Figure 4. Membrane targeting strategies employed by multi-domain RGS proteins.
(C) The AtRGS1 protein of Arabidopsis thaliana (thale cress) has a unique structure for an RGS protein: an N-terminus resembling a 7TM receptor and a C-terminal RGS-box [64]. Although a ligand is not known for the 7TM portion of AtRGS1, a simple sugar is most likely [66]. (D) The transmembrane receptor Plexin-B1 couples binding of the membrane-bound semaphorin Sema4D to RhoA activation via an interaction with the PDZ domain of PDZ-RhoGEF (and of the related RGS-RhoGEF LARG) [88]. Domain abbreviations [35]: IPT, immunoglobulin-like fold found in plexins, Met and Ron tyrosine kinase receptors, and intracellular transcription factors; PSI, domain found in plexins, semaphorins, and integrins; Sema, semaphorin domain.

           Our recent discovery of AtRGS1, the first plant RGS protein (from Arabidopsis thaliana), has given the clearest demonstration yet of functional linkage between an RGS-box and a particular 7TM receptor. Indeed, AtRGS1 is an amalgam of the two, with an N-terminal region predicted to have the topology and transmembrane domains of a 7TM receptor, along with a C-terminal intracytosolic RGS-box (Fig. 4C). Genetic evidence is consistent with a model that the action of the AtRGS1 RGS-box opposes that of the activated plant G-alpha (AtGPA1) in increasing cell elongation in hypocotyls in darkness and increasing cell production in roots grown in light [64]. In support of this model, we have shown AtRGS1 to be a potent GAP for AtGPA1 [64, 65]. The presence of both GEF-like (7TM) and GAP-like (RGS-box) domains within AtRGS1 may seem paradoxical; however, our prevailing hypothesis is that AtRGS1 represents a ligand-operated GAP, given that AtGPA1 exhibits a high rate of spontaneous nucleotide exchange, as well as slow intrinsic GTPase activity, in comparison to mammalian G-alpha subunits [65]. A definitive test of this "ligand-operated GAP" hypothesis awaits the identification of an agonist for the 7TM portion of AtRGS1. Although no ligand has yet been identified, a simple sugar appears the most likely candidate [66].

2.c. RGS-box-containing RhoGEFs as G-alpha effectors

            Most RGS proteins are considered negative regulators of 7TM receptor signaling, either via Galpha-directed GAP activity or by "effector antagonism" (i.e., binding activated Galpha·GTP in competition with effectors; e.g., [67, 68, 69, 70]). In contrast, the three members of the F/GEF subfamily of RGS proteins (Figs. 2&3), p115-RhoGEF, PDZ-RhoGEF, and leukemia-associated RhoGEF (LARG), represent positive regulators of 7TM receptor signaling – specifically, as true effectors that couple Galpha-q, Galpha-12, and/or Galpha-13 subunits to activation of the GTPase RhoA. Thus, these RGS-box effectors link Gq-, G12-, and G13-coupled 7TM receptors to the panoply of cytoskeletal and transcriptional responses mediated by RhoA·GTP-dependent effectors [71, 72]. All three proteins possess an RGS-box N-terminal to DH (Dbl-homology) and PH (pleckstrin-homology) domains that, as a tandem, are responsible for catalyzing guanine nucleotide exchange necessary to convert inactive RhoA·GDP into active RhoA·GTP [73]. Kozasa et al. first demonstrated in 1998 that the RGS-box of p115-RhoGEF is a potent GAP for both Galpha12 and Galpha13 subunits [74]; more importantly, the same group found that interaction between the RGS-box and Galpha13·GTP, but not Galpha12·GTP, serves to trigger exchange activity by the C-terminal DH/PH cassette [75]. In contrast with p115-RhoGEF, LARG serves as a Galpha-responsive RhoGEF not only for Galpha-13, but also for Galpha-12 and Galpha-q [76, 77]. The ability of LARG GEF activity to be stimulated by Galpha12·GTP appears dependent on tyrosine phosphorylation of LARG ([78]), ostensibly by Tec-family kinases or focal adhesion kinase (FAK; [79]).

            Recent work in PC-3 prostate cancer cells by Wang et al. suggests that these three RGS-RhoGEFs each couple distinct receptors to RhoA activation. RNAi-mediated knockdown of LARG specifically inhibited thrombin signaling via the 7TM protease activated receptor-1 (PAR1), whereas RNAi knockdown of PDZ-RhoGEF specifically inhibited lysophosphatidic acid receptor signaling; reduction of p115-RhoGEF levels had no effect on either response [80]. Of the three RhoGEFs, two of them possess an N-terminal PDZ domain: LARG and the eponymic PDZ-RhoGEF (Fig. 2). The mechanism whereby RGS-RhoGEF signaling specificity is engendered in PC‑3 cells remains unknown, but might entail direct 7TM receptor tail/PDZ domain interactions. It is known that LARG associates via its PDZ domain with the C-terminal tail of the insulin-like growth factor-1 (IGF-1) receptor, providing functional linkage between extracellular IGF-1 and RhoA-mediated cytoskeletal rearrangements [81, 82]. Both LARG and PDZ-RhoGEF also use their PDZ domains to bind plexin-B1 (Fig. 4D), a transmembrane receptor for the semaphorin Sema4D (a.k.a. "cluster of differentiation antigen 100" or CD100) [83]. Binding of Sema4D to plexin-B1 stimulates RhoA activation via LARG/PDZ-RhoGEF [84, 85, 86, 87, 88]. However, the role of the RGS-box in LARG/PDZ-RhoGEF signaling to RhoA activation via these latter, non-7TM receptors remains ill-defined.