Identification of Guanine Nucleotide Exchange Factors (GEFs) for the Rap1 GTPase
2000; Elsevier BV; Volume: 275; Issue: 45 Linguagem: Inglês
10.1074/jbc.m005327200
ISSN1083-351X
AutoresJohn F. Rebhun, Ariel F. Castro, Lawrence A. Quilliam,
Tópico(s)Ion channel regulation and function
ResumoAlthough the Ras subfamily of GTPases consists of ∼20 members, only a limited number of guanine nucleotide exchange factors (GEFs) that couple extracellular stimuli to Ras protein activation have been identified. Furthermore, no novel downstream effectors have been identified for the M-Ras/R-Ras3 GTPase. Here we report the identification and characterization of three Ras family GEFs that are most abundantly expressed in brain. Two of these GEFs, MR-GEF (M-Ras-regulated GEF, KIAA0277) and PDZ-GEF (KIAA0313) bound specifically to nucleotide-free Rap1 and Rap1/Rap2, respectively. Both proteins functioned as Rap1 GEFs in vivo. A third GEF, GRP3 (KIAA0846), activated both Ras and Rap1 and shared significant sequence homology with the calcium- and diacylglycerol-activated GEFs, GRP1 and GRP2. Similarly to previously identified Rap GEFs, C3G and Smg GDS, each of the newly identified exchange factors promoted the activation of Elk-1 in the LNCaP prostate tumor cell line where B-Raf can couple Rap1 to the extracellular receptor-activated kinase cascade. MR-GEF and PDZ-GEF both contain a region immediately N-terminal to their catalytic domains that share sequence homology with Ras-associating orRalGDS/AF6 homology (RA) domains. By searching for in vitro interaction with Ras-GTP proteins, PDZ-GEF specifically bound to Rap1A- and Rap2B-GTP, whereas MR-GEF bound to M-Ras-GTP. C-terminally truncated MR-GEF, lacking the GEF catalytic domain, retained its ability to bind M-Ras-GTP, suggesting that the RA domain is important for this interaction. Co-immunoprecipitation studies confirmed the interaction of M-Ras-GTP with MR-GEF in vivo. In addition, a constitutively active M-Ras(71L) mutant inhibited the ability of MR-GEF to promote Rap1A activation in a dose-dependent manner. These data suggest that M-Ras may inhibit Rap1 in order to elicit its biological effects. Although the Ras subfamily of GTPases consists of ∼20 members, only a limited number of guanine nucleotide exchange factors (GEFs) that couple extracellular stimuli to Ras protein activation have been identified. Furthermore, no novel downstream effectors have been identified for the M-Ras/R-Ras3 GTPase. Here we report the identification and characterization of three Ras family GEFs that are most abundantly expressed in brain. Two of these GEFs, MR-GEF (M-Ras-regulated GEF, KIAA0277) and PDZ-GEF (KIAA0313) bound specifically to nucleotide-free Rap1 and Rap1/Rap2, respectively. Both proteins functioned as Rap1 GEFs in vivo. A third GEF, GRP3 (KIAA0846), activated both Ras and Rap1 and shared significant sequence homology with the calcium- and diacylglycerol-activated GEFs, GRP1 and GRP2. Similarly to previously identified Rap GEFs, C3G and Smg GDS, each of the newly identified exchange factors promoted the activation of Elk-1 in the LNCaP prostate tumor cell line where B-Raf can couple Rap1 to the extracellular receptor-activated kinase cascade. MR-GEF and PDZ-GEF both contain a region immediately N-terminal to their catalytic domains that share sequence homology with Ras-associating orRalGDS/AF6 homology (RA) domains. By searching for in vitro interaction with Ras-GTP proteins, PDZ-GEF specifically bound to Rap1A- and Rap2B-GTP, whereas MR-GEF bound to M-Ras-GTP. C-terminally truncated MR-GEF, lacking the GEF catalytic domain, retained its ability to bind M-Ras-GTP, suggesting that the RA domain is important for this interaction. Co-immunoprecipitation studies confirmed the interaction of M-Ras-GTP with MR-GEF in vivo. In addition, a constitutively active M-Ras(71L) mutant inhibited the ability of MR-GEF to promote Rap1A activation in a dose-dependent manner. These data suggest that M-Ras may inhibit Rap1 in order to elicit its biological effects. guanine nucleotide exchange factor extracellular receptor activated kinase mitogen-activated protein kinase Ras-association or RalGDS/AF6 homology domain trypsin inhibitory units kilobase pair glutathioneS-transferase polyacrylamide gel electrophoresis phenylmethylsulfonyl fluoride wild type hemagglutinin guanosine 5′-3-O-(thio)triphosphate structurally conserved regions Ras exchanger motif diacylglycerol cytomegalovirus Ras is the prototype for a large superfamily of GTPases that regulate multiple cellular processes (1Campbell S.L. Khosravi-Far R. Rossman K.L. Clark G.J. Der C.J. Oncogene. 1998; 17: 1395-1413Crossref PubMed Scopus (918) Google Scholar, 2Macara I.G. Lounsbury K.M. Richards S.A. McKiernan C. Bar-Sagi D. FASEB J. 1996; 10: 625-630Crossref PubMed Scopus (210) Google Scholar). These include intracellular signal transduction for cell growth and differentiation (Ras subfamily), regulation of the actin cytoskeleton (Rho subfamily), membrane trafficking (Rab subfamily), vesicle trafficking (ARF subfamily), and nuclear transport (Ran) (1Campbell S.L. Khosravi-Far R. Rossman K.L. Clark G.J. Der C.J. Oncogene. 1998; 17: 1395-1413Crossref PubMed Scopus (918) Google Scholar, 2Macara I.G. Lounsbury K.M. Richards S.A. McKiernan C. Bar-Sagi D. FASEB J. 1996; 10: 625-630Crossref PubMed Scopus (210) Google Scholar). Each protein functions as a GTP/GDP-regulated switch that cycles between inactive GDP- and active GTP-bound states. This cycle is tightly controlled in vivo by two classes of regulatory proteins. Guanine nucleotide exchange factors (GEFs)1serve as Ras activators by promoting acquisition of the active GTP-bound state and are the key link between cell surface receptors and Ras activation (3Quilliam L.A. Khosravi-Far R. Huff S.Y. Der C.J. BioEssays. 1995; 17: 395-404Crossref PubMed Scopus (193) Google Scholar). Meanwhile, GTPase-activating proteins promote rapid GTP hydrolysis, returning Ras back to its inactive GDP-bound state (4Wittinghofer A. Biol. Chem. Hoppe-Seyler. 1998; 379: 933-937PubMed Google Scholar).Presently ∼20 members of the Ras subfamily of GTPases have been identified in mammalian cells that include R-Ras, TC21/R-Ras2, M-Ras/R-Ras3, Rap1A, -1B, -2A, and -2B, Ral A and B, Rit, Rin, Dex-Ras, Rheb, Rhes, NOEY2, κB-Ras 1 and 2, as well as the classic (Ha-, K-, and N-) Ras proteins (1Campbell S.L. Khosravi-Far R. Rossman K.L. Clark G.J. Der C.J. Oncogene. 1998; 17: 1395-1413Crossref PubMed Scopus (918) Google Scholar). Rap1A (Krev-1) was originally described as a competitive antagonist of K-Ras-induced transformation (5Kitayama H. Sugimoto Y. Matsuzaki T. Ikawa Y. Noda M. Cell. 1989; 56: 77-84Abstract Full Text PDF PubMed Scopus (757) Google Scholar). The ability of Rap1 to block transformation is likely due to its ability to form non-productive complexes with Ras effectors such as c-Raf-1 (6Cook S.J. Rubinfeld B. Albert I. McCormick F. EMBO J. 1993; 12: 3475-3485Crossref PubMed Scopus (332) Google Scholar). Although it fails to activate c-Raf-1, the Raf isomer, B-Raf is a major known effector of Rap1 and has been shown to elicit the ability of Rap1 to promote ERK/MAPK activation in PC12 cells as well as the LNCaP prostate carcinoma cell line (7Vossler M.R. Yao H. York R.D. Pan M.G. Rim C.S. Stork P.J. Cell. 1997; 89: 73-82Abstract Full Text Full Text PDF PubMed Scopus (942) Google Scholar, 8Chen T. Cho R.W. Stork P.J. Weber M.J. Cancer Res. 1999; 59: 213-218PubMed Google Scholar). Other putative Rap1 effectors include AF6 (9Linnemann T. Geyer M. Jaitner B.K. Block C. Kalbitzer H.R. Wittinghofer A. Herrmann C. J. Biol. Chem. 1999; 274: 13556-13562Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar), RalGDS (10Zwartkruis F.J. Wolthuis R.M. Nabben N.M. Franke B. Bos J.L. EMBO J. 1998; 17: 5905-5912Crossref PubMed Scopus (191) Google Scholar), and Krit1 (11Serebriiskii I. Estojak J. Sonoda G. Testa J.R. Golemis E.A. Oncogene. 1997; 15: 1043-1049Crossref PubMed Scopus (181) Google Scholar), a molecule recently implicated in the formation of cavernous angiomas (12Sahoo T. Johnson E.W. Thomas J.W. Kuehl P.M. Jones T.L. Dokken C.G. Touchman J.W. Gallione C.J. Lee-Lin S.Q. Kosofsky B. Kurth J.H. Louis D.N. Mettler G. Morrison L. Gil-Nagel A. Rich S.S. Zabramski J.M. Boguski M.S. Green E.D. Marchuk D.A. Hum. Mol. Genet. 1999; 8: 2325-2333Crossref PubMed Scopus (292) Google Scholar, 13Laberge-le Couteulx S. Jung H.H. Labauge P. Houtteville J.P. Lescoat C. Cecillon M. Marechal E. Joutel A. Bach J.F. Tournier-Lasserve E. Nat. Genet. 1999; 23: 189-193Crossref PubMed Scopus (388) Google Scholar).M-Ras/R-Ras3 was identified as a Ras-related sequence in the EST data base (14Quilliam L.A. Graham K.R. Castro A.F. Martin C.B. Der C.J. Bi C. J. Biol. Chem. 1999; 274: 23850-23857Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 15Kimmelman A. Tolkacheva T. Lorenzi M.V. Osada M. Chan A.M. Oncogene. 1997; 15: 2675-2685Crossref PubMed Scopus (78) Google Scholar, 16Ehrhardt G.R. Leslie K.B. Lee F. Wieler J.S. Schrader J.W. Blood. 1999; 94: 2433-2444Crossref PubMed Google Scholar) and independently by others (17Matsumoto K. Asano T. Endo T. Oncogene. 1997; 15: 2409-2417Crossref PubMed Scopus (80) Google Scholar, 18Louahed J. Grasso L. De Smet C. Van Roost E. Wildmann C. Nicolaides N.C. Levitt R.C. Renauld J.-C. Blood. 1999; 94: 1701-1710Crossref PubMed Google Scholar) due to its expression in muscle cells and interleukin 9-induced T helper cells. M-Ras is highly abundant in brain but is also expressed in other tissues and a broad variety of cultured cell lines (14Quilliam L.A. Graham K.R. Castro A.F. Martin C.B. Der C.J. Bi C. J. Biol. Chem. 1999; 274: 23850-23857Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 15Kimmelman A. Tolkacheva T. Lorenzi M.V. Osada M. Chan A.M. Oncogene. 1997; 15: 2675-2685Crossref PubMed Scopus (78) Google Scholar, 16Ehrhardt G.R. Leslie K.B. Lee F. Wieler J.S. Schrader J.W. Blood. 1999; 94: 2433-2444Crossref PubMed Google Scholar). Although it is activated by Ras GEFs and can bind/activate some Ras protein effectors such as AF-6 and the Raf/ERK cascade (14Quilliam L.A. Graham K.R. Castro A.F. Martin C.B. Der C.J. Bi C. J. Biol. Chem. 1999; 274: 23850-23857Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 15Kimmelman A. Tolkacheva T. Lorenzi M.V. Osada M. Chan A.M. Oncogene. 1997; 15: 2675-2685Crossref PubMed Scopus (78) Google Scholar), unique sequences surrounding its effector binding domain suggest that, like other Ras-related GTPases, it will interact with novel downstream targets. Despite this prediction none have been identified so far.Although over 20 Ras family GTPases have been identified, only a handful of the GEFs that regulate them have been described as follows: Sos, GRF and GRP isoforms for Ras, the RalGDS family for Ral, and C3G, RapGRP/CalDAG GEF, Epac/cAMP-GEFs for Rap1 (3Quilliam L.A. Khosravi-Far R. Huff S.Y. Der C.J. BioEssays. 1995; 17: 395-404Crossref PubMed Scopus (193) Google Scholar, 19Ebinu J.O. Bottorff D.A. Chan E.Y. Stang S.L. Dunn R.J. Stone J.C. Science. 1998; 280: 1082-1086Crossref PubMed Scopus (545) Google Scholar, 20Kawasaki H. Springett G.M. Toki S. Canales J.J. Harlan P. Blumenstiel J.P. Chen E.J. Bany I.A. Mochizuki N. Ashbacher A. Matsuda M. Housman D.E. Graybiel A.M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13278-13283Crossref PubMed Scopus (309) Google Scholar, 21de Rooij J. Zwartkruis F.J. Verheijen M.H. Cool R.H. Nijman S.M. Wittinghofer A. Bos J.L. Nature. 1998; 396: 474-477Crossref PubMed Scopus (1602) Google Scholar). One unique GEF, Smg GDS (small molecular weightGTP-binding protein guanine nucleotidedissociation stimulator), catalyzes exchange on Rho family GTPases as well as Rap1 and K-Ras in vitro (22Mizuno T. Kaibuchi K. Yamamoto T. Kawamura M. Sakoda T. Fujioka H. Matsuura Y. Takai Y. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 6442-6446Crossref PubMed Scopus (167) Google Scholar,23Chuang T.H. Xu X. Quilliam L.A. Bokoch G.M. Biochem. J. 1994; 303: 761-767Crossref PubMed Scopus (36) Google Scholar). Smg GDS is composed of 11 armadillo repeats similar to those found in APC and catenins (24Peifer M. Berg S. Reynolds A.B. Cell. 1994; 76: 789-791Abstract Full Text PDF PubMed Scopus (545) Google Scholar) and does not share significant homology with CDC25 (25Boguski M.S. McCormick F. Nature. 1993; 366: 643-654Crossref PubMed Scopus (1752) Google Scholar).The catalytic domains of all other Ras subfamily GEFs share ∼30% homology with each other and the Saccharomyces cerevisiaeprotein, CDC25. Conservation between "CDC25 homology" domains is greatest within structurally conserved regions (SCR) 1–3 that were first noted by Boguski and McCormick (25Boguski M.S. McCormick F. Nature. 1993; 366: 643-654Crossref PubMed Scopus (1752) Google Scholar), whereas additional C-terminal regions (SCR 4 and 5) have subsequently become evident (Fig.1) (26Rebhun J.F. Chen H. Quilliam L.A. J. Biol. Chem. 2000; 275: 13406-13410Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar). A region outside of the core catalytic domain, referred to as Ras exchanger motif (REM), conserved non-catalytic, or SCR 0 has also been noted (27Lai C.C. Boguski M. Broek D. Powers S. Mol. Cell. Biol. 1993; 13: 1345-1352Crossref PubMed Scopus (138) Google Scholar, 28Fam N.P. Fan W.T. Wang Z. Zhang L.J. Chen H. Moran M.F. Mol. Cell. Biol. 1997; 17: 1396-1406Crossref PubMed Scopus (133) Google Scholar, 29Boriack-Sjodin P.A. Margarit S.M. Bar-Sagi D. Kuriyan J. Nature. 1998; 394: 337-343Crossref PubMed Scopus (608) Google Scholar). Based on the Sos1 Ras GEF x-ray crystal structure, REM/SCR 0 is a structural component that binds to SCR4 and is not involved in Ras interaction (29Boriack-Sjodin P.A. Margarit S.M. Bar-Sagi D. Kuriyan J. Nature. 1998; 394: 337-343Crossref PubMed Scopus (608) Google Scholar).Besides the common "CDC25 homology" catalytic domain, GEFs possess a wide variety of domains that are important for the regulation of their function. For instance, Sos contains proline-rich clusters that interact with the SH3 domains of adapter proteins such as Grb2 bringing Sos to the membrane after receptor tyrosine kinase activation (3Quilliam L.A. Khosravi-Far R. Huff S.Y. Der C.J. BioEssays. 1995; 17: 395-404Crossref PubMed Scopus (193) Google Scholar). Translocation to the membrane is assumed to be critical in the activation of GEFs since it brings them into contact with membrane-bound GTPases. PH domains are also found in several GEFs, including Sos and GRFs, where this domain is also important for membrane interaction (30Chen R.H. Corbalan-Garcia S. Bar-Sagi D. EMBO J. 1997; 16: 1351-1359Crossref PubMed Scopus (115) Google Scholar). The GRP or CalDAG GEFs for Ras and Rap each possess EF hands and C1 domains and are regulated by calcium and/or DAG (19Ebinu J.O. Bottorff D.A. Chan E.Y. Stang S.L. Dunn R.J. Stone J.C. Science. 1998; 280: 1082-1086Crossref PubMed Scopus (545) Google Scholar, 20Kawasaki H. Springett G.M. Toki S. Canales J.J. Harlan P. Blumenstiel J.P. Chen E.J. Bany I.A. Mochizuki N. Ashbacher A. Matsuda M. Housman D.E. Graybiel A.M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13278-13283Crossref PubMed Scopus (309) Google Scholar, 31Tognon C.E. Kirk H.E. Passmore L.A. Whitehead I.P. Der C.J. Kay R.J. Mol. Cell. Biol. 1998; 18: 6995-7008Crossref PubMed Scopus (203) Google Scholar). Furthermore, the Rap GEFs, Epac/cAMP-GEF 1 and 2, are regulated by cyclic AMP (21de Rooij J. Zwartkruis F.J. Verheijen M.H. Cool R.H. Nijman S.M. Wittinghofer A. Bos J.L. Nature. 1998; 396: 474-477Crossref PubMed Scopus (1602) Google Scholar, 32Kawasaki H. Springett G.M. Mochizuki N. Toki S. Nakaya M. Matsuda M. Housman D.E. Graybiel A.M. Science. 1998; 282: 2275-2279Crossref PubMed Scopus (1165) Google Scholar) indicating that Ras function can be regulated by receptors others than tyrosine kinases. Finally, binding to Ras activates the RalGDS family of Ral GEFs (1Campbell S.L. Khosravi-Far R. Rossman K.L. Clark G.J. Der C.J. Oncogene. 1998; 17: 1395-1413Crossref PubMed Scopus (918) Google Scholar). These proteins contain a Ras-associating or RalGDS/AF6-homology (RA) domain responsible for the interaction with Ras-GTP (33Ponting C.P. Benjamin D.R. Trends Biochem. Sci. 1996; 21: 422-425Abstract Full Text PDF PubMed Scopus (177) Google Scholar, 34Hofer F. Fields S. Schneider C. Martin G.S. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 11089-11093Crossref PubMed Scopus (245) Google Scholar).Since the increasing number of Ras family members suggested the existence of additional Ras GEFs to regulate them, we searched the NCBI data bases for novel CDC25 homology domain-containing proteins. Three of the putative GEFs that we identified promote the formation of Rap1-GTP in vivo, and splice variants of a fourth activate Ral (26Rebhun J.F. Chen H. Quilliam L.A. J. Biol. Chem. 2000; 275: 13406-13410Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar). Two of the Rap GEFs contain RA domains. Just as the Ral family GEFs, RalGDS, RGL and RGL2/Rlf, are downstream effectors of Ras, one of the novel GEFs, MR-GEF (M-Ras-regulated Rap GEF), appears to be regulated by interaction with M-Ras in a GTP-dependent manner. Since Rap1 can antagonize the activation of Raf-1 by other GTPases, negative regulation of Rap1 by M-Ras via MR-GEF may help facilitate M-Ras-induced activation of Raf-1.DISCUSSIONCurrently 14 members of the Rho family of GTPase have been identified for which there are more than 35 GEFs (the Dbl/PH proteins) responsible for loading them with GTP (40Zohn I.M. Campbell S.L. Khosravi-Far R. Rossman K.L. Der C.J. Oncogene. 1998; 17: 1415-1438Crossref PubMed Scopus (319) Google Scholar). Although there are an increasing number of Ras family members, until recently only a limited number of CDC25 homology domain-containing GEFs existed to mediate Ras protein activation. A sequence homology search of the DNA data bases has now revealed several additional members. We report here that putative exchange factors MR-GEF and PDZ-GEF activate Rap1, and a third, GRP3, activates Rap and Ras. An additional Ral GEF family "RalGPS" has been described elsewhere (26Rebhun J.F. Chen H. Quilliam L.A. J. Biol. Chem. 2000; 275: 13406-13410Abstract Full Text Full Text PDF PubMed Scopus (65) Google Scholar). We found that MR-GEF and PDZ-GEF contain RA domains. Importantly, the RA domain present in MR-GEF is regulated by interaction with M-Ras in a GTP-dependent manner and is conserved in other Rap GEFs. In agreement with the results presented here, several other recent reports have described these new GEFs (41de Rooij J. Boenink N.M. van Triest M. Cool R.H. Wittinghofer A. Bos J.L. J. Biol. Chem. 1999; 274: 38125-38130Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar, 42Liao Y. Kariya K. Hu C.D. Shibatohge M. Goshima M. Okada T. Watari Y. Gao X. Jin T.G. Yamawaki-Kataoka Y. Kataoka T. J. Biol. Chem. 1999; 274: 37815-37820Abstract Full Text Full Text PDF PubMed Scopus (77) Google Scholar, 43Ichiba T. Hoshi Y. Eto Y. Tajima N. Kuraishi Y. FEBS Lett. 1999; 457: 85-89Crossref PubMed Scopus (26) Google Scholar, 44Ohtsuka T. Hata Y. Ide N. Yasuda T. Inoue E. Inoue T. Mizoguchi A. Takai Y. Biochem. Biophys. Res. Commun. 1999; 265: 38-44Crossref PubMed Scopus (92) Google Scholar, 45de Rooij J. Rehmann H. van Triest M. Cool R.H. Wittinghofer A. Bos J.L. J. Biol. Chem. 2000; 275: 20829-20836Abstract Full Text Full Text PDF PubMed Scopus (319) Google Scholar, 46Ohba Y. Mochizuki N. Yamashita S. Chan A.M. Schrader J.W. Hattori S. Nagashima K. Matsuda M. J. Biol. Chem. 2000; 275: 20020-20026Abstract Full Text Full Text PDF PubMed Scopus (127) Google Scholar, 47Pham N. Cheglakov I. Koch C.A. de Hoog C.L. Moran M.F. Rotin D. Curr. Biol. 2000; 10: 555-558Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). However, this study is the first to report the interaction of activated M-Ras with MR-GEF and to demonstrate M-Ras regulation of the Rap1 exchange activity of MR-GEF.The identification of the above proteins brings the total number of known Rap1 GEFs to nine (SmgGDS, C3G, GRP2 and -3, Epac 1 and 2, MR-GEF, PDZ-GEFs 1 and 2). Within the group of Rap GEFs, Smg GDS presents unusual characteristics. It does not possess the CDC25 homology domain found in other Ras GEFs and has previously been reported to activate Rap1, Rac, Rho, and K-Ras in vitro (22Mizuno T. Kaibuchi K. Yamamoto T. Kawamura M. Sakoda T. Fujioka H. Matsuura Y. Takai Y. Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 6442-6446Crossref PubMed Scopus (167) Google Scholar,23Chuang T.H. Xu X. Quilliam L.A. Bokoch G.M. Biochem. J. 1994; 303: 761-767Crossref PubMed Scopus (36) Google Scholar). Here we showed that Smg GDS acts as a Rap GEF in vivoand further demonstrate its ability to activate Ral. This is consistent with the observed interaction of Xenopus Ral with a frog Smg GDS-related protein in a yeast two-hybrid screen (48Iouzalen N. Camonis J. Moreau J. Biochem. Biophys. Res. Commun. 1998; 250: 359-363Crossref PubMed Scopus (11) Google Scholar) and the presence of a polybasic C terminus in Ral similar to that found in other Smg GDS substrates. Why a single GEF is responsible for activating so many GTPases and how it is regulated will require further investigation. However, our in vivo studies confirm previous in vitro and functional assays that suggested that Smg GDS would act as a Rap1 GEF.It is interesting to note how divergent the mechanisms of Rap GEF regulation are. C3G is activated downstream of tyrosine kinases, being recruited by adapter proteins such as Crk and p130Cas (49Matsuda M. Kurata T. Cell. Signal. 1996; 8: 335-340Crossref PubMed Scopus (50) Google Scholar). SHEP1/NSP3, a putative GEF that can bind Rap1 in vitro, is also downstream of protein tyrosine kinases (50Dodelet V.C. Pazzagli C. Zisch A.H. Hauser C.A. Pasquale E.B. J. Biol. Chem. 1999; 274: 31941-31946Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 51Lu Y. Brush J. Stewart T.A. J. Biol. Chem. 1999; 274: 10047-10052Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar). Epac/cAMP-GEFs 1 and 2 are activated by the second messenger cAMP, whereas GRP2 (and potentially GRP3) is activated by phorbol esters/diacylglycerol and/or calcium (19Ebinu J.O. Bottorff D.A. Chan E.Y. Stang S.L. Dunn R.J. Stone J.C. Science. 1998; 280: 1082-1086Crossref PubMed Scopus (545) Google Scholar, 20Kawasaki H. Springett G.M. Toki S. Canales J.J. Harlan P. Blumenstiel J.P. Chen E.J. Bany I.A. Mochizuki N. Ashbacher A. Matsuda M. Housman D.E. Graybiel A.M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13278-13283Crossref PubMed Scopus (309) Google Scholar, 21de Rooij J. Zwartkruis F.J. Verheijen M.H. Cool R.H. Nijman S.M. Wittinghofer A. Bos J.L. Nature. 1998; 396: 474-477Crossref PubMed Scopus (1602) Google Scholar, 32Kawasaki H. Springett G.M. Mochizuki N. Toki S. Nakaya M. Matsuda M. Housman D.E. Graybiel A.M. Science. 1998; 282: 2275-2279Crossref PubMed Scopus (1165) Google Scholar). The discovery of PDZ-GEF and MR-GEF identifies novel domains/mechanisms to direct Rap GEF activation. PDZ-GEF has a number of putative domains that may regulate its activity. The PDZ domain regulates the attachment of PDZ-GEFs to the plasma membrane of HEK cells (47Pham N. Cheglakov I. Koch C.A. de Hoog C.L. Moran M.F. Rotin D. Curr. Biol. 2000; 10: 555-558Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar), whereas a putative C-terminal (SAV) PDZ-binding motif may be involved in the interaction of PDZ-GEF with a six PDZ domain-containing synaptic scaffolding protein, S-SCAM, described by Takai and colleagues (44Ohtsuka T. Hata Y. Ide N. Yasuda T. Inoue E. Inoue T. Mizoguchi A. Takai Y. Biochem. Biophys. Res. Commun. 1999; 265: 38-44Crossref PubMed Scopus (92) Google Scholar).PDZ-GEF also possesses a cAMP-binding domain, sharing weak homology with those of Epac and the protein kinase A regulatory subunits (21de Rooij J. Zwartkruis F.J. Verheijen M.H. Cool R.H. Nijman S.M. Wittinghofer A. Bos J.L. Nature. 1998; 396: 474-477Crossref PubMed Scopus (1602) Google Scholar,32Kawasaki H. Springett G.M. Mochizuki N. Toki S. Nakaya M. Matsuda M. Housman D.E. Graybiel A.M. Science. 1998; 282: 2275-2279Crossref PubMed Scopus (1165) Google Scholar). The binding of cAMP to PDZ-GEF is not required for its activation of Rap, although the presence of the cAMP-binding domain does have a slight inhibitory role (41de Rooij J. Boenink N.M. van Triest M. Cool R.H. Wittinghofer A. Bos J.L. J. Biol. Chem. 1999; 274: 38125-38130Abstract Full Text Full Text PDF PubMed Scopus (105) Google Scholar), analogous to the domain in Epac, which cannot activate Rap1A unless cAMP is present (21de Rooij J. Zwartkruis F.J. Verheijen M.H. Cool R.H. Nijman S.M. Wittinghofer A. Bos J.L. Nature. 1998; 396: 474-477Crossref PubMed Scopus (1602) Google Scholar). During the completion of this manuscript it was reported that the presence of cAMP or cGMP allowed PDZ-GEF to activate Ha-Ras in vivo without affecting Rap1 activation (47Pham N. Cheglakov I. Koch C.A. de Hoog C.L. Moran M.F. Rotin D. Curr. Biol. 2000; 10: 555-558Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar). We saw no binding of PDZ-GEF to Ha-Rasin vitro and no activation of Ha-Ras by PDZ-GEF in the absence of cAMP in vivo, suggesting that Ha-Ras activation by PDZ-GEF is highly dependent on cyclic nucleotides and that they must act by permitting/promoting Ha-Ras-GEF association. PDZ-GEF is not the only example of a Rap1 GEF that also works on Ras. We found that GRP3 also activates both Rap1 and Ha-Ras and is consistent with reports indicating that Ras and Rap1 are concomitantly activated by mitogens (10Zwartkruis F.J. Wolthuis R.M. Nabben N.M. Franke B. Bos J.L. EMBO J. 1998; 17: 5905-5912Crossref PubMed Scopus (191) Google Scholar, 52Posern G. Weber C.K. Rapp U.R. Feller S.M. J. Biol. Chem. 1998; 273: 24297-24300Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). Like GRP1 and GRP2, GRP3 contains EF hands and a C1 domain suggesting that it is regulated by calcium and/or diacylglycerol. Although we have seen strong activation of Ha-Ras and Rap1 by GRP3 under serum-starved conditions, it will be interesting to study whether activation by calcium and/or diacylglycerol regulates the specificity of GRP3 for Ha-Ras and Rap1 as cAMP appears to do with PDZ-GEF.The presence of RA domains in PDZ-GEF and MR-GEF suggests a novel mechanism of regulating the activity and/or specificity of Rap GEFs. The binding of Rap1 or -2 to the RA domain of PDZ-GEF may modulate its activity as part of a regulatory feedback mechanism. To date, regulation of PDZ-GEF activity by Rap1-GTP has not been reported. The C terminus of the MR-GEF RA domain is poorly conserved with the "classical" RA domains of RalGDS and AF-6 (Fig. 1 C). However, the ability of M-Ras to bind the N-terminal regulatory domain of MR-GEF in a GTP-dependent manner together with its ability to inhibit the activation of Rap1 by MR-GEF strongly suggest that association with the RA domain inhibits MR-GEF activity. Additionally, it was interesting to find the putative RA domain of MR-GEF is conserved in two other GEFs. In fact, the Rap1 GEF Epac2 and the putative exchange factor Link GEFII contain considerable homology to MR-GEF beyond their RA and CDC25 homology domains (Fig.1 C). This suggests that there is a family of Rap GEFs sensitive to M-Ras or other Ras protein regulations. As mentioned above, association of M-Ras with MR-GEF appears to inhibit Rap activation. Such negative regulation seems logical given the previous studies on these two GTPases. M-Ras, like the classical Ras proteins, induces cellular transformation of 3T3 fibroblasts (14Quilliam L.A. Graham K.R. Castro A.F. Martin C.B. Der C.J. Bi C. J. Biol. Chem. 1999; 274: 23850-23857Abstract Full Text Full Text PDF PubMed Scopus (99) Google Scholar, 15Kimmelman A. Tolkacheva T. Lorenzi M.V. Osada M. Chan A.M. Oncogene. 1997; 15: 2675-2685Crossref PubMed Scopus (78) Google Scholar), whereas Rap1 was initially discovered as a suppressor of K-Ras-induced transformation, thus antagonizing Ras functions. It has similarly been shown to antagonize T cell activation (53Boussiotis V.A. Freeman G.J. Berezovskaya A. Barber D.L. Nadler L.M. Science. 1997; 278: 124-128Crossref PubMed Scopus (394) Google Scholar). Under this scenario, it is reasonable to speculate that activation of M-Ras could be responsible for the negative regulation of the Rap1 pathway through its association with the RA domain of Rap GEFs. The lack of activated Rap1 would lead to the disassociation of the inactive Rap-Raf-1 complex leaving Raf-1 free for activation by M-Ras or other Ras proteins. As M-Ras is widely expressed and in some tissues is present at higher levels than classical Ras proteins (16Ehrhardt G.R. Leslie K.B. Lee F. Wieler J.S. Schrader J.W. Blood. 1999; 94: 2433-2444Crossref PubMed Google Scholar), such a mechanism may apply to several systems where Rap-Ras pathways result in different physiological outcomes. Currently, this hypothesis is under investigation.Regulation of downstream GTPases by Ras was originally described for a family of Ral exchange factors (RalGDS) that contain C-terminal RA domains (34Hofer F. Fields S. Schneider C. Martin G.S. Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 11089-11093Crossref PubMed Scopus (245) Google Scholar). This interaction serves to recruit Ral GEFs to the membrane where they contact and activate the Ral GTPase. It is interesting to note that MR-GEF and PDZ-GEF have RA domains immediately N-terminal to their catalytic domains. Such insertions of regulatory domains within their catalytic domains may be a means of modulating the activity as well as the location of the catalytic domain. Although the inhibitory role of M-Ras in Rap1 activation by MR-GEF supports this idea, further studies will be required to clarify their role. It will also be interesting to determine whether the Rap1/2 interaction with the RA domain of PDZ-GEF results in a similar effect. MR-GEF and PDZ-GEF join a growing list of proteins containing putative Ras associating doma
Referência(s)