Type 5 G Protein β Subunit (Gβ5) Controls the Interaction of Regulator of G Protein Signaling 9 (RGS9) with Membrane Anchors*

The R7 family of regulators of G protein signaling (RGS) proteins, comprising RGS6, RGS7, RGS9, and RGS11, regulate neuronal G protein signaling pathways. All members of the R7 RGS form trimeric complexes with the atypical G protein β subunit, Gβ5, and membrane anchor R7BP or R9AP. Association with Gβ5 and membrane anchors has been shown to be critical for maintaining proteolytic stability of the R7 RGS proteins. However, despite its functional importance, the mechanism of how R7 RGS forms complexes with Gβ5 and membrane anchors remains poorly understood. Here, we used protein-protein interaction, co-localization, and protein stability assays to show that association of RGS9 with membrane anchors requires Gβ5. We further establish that the recruitment of R7BP to the complex requires an intact interface between the N-terminal lobe of RGS9 and protein interaction surface of Gβ5. Site-directed mutational analysis reveals that distinct molecular determinants in the interface between Gβ5 and N-terminal Dishevelled, EGL-10, Pleckstrin/DEP Helical Extension (DEP/DHEY) domains are differentially involved in R7BP binding and proteolytic stabilization. On the basis of these findings, we conclude that Gβ5 contributes to the formation of the binding site to the membrane anchors and thus is playing a central role in the assembly of the proteolytically stable trimeric complex and its correct localization in the cell.

Regulators of G protein signaling (RGS) 2 proteins are key components of G protein-coupled receptor (GPCR) signaling pathways (1). They limit the duration that heterotrimeric G proteins spend in their active state and thus serve as powerful negative regulators of GPCR signal transmission. Biochemically, RGS proteins function as GTPase-accelerating proteins that facilitate GTP hydrolysis on G protein ␣ subunits (G␣) thereby promoting their inactivation (2,3). There are more than 30 RGS proteins that are grouped into six families according to the homology and structural features (3,4). Cumulatively, RGS proteins are thought to be ubiquitously expressed and have been demonstrated to regulate a range of the vital physiological processes (5)(6)(7)(8).
The R7 family of RGS proteins comprises four members: RGS6, RGS7, RGS9, and RGS11 and plays prominent roles in the nervous and cardiovascular systems (9,10). The best studied member of this family, RGS9, is involved in regulating diverse GPCR signaling cascades in mammals (11,12). Its short splice isoform, RGS9-1, regulates phototransduction cascade in rod and cone photoreceptors of the retina and is essential for the high temporal resolution of our vision (13)(14)(15). The long splice isoform, RGS9-2, is predominantly expressed in the striatum, a region of the brain involved in movement control and reward behavior (16 -18). RGS9-2 is thought to control G protein signaling downstream from the -opioid and D2 dopamine receptors and has been implicated in drug addiction and movement disorders (19 -23).
Both RGS9 splice isoforms exist as constitutive heterotrimers with two other proteins. In the retina, RGS9-1 is found in the complex with type 5 G protein ␤ subunit (G␤5) (24) and RGS9 anchor protein (R9AP) (25). Similarly, in the striatum, RGS9-2 is complexed with G␤5 and R9AP-like protein called R7-binding protein (R7BP) (26). Both R9AP and R7BP are small SNARE-like membrane proteins. Their binding dictates localization of both RGS9 isoforms in photoreceptors and striatal neurons, respectively (27)(28)(29)(30)(31). In addition, both R7BP/R9AP and G␤5 are critical for regulating the expression level of RGS9. Knock out of G␤5 (32), R7BP (29), or R9AP (33) leads to a dramatic reduction in RGS9 protein levels by increasing its susceptibility to proteolytic degradation. Conversely, overexpression of R7BP (31,34) or R9AP (35) results in an increase in RGS9 levels. Interestingly, reciprocal effects on the stability of RGS9-containing complexes have been also described: elimination of RGS9 in photoreceptors leads to down-regulation in the levels of G␤5 (14), and knock-out of G␤5 compromises the stability of R7BP (29,36). Furthermore, association of RGS9 with the binding partners and consequently its expression level is modulated by oxygen concentration and neuronal activity, suggesting that reorganization in RGS9 complexes could contribute to the neuronal plasticity mechanisms (37). These observations lead to a general view that RGS9 forms an obligatory trimeric complex with its bona fide subunits R9AP/R7BP and G␤5 and that the interactions between the subunits are responsible for setting an appropriate expression level of the complex. However, the mechanisms governing complex assembly and intrasubunit interactions are far from being understood.
The recently solved crystal structure of the RGS9-G␤5 complex reveals tight integration of the G␤5 subunit, which interacts with multiple domains of RGS9 (38). One critical focal contact point is established between the so-called "protein interaction interface" of G␤5 and the N-terminal segment of RGS9 containing the DEP domain. Interestingly, the same DEP domain of the molecule is also indispensable for the recruitment of R7BP and R9AP membrane anchors to the RGS9-G␤5 complex (26,28,30,34). However, the molecular determinants mediating this interaction and mechanisms by which proteolytic stabilization of the trimeric complex is achieved are unknown.
In this study we examined assembly of the RGS9-G␤5-R7BP complex and mechanisms by which interactions between individual subunits regulate its expression and localization. We report an unexpected observation that G␤5 is required for the interaction of RGS9 with its membrane anchors R7BP and R9AP, which was previously thought to be mediated exclusively by the DEP domain of RGS9. We further established that the binding site for the membrane anchors is formed by the interface between the protein interaction surface of G␤5 and the N-terminal lobe of RGS9. Based on our site-directed mutagenesis results we propose a model where R7BP acts to remodel the interface that is essential for the stability of the complexes and their membrane recruitment.
Cell Culture and Transfection-HEK293T/17 cells were chosen because of their high transfectability (41). The cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS), MEM nonessential amino acids, 1 mM sodium pyruvate, and antibiotics (100 units/ml penicillin and 100 g/ml streptomycin) at 37°C in a humidified incubator containing 5% CO 2 . For transfection, cells were seeded into 6-cm dishes at a density of 1 ϫ 10 6 cells/ dish. After 24 h, expression constructs (0.83 g/construct, total 2.5 g/dish) were transfected into the cells using Lipofectamine LTX (8 l/dish) and PLUS (2.5 l/dish) reagents. Empty vector was used to normalize the amount of transfected DNA. The cells were used for experiments at 48 h after transfection.
Selection of Stable Cell Lines-R7BP stable cell lines were generated by transfecting linearized R7BP in pcDNA4/ HisMax-TOPO into HEK293T/17 cells. Cells were selected in 100 g/ml zeocin (Invitrogen). Resistant colonies were expanded, and the expression of R7BP protein was evaluated by Western blotting and immunocytochemistry.
Immunoprecipitation Assay-HEK293T/17 cells in 6-cm plates were transfected with the indicated constructs. Fortyeight hours after transfection, cells were washed once with icecold PBS and lysed with 0.5 ml of lysis buffer (50 mM Tris, pH 7.4, 0.5% n-dodecanoylsucrose, 300 mM NaCl, and Complete Mini Protease Inhibitor Mixture (Roche Applied Science)) by sonication on ice. The resultant whole cell lysates were incubated for 30 min at 4°C with rotary agitation to solubilize membrane proteins. After lysis, cell lysates were centrifuged at 14,000 rcf for 15 min at 4°C. A 5 g/sample of the indicated antibody and 20 l of Dynabeads Protein G (Invitrogen) were added, and the supernatants were tumbled for 1 h at 4°C. After three washes with 1 ml of ice-cold wash buffer (50 mM Tris, pH 7.4, 0.5% n-dodecanoylsucrose, 300 mM NaCl, and 0.5 mM phenylmethanesulfonyl fluoride), proteins bound to the beads were eluted with SDS-sample buffer (50 mM Tris, pH 6.8, 1% SDS, 143 mM ␤-mercaptoethanol, 0.08 mg/ml bromphenol blue, 10% glycerol). Immunoprecipitation from the mouse brain extracts was performed as described previously (42). Immunoprecipitates were subjected to SDS-PAGE, transferred to polyvinylidene difluoride membranes, and probed with the indicated antibodies.
Quantitative Western Blot Analysis-Fluorescent Western blot analysis was performed using the Odyssey Infrared Imaging System according to the manufacturer's instructions (LI-COR Biosciences), which is reported to have a 16 -250-fold wider quantifiable linear range with increased reproducibility than a chemiluminescence Western blot detection system. For detection, blots were blocked with Odyssey Blocking Buffer for 1 h at room temperature followed by a 1.5-h incubation with specific antibodies diluted in Odyssey Blocking Buffer containing 0.1% Tween 20. Blots were washed in PBS containing 0.1% Tween 20 and incubated for 45 min with a 1:5,000 dilution of secondary antibodies labeled with IRDye 800CW in Odyssey buffer containing 0.1% Tween 20 and 0.01% SDS. Proteins were visualized using the Odyssey Infrared Imager, and molecular analysis of band intensities was quantified using the Odyssey Application software by measuring the integrated intensity. The integrated intensity is proportional to the amount of dyelabeled antibodies bound to the blot.
Immunocytochemistry-HEK293T/17 cells were seeded onto poly-L-lysine-coated coverslips. After 24 h, expression constructs (total, 5 g/well) were transfected into the cells using Lipofectamine LTX (l/well) and PLUS (0.5 l/well) reagents. The cells were used for immunocytochemistry at 48 h after transfection. Cells were fixed with 4% paraformaldehyde in PBS for 20 min and permeabilized by incubation in 0.5% Triton X-100 in PBS for 5 min. Cells were blocked with 10% goat serum in PBS containing 0.1% Triton X-100 (0.1% PBS-T) for 1 h and were then incubated with primary antibody in 2% goat serum in 0.1% PBS-T for 90 min. The cells were washed with 0.1% PBS-T, and the antibody was visualized with fluorophore-conjugated secondary antibody in 2% goat serum in 0.1% PBS-T for 45 min. The coverslips were washed three times with 0.1% PBS-T and twice with PBS and mounted on glass slides with Fluoromount (Sigma). Microscopy was performed with an Olympus Fluoview 1000 confocal microscope. The following concentrations of the antibodies were used: 2.5 g/ml anti-HA (12CA5), 1.2 g/ml anti-R7BP, and 1.4 g/ml anti-G␤5 antibodies.

Proteolytic Stability of RGS9-G␤5-R7BP Complex Is Achieved by Mutually Reciprocal Contribution of Individual Subunits-
To begin examining the relationship between subunits in RGS9-G␤5-R7BP we compared their individual contributions toward establishing a proteolytically stable complex. We have previously determined that expression levels of RGS9 in the cells are limited by its proteolytic degradation rate as measured by the pulse-chase method (34,39). Therefore, proteolytic stability of the complex could be approximated by monitoring changes in protein expression levels. To facilitate these studies we chose RGS9-2 as a model isoform due to its relatively short lifetime in transfected cells and large degree of stabilization by R7BP (34). Mammalian HEK293 cells were transfected with RGS9-2, G␤5, and R7BP subunits of the complex in all possible combinations, and the effects of co-expression on modulating protein levels were evaluated by quantitative Western blotting. We found no change in RGS9-2 protein levels when G␤5 or R7BP was added separately (Fig. 1, A and B). In contrast, co-expression of RGS9-2 with both G␤5 and R7BP produced a substantial increase in the amount of RGS9-2 (Fig. 1, A and B). We further found that the levels of G␤5 increase only upon co-expression with both RGS9-2 and R7BP, and neither of the pro-teins alone affects G␤5 expression (Fig. 1, C and D). Similarly, R7BP levels are elevated only when it is expressed together with both G␤5 and R7BP (Fig. 1, E and F). The degree of the effect on the protein expression level by co-expressing all components was found to be similar for all three subunits. These results indicate that stabilizing effects of the individual proteins in the RGS9-G␤5-R7BP complex are not additive but rather require synergistic contributions from all three subunits to achieve proteolytic stabilization.
G␤5 Is Required for Interaction of RGS9 with Membrane Anchors-RGS9 binds to G␤5 directly via the G␥-like domain, and this association was shown not to require contributions from membrane anchors R7BP/R9AP (29,38). On the other hand, the complex formation of RGS9 with R7BP/R9AP has not been fully characterized. Therefore, we have next analyzed the impact of G␤5 on the association between RGS9 and R7BP (Fig.  2). In these studies, we used the short splice isoform of RGS9, RGS9-1, as the absence of intrinsically disordered C-terminal domain unique to RGS9-2 makes this isoform more proteolytically stable and therefore less dependent on G␤5 and R7BP for achieving high expression levels in transfected cells (42). Indeed, we observed robust expression of RGS9-1 in HEK293 cells line stably transfected with R7BP alone (Fig. 2A). Coexpression with G␤5 in this cell line had only minimal effect on the stability of RGS9-1 ( Fig. 2A). Strikingly, co-immunoprecipitation assays showed that RGS9-1 can form complexes with R7BP only when G␤5 is present ( Fig. 2A). Similar observations were made when cells were transfected with R9AP instead of R7BP (Fig. 2B). Again, R9AP failed to coprecipitate with RGS9-1 in the absence of G␤5 (Fig. 2B). Nevertheless, RGS9-1 expressed without G␤5 retained its ability to interact with the AlF 4 Ϫ -activated G␣ o , arguing that monomeric RGS9-1 retains the ability to associate with its functional partners (Fig. 2C). We further extended these observations and analyzed binding of RGS9 with its subunits in the native striatal tissues derived from mouse knock-out strains lacking individual proteins of the complex (Fig. 2D). We observed that knock-out of R7BP or G␤5 resulted in the similar reduction in RGS9-2 levels. However, although RGS9-2 effectively co-precipitated with G␤5 in the absence of R7BP, it failed to pull down R7BP in the absence of G␤5 (Fig. 2D). Collectively, these findings demonstrate that the interaction between RGS9 and R7BP is depen-dent on G␤5, both in transfected cells and in native striatal neurons.
Plasma Membrane Recruitment of RGS9 by R7BP Is G␤5-dependent-At the cellular level, one of the central effects elicited by the association with R7BP is the recruitment of RGS proteins to the plasma membrane which substantially augments their ability to regulate GPCR signaling (27,31). Therefore, we next examined the effect of R7BP on the subcellular localization of RGS9 in the absence or presence of G␤5. In the absence of its binding partners, monomeric RGS9-1 was found to be evenly distributed through the cytoplasm (Fig. 3A) as indi- Ϫ , and RGS9-1 complexes were immunoprecipitated with anti-HA antibodies. Protein content in the lysate and eluate fractions was analyzed with indicated specific antibodies. D, in native tissues, RGS9-2 co-immunoprecipitates with R7BP only in the presence of G␤5. Striatal protein extracts were prepared from wild-type (WT) or knock-out mice lacking RGS9 (RGS9 Ϫ/Ϫ ), G␤5 (G␤5 Ϫ/Ϫ ), R7BP (R7BP Ϫ/Ϫ ), and used for co-IP studies with anti-RGS9-2 antibody. Lysates and IP eluates were probed with indicated specific antibodies.

FIGURE 3. Expression of G␤5 induces R7BP-dependent plasma membrane localization of RGS9.
Effect of co-expression with R7BP on RGS9-1 localization was studied in the absence (A) or presence (B) of G␤5. HEK293 or an R7BP-stable cell line was transiently transfected with the indicated constructs encoding RGS9-1-HA or G␤5. At 48 h after transfection, cells were fixed and subjected to immunostaining for RGS9-1 (red) and R7BP (green) followed by the image acquisition on a confocal microscope. Green and red images were merged to determine co-localization, and overlapping pixels appear as yellow pseudofluorescence. Representative images of the intracellular distribution for each condition are shown. cated by the presence of a nuclear shadow and the decrease in immunofluorescence toward the cell periphery. Co-expression with G␤5 did not change this localization pattern, and the dimeric RGS9-1-G␤5 complex remained mainly cytoplasmic (Fig. 3B). Co-transfection of RGS9-1 with G␤5 into a R7BPstable cell line resulted in the marked recruitment of the complex to the plasma membrane, as visualized by the sharp staining in the cell periphery and co-localization with R7BP immunofluorescence (Fig. 3B), as noted before (28,43). In contrast, in the absence of G␤5, RGS9-1 failed to be recruited to the plasma membrane by R7BP and remained mostly cytosolic (Fig.  3A). We therefore conclude that R7BP-driven recruitment of RGS9-1 to the plasma membrane requires G␤5. These observations support a critical role of G␤5 in the interaction of RGS9 isoforms with membrane anchors.
G␤5 and an N-terminal DEP/DHEX Module Are Not Sufficient for Binding to R7BP-The observation that RGS9 can bind to R7BP only in the presence of G␤5 could be explained if G␤5 binds directly to R7BP. Because this possibility has not been addressed before, we have next investigated potential proteinprotein interaction between G␤5 and R7BP. Using a co-immunoprecipitation approach we did not detect any appreciable interaction of G␤5 with R7BP in the absence of RGS9-1 (Fig.  4A). Accordingly, when protein localization was studied in the intact cells by immunostaining, G␤5 failed to be recruited to the plasma membranes by R7BP when RGS9-1 was not present in the cells (Fig. 4B). In contrast, addition of RGS9-1 induced G␤5 binding to R7BP (Fig. 4A) and resulted in its translocation to the plasma membrane (Fig. 4B). These data indicate that G␤5 by itself does not have any significant R7BP binding affinity.
The association of RGS9-G␤5 with R7BP has been also shown to require the presence of the intact DEP/DHEX domain of RGS9 (30,34) that forms direct contacts with G␤5 (38,44). Therefore, we have next tested whether R7BP could bind to the complex between DEP/DHEX domains and G␤5 reconstituted in trans. Although the DEP/DHEX domain was able to pull down G␤5 efficiently upon co-expression in the cells, it failed to recruit R7BP to the complex as evidenced by the immunopre-cipitation data (Fig. 4C). Neither did we find evidence for an appreciable interaction between R7BP and the DEP/DHEX domain in the absence of G␤5. These data indicate that neither G␤5 nor the N-terminal DEP/DHEX module alone or in combination is sufficient to enable binding of R7BP to the complex.
Interaction of R7BP with RGS9-G␤5 Complex Requires Intact Conformation of the Interface between G␤5 and N Terminus of RGS9-The inability to reconstitute R7BP binding to the complex between the DEP/DHEX domains and G␤5 suggests that recruitment of R7BP is sensitive to the conformational orientation of the elements involved in the establishment of interface between G␤5 and DEP/DHEX module. Analysis of the crystal structure of RGS9-1-G␤5 complex reveals that contacts with G␤5 mainly mediated by either the DEP domain with helixes ␣2-␣3 contributing the most, and the linker loop connecting ␣4 helix and short ␤1 in the DHEX domain (Fig. 5A,B) (38). To study the role of these regions in mediating the interaction with R7BP, we have introduced point mutations predicted to disrupt the conformation of the interface between G␤5 and the N-terminal lobe of RGS9. The F16E/F76E double mutation was introduced to the DEP domain of RGS9 (DEP mutant), and the amino acids 202 DYGL 205 and 207 RVT 209 were replaced to alanines in the DHEX linker loop (Linker mutant). We have next examined the effects of these mutations on the proteolytic stability and G␤5-induced interaction with R7BP. When transfected in cells without G␤5, both mutants showed expression levels similar to wild-type RGS9-1, indicating that their basal stability is uncompromised (Fig. 5, C and D). However, the ability of G␤5 to increase the levels of RGS9-1 mutants in R7BPdependent fashion was markedly compromised (Fig. 5, C and  E). Although the R7BP/G␤5-induced stabilization of the Linker mutant was reduced by ϳ70%, the protective effect of G␤5-R7BP was completely lost in case of the DEP mutant.
We next studied the ability of the mutants to interact with G␤5 and R7BP by co-immunoprecipitation (Fig. 5, F-H). Both DEP and Linker mutants showed normal interaction with G␤5. In contrast, their association with R7BP was substantially compromised. Although both mutants retained some ability to recruit R7BP, the strength of this association was much lower than observed for the wild-type RGS9-1. Interestingly, the DEP mutant was affected half as much as the Linker mutant. These differences did not correlate with the differences seen in the stabilizing effects of R7BP on the expression level of RGS9 mutants. These observations indicate that interaction interface between G␤5 and the N-terminal lobe of RGS9 controls association of the RGS9-G␤5 complex with R7BP. Our results also suggest that R7BP recruitment is separate from the proteolytic stabilization effects that likely require additional conformational changes.

DISCUSSION
The main observation of this study is that the formation of the RGS9-G␤5-R7BP (R9AP) complex occurs sequentially and there is a distinct hierarchy of RGS9 association with its sub- units. First, RGS9 forms a complex with G␤5, and then this dimer recruits membrane-targeting subunit R7BP or R9AP. Interestingly, we find that neither RGS9 nor G␤5 alone can bind to R7BP(R9AP). Thus, our study for the first time establishes the assembly relationship between the subunits of the trimeric complex and demonstrates that the G␤5 subunit is required for the interaction of RGS9-containing complex with R7BP or R9AP.
Previous studies have shown that the N-terminal DEP domain of RGS9 is necessary for the interaction of the RGS9-G␤5 complex with both R9AP and R7BP (26,30) and that it acts synergistically with the DHEX domain to establish the selectivity of R7BP/R9AP binding (34). Furthermore, it was reported that the N termini of RGS9 homologues, RGS7 and Egl-10, can interact directly with G␤5 (44,45). The crystal structure of the RGS9-G␤5 complex confirmed that DEP and DHEX domains form a structurally integrated module that makes direct contacts with the protein interaction interface of G␤5 (38). Therefore, to account for the dual requirement for both G␤5 and the DEP domain we suggest that the binding site for R7BP and R9AP is formed by the interface between the DEP/DHEX module and G␤5. This model is consistent with our site-directed mutagenesis data showing that substitution of amino acids in RGS9 involved in direct contacts with G␤5 substantially reduces R7BP recruitment to the complex (39).
Unexpectedly, we found that there was no correlation between the ability of the "interface" mutations to disrupt R7BP binding and affect the stability of RGS9-G␤5 complexes. For example, mutating the linker loop in DHEX domain severely compromises R7BP binding but does not eliminate its effects on the proteolytic stability of the complexes. Conversely, mutations of the DEP domain amino acids involved in contacts with G␤5 completely eliminate proteolytic stabilization by R7BP but exert less severe effects on R7BP binding compared with the DHEX linker mutant. The latter observation is in line with the report that reciprocal mutations in G␤5 that disrupt interaction with the DEP domain also completely abolish stabilizing effects of R7BP while preserving its recruitment to the complex (39). Taken together, these observations indicate that even though binding of R7BP leads to the proteolytic stabilization of the complex, these are two mechanistically separate events that are not firmly hardwired together. Furthermore, our data pinpoint the underlying molecular determinants: the contacts of G␤5 with the DEP domain appear to regulate the proteolytic stability of the complexes, whereas coordination of the DHEX linker region seems to be primarily engaged in R7BP (R9AP) binding. Interestingly, G␤5 uses distinct clusters on its surface for establishing contacts with the DEP and DHEX linker regions (38) (Fig. 5B). Although understanding precise details of R7BP interaction with RGS9-G␤5 would certainly require determination of the RGS9-G␤5-R7BP complex structure at the atomic level, it is tempting to speculate that R7BP binds to the interdomain cleft between the DHEX and the G␤5 in the close vicinity to the DHEX loop region. Supporting this model is the observation that RGS7, a close homologue of RGS9 that also binds G␤5, has a markedly different composition of the DHEX linker region (containing ϳ25 additional unique amino acids). This correlates with drastically lower affinity of RGS7 for R7BP binding (37). Furthermore, the RGS7-G␤5 complex does not require R7BP binding to achieve proteolytic stabilization (29,34).
The recruitment of G␤5 to the R7 RGS proteins is primarily mediated by the G␥-like domain of the RGS proteins. This interaction resembles classical stable pairing of the G protein ␤␥ subunits (38) and is mediated by the specialized CCT chaperonin complex that permanently locks the subunits together (46). In contrast, the association of the G␤5 at the opposite interface with the DEP/DHEX module of the RGS proteins appears to be dynamic. Narayanan et al. (44) reported that G␤5 can bind to the DEP domain of RGS7 and that this interaction is dynamic and is modulated by the R7BP. In addition, the same group has shown that the DEP domain of RGS7 can also bind to the third intracellular loop of M3 receptor and that this interaction is negatively modulated by R7BP (47). On the basis of these observations it was proposed that the interaction of G␤5 with the DEP/DHEX module is dynamic. Our findings suggest that a similar mechanism modulating the interface between DEP/DHEX and G␤5 is involved in controlling the proteolytic stability of the R7 RGS complexes. We propose that the opening of the interface at the DEP/G␤5 site exposes degradation determinants located in the DEP/DHEX module and facilitates proteolysis of the complex. Binding of R7BP in the vicinity of the DHEX linker located on the other side of the interface serves as a "prop" to restrict the dynamics of the interactions and to promote closed conformation, thus resulting in shielding of the degradation determinants and proteolytic stabilization (Fig. 6). FIGURE 6. Model of the heterotrimeric RGS9-G␤5-R7BP complex formation. RGS9 forms tight complex with G␤5 via its G␥-like (GGL) domain. In addition, the N-terminal DEP/DHEX module transiently interacts with the opposite surface of G␤5 through two points of contact involving DHEX linker (red circles) and DEP domain (blue circles). The association of membrane anchors R7BP or R9AP with the complex is mediated by the interface between G␤5 and the DEP/DHEX module, primarily involving DHEX linker region. Binding of R7BP/R9AP changes the conformation of the interface stabilizing the DEP contacts that contain instability determinants, thus providing proteolytic protection.
In summary, our results indicate that intramolecular interactions between G␤5 and DEP/DHEX module of RGS9 determine protein expression of the complex and control its localization in cells, revealing a powerful regulatory hot-spot in the R7 RGS proteins.