Mutational Analysis of the Asn Residue Essential for RGS Protein Binding to G-proteins*

Members of the RGS family serve as GTPase-activat-ing proteins (GAPs) for heterotrimeric G-proteins and negatively regulate signaling via G-protein-coupled re-ceptors. The recently resolved crystal structure of RGS4 bound to G i a 1 suggests two potential mechanisms for the GAP activity of RGS proteins as follows: stabilization of the G i a 1 switch regions by RGS4 and the catalytic action of RGS4 residue Asn 128 . To elucidate a role of the Asn residue for RGS GAP function, we have investigated effects of the synthetic peptide corresponding to the G a binding domain of human retinal RGS (hRGSr) containing the key Asn at position 131, and we have carried out mutational analysis of Asn 131 . Synthetic peptide hRGSr-(123–140) retained its ability to bind the AlF 4 2 -complexed transducin a -subunit, G t a z AlF 4 2 , but failed to elicit stimulation of Gt a GTPase activity. Wild-type hRGSr stimulated G t a GTPase activity by ; 10-fold with an EC 50 value of 100 n M . Mutant hRGSr proteins with substitutions of Asn 131 by Ser and Gln had a significantly reduced affinity for G t a but were capable of substantial stimulation of G t a GTPase activity, 80 and 60% of V max , respectively. Mutants hRGSr-Leu 131 , hRGSr-Ala 131 , and hRGSr-Asp 131 were able to accelerate G t a GTPase activity only at very high concentrations ( > 10 m M ) which appears to correlate with a further decrease of their affinity for transducin. Two mutants, hRGSr-His 131 and hRGSr- D 131 ,

spectively (1)(2)(3). A novel class of GTPase-activating proteins (GAPs) for heterotrimeric G-proteins called RGS has been identified (4 -6). Strong evidence suggests that members of this family, GAIP, RGS4, RGS1, RGS10 and others, negatively regulate G-protein signaling by stimulating GTPase activity of G-proteins, particularly those from G i and G q families (7)(8)(9). RGS proteins from yeast to mammals share a highly conserved RGS domain that provides relatively broad specificity of different RGS proteins toward members of the two G-protein classes in vitro. Tissue expression patterns and diverse domains outside the RGS segment may play an important role in determining specificity of RGS proteins in vivo (10 -12). Precise mechanisms of RGS GAP activity are not yet clear. The transition state during GTP hydrolysis is thought to be mimicked by the AlF 4 Ϫ -bound conformation of G␣ subunits (13,14). It has been demonstrated that many RGS proteins interact preferentially with the AlF 4 Ϫ -bound conformation of G␣ subunits and thus may accelerate GTP hydrolysis through stabilization of the transitional state of G-proteins (8,15,16).
Recently, the crystal structure of RGS4 bound to G i ␣ 1 ⅐AlF 4 Ϫ has been solved at a resolution of 2.8 Å (17). This structure provides the first structural insights into the mechanism of RGS protein action. The conserved RGS core forms three distinct sites of interaction with the three switch regions of G i ␣ 1 suggesting that stabilization of the switch regions and G ␣ residues directly involved in GTP hydrolysis may be a major component of RGS GAP activity (17). Furthermore, RGS proteins could also contribute catalytic residues to the active site and thus enhance the GTPase rate constant. The conserved residue Asn 128 of RGS4 makes a contact with the side chain of Gln 204 of G i ␣ 1 which stabilizes and orients the hydrolytic water molecule in the transitional state of G i ␣ 1 (17). Asn 128 also may be localized within hydrogen-bonding distance of the hydrolytic water molecule for nucleophilic attack on the GTP ␥-phosphate (17). In this study we evaluate a potential catalytic role of the Asn residue for G ␣ GTPase acceleration by RGS proteins using the interaction between human retinal RGS (hRGSr) protein and transducin as a model system and mutational analysis of Asn 131 of hRGSr which is equivalent to Asn 128 of RGS4.  19. The G t ␣⅐GTP␥S was extracted from ROS membranes using GTP␥S and purified by chromatography on Blue-Sepharose CL-6B by the procedure described in Ref. 20. G t ␣⅐GDP was prepared and purified according to protocols in Ref. 21. hRGSr was prepared and purified as described previously (22). The purified proteins were stored in 40% glycerol at Ϫ20°C or without glycerol at Ϫ80°C. Site-directed Mutagenesis of hRGSr-Mutagenesis of Asn 131 residue of hRGSr was performed using PCR amplifications from the pGEX-KG-hRGSr template (22) with 3Ј-antisense primer ATGCCTCGAGACT-CAGGTGTGTGAGG (unique XhoI site is underlined) and the 5Ј primers: XXXATTGACCATGAGACCCGCGAGC. XXX indicates nucleotides that generate substitutions of Asn 131 (AAC) in hRGSr cDNA by the following amino acid residues: Ala (GCG), Asp (GAT), His (CAT), Leu (CTG), Gln (CAG), Ser (AGC), and deletion mutant (-). PCR reactions were performed in 100 l of reaction mixture containing 1 ng of the pGEX-KG-hRGSr plasmid, 3 units of AmpliTaq DNA polymerase (Perkin-Elmer), 25 mM Tris-(hydroxymethyl)-methylaminopropane sulfonic acid, pH 9.3, 2 mM MgCl 2 , 1 mM 2-mercaptoethanol, 200 M of dNTPs, and 0.5 M primers. Conditions for PCR were as follows: 94°C for 3 min, 30 cycles of 94°C for 1 min, 64°C for 30 s and 72°C for 30 s, and a final extension at 72°C for 3 min. The PCR products (ϳ220 base pairs) were blunt-ended with Klenow fragment and digested with XhoI. Wild-type hRGSr cDNA was subcloned into XbaI/XhoI sites of pBluescript polylinker. The resulting construct was digested with HincII and XhoI and ligated with the XhoI-digested PCR products carrying mutations. The mutant sequences were verified by automated DNA sequencing at the University of Iowa DNA Core Facility using the T7 primer and subcloned into the XbaI/XhoI sites of pGEX-KG vector for protein expression. Mutant GST-hRGSr proteins were expressed in DH5␣ Escherichia coli cells, and the GST portion was removed as described earlier (22). Typical yields of purified hRGSr and hRGSr mutants, except for a mutant with deletion of Asn 131 , were 5-6 mg/liter of culture. Deletion of Asn 131 led to an ϳ4 -5-fold reduction in expression of soluble recombinant protein suggesting that the residue at position 131 may be important to the stability and proper folding of RGS proteins.

Materials-GTP
Binding of Transducin to GST-hRGSr and Mutants-G t ␣⅐GTP␥S or G t ␣⅐GDP (10 g) were incubated with hRGSr or its mutants (50 g) immobilized on glutathione-agarose in 100 l of 20 mM Tris-HCl buffer (pH 8.0), containing 100 mM NaCl, 2 mM MgSO 4 , 6 mM 2-mercaptoethanol, and 5% glycerol (buffer A). Where indicated, the buffer contained 30 M AlCl 3 and 10 mM NaF. After incubation for 20 min at 25°C, the agarose beads were spun and washed twice with 1 ml of buffer A, and the bound proteins were eluted with a sample buffer for SDS-polyacrylamide gel electrophoresis.
Single Turnover GTPase Assay-Single turnover GTPase activity measurements were carried out in suspensions of dROS membranes containing 5 M rhodopsin and 0.4 M transducin essentially as described in Refs. 22 and 23. Transducin concentration of 0.4 M was determined using the [ 35 S]GTP␥S binding assay as described previously (22). Bleached dROS membranes were mixed with different concentrations of the tested peptides, hRGSr or hRGSr mutants, and preincubated for 5 min at 25°C. The GTPase reaction was initiated by addition of 100 nM [␥-32 P]GTP (ϳ5 ϫ 10 4 dpm/pmol) in a total volume of 20 l. At 5, 10, 20, 40, and 60 s aliquots of the reaction mixture were withdrawn and quenched with 7% perchloric acid. Nucleotides were then precipitated using activated Norit A charcoal (10% w/v) in 50 mM sodium phosphate buffer (pH 7.5), and 32 P i formation was measured by liquid scintillation counting. The GTPase rate constants were calculated by fitting the experimental data to an exponential function: % GTP hydrolyzed ϭ 100 (1 Ϫ e Ϫkt ), where k is a rate constant for GTP hydrolysis.
Peptide Synthesis-A peptide, CSEAPKEVNIDHETRELT, corresponding to residues 123-140 of hRGSr was custom made by Genosys Biotechnologies Inc. The N and C termini of the peptide were acetylated and amidated, respectively. The peptide was purified by reverse-phase high pressure liquid chromatography on a preparative Dynamax-300A column (Rainin). The purity and chemical formula of the peptide were confirmed by fast atom bombardment-mass spectrometry and analytical high pressure liquid chromatography. Preparation of synthetic peptides corresponding to residues 21-31, 461-491, 492-516, and 517-541 of rod PDE ␣-subunit was described previously (24).
Miscellaneous-Protein concentrations were determined by the method of Bradford (25) using IgG as a standard or using calculated extinction coefficients at 280 nm. SDS-polyacrylamide gel electrophoresis was performed by the method of Laemmli (26) in 12% acrylamide gels. Rhodopsin concentrations were measured using the difference in absorbance at 500 nm between "dark" and bleached ROS preparations. Fitting of the experimental data was performed with nonlinear least squares criteria using GraphPad Prizm (version 2) software. The results are expressed as the mean Ϯ S.E. of triplicate measurements.

Effects of Synthetic Peptide hRGSr-(123-140)-
The G i ␣ binding region of RGS4 containing Asn 128 resides in the loop ␣5-␣6 of RGS4 and contains 7 amino acid residues making contacts with all three switch regions of G-protein (17). The number of interactions is sufficient to ensure a relatively high affinity between the corresponding synthetic peptide and G␣, provided that the peptide is able to adopt a functional conformation. For the preliminary testing of the catalytic role of Asn 131 of hRGSr, the hRGSr-(123-140)-peptide was synthesized. The length of the peptide was chosen to allow the G␣ contact residues to be flanked by at least 3 terminal residues. hRGSr-(123-140) was first examined for its ability to stimulate GTPase activity of transducin in suspensions of dROS membranes containing 5 M rhodopsin and 0.4 M transducin. dROS membranes lacked intrinsic catalytic PDE␣␤ and inhibitory PDE␥ subunits. Use of such ROS avoided interference of PDE␥ effects with effects of RGS protein or RGS peptide (22,27,28). The peptide at concentrations of up to 2 mM had no effect on GTPase activity of transducin (not shown). To determine if hRGSr-(123-140) is capable of binding to transducin, we investigated effects of the hRGSr peptide on the stimulation of GTPase activity of transducin by hRGSr. Fig. 1 shows that hRGSr-(123-140) was able to compete with hRGSr for binding to G t ␣ resulting in a dosedependent (IC 50 ϭ 1.6 Ϯ 0.3 mM) decrease of the stimulated GTPase activity of transducin. hRGSr-(123-140) in the same range of concentrations had no notable effect on the basal transducin GTPase activity (Fig. 1). Because the competition experiments were carried out at a concentration of hRGSr causing half-maximal stimulation of the GTPase activity, the affinity of hRGSr-(123-140) for G t ␣ can be estimated as 0.8 mM.
Binding of hRGSr Mutants with Substitutions of Asn 131 to Different Conformations of G t ␣-Recently, we have shown that similar to other characterized RGS proteins, hRGSr binds with high affinity to the AlF 4 Ϫ conformations of transducin and very weakly to the GTP␥S and GDP-bound conformations (22). We evaluated the interaction between hRGSr mutants with substitutions of Asn 131 by Ser, Gln, Ala, Leu, His, Asp as well as the mutant with deletion of Asn 131 and transducin using precipitation of G t ␣ with the GST-hRGSr mutant proteins immobilized on glutathione-agarose beads. Mutations hRGSr-Ser 131 and hRGSr-Gln 131 led to a reduction in affinity of the corresponding GST fusion proteins for G t ␣⅐AlF 4 Ϫ (Fig. 2A). Mutants hRGSr-Leu 131 , hRGSr-Asp 131 , and hRGSr-Ala 131 showed a more significant decrease in their affinity for the G t ␣ conformation ( Fig. 2A). hRGSr-His 131 and hRGSr-⌬ 131 failed to coprecipitate G t ␣⅐AlF 4 Ϫ . Mutations of Asn 131 could potentially alter hRGSr interaction with G t ␣⅐GTP␥S and G t ␣⅐GDP since the RGS4 Asn residue makes contact with the switch I and II regions of G i ␣ 1 (17). We have tested this possibility by preincubating mutant GST-hRGSr containing beads with both conformations of G t ␣. None of the seven hRGSr mutants has demonstrated enhanced affinity for either conformation of G t ␣ compared with the native GST-hRGSr (Fig. 2, B and C).
Stimulation of GTPase Activity of Transducin by Mutant hRGSr-Effects of hRGSr mutants with substitutions of Asn 131 were tested in dROS membranes containing 5 M rhodopsin and 0.4 M transducin. Under these conditions, the calculated rate of GTP hydrolysis by transducin was 0.025 Ϯ 0.004 s Ϫ1 (Fig. 3). The rates of transducin GTPase activity were then determined in the presence of increasing concentrations of hRGSr or individual hRGSr mutants and plotted as a function of their concentration. Wild-type hRGSr purified after cleavage of GST-hRGSr with thrombin stimulated GTPase activity of transducin by ϳ10-fold to a maximal rate k ϭ 0.27 Ϯ 0.01 s Ϫ1 with an EC 50 value of 101 Ϯ 14 nM (Fig. 3). All hRGSr mutants had substantially reduced ability to stimulate the GTPase activity of transducin. The tested mutants can be arbitrarily separated into three groups. Two of the mutants, hRGSr-Ser 131 and hRGSr-Gln 131 , were relatively potent, and saturation of their GAP effect could be achieved at 10 -40 M concentration of mutant. hRGSr-Ser 131 mutant was the most effective and stimulated G t ␣ GTPase activity with an EC 50 value of 1.34 Ϯ 0.17 M and V max ϳ80% (k ϭ 0.22 Ϯ 0.01 s Ϫ1 ). The mutant hRGSr-Gln 131 was capable of accelerating the G t ␣ GTPase activity to V max of 60% (k ϭ 0.16 Ϯ 0.01 s Ϫ1 ) with an EC 50 value of 3.9 Ϯ 1.1 M (Fig. 3). Three mutants, hRGSr-Leu 131 , hRGSr-Ala 131 , and hRGSr-Asp 131 , began to cause acceleration of G t ␣ GTPase activity only at very high concentrations (Ͼ10 M) (Fig.  3). We were unable to practically achieve saturation of the GAP activity by these mutants due to the very high protein concentrations required. Two mutants, hRGSr-His 131 and hRGSr-⌬ 131 , did not show GAP activity at the concentration tested (40 M). Interestingly, the potency of hRGSr mutants in stimulating G t ␣ GTPase activity (Fig. 3) appears to correlate well with their ability to bind and precipitate G t ␣⅐AlF 4 Ϫ ( Fig. 2A), Fig. 2 have suggested that hRGSr mutants with substitutions of Asn 131 have impaired binding to G t ␣⅐AlF 4

Competition between hRGSr and hRGSr Mutants in Stimulation of G t ␣ GTPase Activity-Experiments in
Ϫ . The binding assay may, however, not be sufficiently sensitive to detect relatively weak interactions. To determine if the drastically reduced ability of some RGS mutants to stimulate the GTPase activity of transducin correlates with the lack of mutant binding to transducin, we carried out competition experiments. The hRGSr mutants incapable of accelerating G t ␣ GTPase activity were examined for their ability to block stimulation of GTPase activity of transducin by hRGSR. Fig. 4 demonstrates that none of the tested mutants, hRGSr-Ala 131 , hRGSr-His 131 , and hRGSr-⌬ 131 , at 5 M concentration, had any effect on stimulation of GTPase activity of transducin by 50 nM hRGSr. These data suggest that the hRGSr mutants that produced no stimulation of G t ␣ GTPase activity lost their binding to G t ␣.

DISCUSSION
Molecular mechanisms of RGS protein action as GAP for heterotrimeric GTP-binding proteins are not well understood. Studies on the Ras-specific p120 GAP suggest that the Ras GAP donates conserved Arg 789 residue to the RAS catalytic site (29,30) thus providing the catalytic mechanism for p120 GAP activity. The crystal structure of RGS4 bound to G i ␣ 1 ⅐AlF 4 Ϫ has suggested two mechanisms for RGS GAP activity toward heterotrimeric G-proteins (17). Interaction of RGS protein with the G-protein switch regions indicates that the mechanism of the GTPase activation by RGS may primarily be a reduction in the free energy of the transitional state via stabilization of G␣ switch regions and residues directly involved in GTP hydrolysis (17). An additional putative mechanism for the RGS GAP activity would be a donation of the catalytic residue to the active site of G␣. The only residue that RGS4 introduces into the active site of G i ␣ 1 is Asn 128 . Although Asn 128 , in contrast to the Ras GAP Arg 789 or an intrinsic Arg in G␣ subunits, does not directly interact with GDP and AlF 4 Ϫ (13, 14, 17, 30), it makes a contact with the side chain of Gln 204 of G i ␣ 1 , which stabilizes and orients the hydrolytic water molecule in the transitional state of G i ␣ 1 . Conceivably, Asn 128 is within hydrogen-bonding distance of the hydrolytic water molecule and may bind and orient it for nucleophilic attack of the ␥-phosphate of GTP (17).
To probe the role of Asn 131 of hRGSr for the mechanism of RGS protein GAP activity, we initially synthesized a peptide of hRGSr corresponding to the region of interaction between G i ␣ 1 ⅐AlF 4 Ϫ and RGS4 containing Asn 128 . We reasoned that if the catalytic role of the Asn residue is a major component of RGS GAP activity, then perhaps a peptide containing the catalytic residue would alone be capable of eliciting the stimulation of GTPase activity. Our data demonstrated that hRGSr peptide-(123-140) containing catalytic Asn 131 retained the ability to bind hRGSr but failed to accelerate the GTPase activity of transducin. This indicates that the interaction of at least two and likely all three G␣ binding regions of RGS protein is required to stimulate G␣ GTPase activity. Consistent with this conclusion is the recent finding that even short deletions within the RGS domain of RGS4 destroyed its GAP activity (31).
Further analysis of the role of Asn 131 of hRGSr was carried out using mutational substitutions of this residue. The major result from testing all hRGSr mutants is that replacement of Asn 131 with other residues dramatically decreases the affinity of mutant hRGSr binding to G t ␣. Substitution of Asn 131 by Ser was intriguing because the Asn residue is not absolutely conserved in RGS proteins, and some RGS proteins, including GAIP, have a Ser at this position (5,17,31). Serine has proven to be the best substitution for Asn with respect of retaining the GAP activity of hRGSr protein. The hRGSr-Ser 131 mutant had more than 10-fold lower affinity for G t ␣ but can stimulate its GTPase activity nearly as well as native hRGSr (V max ϳ80%). Therefore, it is not surprising that this residue was evolutionary selected instead of Asn for some RGS proteins. Interestingly, the reported concentrations of the Ser containing RGS domain of retina-specific RET-RGS1 (1 M) and GAIP (5 M) required for the half-maximal GTPase stimulation of transducin are comparable to the EC 50 value for hRGSr-Ser 131 (28,32).
An indication that the Asn residue may indeed serve to some extent as a catalytic residue was provided by the hRGSr-Gln 131 mutant. The effects of this mutant on GTPase activity of transducin nearly reached a plateau at only ϳ60% V max of that observed with hRGSr. However, effects of other hRGSr mutants argue against a catalytic role of the Asn residue as the key component of the RGS GAP activity. The hRGSr-Leu 131 and hRGSr-Ala 131 mutants at very high concentrations started to have a stimulatory effect on G t ␣ GTPase activity even though these residues are not expected to form hydrogen bonds which are made by the Asn residue. Our mutational analysis suggests that although Asn 131 of hRGSr may play a catalytic role in the RGS GAP activity, stabilization of the switch regions of G-protein and reduction of the energy of the transition state appear to be the major components of the RGS GAP function. The Asn residue is absolutely essential for the stabilization of the transition state for GTP hydrolysis because its replacement or deletion leads to a drastic reduction in hRGSr affinity for G t ␣.
In addition to their role as GAPs, RGS proteins may act as antagonists for some G-protein effectors, particularly for phospholipase C␤. RGS4 has been shown to block activation of phospholipase C␤ by G q ␣GTP␥S (33). In another study, RGS4 inhibited inositol phosphate synthesis activated by AlF 4 Ϫ in COS-7 cells overexpressing G q (34). Tesmer et al. (17) have suggested that the RGS proteins lacking the Asn residue may better serve as inhibitors of effector binding than as GAPs. This would appear to be a likely scenario if replacements of the Asn residue resulted in a loss of GAP activity without a concurrent reduction of the RGS protein affinity for activated G ␣ subunits. The results of this work suggest that the main consequence of Asn replacement is an impairment of binding between mutated hRGSr protein and G t ␣. Furthermore, none of the hRGSr mutants have shown enhanced affinity to the active G t ␣⅐GTP␥S conformation which could be indicative of the potential of such a mutant to serve as an antagonist for the G-protein effector.
This study only begins to address the questions, introduced by the first crystal structure between G-protein and RGS protein, about the mechanism of RGS protein GAP activity (17). Further biochemical analysis coupled with resolution of other crystal structures between activated G␣ subunits and RGS proteins would ultimately define a role of the critical Asn residue.