Direct Stimulation of the Guanine Nucleotide Exchange Activity of p115 RhoGEF by Gα13

See allHide authors and affiliations

Science  26 Jun 1998:
Vol. 280, Issue 5372, pp. 2112-2114
DOI: 10.1126/science.280.5372.2112


Signaling pathways that link extracellular factors to activation of the monomeric guanosine triphosphatase (GTPase) Rho control cytoskeletal rearrangements and cell growth. Heterotrimeric guanine nucleotide–binding proteins (G proteins) participate in several of these pathways, although their mechanisms are unclear. The GTPase activities of two G protein α subunits, Gα12 and Gα13, are stimulated by the Rho guanine nucleotide exchange factor p115 RhoGEF. Activated Gα13 bound tightly to p115 RhoGEF and stimulated its capacity to catalyze nucleotide exchange on Rho. In contrast, activated Gα12 inhibited stimulation by Gα13. Thus, p115 RhoGEF can directly link heterotrimeric G protein α subunits to regulation of Rho.

Certain heterotrimeric G proteins, particularly G12 and G13, regulate cell function through Rho-dependent pathways (1). One or both of these G proteins cause cytoskeletal rearrangements, Na+/H+ exchanger activation, cell transformation, stimulation of phospholipase D activity, and apoptosis in a Rho-dependent manner (2). Mouse fibroblasts deficient in Gα13 fail to display thrombin-stimulated cell migration, a phenomenon thought to involve activation of Rho (3).

Activation of Rho can occur by stimulation of guanosine triphosphate (GTP) binding or by inhibition of its intrinsic GTPase activity (4). The S19N variant of Rho, in which Ser19(S19) is changed to Asn (N), is thought to inhibit endogenous Rho function by forming nonproductive complexes with guanine nucleotide exchange factors (GEFs) for Rho (5). Thus, attenuation of heterotrimeric G protein–mediated pathways by expression of S19N Rho suggests that stimulation of GTP binding is a prominent mechanism for activation of Rho by relevant stimuli. Potential mediators of such stimulation are GEFs with specificity for Rho, including Dbl, Lbc, Lfc, Lsc, and p115 RhoGEF (6, 7). The NH2-terminal region of p115 RhoGEF contains a domain characteristic of RGS (regulator of G protein signaling) proteins (8, 9) and specifically stimulates the intrinsic GTPase activity of the α subunits of G13 and G12 (8). We therefore tested whether these two G proteins might alter the function of the exchange factor, thereby providing a mechanism for regulation of Rho by heterotrimeric G proteins.

Physical interaction between Gα13 and p115 RhoGEF was detected by immunoprecipitation (Fig. 1) (10, 11). Antibody to Myc (anti-Myc) immunoprecipitated Gα13 from a lysate of transfected COS cells expressing Myc-tagged p115 RhoGEF. This interaction was observed in the presence of AlF4 , an activator of Gα13. The conformation of guanosine diphosphate (GDP)–bound, AlF4 -activated G protein α subunits resembles that of the transition state for GTP hydrolysis, and RGS proteins bind with highest affinity to Gα proteins in this conformation (12). Gα13 did not coprecipitate with a truncated form of p115 RhoGEF that lacks the NH2-terminal RGS domain [required for GTPase-activating protein (GAP) activity], but it did bind to p115 RhoGEF lacking the Dbl homology (DH) domain, which is required for GEF activity. Similar interactions were observed when antibodies to Gα13 were used to immunoprecipitate proteins from cells expressing Myc-tagged p115 RhoGEF (Fig. 1C). Weaker interactions were detected between Gα12 and Myc-tagged p115 RhoGEF, but other G protein α subunits (Gαs, Gαi, Gαq, and Gαz) were not detected in the anti-Myc immunoprecipitates, even though they were present in the lysates (13). These interactions are consistent with the specificity of p115 RhoGEF to stimulate the GTPase activity of only Gα13 and Gα12 (8).

Figure 1

Binding of Gα13 to p115 RhoGEF. (A) Expression of p115 RhoGEF in COS cells. Myc-epitope–tagged versions (10) of full-length p115 RhoGEF (FL115), p115 RhoGEF with a deletion of residues 466 to 547 in the Dbl homology region (FL115-ΔDH), and p115 RhoGEF with a deletion of residues 1 to 249 (ΔN-115) were expressed in COS cells, and lysates were prepared in the presence or absence of AMF. Portions (20 μl) of lysates were separated by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) (8% gels) and immunoblotted with a monoclonal antibody to the Myc epitope (BAbCO). Molecular weight markers are shown at left. (B) Coimmunoprecipitation of p115 RhoGEF and Gα13 with anti–Myc epitope. Lysates (1 ml) were prepared as described (A), and proteins were immunoprecipitated with anti–Myc epitope. Immunoprecipitates were washed, separated by SDS-PAGE, and immunoblotted with the Gα13 antiserum, B860 (11). Left lane, lysate (20 μl) from control COS. (C) Coimmunoprecipitation with Gα13antiserum. Lysates (1 ml) of COS cells transfected with FL115 or ΔN-115 as in (A) were prepared in the presence of AMF, and proteins were immunoprecipitated with vehicle or antiserum B860 to Gα13 as indicated. Proteins were separated by SDS-PAGE and immunoblotted with either B860 antiserum (anti-Gα13) or anti–Myc epitope. Portions (20 μl) of lysates were included for reference; pAb, polyclonal antibody. Molecular weight markers are shown at right.

Association of purified Gα13 with recombinant p115 RhoGEF stimulated the capacity of the latter protein to facilitate dissociation of GDP from Rho (Fig. 2, A and B) (14, 15). In contrast, several other G protein α subunits, including Gα12, did not influence the nucleotide exchange activity of p115 RhoGEF (Fig. 2, A and C). Stimulation of p115 RhoGEF activity required activation of Gα13 (Fig. 2D). Thus, treatment of Gα13with AlF4 [AMF (15)] or binding of GTP-γ-S to Gα13 supported p115-mediated dissociation of GDP from Rho, but binding of GDP-β-S to Gα13 did not. Thus, stimulation of p115 RhoGEF exchange activity by Gα13 presumably depends on receptor-mediated activation of the G protein.

Figure 2

Stimulation of Rho-dependent nucleotide exchange activity of p115 RhoGEF by Gα13. (A) Dissociation of bound GDP from 100 nM RhoA (14, 15) was examined after 10 min at 30°C in the absence or presence of 100 nM AMF-activated Gα13 or Gα12 and various concentrations of p115 RhoGEF, as indicated. (B) Dissociation of GDP from 100 nM RhoA after 10 min at 30°C in the presence of 25 nM p115 RhoGEF and the indicated concentrations of AMF-activated Gα13 or Gα12. The dashed lines indicate dissociation of GDP from RhoA alone or in the presence of p115 RhoGEF but no Gα. (C) Dissociation of GDP from 100 nM RhoA after 10 min of incubation at 30°C with p115 (25 nM) and various Gα subunits (100 nM), as indicated. (D) Dissociation of GDP from 100 nM RhoA after 10 min of incubation at 30°C with p115 and Gα13 that had been treated as indicated with AMF, GTP-γ-S, or GDP-β-S.

The NH2-terminal RGS domain of p115 RhoGEF is apparently required for binding Gα13, and this domain might inhibit nucleotide exchange activity—a restraint that could be relieved by interaction with Gα13. Consistent with this hypothesis, truncated p115 RhoGEF lacking the RGS domain was more active than the full-length protein (Fig. 3A). Furthermore, addition of the isolated RGS domain [as a glutathioneS-transferase (GST) fusion protein] caused partial inhibition of Gα13-stimulated p115 RhoGEF activity (Fig. 3B), presumably by sequestration of the α subunit. Although these data do not preclude additional sites of interaction of Gα13 with p115 RhoGEF, they are consistent with a primary mode of action through the RGS domain.

Figure 3

Effects of domains of p115 and activated Gα12 on the nucleotide exchange activity of p115 RhoGEF. (A) Binding of 1 nM [33P]GTP to 100 nM RhoA in the presence of the indicated concentrations of truncated or full-length p115 RhoGEF was measured by filtration of a 50-μl reaction after 30 min at 30°C (7). (B) Dissociation of [3H]GDP from 100 nM RhoA was measured for 10 min at 30°C in the absence or presence of 25 nM p115, 20 nM AMF-activated Gα13, and 300 nM RGS-p115, as indicated. RGS-p115, a fusion protein composed of GST ligated to the NH2-terminus of residues 1 to 246 of p115 RhoGEF, was prepared as described (8). (C) Dissociation of [3H]GDP from 100 nM RhoA was measured for 10 min at 30°C with 25 nM p115 RhoGEF and in the absence or presence of 25 nM AMF-activated Gα13 and the indicated concentrations of AMF-activated Gα12. Assays for (B) and (C) were performed as described (15).

The inability of Gα12 to stimulate nucleotide exchange on Rho is puzzling, particularly because p115 RhoGEF stimulates the GTPase activity of both Gα12 and Gα13, and the AMF-stimulated forms of both α subunits are equipotent in attenuating this action (8). Conversely, Gα12 blocked stimulation of p115 RhoGEF activity by Gα13 (Fig. 3C). Thus, Gα12 may compete with Gα13 for binding to the RGS domain of p115 RhoGEF. However, this binding of Gα12 to p115 RhoGEF is clearly not sufficient to stimulate Rho exchange activity. This ineffectiveness of Gα12 may relate to the lower efficacy of p115 RhoGEF to activate the GTPase activity of Gα12 compared with Gα13 or may indicate an additional site of interaction between Gα13 and p115 RhoGEF. Differential regulation of p115 RhoGEF by Gα13 and Gα12 indicates that signals transduced through Gα13 might be modulated by Gα12. The inability of Gα12 to stimulate p115 RhoGEF raises the possibility that this G protein activates Rho through another RhoGEF.

The p115 RhoGEF protein appears to serve as a direct link between Rho GTPases and heterotrimeric G proteins. A correlate inDrosophila is DRhoGEF2, which transduces developmental signals mediated by the G protein Cta, a homolog of Gα12and Gα13 (8, 16). Furthermore, p115 RhoGEF is the first protein with an RGS domain to function as an effector of G protein action. Other effectors for G proteins can also serve as GAPs for their activators and thus stimulate their deactivation (17,18). This coupling of effector and deactivator may provide mechanisms for precise control by permitting rapid attenuation of signaling upon removal of the stimulus. Furthermore, Biddlecomeet al. (18) have suggested that rapid deactivation of Gαq by phospholipase Cβ provides the opportunity for receptor-mediated reactivation of Gα while still present in a complex of G protein and effector. Thus, rapid GTPase cycles may permit the continued existence of receptor–G protein–effector complexes capable of generating higher amplitude signals because the signaling complex does not undergo dissociation and reformation.

The Gα13-Rho pathway participates in phenomena such as cell migration, angiogenesis, and apoptosis (2, 3). The identification of p115 RhoGEF as a critical link in this pathway will facilitate mechanistic understanding of these functions. Furthermore, RGS domains in other proteins may also impart sensitivity to regulation by G protein α subunits.

  • * These authors contributed equally to the work.

  • To whom correspondence should be addressed. E-mail: sternwei{at} (P.C.S.) and bollag{at}


Stay Connected to Science

Navigate This Article