Research CommentariesSignal Transduction

G Proteins and Small GTPases: Distant Relatives Keep in Touch

See allHide authors and affiliations

Science  26 Jun 1998:
Vol. 280, Issue 5372, pp. 2074-2075
DOI: 10.1126/science.280.5372.2074

Cells use all sorts of tricks to make the signal transduction pathways that tailor the cells' physiology to the changing environment. One feature used repeatedly is the protein switch, flicked on and off by the nucleotide guanosine 5′-triphosphate (GTP). When GTP is bound, two families of proteins—heterotrimeric guanine nucleotide-binding proteins (G proteins) and their distant relatives, the small molecular weight guanosine triphosphatases (GTPases)—are “on” and can activate the element immediately downstream to send a signal further down the line. But each of these proteins is also a GTPase, containing within the molecule itself the ability to hydrolyze GTP to guanosine diphosphate (GDP) and so turn off the switch.

Small GTPases control fundamental cell properties—polarity, shape, and the commitment to divide or differentiate. The larger G proteins usually regulate more specialized signals—the production of second messengers like cyclic AMP and calcium. Two members of the G protein family, G12 and G13, are unusual in that they promote cell cycle progression and reorganization of the actin cytoskeleton, changes that are typically associated with the small GTPases. Now an impressive piece of detective work, described on pages 2109 and 2112 of this issue, unites the two distantly related families through these unique G proteins. Kozasa et al. and Hart et al. show that G13 directly activates a guanine nucleotide exchange factor, which in turn promotes GDP dissociation from the small GTPase Rho, allowing it to be activated again by GTP (1, 2). At least in this instance, a G protein triggers action in its distant cousin, the small GTPase Rho.

Small, monomeric GTPases of the Rho-Rac family control the assembly of filamentous actin structures in response to signals from outside the cell (3). Rho, the founder member of this family, interacts with effector (downstream) proteins to cause the assembly of contractile actin:myosin filaments. Although the most clearly visible of these filaments are the stress fibers seen in fibroblasts adhering to a surface, actin:myosin structures actually play a fundamental role in all cell types. Consequently, Rho controls such diverse processes as smooth muscle contraction, growth cone collapse, embryonic wound healing, and cell shape changes during morphogenesis (4, 5). Rho is activated by some members of a large family of guanine nucleotide exchange factors (RhoGEFs), each of which has a Dbl homology (DH) domain followed immediately by a pleckstrin homology (PH) domain (6). In addition, RhoGEFs have a variety of other motifs and domains unique to each member—one of these, an RGS (regulators of G protein signaling) domain in Lsc/p115RhoGEF, a GEF specific for Rho, is identified in the two reports in this issue.

The RGS domain was first detected in a yeast protein, Sst2p, which stimulates the intrinsic GTPase activity of the single G protein present in Saccharomyces cerevisiae (7). Since then a family of mammalian RGS-containing proteins (with over 19 members) has been identified and, as predicted from the yeast results, most stimulate the GTPase activity of mammalian G proteins. The RGS sequence therefore defines a family of GTPase-activating proteins (GAPs) capable of down-regulating heterotrimeric G proteins. Kozasa et al., using a database search, observed an RGS domain in mammalian Lsc/p115RhoGEF and in Drosophila DRhoGEF2 (1). By screening various G proteins, they found unexpectedly that the RGS in Lsc/p115RhoGEF interacts specifically with the GTP-bound α subunits of G12 and G13 and that it acts as an activating protein (a GAP) for both GTPases.

In 1992, phospholipase C β was reported to be both a specific GAP and an effector for the heterotrimeric G protein, Gq, raising the possibility that GAPs might in general also be targets of G proteins (8). Now it seems that in addition to acting as GAPs for G proteins, some RGS-containing proteins might also be effectors. Hart et al. therefore examined whether the RGS-containing Lsc/p115RhoGEF could be a target of G12 or G13 (2). Indeed, the ability of Lsc/p115RhoGEF to stimulate GDP/GTP exchange on Rho is significantly greater in the presence of the GTP-bound α subunit of G13, but not G12, demonstrating that Lsc/p115RhoGEF is both a GAP and a target for G13 (see the figure).

Family reunion.

Interaction of Lsc/p115RhoGEF with Gα13-GTP, but not Gα12-GTP, stimulates its ability to catalyze guanine nucleotide exchange on Rho, thereby providing a direct biochemical link between the heterotrimeric G protein and the small GTPase. The RGS domain of Lsc/p115RhoGEF functions as a GAP toward both Gα12 and Gα13, but its preferred substrate is Gα13. Extracellular ligands activate Rho-lysophosphatidic acid (LPA), sphingosine-1-phosphate (S-1-P), bombesin, thrombin, and the chemotactic agents formyl-methionyl-leucyl-phenylalanine (fMLP) and interleukin-8 (IL-8). All act through heptahelical receptors and therefore activate G proteins. Once activated, Rho-GTP interacts with effectors leading to the assembly of contractile actin:myosin filaments and integrin-containing focal adhesion complexes. It may also control other cellular activities such as the transcription factors SRF and NF-κB, the JNK MAP kinase pathway, phospholipase D, and the sodium/proton exchanger.

The new results point to the Rho exchange factor as a target for G13. They also offer a mechanism for the observation that both G12 and G13 can induce Rho-dependent stress fiber formation. Because Rho appears to be activated exclusively by ligands that act through heptahelical receptors (see the figure), it seems likely that these receptors activate Rho through G12 or G13 (6, 9). Kozasa et al. and Hart et al. illuminate how this might work: Activated G13 interacts with and stimulates the catalytic activity of Lsc/p115RhoGEF. The new work also meshes well with the genetic analysis of gastrulation in Drosophila, which is driven by an extracellular ligand, fog, that activates a G protein, concertina. Two other components of this pathway are a Drosophila GEF (DRhoGEF2) and Drosophila Rho. Fog-mediated activation of Rho leads to an actin:myosin-dependent constriction at the apical surface of epithelial cells to drive this morphogenetic process (5). It now seems likely that concertina, which belongs to the G12/13 family, interacts directly with DRhoGEF2 through an RGS domain.

Is this the only way Rho can be activated? Almost certainly not; other GEFs for Rho (for example, Lbc) lack an RGS domain, and it is still unclear how G12 activates Rho. Even for Lsc/p115RhoGEF, there is likely more to the story. First, the PH domain is essential for full activity of many GEFs, although it is not known why. Some PH domains interact with phosphoinositides, but so far the only lipids implicated in Rho activation are derivatives of arachidonic acid (10, 11). Interestingly, however, activation of the yeast RhoGEF, ROM2, is mediated by TOR2, a phosphatidylinositol kinase-related protein (12). Second, G13-but not G12-induced activation of Rho is inhibited by tyrosine kinase inhibitors and so perhaps phosphorylation of Lsc/p115RhoGEF is required for exchange activity, as has been reported for another RhoGEF, Vav (13). Alternatively, phosphorylation of another protein might be required to initiate Rho signaling. Whatever the explanation, the analysis of gastrulation in Drosophila provides further support for an additional signal contributing to exchange factor activation, because deletion of DRhoGEF2 produces a more severe phenotype than deletion of concertina (5).

Kozasa et al. and Hart et al. have identified the first target for the G12/13 family of G proteins and in so doing provide a biochemical link between heptahelical receptors and activation of the small GTPase Rho. Actin:myosin filament assembly underlies many fundamental biological processes, and this work is an important step in understanding its control.


View Abstract

Stay Connected to Science

Navigate This Article