PerspectiveProtein Interactions

RGS Proteins: Swiss Army Knives in Seven-Transmembrane Domain Receptor Signaling Networks

See allHide authors and affiliations

Science's STKE  23 Jan 2007:
Vol. 2007, Issue 370, pp. pe2
DOI: 10.1126/stke.3702007pe2


Coordinated regulation of heterotrimeric guanine nucleotide–binding protein (G protein) activity is critical for the integration of information from multiple intracellular signaling networks. The human regulator of G protein signaling (RGS) protein family contains more than 35 members that are well suited for this purpose. Although all RGS proteins contain a core ~120–amino acid Gα-interacting domain (called the RGS domain), they differ widely in size and organization of other functional domains. Architecturally complex RGS proteins contain multiple modular protein-protein interaction domains that mediate their interaction with diverse signaling effectors. Architecturally simple RGS proteins contain small amino-terminal domains; however, they show surprising versatility in the number of intracellular partners with which they interact. This Perspective focuses on RGS2, a simple RGS protein with the potential to integrate multiple signaling networks. In three recent studies, the amino-terminal domain of RGS2 was shown to interact with and regulate three different effector proteins: adenylyl cyclase, tubulin, and the cation channel TRPV6. To explain this growing list of RGS2-interacting partners, we propose two models: (i) The amino-terminal domain of RGS2 comprises several short effector protein interaction motifs; (ii) the amino-terminal domain of RGS2 adopts distinct structures to bind various targets. Whatever the precise mechanism controlling its target interactions, these studies suggest that RGS2 is a key point of integration for multiple intracellular signaling pathways, and they highlight the role of RGS proteins as dynamic, multifunctional signaling centers that coordinate a diverse range of cellular functions.

Signal transduction by seven-transmembrane domain receptors (7TMRs) controls a vast array of physiological processes in humans and other eukaryotic organisms. Biological responses triggered by 7TMRs are elicited by the action of complex intracellular networks of signaling proteins. Regulatory proteins that modulate and integrate these networks are critically important for controlling cellular responses by governing signaling magnitude, kinetics, and fidelity. For example, arrestins bind activated receptors and trigger G protein uncoupling, receptor endocytosis, or extracellular signal–regulated kinase (ERK) activation, thereby regulating such diverse processes as nociception, cell migration, and apoptosis (1).

Since their discovery a decade ago, RGS (regulator of G protein signaling) proteins have emerged as crucial regulators and integrators of 7TMR signaling. RGS proteins possess a ~120–amino acid RGS domain that usually exhibits guanosine triphosphatase (GTPase)–activating protein (GAP) activity toward G protein α subunits (24), allowing them to accelerate 7TMR signaling activation and deactivation rates. Many RGS proteins are not just GAPs, however. Architecturally complex RGS proteins, including the RGS7-like, RGS12-like, RhoGEF-containing, and G protein–coupled receptor kinase (GRK)–like classes, possess several protein-protein interaction domains that transduce signals to downstream effectors, coordinate signaling between intracellular signaling networks, or mediate shuttling between intracellular compartments (5, 6). In contrast, architecturally simple RGS proteins, such as the RGSZ-like and RGS4-like subfamilies, consist of little more than an RGS domain flanked by short (typically 10 to 70 residues) N- and C-terminal extensions, and thus they were thought until recently to function simply as GAPs. However, even such simple RGS proteins now are proving to be remarkably versatile integrators and regulators of 7TMR signaling by interacting with unexpectedly diverse classes of protein partners. We highlight this emerging concept by focusing on three studies of the RGS4-like family member RGS2 (79) that raise intriguing questions about the molecular mechanisms used by RGS2 to recognize a wide variety of targets and the biological roles of these interactions in 7TMR signaling.

In the first of these investigations, Roy, Chidiac, and colleagues describe a novel feedback mechanism whereby RGS2 co-regulates Ca2+ and adenosine 3′,5′-monophosphate (cAMP) signaling (7). The Chidiac team began with the appreciation that (i) RGS2 is a potent GAP for Gαq (10, 11), the G protein that activates phospholipase Cβ and consequently triggers inositol 1,4,5-trisphosphate (IP3) production and Ca2+ release from intracellular stores; (ii) RGS2 lacks GAP activity toward Gαs [reviewed in (12)], the direct activator of adenylyl cyclase; (iii) the N-terminal non-RGS domain of RGS2 can directly inhibit types III, V, and VI of adenylyl cyclase (13, 14); and (iv) RGS2 mRNA is up-regulated as an immediate-early response in osteoblasts (1518) and many other cell types (19). Despite such understanding, the biological and mechanistic functions of RGS2 up-regulation have remained elusive. To address these questions, Chidiac and colleagues examined osteoblasts, where they found that the abundance of endogenously expressed RGS2 protein was strikingly increased in response to activation of effector molecules that function downstream of Gq or Gs. Under conditions that increased RGS2 abundance in osteoblasts, the Chidiac team found that signaling by Gq- or Gs-coupled 7TMRs was markedly attenuated, whereas such inhibitory effects failed to occur in osteoblasts lacking RGS2. In contrast, under control conditions where RGS2 was not increased, wild-type and RGS2-deficient osteoblasts exhibited indistinguishable Gq- and Gs-mediated signaling. Therefore, in osteoblasts, up-regulated rather than basally expressed RGS2 coordinately inhibits or desensitizes Gs and Gq signaling. Although the precise mechanisms remain to be established, up-regulated RGS2 probably inhibits Gq signaling through the GAP activity of its RGS domain and attenuates Gs-mediated signaling through its non-GAP N-terminal domain to inhibit adenylyl cyclase. Because both Gs and Gq signaling regulate both osteoblast differentiation and function, RGS2 knockout mice may exhibit defects in these processes under conditions where RGS2 is normally up-regulated.

In the second study, Schoeber, Hoenderop, and colleagues describe a novel mechanism whereby RGS2 regulates Ca2+ homeostasis independently of its GAP activity toward Gαq (8). The Hoenderop group found that RGS2 bound the epithelial Ca2+ channel TRPV6 in two-hybrid and cell pulldown assays. Overexpressed RGS2 strikingly inhibited the Na+ and Ca2+ currents of TRPV6, but not the closely related transporter TRPV5, by a mechanism requiring the N-terminal region flanking the RGS domain of RGS2. Thus, RGS2 regulation of TRPV6 apparently occurs by a GAP-independent mechanism that regulates channel gating. Analyzing TRPV6 activity in wild-type and RGS2 knockout cells may reveal a role of this mechanism in epithelial cell biology. The question also remains as to whether other RGS proteins regulate TRPV6 or related channels.

In the third paper, Heo, Shu, and colleagues identified tubulin as an RGS2-interacting protein (9). The N-terminal non-GAP domain of RGS2 bound directly to tubulin and promoted microtubule polymerization in vitro. In contrast, the related RGS4-like family member RGS20 (also known as RGSZ1) promotes microtubule assembly indirectly by inhibiting SCG10 (20), a microtubule destabilizing protein. The Shu team also reported that RGS2 overexpression in PC12 cells enhanced nerve growth factor (NGF)–mediated neurite outgrowth by promoting microtubule assembly. Accordingly, they hypothesized that loss of RGS2-promoted microtubule stabilization may contribute to the observed decrease in apical and basilar spines of CA1 hippocampal neurons in RGS2 knockout mice (21), a possibility that warrants further investigation. It is interesting that G proteins and their regulators have been implicated as functionally important modulators of microtubule assembly, function, and spindle pole positioning during cell division (2225). Although many details of this system remain to be worked out, RGS2 apparently provides another means by which G protein signaling pathways and microtubules are functionally linked.

What new mechanisms of 7TMR signaling regulation are suggested by these investigations of RGS2? We suggest two models—which are not mutually exclusive—to explain how the N-terminal domain (NTD) of RGS2 regulates a diverse collection of target proteins. In one model, we suggest that the NTD contains several peptide sequences that provide membrane-targeting activity and regulatory ligands for various signaling targets. Indeed, a short peptide within the NTD of RGS2 is sufficient to inhibit type V adenylyl cyclase (14) and does not overlap with the plasma membrane–targeting signal in the NTD located more proximally to the RGS domain (26). Furthermore, the NTD of RGS2 binds several other non–G protein targets, including the third intracellular loops of certain 7TMRs (2729), type Iα guanosine 3′,5′-monophosphate (cGMP)–dependent protein kinase (30), and spinophilin (31), a scaffolding protein. Thus, distinct peptides within the NTD potentially direct the interaction of RGS2 with membranes and various signaling partners. If such interactions occur noncompetitively, then the NTD of RGS2 could interact simultaneously with several signaling proteins to coordinately regulate the transport, compartmentalization, or activity of proteins involved 7TMR signaling or other processes. Indeed, results of bioluminescence resonance energy transfer (BRET) experiments indicate that RGS2, when overexpressed, can associate in signaling complexes containing a 7TMR (β2-adrenergic receptor), Gs, and type IV or VI adenylyl cyclase (29). In a second model, we propose that the NTD of RGS2 can adopt several conformations that are recognized or stabilized differentially by various targets. In support of this idea, the NTD is disordered in the crystal structure of the related R4 family member RGS4 (32). Furthermore, peptides from the NTDs of RGS2 or RGS4 are disordered in solution but adopt α-helical conformations when bound to lipid bilayers (26, 33). Thus, the NTD of RGS2 and other R4-class RGS proteins may sample conformational space to adopt structures that are stabilized upon binding to membranes or target proteins. Indeed, conformational flexibility has been hypothesized to be a key feature of signal transfer domains of other signaling proteins (34). Either model suggests that RGS2 provides a nodal point for regulating and integrating 7TMR signaling by using its NTD to regulate various classes of signaling partners while keeping its conformationally rigid RGS domain poised for G protein binding and regulation (Fig. 1).

Fig. 1.

RGS2, a master integrator. RGS2 has a structured RGS domain and an unstructured N-terminal domain (NTD), which can adopt an α-helical conformation when bound to lipids. Through the RGS domain, RGS2 enhances the GTPase activity of Gα subunits. Through the NTD, RGS2 influences the activity of channels, scaffolds, and kinases, and stabilizes microtubules. cGKIα, cGMP-dependent protein kinase α.

It is remarkable how far our understanding of RGS proteins has evolved. Appreciated a decade ago simply as GAPs for Gα subunits, RGS proteins now are recognized as dynamic, multifunctional signaling centers that coordinate a surprisingly diverse range of molecular and cellular functions. What lies ahead? Perhaps, as exemplified by the NTD of RGS2, RGS domains are more versatile than currently appreciated. RGS domains may possess several functionally and structurally distinct protein-protein interaction surfaces, one that confers GAP activity and others that interact with novel signaling proteins. This hypothesis is suggested by structural studies of the RGS domains of axin and GRK2 (3537), which use surfaces distinct from those that mediate GAP activity in other RGS proteins to bind the tumor suppressor APC and Gαq, respectively. Other major challenges include determining which regulatory mechanisms indicated by biochemical or overexpression studies are biologically important at physiological concentrations of RGS proteins, establishing how these mechanisms are controlled, and investigating their roles in human disease and as targets for therapy (38). Answers to such questions promise to reveal that even the simplest RGS proteins will continue to surprise us with their versatility, extending the reach of 7TMR signaling networks beyond currently established paradigms to mediate intracellular communication with a host of targets, organelles, and compartments (39, 40).


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
View Abstract

Stay Connected to Science Signaling

Navigate This Article