen EY, Ando ME, Harkins RN, Wilk-Blaszczak MA, Singer WD, Gutowski S, Sternweis PC Francke U. et al. (1986). Nature, 323, 226 ± 232. and Belardetti F. (1994). Neuron, 13, 1215 ± 1224. Young D, Waitches G, Birchmeier C, Fasano O and Wigler Wilkie TM, Gilbert DJ, Olsen AS, Chen XN, Amatruda TT, M. (1986). Cell, 45, 711 ± 719. Korenberg JR, Trask BJ, de Jong P, Reed RR, Simon MI. Zohn IE, Klinger M, Karp X, Kirk H, Symons M, et al. (1992). Nat. Genet., 1, 85 ± 91. Chrzanowska-Wodnicka M, Der CJ and Kay RJ. (2000). Xu N, Bradley L, Ambdukar I and Gutkind JS. (1993). Proc. Oncogene, 19, 3866 ± 3877. Natl. Acad. Sci. USA, 90, 6741 ± 6745.

Oncogene