Allosteric inhibition of the regulator of G protein signaling-Galpha protein-protein interaction by CCG-4986

Abstract

Regulator of G protein signaling (RGS) proteins act to temporally modulate the activity of G protein subunits after G protein-coupled receptor activation. RGS proteins exert their effect by directly binding to the activated Galpha subunit of the G protein, catalyzing the accelerated hydrolysis of GTP and returning the G protein to its inactive, heterotrimeric form. In previous studies, we have sought to inhibit this GTPase-accelerating protein activity of the RGS protein by using small molecules. In this study, we investigated the mechanism of CCG-4986 [methyl-N-[(4-chlorophenyl)sulfonyl]-4-nitro-benzenesulfinimidoate], a previously reported small-molecule RGS inhibitor. Here, we find that CCG-4986 inhibits RGS4 function through the covalent modification of two spatially distinct cysteine residues on RGS4. We confirm that modification of Cys132, located near the RGS/Galpha interaction surface, modestly inhibits Galpha binding and GTPase acceleration. In addition, we report that modification of Cys148, a residue located on the opposite face of RGS4, can disrupt RGS/Galpha interaction through an allosteric mechanism that almost completely inhibits the Galpha-RGS protein-protein interaction. These findings demonstrate three important points: 1) the modification of the Cys148 allosteric site results in significant changes to the RGS interaction surface with Galpha; 2) this identifies a "hot spot" on RGS4 for binding of small molecules and triggering an allosteric change that may be significantly more effective than targeting the actual protein-protein interaction surface; and 3) because of the modification of a positional equivalent of Cys148 in RGS8 by CCG-4986, lack of inhibition indicates that RGS proteins exhibit fundamental differences in their responses to small-molecule ligands.

Fig. 1.

Ribbon diagram of the RGS homology domain of RGS4 (left) and RGS8 (right) extracted from Protein Data Bank entries 1agr and 2ihd, respectively, with cysteine residues indicated by magenta space-filling spheres.

Fig. 2.

Reversibility of CCG-4986 inhibition in the presence of the reducing agent dithiothreitol. Immobilized, biotinylated RGS4 on avidin microspheres was treated with CCCG-4986 and subsequently washed with buffer (●) or buffer containing DTT (■). Binding to increasing concentrations of Alexa Fluor 532-labeled Gα_o was measured by FCPIA. RGS4 binding to Gαo before washing is also shown (▲). Data points are from duplicate measures from three independent experiments (n = 3).

Fig. 3.

Inhibition of RGS/Gα_o interaction by CCG-4986 at cysteine point mutants. Dose-response curves for CCG-4986 at LumAvidin microspheres were coupled to 10 nM concentration of biotinylated constructs preincubated with varying concentrations of CCG-4986 before the addition of 50 nM AMF-activated Gα_o. Indicated values are IC₅₀ ± S.E.M., with measurements made in duplicate in three independent experiments (n = 3).

Fig. 4.

Inhibition of RGS/Gα_o interaction by CCG-4986 at cysteine “add-back” constructs. These mutants have had specific cysteine residues added back to the Cysless RGS4 construct. Dose-response curves for CCG-4986 at LumAvidin microspheres were coupled to 10 nM concentration of biotinylated constructs preincubated with varying concentrations of CCG-4986 before the addition of 50 nM AMF-activated Gα_o.

Fig. 5.

Saturation binding of Gα_o to wild-type RGS4 and point mutants in the presence and absence of 10 μM CCG-4986. Biotinylated RGS4 proteins were incubated with avidin-coated beads and incubated for 30 min in the presence of varying concentrations of AF523-labeled Gα_o with or without the addition of 10 μM CCG-4986. The y-axis indicates median fluorescence intensity (MFI) in the AF532 channel of 100 bead-events collected with the Luminex Cytometer. Specific binding is shown. Nonspecific binding was determined by using GDP instead of AMF/GDP. The K_D and B_max values, respectively, for each data set are: RGS4, 42 nM, 538 MFI; RGS4 + 4986, 195 nM, 51 MFI; A132C, 19 nM, 149 MFI; A132C + 4986, 49 nM, 135 MFI; A148C, 13 nM, 62 MFI; and A148C + 4986, 19 nM, 7 MFI.

Fig. 6.

LC/MS analysis of CCG-4986-treated RGS8. Purified human RGS8 was incubated with 100 μM CCG-4986 for 1 h at room temperature. RGS8 (50 pmol) was subject to reversed-phase high-performance LC using a C18 column. The deconvoluted chromatogram shows unmodified protein, a singly adducted RGS8 (+153 mol. wt.), and the doubly modified RGS8 (+306 mol. wt.), indicating that both of the cysteines present in RGS8 can be modified by CCG-4986.