p115 RhoGEF, a GTPase Activating Protein for Gα12 and Gα13

See allHide authors and affiliations

Science  26 Jun 1998:
Vol. 280, Issue 5372, pp. 2109-2111
DOI: 10.1126/science.280.5372.2109


Members of the regulators of G protein signaling (RGS) family stimulate the intrinsic guanosine triphosphatase (GTPase) activity of the α subunits of certain heterotrimeric guanine nucleotide–binding proteins (G proteins). The guanine nucleotide exchange factor (GEF) for Rho, p115 RhoGEF, has an amino-terminal region with similarity to RGS proteins. Recombinant p115 RhoGEF and a fusion protein containing the amino terminus of p115 had specific activity as GTPase activating proteins toward the α subunits of the G proteins G12 and G13, but not toward members of the Gs, Gi, or Gqsubfamilies of Gα proteins. This GEF may act as an intermediary in the regulation of Rho proteins by G13 and G12.

G proteins transduce signals from a large number of cell surface heptahelical receptors to various intracellular effectors, including adenylyl cyclases, phospholipases, and ion channels. Each heterotrimeric G protein is composed of a guanine nucleotide–binding α subunit and a high-affinity dimer of β and γ subunits. Gα subunits are commonly grouped into four subfamilies (Gs, Gi, Gq, and G12) on the basis of their amino acid sequences and function (1). The G12 subfamily has only two members, α12 and α13(2). Gα12 and Gα13 participate in cell transformation and embryonic development, but the signaling pathways that are regulated by these proteins have not been identified (3). However, the small GTPase Rho mediates the formation of actin stress fibers and the assembly of focal adhesion complexes induced by the expression of constitutively active forms of Gα12 or Gα13 (4).

Members of the RGS family of proteins negatively regulate G protein signaling (5). The family includes at least 19 members in mammals and is defined by a core domain called the RGS box. Several RGS proteins act as GTPase activating proteins (GAPs) for α subunits in the Gi or Gqsubfamilies (6, 7). The crystal structure of a complex between RGS4 and AlF4 -activated Gαi1 revealed that the functional core of RGS4 (the RGS box), which is sufficient for GAP activity (8), contains nine α helices that fold into two small subdomains (9). The residues of the box that form its hydrophobic core are conserved, and they are important for the stability of structure and GAP activity (9, 10). RGS4 stimulates the GTPase activity of Gαi1 predominantly by interacting with its three mobile switch regions, thereby stabilizing the transition state for GTP hydrolysis (9, 11).

The activities of members of the Rho family of monomeric GTPases are regulated by guanine nucleotide exchange factors (GEFs) that contain a dbl homology (DH) domain (12). Examination of the sequence of p115 (13), a GEF specific for Rho, reveals an NH2-terminal region with similarity to the conserved domain of RGS proteins (Fig. 1). Most of the hydrophobic residues that form the core of this domain (17 of 23) are conserved in p115 RhoGEF. The positions of breaks in the alignment correspond to the loops between α helices in the RGS domain structure. This suggests that p115 RhoGEF may have a similar structural domain and GAP activity. However, the residues of RGS4 that make contact with the switch regions of Gαi1-GDP-AlF4 (GDP, guanosine diphosphate) are not well conserved in p115 RhoGEF, suggesting differences in the mechanism of action or specificity of p115 RhoGEF relative to other RGS proteins.

Figure 1

Sequence alignment of the NH2-terminal region of p115 RhoGEF with selected RGS proteins. Sequences were aligned with the program Clustal W and secondary structure prediction, based on the structure of RGS4, to assign penalties for gaps. The sequences of Lsc, KIAA380, and DRhoGEF2 were added to this alignment with Clustal W and manual adjustments. The thick lines above the RGS4 sequence indicate the positions of α helices in RGS4 (9). Dark shaded boxes indicate conserved residues of the hydrophobic core of the RGS box. Lightly shaded boxes show other conserved residues. Asterisks mark the residues of RGS4 that contact Gαi1. Primary sequences used in the alignment: rat RGS4 (accession number P49799), mouse RGS2 (O08849), human GAIP (P49795), rat RGS12 (O08774), rat RGS14 (O08773), human p115 RhoGEF (U64105), mouse Lsc (U58203), human KIAA380 (AB002378), andDrosophila DRhoGEF2 (AF032870). Abbreviations for the amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.

The capacity of Gα12 and Gα13 to activate Rho in vivo suggested a relationship with p115 RhoGEF, and a physical interaction between p115 and Gα13 was detected (14). Moreover, p115 RhoGEF stimulated hydrolysis of [γ-32P]GTP bound to either Gα13 or Gα12 (Fig. 2) (15). At 15°C, 10 nM p115 RhoGEF increased thek cat's for hydrolysis of GTP by Gα12 (0.07 min–1) and Gα13(0.24 min–1) by factors of 5 and 10, respectively (Fig. 2A). Similar results were obtained with several preparations of Gα12 and Gα13, and treatment of p115 RhoGEF at 90°C for 5 min inactivated its GAP activity. Because of rapid hydrolysis of GTP by Gα13, assays were performed at 4°C to estimate more accurately the effect of p115 RhoGEF on the rate of GTP hydrolysis by the G protein (Fig. 2B). Under these conditions, 100 nM p115 RhoGEF increased the GTPase activities of Gα13 and Gα12 by factors of 80 and 6, respectively. Thus, although stimulation of the GTPase activity of both proteins was observed at concentrations of p115 RhoGEF as low as 1 nM, p115 RhoGEF was a substantially more efficacious GAP for Gα13 than for Gα12.

Figure 2

The p115 RhoGEF protein stimulates GTPase activity of Gα13 and Gα12. (A) Hydrolysis of GTP bound to Gα13 and Gα12 at 15°C either with (circles) or without (squares) 10 nM p115 RhoGEF (15). (B) Hydrolysis of GTP bound to Gα13 (•) or Gα12 (○) was measured at 4°C in the presence of the indicated concentrations of p115 RhoGEF. The initial rates of the reactions are plotted as a function of the concentration of p115 RhoGEF.

In the absence of a receptor, the rate-limiting step for binding of the nonhydrolyzable GTP analog guanosine 5′-O-(3′-thiotriphosphate) (GTP-γ-S) to Gα proteins and for their steady-state hydrolysis of GTP is the release of GDP. The p115 RhoGEF protein did not affect the rate of GTP-γ-S binding to Gα12 or Gα13, nor did it affect the steady-state GTPase activity of either subunit (16). Therefore, p115 RhoGEF stimulates only the intrinsic GTPase activity of Gα12 and Gα13 without affecting their rates of nucleotide exchange.

The conserved RGS box of RGS proteins is sufficient for GAP activity in vitro (8). We thus tested a fusion protein (17) containing glutathione S-transferase (GST) and the NH2-terminal region of p115 RhoGEF for GAP activity. The fusion protein, which contains the RGS similarity region (Fig. 1) but not the DH or PH domains of p115 RhoGEF, was almost as active as full-length p115 (Fig. 3). In contrast, p115 RhoGEF truncated at the NH2-terminus to eliminate the RGS box was ineffective. Thus, the RGS homology region mediated the GAP activity of p115 RhoGEF.

Figure 3

Stimulation of the GTPase activity of Gα13 and Gα12 by the NH2-terminal region of p115 RhoGEF. Hydrolysis of GTP bound to Gα13 and Gα12 was measured at 15°C without (▾) or with (•) 10 nM p115 RhoGEF, 10 nM RGS-p115 (▴), or 50 nM ΔN-p115 RhoGEF (▪).

The p115 RhoGEF protein (100 nM) did not stimulate the GTPase activity of Gαi1, Gαz, or Gαq under conditions where RGS4 did act as a GAP for these Gα subunits (16). Similarly, p115 RhoGEF did not accelerate the GTPase activity of Gαs, RhoA, or Rac1 (16). Thus, p115 RhoGEF is a GAP with specificity for Gα12 and Gα13. However, it is possible that this specificity could be different in the presence of an activated receptor. The GAP activity of RGS2 toward Gαi1 was observed only after reconstitution of the proteins into phospholipid vesicles containing M2 muscarinic cholinergic receptors (18).

RGS proteins have relatively high affinity for the GDP-AlF4 –bound forms of G protein α subunits, whose conformation is similar to that of the transition state for GTP hydrolysis (11). Therefore, the GDP-AlF4 –bound forms of appropriate Gα proteins should compete with Gα-GTP for interaction with p115 RhoGEF. GDP-AlF4 –bound Gα12 and Gα13 inhibited GAP activity of p115 RhoGEF on Gα12, whereas similar forms of Gαs, Gαi1, and Gαq did not (Fig. 4A); these results are indicative of the selectivity of p115 RhoGEF. Furthermore, the GDP-AlF4 –bound forms of Gα12and Gα13 are equipotent inhibitors of the GAP activity of p115 RhoGEF toward Gα13 (Fig. 4B).

Figure 4

Selective inhibition of p115 RhoGEF GAP activity by AlF4 -activated Gα subunits. (A) p115 RhoGEF (400 nM) was incubated on ice for 15 min with various Gα subunits (400 nM) in the presence of AMF (30 μM AlCl3, 10 mM NaF, and 10 mM MgSO4). The mixtures were diluted with 19 volumes of buffer A (15) and mixed with 0.3 nM GTP-Gα12. The hydrolysis of bound GTP was measured after incubation at 15°C for 2 min in the presence of AMF. (B) p115 RhoGEF (400 nM) was incubated with various concentrations of GDP-AlF4 -Gα12(○) or GDP-AlF4 -Gα13 (□) as in (A). The mixtures were diluted with 19 volumes of buffer A and mixed with 1 nM GTP-Gα13; the hydrolysis of bound GTP was assessed over time at 4°C in the presence of AMF. The initial rate of GTP hydrolysis by Gα13 was plotted against the final concentration of GDP-AlF4 –bound α subunit (▵ indicates the rate of GTP hydrolysis by Gα13 in the absence of p115 RhoGEF).

Although we assume that the structure of the GAP domain of p115 RhoGEF is similar to that of the RGS box of RGS4 and that it interacts with the switches of Gα12 and Gα13, the amino acid sequences of these switch regions differ between the G12 and Gi subfamilies (1, 2). Six of the seven residues in RGS4 that interact with Thr182 of Gαi1, a residue of switch I that is in the center of the RGS4-Gαi1 interface (9), are not conserved in p115 RhoGEF. Thus, the surface between p115 and Gα12 or Gα13 is likely quite different from that observed with RGS4 and Gαi1, and this amino acid diversity must contribute to the different specificity of the GAP activity of p115 RhoGEF relative to other members of the RGS family.

Three other RhoGEF proteins (Lsc, KIAA380, and DRhoGEF2) contain regions of similarity to the RGS domain of p115 RhoGEF (Fig. 1). Lsc appears to be the mouse homolog of p115 RhoGEF, and KIAA380 is likely the human homolog of Drosophila DRhoGEF2 (19,20). The latter provides a biological correlate for the biochemical relationships defined for Gα13 and p115 RhoGEF. DRhoGEF2 is a mediator of a signal critical for gastrulation (20), and genetic evidence suggests that DRhoGEF2 functions downstream of a G protein α subunit, concertina (Cta), that is most similar to the mammalian Gα12 and Gα13(21). The four RhoGEFs may define a new subset of RGS proteins that not only have GAP activities but also couple RhoGEF activity to G protein α subunits. This coupling was observed in the stimulation of the exchange activity of p115 RhoGEF by Ga13(14).

  • * To whom correspondence should be addressed. E-mail: tohru.kozasa{at} (T.K.), bollag{at} (G.B.), and sternwei{at} (P.C.S.).


View Abstract

Navigate This Article