A retinal-specific regulator of G-protein signaling interacts with Galpha(o) and accelerates an expressed metabotropic glutamate receptor 6 cascade

Abstract

G(o) is the most abundant G-protein in the brain, but its regulators are essentially unknown. In retina, Galpha(o1) is obligatory in mediating the metabotropic glutamate receptor 6 (mGluR6)-initiated ON response. To identify the interactors of G(o), we conducted a yeast two-hybrid screen with constituitively active Galpha(o) as a bait. The screen frequently identified a regulator of G-protein signaling (RGS), Ret-RGS1, the interaction of which we confirmed by coimmunoprecipitation with Galpha(o) in transfected cells and in retina. Ret-RGS1 localized to the dendritic tips of ON bipolar neurons, along with mGluR6 and Galpha(o1). When Ret-RGS1 was coexpressed in Xenopus oocytes with mGluR6, Galpha(o1), and a GIRK (G-protein-gated inwardly rectifying K+) channel, it accelerated the deactivation of the channel response to glutamate in a concentration-dependent manner. Because light onset suppresses glutamate release from photoreceptors onto the ON bipolar dendrites, Ret-RGS1 should accelerate the rising phase of the light response of the ON bipolar cell. This would tend to match its kinetics to that of the OFF bipolar that arises directly from ligand-gated channels.

Figure 1.

Gα_o interacts with Ret-RGS1. A, Alignment of amino acid residues of mouse (Mo), bovine (Bo), and human (Hu) Ret-RGS1 protein N termini (corresponding to the N1 bovine peptide used to generate the antibody against Ret-RGS1). B, HEK 293 cells transfected with bovine Ret-RGS1 expression plasmid (top) and mouse Ret-RGS1 EST clone (bottom), immunostained with N1 antibody (red) and counterstained with the nuclear dye SYTO 13 (green). This antibody recognized the bovine protein (top) but not the mouse protein (bottom). C, A Western blot stained with N1 antibody shows two bands of 45 and 40 kDa in bovine retina and a 45 kDa band in HEK cells transfected with Ret-RGS1 expression plasmid. These bands were eliminated by preabsorbing the antibody with the N1 peptide (preab). D, Coimmunoprecipitation of Gα_o1 and Ret-RGS in HEK cells transfected with Gα_o1 and Ret-RGS1 expression vectors. Gα_o was immunoprecipitated in the presence or absence of AlF₄^- (as indicated). The immunoprecipitates were subjected to Western blotting with Gα_o (top) or Ret-RGS1 (bottom) antibodies in parallel. 1, Homogenate; 2 and 3, immunoprecipitation (IP) with anti-Gα_o; 4 and 5, mock IP without any antibody. In bottom blot, an empty lane next to lane 2 was cut out for alignment. For both D and E, 0.3% of total sample was applied for the homogenate lane, and 5% was applied for the rest of the lanes. E, Coimmunoprecipitation of Gα_o and Ret-RGS1 from bovine retina. IP was done with anti-Gα_o in the presence of AlF₄^-. h, Homogenate.

Figure 2.

Ret-RGS1 is localized to ON bipolar somas and dendrites. A, Radial section of bovine retina stained for Ret-RGS1. Staining is present in all retinal layers, most strongly in inner segments and then OPL. Staining in IPL and ganglion cell somas was somewhat variable between tissues. IS, Inner segments; ONL, outer nuclear layer; OPL, outer plexiform layer; INL, inner nuclear layer; IPL, inner plexiform layer; GCL, ganglion cell layer; NFL, nerve fiber layer. (Abbreviations apply to all Figures.) B, Double staining for Ret-RGS1 (red) and PKC (green). Rod bipolar cells marked with anti-PKC (arrows) were also stained for Ret-RGS1. Certain PKC-negative somas were Ret-RGS1 positive (arrowhead). C, Double staining for Ret-RGS1 (red) and Gα_o (green). All ON bipolar cells marked by Gα_o express Ret-RGS1 (arrows). D, Electron micrograph of a rod synaptic complex. The invaginating rod bipolar dendritic tip (RB) is stained for Ret-RGS1. r, Ribbon. E, Electron micrograph of a cone synaptic complex. The central element of the postsynaptic triad (CB) was stained for Ret-RGS1. H, Horizontal cell process.

Figure 3.

Ret-RGS1 is expressed in presynaptic terminals. Top row, Immunostaining for Ret-RGS1 (red) and horizontal cell marker, calbindin (green). Horizontal cell processes are unstained for Ret-RGS1. Middle row, Immunostaining for Ret-RGS1 (red) and synaptophysin (green). Ret-RGS1 and synaptophysin colocalize in presynatic photoreceptor terminals (PT) (arrows). Note that Ret-RGS1 staining forms clusters, which often outline the synaptophysin-positive terminals (arrowheads). Bottom row, Immunostaining for Ret-RGS1 (red) and PKC (green). Staining for Ret-RGS1 forms large puncta throughout the IPL (arrowheads). Rod bipolar axon terminals (rba) (identified by PKC) are also stained (arrows).

Figure 4.

Ret-RGS1 associates with plasma membrane and synaptic vesicles in presynaptic terminals. A-C, Electron micrographs of rod terminals. A, Diffused stain appears throughout the terminal; B, C, clusters of gold particles (denoted within the dashed pink ellipses) appear ∼100-200 nm from plasma membrane (outlined in blue). D, Western analysis of subfractions from bovine retina with Ret-RGS1 antibody. Fifty micrograms of total proteins of each fraction were loaded on a 12% acrylamide gel. Ret-RGS1 is present in plasma membrane fraction and synaptic vesicles. Hom, Homogenate; crude synapt, crude synaptosomes. E, Fractions of synaptic vesicles on a sucrose gradient analyzed by Western blotting. Ret-RGS1 is co-present with synaptophysin in medium weight fractions; Na/K ATPase α-subunit (96 kDa; bottom band), a marker of a plasma membrane, settles in the heavier fractions. The 120 kDa (top band) is an aggregate of the α-subunit, which is commonly seen after sample boiling. As for Ret-RGS1, its migration was not perturbed by sample boiling.

Figure 5.

Ret-RGS1 is strongly bound to the plasma membrane. Retinal membranes from 25,000 × g pellets were washed with different treatments: hypotonic (hypo), hypertonic (hyper), alkaline (Na₂CO₃), octyl-glucoside (OG), urea, and NH₂OH. For most treatments, Ret-RGS was recovered only in the pellet fraction (P); a small amount of Ret-RGS1 in supernatant (S) was recovered only after octyl-glucoside treatment. This suggests that Ret-RGS1 is tightly associated with the plasma membrane.

Figure 6.

Ret-RGS1 modulates acetylcholine-evoked response. A, Experimental protocol (current responses to 10 μm acetylcholine). Bars above the records denote application of high K⁺ and acetylcholine solutions. Before and after application of high K⁺, oocytes were perfused with ND96 solution. Basal and evoked currents are defined on the left. Gray and black records correspond to records without (-) or with (+) 1 ng Ret-RGS RNA; current scale: 4 μA for - and 8 μA for +. Oocytes were injected with RNA for (in nanograms) m2R (0.24), Gα_o1 (1), and GIRK1/2 channels (0.8 each). For all RNAs in this and the following figures, the amounts are given in nanograms. B, C, Same experiment as in A showing mean and SEM for one experiment. Ret-RGS1 (1) and RGS4 (1) reduced deactivation half-time (t_50% deact) (B) and increased ACh-evoked current (C) significantly. Above each bar is the number of oocytes. For all figures, a single asterisk denotes a significant difference from control (p < 0.05) and double asterisks denote a highly significant difference (p < 0.001). D, Ret-RGS1 increased the evoked current when interacting with the wild-type Gα_o [normalized average of 3 experiments; oocytes were injected with RNA (in nanograms) for m2R (0.2) or mGluR6 (1), Gα_o (0 or 1), GIRK1/2 channels (0.8 each), and Ret-RGS1 (0 or 1)].

Figure 7.

Ret-RGS1 accelerates deactivation of mGluR6 response in a concentration-dependent manner. Pertussis toxin (200 ng) was injected ∼16 hr before the recordings. A, PTX completely diminished responses carried by endogenous Gα_i/o protein. B, Current evoked by 1 mm glutamate versus quantity of injected mGluR6 RNA. Responses first grow linearly and then saturate ∼1 ng mRNA. Oocytes were injected with RNAs (in nanograms) for mGluR6 (as indicated), pertussis toxin-insensitive Gα_o1 C351V (1) and GIRK1/2 channels (0.8 each). C, Ret-RGS1 (0.25 ng) accelerated deactivation. Currents were normalized to the same amplitude to show effect of time course. D, E, Responses to glutamate with increasing amounts of Ret-RGS1 (data pooled from 2 experiments). The parameters quantified are denoted above the bar graphs. Act, activation. Cells were injected with mRNAs for mGluR6 (2), pertussis toxin-insensitive Gα_o1 C351V (1), GIRK1/2 channels (0.8 each), and the indicated amount of Ret-RGS1. For activation and deactivation half-time, data were simply averaged. For currents, data were normalized to responses at 0 Ret-RGS1 as explained in Materials and Methods. Numbers of oocytes denoted in D apply also to E.

Figure 8.

Ret-RGS1 is more potent than RGS4 and RGS7. A, B, Current responses (A) and deactivation half-time (B) in the absence or presence of RGS4, RGS7, or Ret-RGS1. Oocytes were injected with RNAs (in nanograms) for mGluR6 (1), pertussis toxin-insensitive Gα_o1 C351V (1), GIRK1/2 channels (0.8 each), and Ret-RGS1 (1), RGS4 (1), or RGS7 (1).