Oncogene (2007) 26, 3122–3142 & 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc REVIEW G Protein regulation of MAPK networks ZG Goldsmith and DN Dhanasekaran Fels Institute for Cancer Research and Molecular Biology, Temple University School of Medicine, Philadelphia, PA, USA G proteins provide signal-coupling mechanisms to hepta- the a-subunits has been used as a basis for the helical cell surface receptors and are criticallyinvolved classification of G proteins into Gs,Gi,Gq and G12 in the regulation of different mitogen-activated protein families in which the a-subunits that show more than kinase (MAPK) networks. The four classes of G proteins, 50% homology are grouped together (Simon et al., defined bythe G s,Gi,Gq and G12 families, regulate 1991). In G-protein-coupled receptor (GPCR)-mediated ERK1/2, JNK, p38MAPK, ERK5 and ERK6 modules by signaling pathways, ligand-activated receptors catalyse different mechanisms. The a- as well as bc-subunits are the exchange of the bound GDP to GTP in the a-subunit involved in the regulation of these MAPK modules in a following which the GTP-bound a-subunit disassociate context-specific manner. While the a- and bc-subunits from the receptor as well as the bg-subunit. The GTP- primarilyregulate the MAPK pathwaysvia their respec- bound a-subunit and the bg-subunit stimulate distinct tive effector-mediated signaling pathways, recent studies downstream effectors including enzymes, ion channels have unraveled several novel signaling intermediates and small GTPase, thus regulating multiple signaling including receptor tyrosine kinases and small GTPases pathways including those involved in the activation of through which these G-protein subunits positivelyas well mitogen-activated protein kinase (MAPK) modules as negativelyregulate specific MAPK modules. Multiple (Johnson and Dhanasekaran, 1989; Hepler and Gilman, mechanisms together with specific scaffold proteins that 1992; Dhanasekaran et al., 1995; Dhanasekaran and can link G-protein-coupled receptors or G proteins to Prasad, 1998; Rozengurt, 1998). distinct MAPK modules contribute to the context-specific MAPK modules regulate many critical signaling and spatio-temporal regulation of mitogen-activated pathways including cell proliferation, differentiation protein signaling networks byG proteins. and apoptosis (Fanger et al., 1997; Minden and Karin, Oncogene (2007) 26, 3122–3142. doi:10.1038/sj.onc.1210407 1997; Dhanasekaran and Premkumar Reddy, 1998; Gutkind, 1998; Garrington and Johnson, 1999; Davis, Keywords: GPCR; signal transduction; G protein; 2000; Chang and Karin, 2001). Each of these modules MAPK; oncogene; cancer consists of at least three distinct kinases, namely an upstream mitogen-activated protein (MAP) kinase kinase kinase (MAP3K), a MAP kinase kinase (MAP2K) and a downstream MAP kinase (MAPK). In a typical signaling pathway, growth factor stimulated Introduction small GTPases such as Ras, CDC42, and Rac activate an upstream MAP kinase kinase kinase kinase Heterotrimericguanine nucleotide-binding proteins, (MAP4K), following which the MAP kinase cascades or G proteins, consisting of a-, b-andg-subunits, proceed through the sequential phosphorylation of transduce signals from heptahelical cell surface receptors constituent MAP3K, MAP2K and MAPK (Minden to intracellular effectors (Simon et al., 1991; Hepler and Karin, 1997; Dhanasekaran and Premkumar and Gilman, 1992; Conklin and Bourne, 1993). The Reddy, 1998; Davis, 2000; Chang and Karin, 2001). a-subunit is a 37–42 kDa protein, which contains Activated MAP kinases translocate from the cytosol to the guanine nucleotide-binding pocket along with the the nucleus where they regulate the activities of various intrinsicGTPase activity.The 35-kDa b-subunit and the transcription factors through phosphorylation (Khokh- 8–11-kDa g-subunit are tightly associated with each latchev et al., 1998; Brunet et al., 1999). GPCRs, other and are often referred to as a single entity, the primarily through their cognate G proteins, regulate bg-subunit (Simon et al., 1991; Hepler and Gilman, these three-tier MAPK phospho-relay systems to 1992; Conklin and Bourne, 1993). To date, 17 different modulate the expression of specific primary as well as a-subunits, five b-subunits and 12 g-subunits have secondary response genes involved in cell proliferation, been identified. However, the amino acid identity of differentiation and apoptosis (Gutkind, 2000; Kranen- burg and Moolenaar, 2001; Werry et al., 2005). G proteins have been shown to regulate diverse transcrip- Correspondence: Dr DN Dhanasekaran, Fels Institute for Cancer Research and Molecular Biology, Temple University School of tion factors such as AP-1 (Araki et al., 1997), NF-kB, Medicine, Philadelphia, PA 19140, USA. CRE, SRE (Takeuchi and Fukunaga, 2003), ATF1 E-mail: [email protected] (Ghosh et al., 2002), STAT3 (Sellers et al., 1999), ELK1 G Proteins and MAPKs ZG Goldsmith and DN Dhanasekaran 3123 (Todisco et al., 1997), Egr-1 (Gerasimovskaya et al., studies focused on defining the oncogenic pathways 2002), HIF-1a (Sodhi et al., 2000), MEF2 (Fukuhara regulated by the gsp oncogene that encodes the et al., 2000a, b) via ERK-, JNK-, p38MAPKs- or ERK5 constitutively activated mutants of Gas (Landis et al., modules. However, only recently the dynamic and 1989; Lyons et al., 1990). Functional characterization of multilayered mechanisms by which G proteins regulate these activating mutations indicated that the ectopic these MAPK networks have begun to be unraveled. This expression of Gas could stimulate ERK1/2 in different review is focused on analysing the mechanisms by which cell types (Faure et al., 1994). Using mutant S49 G proteins regulate different MAPK networks in a lymphoma cell lines that lack Gas or PKA, it was context-specific manner. further established that Gas and PKA are required to couple b2-AR signaling to ERK1/2 module (Wan and Huang, 1998). Analysis of such Gas-mediated stimula- tion of ERK1/2 indicated that it was mediated by a Regulation of MAPKs byG s pathway involving Rap-1 and its activation of B-Raf, one of the isoforms of Raf (Vossler et al., 1997; Wan G , defined by the a-subunit Ga , transduces signals s s and Huang, 1998; Schmitt and Stork, 2000; Weissman from G -coupled heptahelical receptors to adenylyl s et al., 2002; Dumaz and Marais, 2005; Houslay, 2006; cyclase that converts ATP to cAMP and subsequently Wang et al., 2006b). One of the mechanisms through to cAMP-mediated activation of protein kinase A which Ga activates Rap-1 involves cAMP-mediated (PKA) (Gilman, 1987; Hepler and Gilman, 1992). s direct activation of EPAC (exchange protein directly Consequently, as depicted in Figure 1, cAMP as well activated by cAMP), a guanine nucleotide exchange as cAMP-activated PKA have been shown to play a factor (GEF) specific for Rap-1 (de Rooij et al., 1998; major role in Ga -mediated stimulation as well as inhi- s Quilliam et al., 2002; Bos, 2006). In rat kidney cortical bition of MAPK modules such as those of ERK1/2, collecting duct cells, it has been shown that the ERK5 and p38MAPK (Cook and McCormick, 1993; stimulation of Ga and subsequent generation of cAMP Graves et al., 1993; Wu et al., 1993; Wan and s leads to the activation of EPAC, which stimulates Huang, 1998; Zheng et al., 2000; Laroche-Joubert GDP–GTP exchange in Rap-1. The GTP-bound Rap-1, et al., 2002; Stork and Schmitt, 2002; Dumaz and thus stimulated, activates B-Raf and the downstream Marais, 2005; Houslay, 2006; Pearson et al., 2006). ERK1/2 modules (Laroche-Joubert et al., 2002). The observation that the introduction of antibodies to Epac-1, Gas stimulation of ERK1/2 Rap-1 and B-Raf into these cells inhibited Gs-mediated The initial evidence that Gas is involved in the activation of ERK1/2 module established the linear regulation of ERK1/2 signaling module consisting pathway leading from cAMP to the activation of of a MAP3K–MAP2K–MAPK module defined by ERK1/2 module involving Epac-1, Rap-1 and B-Raf Raf, MEK1/2 and ERK1/2, respectively, came from (Laroche-Joubert et al., 2002). Adenylyl Cylclase P s P cAMP Ras EPAC PKA Src C-Raf C3G MEK-1/2 ERK-1/2 Rap1 MEKK2/3 MAP3K MEK5 MKK3/6 ERK5 B-Raf p38MAPK MEK-1/2 ERK-1/2 Figure 1 Gs regulation of ERK1/2. Gas stimulates the B-Raf-mediated activation of ERKs via Rap-1 or Ras and inhibits C-Raf- mediated activation of ERKs by phosphorylating C-Raf through PKA. Gas has also been shown to stimulate p38MAPK and inhibit ERK-5 modules via PKA-dependent mechanism. The dashed lines indicate the pathways that have not been fully resolved. Oncogene G Proteins and MAPKs ZG Goldsmith and DN Dhanasekaran 3124 Although further studies confirmed the role of Rap-1 ERK pathway involves PKA-mediated phosphorylation in Gas-stimulated activation of ERK1/2 module, they of a specific isoform of Raf, known as Raf-1 or C-Raf also indicated that Gas stimulation of ERK1/2 module (Wu et al., 1993; Dhillon et al., 2002). C-Raf contains via Rap-1 involves a Rap-1 GEF other than EPAC in three PKA-target Serines, namely Ser 43, 259 and 621 many cell types (Stork and Schmitt, 2002). The cAMP (Morrison et al., 1993), all of which can be phosphory- analog, 8-(4-chloro-phenylthio)-20-O-methyladenosine- lated in vitro by PKA (Wu et al., 1993; Hafner et al., 30,50-cyclic monophosphate (8CPT-2Me-cAMP) that 1994; Mischak et al., 1996). Mutational analyses of can activate EPAC, but not PKA, proved useful in C-Raf indicated that the Gas-mediated cAMP–PKA identifying the alternate pathway through which Gas signaling axis inhibits C-Raf through the phosphoryla- activated ERK1/2 via Rap-1 (Enserink et al., 2002). The tion of Ser-259 (Dhillon et al., 2002). The PKA observations that 8CPT-2Me-cAMP stimulated EPAC phosphorylation of Ser-259 of C-Raf attenuates the but failed to activate Rap-1-dependent activation of phosphorylation of Ser-338, which is involved in the ERK suggested that EPAC might not be involved in activation of C-Raf (Dhillon et al., 2002), thus providing Rap-1-mediated activation of ERK1/2 module in all a molecular basis for Gas-mediated inhibition of C-Raf– the cells (Enserink et al., 2002).