Phosphorylation and Nuclear Translocation of a Regulator of G Protein Signaling (RGS10)*

Heterotrimeric G proteins are involved in the trans-duction of hormonal and sensory signals across plasma membranes of eukaryotic cells. Hence, they are a critical point of control for a variety of agents that modulate cellular function. Activation of these proteins is dependent on GTP binding to their (cid:1) (G (cid:1) ) subunits. Regulators of G protein signaling (RGS) bind specifically to activated G (cid:1) proteins, potentiating the intrinsic GTPase activity of the G (cid:1) proteins and thus expediting the termination of G (cid:1) signaling. Although there are several points in most G protein controlled signaling pathways that are affected by reversible covalent modification, little evidence has been shown addressing whether or not the functions of RGS proteins are themselves regulated by such modifications. We report in this study the acute functional regulation of RGS10 thru the specific and inducible phosphorylation of RGS10 protein at serine 168 by cAMP-dependent kinase A. This phosphorylation nullifies the RGS10 activity at the plasma membrane, which controls the G protein-dependent activation of the inwardly rectifying potassium channel. Surprisingly, the phosphorylation-mediated attenuation of RGS10 activity was not manifested in an alteration of its ability to accelerate GTPase activity of G (cid:1) . Rather, the phosphorylation event correlates

Heterotrimeric guanine nucleotide-binding proteins (G proteins) 1 mediate a wide range of fundamental cellular events in response to external stimulation of seven-transmembranespanning receptors. The activity of G proteins is regulated by a cycle of GTP binding and hydrolysis catalyzed by the G␣ subunits. In their inactive form, G␣ subunits are bound to GDP and complexes with G␤␥ subunits. Ligand-mediated activation of seven transmembrane receptors induces a conformational change in G␣ subunits, which results in the dissociation of GDP from the G␣ subunit. The relatively high endogenous concentration of intracellular GTP leads to its rapid binding to nucleotide-free G␣ subunits. This results in the dissociation of GTP-G␣ from both G␤␥ subunit and receptor. Both the dissociated GTP-G␣ and G␤␥ subunits have the potential to regulate specific downstream target molecules and transduce ligandmediated signals. Termination of the signal occurs upon hydrolysis of bound GTP to GDP by the intrinsic GTPase activity of the G␣ subunit, resulting in the deactivation of G␣ and promotion of G␤␥ binding to G␣-GDP to re-form the inactive heterotrimeric complex.
The discovery and characterization of the signaling properties of heterotrimeric G proteins revealed that the intrinsic GTPase activity of G␣ subunits in vitro was up to 100-fold slower than the rate of deactivation of some G protein-mediated signals in vivo. This paradox suggested that other factors were involved in mediating the increased rate of deactivation of the G␣ protein. A novel family of proteins, RGS proteins, that regulate G protein signaling by an increase in the rate of GTP hydrolysis by G␣ has recently been identified (1)(2)(3)(4)(5)(6). The structures of the RGS-G␣ complexes (7,8) provided a mechanistic explanation for the attenuation of G protein signaling by RGS proteins. To date, at least 27 mammalian RGS proteins have been identified on the basis of a conserved "RGS domain" (see reviews in Refs. 9 and 10). The RGS protein family has been further divided into six subgroups based on a phylogenetic analysis (11).
There are many questions about the involvement of RGS proteins in G protein-mediated signaling processes. One important question is whether RGS proteins are themselves regulated, perhaps by some type of dynamic post-translational modification such as protein palmitoylation and/or phosphorylation. Several RGS proteins (GAIP, RGS4, and RGS16) are palmitoylated at the amino terminus, which has been implicated in subcellular localization and membrane association (12)(13)(14), although the details of the relationship are unclear. Phosphorylation of Sst2p (yeast RGS protein) in yeast by Fus3p (mitogen-activated protein kinase) was reported to increase the half-life of Sst2p in yeast (15). In this study, biochemical pathways that might regulate the activities of RGS proteins (RGS10 in particular) in modulating G protein-coupled signaling were investigated. Compelling evidence has been obtained that the activity of RGS10 is regulated by cAMPdependent protein kinase A (PKA) phosphorylation. Surprisingly, this attenuation of RGS10 activity was not manifested as a reduction in its ability to accelerate GTPase activity of G␣ proteins but, instead, correlates with a translocation of the RGS into the nucleus.

EXPERIMENTAL PROCEDURES
Clones-The RGS10-S168A clone was generated by site-directed mutagenesis (16). The mammalian expression vector for PKI was kindly provided by G. S. McKnight. All clones expressed in mammalian cells were subcloned into the mammalian expression vector pcDNA3 (Invitrogen).
Protein Expression and Purification-Bacterial expression plasmids for RGS10 proteins were constructed by subcloning RGS10 cDNA into the His tag fusion vector pRSETA (Invitrogen). Plasmids were transformed into the bacterial strain BL21 DE3(pLysS) and induced with 250 M isopropyl-1-thio-␤-D-galactopyranoside for 2 h during exponential growth phase. Cells were lysed in 50 mM Tris-HCl (pH 7.5), 500 mM NaCl, 25 mM imidazole (IMAC-25 buffer), and 0.1% Triton X-100 in the presence of a mixture of protease (Sigma) inhibitors, and then centrifuged at 100,000 ϫ g for 45 min. The supernatant was loaded on to a Ni 2ϩ -nitrilotriacetic acid column (Amersham Pharmacia Biotech), washed with IMAC-25 buffer, and eluted with a linear gradient from 25 to 250 mM imidazole. The fractions containing the hexahistidine-tagged RGS10 proteins were then dialyzed against 50 mM Tris-HCl, pH 7.5, and 5 mM DTT. The dialyzed protein was then loaded on to an anion exchange column (MonoQ, Amersham Pharmacia Biotech) and eluted with a linear gradient from 0 to 300 mM NaCl in 50 mM Tris-HCl (pH 7.5) and 5 mM DTT. The hexa-His-RGS10 purity was Ͼ95%, as determined by Coomassie Blue-stained protein gels.
In Vitro PKA Phosphorylation-Hexahistidine-tagged RGS10 (0.1 mg/ml) was incubated with 180 units/ml catalytic subunit of PKA (New England Biolabs) at 30°C for 15-120 min in 50 mM Tris-HCl (pH 7.5), 10 mM MgCl 2 , and 200 M ATP. For the autoradiographic studies, the reaction mixture was supplemented with ␥-labeled ATP to a final specific activity of 200 Ci/mol. RGS10 phosphorylation was assayed by gel-shift on a SDS-PAGE (12.5% gel) in the presence of reducing agents. The phospho-RGS10 was then separated from the PKA and nucleotides by MonoQ column using the same protocol described above.
Cell Culture, Immunoprecipitation, and Metabolic Labeling Studies-All tissue culture cells were grown in DMEM supplemented with 10% fetal calf serum and antibiotics unless otherwise indicated at 37°C in 5% CO 2 . A Myc epitope tag (MEQKLISEEDL) was fused to the 5Ј end of a cDNA fragment encoding RGS10 by site-directed mutagenesis (5,16) and confirmed by DNA sequencing. The Myc-RGS10 was inserted into the mammalian expression vector pcDNA3 (Invitrogen) and used to transiently transfect human HEK 293 cells (5). Transfected cells were lysed in buffer (PBS, 6 mM MgCl 2 , 1% Nonidet P-40, 0.25% sodium deoxycholate, and protease inhibitor mixture (Sigma)) as described previously (5). Binding of RGS10 to endogenous G␣ i3 loaded with GDP or GDP and AlF 4 Ϫ required the lysates to be incubated for 30 min at 20°C in the presence of GDP (10 M) or GDP (10 mM) and AlF 4 Ϫ (30 M). Loaded lysates were then precleared and immunoprecipitated as described previously (5). Immunoprecipitated material was analyzed by immunoblotting with rabbit polyclonal anti-G␣ i3 antibody (0.1 g/ml, Santa Cruz Biotechnology) and goat polyclonal anti-RGS10 antibody (0.1 g/ml, Santa Cruz Biotechnology) and developed using enhanced chemiluminescence. Equal RGS10 expression was determined by Western blot of the lysates with an anti-RGS10 antibody (Santa Cruz Biotechnology).
HEK 293 cells transiently transfected with RGS10 or RGS10-S168A were metabolically labeled (3 h at 37°C, 5% CO 2 in phosphate-free DMEM, 1% dialyzed normal calf serum) with [ 32 P]orthophosphate (100 Ci/100 l) 24 h after transfection. The cells were then stimulated with 25 M forskolin for 15 min at 37°C. Whole cell lysates were then immunoprecipitated with an anti-RGS10 antibody, resolved by SDS-PAGE, and analyzed by autoradiography and immunoblotting as described above.
Electrophysiology-Two-electrode voltage clamp was used to measure the currents from individual Xenopus oocytes immersed in Barth's solution at room temperature as described previously (17). Xenopus laevis oocytes were injected with a 50-nl solution containing cRNA of m2 muscarinic receptor, G protein-coupled inwardly rectifying K ϩ (GIRK) 1, and GIRK 4. Twelve hours later, cRNA encoding RGS10wt or RGS10-S168A were also injected. Electrophysiological recordings were made 3 days after injection under two-electrode voltage clamp in a S168A; lane 6, RGS10-S168A and forskolin. wt, wild type; SA, S168A. D, effect of PKA-mediated phosphorylation on RGS10 binding to activated G␣ i3 . HEK 293 cells transiently transfected with Myc-tagged RGS10 and RGS10-S168A. The indicated whole cell lysates were incubated with GDP (Ϫ) or GDP and AlF 4 Ϫ . The whole cell lysates were then immunoprecipitated with an anti-Myc monoclonal antibody and analyzed by immunoblotting with polyclonal antisera specific for the endogenous G␣ i3 (upper panel) and a polyclonal antibody specific for RGS10 (lower panel). wt, wild type; SA, S168A.
continuous flow chamber perfused with Barth's solution at room temperature. Oocytes were clamped at Ϫ60 mV. During each recording carbachol (2 M) was applied to each oocyte for 20 s and then washed out with Barth's solution.
Protein kinase A inhibitory peptide (PKI; Calbiochem; 50 nl of 0.2 M dissolved in water) was injected into oocytes co-expressing the GIRK 1/4 channel combination, m2 muscarinic receptor, and, when indicated, with RGS10 30 min prior to recording. For oocytes stimulated with forskolin (Calbiochem; 25 M dissolved in OR2, 82.5 mM NaCl, 2 mM KCl, 1.8 mM CaCl 2 , 1 mM MgCl 2 , 5 mM HEPES, pH 7.5), they were incubated in the forskolin-containing solution for 30 min prior to recording. All current activation and deactivation time constants were determined by fitting current trace to a Boltzman equation.
Immunocytochemistry-HEK 293 cells transiently expressing RGS10 were trypsinized and plated on sterile polylysine-coated coverslips and incubated overnight. Two hours prior to stimulation, the cells were placed in serum-free DMEM and then stimulated with forskolin (25 M) as indicated. Cells were then washed in ice cold PBS and fixed with 4% paraformaldehyde in PBS on ice for 30 min, followed by three washes of PBS. The fixed cells were then blocked with 10% normal donkey serum in PBS on ice for 60 min followed by a single wash in PBS. The cells were then incubated in anti-RGS10 (1 g/ml, Santa Cruz Biotechnology) in 1.5% normal donkey serum for 1-2 h at 4°C, washed three times in PBS, and then incubated for 60 min with anti-goat IgG-FITC or IgG-rhodamine (1 g/ml, Santa Cruz Biotechnology) with 1.5% normal donkey serum. Coverslips were then washed three times in PBS, mounted on glass slides, and examined with a fluorescence confocal microscope.

RESULTS AND DISCUSSION
Specific and Inducible Phosphorylation of RGS10 -Regulation of signaling pathways by post-translational modifications that result in conformational change/or changes in localization of the signaling components is an almost universal phenomenon. To investigate whether human RGS10 might be regulated by post-translational modifications, the primary sequence of RGS10 was analyzed for potential sites of phosphorylation. RGS10 was found to contain one consensus PKA phosphorylation motif (Ser 168 ) (Fig. 1A).
To explore whether PKA could directly phosphorylate RGS10, in vitro kinase assays were performed. A RGS10 derivative with an amino-terminal hexahistidine tag, expressed in Escherichia coli and purified by immobilized metal affinity chromatography (IMAC), was incubated in the presence of a recombinant catalytic subunit of PKA and [␥-32 P]ATP. This study revealed that RGS10 was indeed phosphorylated by PKA (Fig. 1B, top panel, lane 1). To determine whether the Ser 168 identified by primary amino acid sequence analysis was a unique PKA phosphorylation site, this residue was mutated to alanine by site-directed mutagenesis. RGS10-S168A was not phosphorylated when incubated with the catalytic subunit of PKA (Fig. 1B, top panel, lane 4), indicating that this residue Having found that RGS10 could serve as a substrate for PKA, we then assessed whether this specific phosphorylation could be detected in intact cells. Human embryonic kidney 293 (HEK293) cells were transfected with RGS10 cDNA and subjected to 32 P metabolic labeling. Phosphorylation of RGS10 was observed under these conditions (Fig. 1C). Phosphorylation of RGS10 was induced by incubation of cells expressing RGS10 with forskolin (Fig. 1C, lane 4), providing evidence that the observed phosphorylation was a PKA-dependent event. Furthermore, co-expression of a dominant-negative PKA with RGS10 inhibited the forskolin-induced phosphorylation of RGS10 (Fig. 1C, lane 3). No PKA-inducible phosphorylation of the RGS10-S168A protein was observed, although there was some apparent background phosphorylation of RGS10 at site(s) other than the Ser 168 (Fig. 1C, lanes 5 and 6).
RGS proteins are thought to function primarily as GTPaseactivating proteins for heterotrimeric G␣ subunits by specifically binding to the activated GTP-bound G␣ and stabilizing the transition state for GTP hydrolysis (7,20). We therefore assessed whether RGS10-S168A could physically associate with activated G␣ by performing co-immunoprecipitation experiments using cells transfected with RGS10 or RGS10-S168A. To facilitate immunoprecipitation in these experiments, a 10-residue epitope tag derived from human Myc protein was inserted immediately after the initiation methionine of RGS10. Myc-RGS10 or Myc-RGS10-S168A was expressed by transient transfection of HEK293 cells. Proteins were recovered by immunoprecipitation of whole cell lysates treated with GDP or GDP and AlF 4 Ϫ with an anti-Myc monoclonal antibody, and analyzed by immunoblotting with a polyclonal antibody specific for the endogenous G␣ i3 . The Myc-RGS10 protein co-immunoprecipitated the G␣ i3 protein from the whole cell lysates treated with GDP and AlF 4 Ϫ (Fig. 1D, lane  2), but failed to interact with wild-type G␣ i3 from whole lysates treated with GDP (Fig. 1D, lane 1). Similarly, Myc-RGS10-S168A co-immunoprecipitated only activated G␣ i3 protein (Fig.  1D, lane 4). This finding confirmed that the S168A variant was produced as a functional protein and thus that its inability to serve as a substrate for PKA was not due to misfolding.
Another possible post-translation modification that may impinge on RGS10 function is palmitoylation of Cys 6 of RGS10, as has been reported previously for other RGS proteins, GAIP, RGS4, and RGS16 (12)(13)(14). RGS10 was transiently expressed in HEK 293 cells metabolically labeled with [ 3 H]palmitic acid. There was no evidence that immunoprecipitated RGS10 was palmitoylated (data not shown).

RGS10 Modulates GIRK Activation and Deactivation by m2
Muscarinic Receptor-GIRK channels are activated directly by G␤␥ subunits released from G proteins of the G␣ i family (21,22). However, discrepancies between native and recombinant GIRK channel kinetics activated by G protein-coupled receptors have also been reported (23,24). Recent studies have resolved this discrepancy by the demonstration of an acceleration of both activation and deactivation of GIRK 1/2 and GIRK 1/4 currents by several RGS proteins (25)(26)(27)(28).
We utilized GIRK channels to determine whether RGS10 was also capable of accelerating G protein-mediated modula-  tion of GIRK currents. Copy RNAs encoding the m2 muscarinic receptor and GIRK 1/4 were co-expressed in X. laevis oocytes in the presence or absence of injected RGS10 cRNA. Three days after injection of the cRNAs, electrophysiological recordings were taken under two-electrode voltage clamp in a continuous flow chamber. RGS10 significantly accelerated GIRK 1/4 activation (2-fold) and deactivation (6-fold) kinetics following application of a m2 muscarinic receptor agonist (carbachol, 2 M) to the bath for 20 s (Fig. 2). The absolute time constants for activation (tau on) and deactivation (tau off) determined in this study are similar to those obtained from a rat atrial myocyte GIRK current (25), as well as for the time constants reported for GIRK 1/2 and GIRK 1/4 currents in the presence of RGS1, -2, -3, -4, -5, -7, and -8 (25)(26)(27)(28).
Acceleration of G protein-mediated modulation of GIRK current by RGS proteins provides an ideal model to study potential functional regulation of RGS proteins through post-translation modifications, such as the PKA-mediated phosphorylation of RGS10 observed here. To investigate the functional effect of preventing PKA-mediated phosphorylation of RGS10 on agonist-induced GIRK current kinetics, electrophysiological recordings were taken from oocytes co-expressing RGS10-S168A with m2 muscarinic receptor and GIRK 1/4 ( Fig. 2A, bottom  panel). When compared with control and RGS10wt, RGS10-S168A further accelerated GIRK 1/4 activation (3-fold, p Ͻ 0.05) and deactivation (14-fold, p Ͻ 0.05) kinetics following application of a m2 muscarinic receptor agonist (carbachol, 2 M) to the bath for 20 s (Fig. 2). When compared with cells expressing wild-type RGS10, a significant (p Ͻ 0.05) increase in the rate of GIRK channel activation and deactivation was observed in oocytes expressing RGS10-S168A (Fig. 2B). Western blot analysis of oocyte lysates showed no difference in the expression levels of wild-type RGS10 and the RGS10-S168A species (data not shown). The difference observed between wild-type RGS10 and RGS10-S168A on GIRK channel activation and deactivation kinetics is likely due to the high basal levels of PKA in X. laevis oocytes (29). These data suggest that the phosphorylation status of Ser 168 of RGS10 may modulate RGS10's molecular interaction with activated G␣ i proteins.

Modulation of PKA Activity Affects the Kinetics of m2 Muscarinic Receptor-evoked GIRK Current Recorded from X. laevis
Oocytes Expressing RGS10 -We next examined more directly whether PKA phosphorylation has a functional consequence on RGS10 action on GIRK channel kinetics by modulating PKA activity in oocytes with an activator (forskolin) and an inhibitor (PKI). Stimulation of PKA activity, by preincubation with forskolin of Xenopus oocytes co-expressing m2 muscarinic receptor, GIRK 1/4 channels, and RGS10 resulted in an inhibition of the apparent activity of the RGS as observed by the GIRK current trace (Fig. 3A, middle trace). Inhibition of PKA activity with PKI led to an apparent enhancement of RGS10 activity toward agonist-mediated GIRK current (Fig. 3A, bottom trace). Furthermore, forskolin treatment inhibited the ability of RGS10 to accelerate the deactivation of GIRK channel currents in response to carbachol (Fig. 3C, bar 2 versus bar 5). In contrast, introduction of PKI resulted in an apparent increase in RGS10's activity of accelerating both the activation (tau on) and deactivation (tau off) rates of GIRK current (Fig. 3, B and C, bar 2 versus bar 8). These changes in GIRK channel kinetics were only significantly observed when wild-type RGS10 was expressed in the oocyte. The tau off of GIRK current for oocytes co-expressing m2 muscarinic receptor, GIRK channel, and RGS10-S168A was increased in the presence of forskolin compared with the absence of forskolin (Fig. 3C, bar 3 versus bar 6). However, the rate of deactivation (tau off) was still significantly more rapid (Ͼ3-fold) than in oocytes not expressing RGS10 in the presence if forskolin (Fig. 3C, bars 1 and 4 versus  bar 6). Additionally, there was no significant change in the GIRK channel kinetics of oocytes expressing RGS10-S168A when PKI was introduced into the cell when compared with control cells lacking RGS10 (Fig. 3, B and C, bar 3 versus bar 9). No significant difference was observed between control and RGS10 groups (Fig. 3B, bar 1 versus bar 2). We believe this lack of significance for tau on between control and RGS10 groups in Fig. 3B is reflective of experimental variation between days, frogs, and oocyte injections. This variation may be a result of differences in oocyte endogenous PKA expression. The lack of significance, however, does not detract from the major observation of the effect of modulating of PKA activity on the tau off of m2 muscarinic receptor-evoked GIRK 1/4 current recorded from oocytes expressing wild-type RGS10. These data suggest that modulation of PKA activity directly affects the functional activity of RGS10 proteins on GIRK channel kinetics and that this modulation is mediated through phosphorylation of Ser 168 . Phosphorylation by PKA Does Not Alter GAP Activity of RGS10 -To investigate whether phosphorylation of RGS10 by PKA modifies the ability of RGS10 to accelerate GTP hydrolysis by G␣ proteins, we assayed the ability of phosphorylated RGS10 (in vitro by PKA) to potentiate the GTP hydrolysis rate of a G␣ subunit in vitro (5). Purified RGS10 was incubated with either the catalytic subunit of PKA or heat-inactivated catalytic subunit. Following each reaction, the RGS10 was separated from the PKA by anion-exchange chromatography. Phosphorylation of RGS10 by PKA resulted in a gel-shift on SDS-PAGE (Fig. 4A, lane 2) that was not observed for RGS10 incubated with inactivated PKA (Fig. 4A, lane 3), suggesting that the phosphorylation was stoichiometric. This phosphorylation had essentially no effect on RGS10's ability to accelerate the GTP hydrolysis rate of G␣ (Fig. 4B). To explore further if the ability of RGS10 to accelerate GTP hydrolysis of G␣ was modified by phosphorylation of Ser 168 by PKA, a time-course analysis was conducted (Fig. 4C). Addition of 50 nM RGS10 caused a Ͼ10-fold enhancement of the rate GTP hydrolysis by G␣ o , and again this activity was unaffected by PKA phosphorylation of RGS10. Evidently, the effects of PKA dependentphosphorylation of RGS10 on GIRK channel deactivation are not the result of a change in GAP activity of RGS10.
PKA Activation Induces Translocation of RGS10 to the Nucleus-The subcellular localization of various RGS proteins has been reported recently. RGS3, RGS7, and GAIP have been found to be cytosolic (30 -33), and sst2, GAIP, RGS3, and RGS16 are all reportedly plasma membrane-associated (3,12,14,31). Dulin et al. (34) recently demonstrated that RGS3 translocates from the cytosol to the plasma membrane upon activation of G␣ proteins and that a truncated variant of RGS3 (RGS3T) is localized to the nucleus, which coincides with an induction of apoptosis. Additionally, Chatterjee and Fisher (35) recently reported the accumulation of RGS2 and RGS10 in the nucleus of COS-7 cells transfected with green fluorescent protein constructs of these proteins.
We examined the subcellular localization of RGS10 by immunocytochemistry and confocal microscopy after stable transfection into HEK293 cells. RGS10 was predominately localized in a diffuse manner in the cytosol (Fig. 5). However, treatment of cells with forskolin (25 M for 2 h) resulted in a dramatic translocation of RGS10 from the cytosol to the nucleus (Fig. 5). The time points utilized for each approach were validated through carefully controlled experiments. For both the co-immunoprecipitation and electrophysiology studies, there was no observed difference between the 15-and 30-min time points (data not shown). Variability was observed with the immunolocalization studies, although we did always observe a discernible difference between the control and the 15-min forskolin stimulation time points (data not shown). However, the 2-h time points provided consistent results, and thus these data were chosen for presentation (Fig. 5).
Investigation of the subcellular distribution of RGS10-S168A required transient transfection of HEK293, since numerous attempts to generate a cell line that stably expresses RGS10-S168A were all unsuccessful. Immunostaining for RGS10-S168A in transfected HEK293 cells revealed both cytoplasmic and nuclear staining (Fig. 5); however, no difference in the distribution of RGS10-S168A was observed when the cells were stimulated with forskolin (Fig. 5).
Having found that RGS10 could translocate to the nucleus, we then assessed whether this localization could be detected in cells that endogenously express RGS10. To date endogenous RGS proteins had not been detected by immunostaining of cells, which suggests the endogenous expression of RGS proteins is very low. In H4 cells, there is predominately nuclear staining with some cytosolic staining, suggesting that most of the endogenous RGS10 protein is localized in the nucleus (data not shown; Ref. 35). These data demonstrate that nuclear localization of RGS10 is not a product of overexpression and is further supportive evidence of a functional role for RGS10 in the nucleus.
These results, coupled with the PKA-mediated modulation of GIRK channel kinetics through RGS10 with minimal effect on its GAP activity, suggest that activation of PKA results in the translocation of RGS10 from the cytoplasm into the nucleus where it can no longer participate in regulating G␣ at the plasma membrane. Hence, we propose the model, as supported by our findings, that the phosphorylation-mediated attenuation of RGS10 activity (Figs. 2 and 3) was not manifest in an alteration of its ability to activate the GTPase of G␣ (Fig. 4). Instead, the phosphorylation event triggered translocation of FIG. 5. RGS10 translocation from the cytosol into the nucleus. RGS10 and RGS10-S168A were transiently expressed in HEK 293 cells and grown on polylysine-coated coverslips. Two hours before fixation, the cells were placed in serum-free medium and followed by treatment without or with forskolin (25 M for 120 min). Cells were then fixed, and RGS10 was detected by immunofluorescence cell staining. The stained cells were then analyzed by confocal microscopy with the three rows of panels being different focal planes of representative cells. wt, wild type; SA, S168A. The bar in the bottom right corner of each panel represents 10 m.
RGS10 from the plasma membrane and cytosol into the nucleus (Fig. 5). An alternative model, that PKA phosphorylation blocks export of RGS10 analogous to the nuclear shuttling of the yeast scaffold Ste5, was recently reported by Mahanty et al. (36). However, it still remains unclear as to whether the direct phosphorylation of RGS10 mediates this translocation or if an accessory protein is activated through PKA phosphorylation that is necessary for RGS10 translocation away from the plasma membrane.
The discovery of RGS proteins provided a novel mechanism for the regulation of signaling through heterotrimeric G proteins, and has raised many questions. For example, knowledge of how RGS proteins are themselves regulated is likely to be crucial to the understanding their intrinsic functions. The goal of this work was to explore the potential regulation of RGS10 by covalent modification. We found that at least one such modification does indeed occur on RGS10, specifically phosphorylation by PKA. Even though the phosphorylation of RGS10 had a profound affect on its ability to modulate ion channel activation by G proteins, it did not have any effect on the ability of RGS10 to accelerate the GTPase activity of G␣ subunits. Instead, the data indicate that PKA phosphorylation of RGS10 results in translocation of the proteins from the cytosol into the nucleus, which would result in the sequestration of RGS10 interactions with activated G␣ subunits at the plasma membrane. Although the experiments described here do not address whether RGS10 has specific functions in the nucleus, the inability to generate stable cell lines expressing RGS10-S168A suggests that the phosphorylation of Ser 168 may be critical in the functioning of RGS10. Although exploring potential functions of RGS10 in the nucleus is beyond the scope of this current study, our finding of the inducible translocation of RGS10 from the cytosol into the nucleus does suggest that the modulation of GAP activities alone does not constitute the only important target for biological regulation of RGS proteins.