In smooth muscle of the gut, Gq-coupled receptor agonists activate preferentially PLC-β1 to stimulate phosphoinositide (PI) hydrolysis and inositol 1,4,5-trisphosphate (IP3) generation and induce IP3-dependent Ca2+ release. Inhibition of Ca2+ mobilization by cAMP- (PKA) and cGMP-dependent (PKG) protein kinases reflects inhibition of PI hydrolysis by both kinases and PKG-specific inhibitory phosphorylation of IP3 receptor type I. The mechanism of inhibition of PLC-β1-dependent PI hydrolysis has not been established. Neither Gq nor PLC-β1 was directly phosphorylated by PKA or PKG in gastric smooth muscle cells. However, both kinases 1) phosphorylated regulator of G protein signaling 4 (RGS4) and induced its translocation from cytosol to plasma membrane, 2) enhanced ACh-stimulated association of RGS4 and Gαq·GTP and intrinsic Gαq·GTPase activity, and 3) inhibited ACh-stimulated PI hydrolysis. RGS4 phosphorylation and inhibition of PI hydrolysis were blocked by selective PKA and PKG inhibitors. Expression of RGS4(S52A), which lacks a PKA/PKG phosphorylation site, blocked the increase in GTPase activity and the decrease in PI hydrolysis induced by PKA and PKG. Blockade of PKA-dependent effects was only partial. Selective phosphorylation of G protein-coupled receptor kinase 2 (GRK2), which contains a RGS domain, by PKA augmented ACh-stimulated GRK2:Gαq·GTP association; both effects were blocked in cells expressing GRK2(S685A), which lacks a PKA phosphorylation site. Inhibition of PI hydrolysis induced by PKA was partly blocked in cells expressing GRK2(S685A) and completely blocked in cells coexpressing GRK2(S685A) and RGS4(S52A) or Gαq(G188S), a Gαq mutant that binds GRK2 but not RGS4. The results demonstrate that inhibition of PLC-β1-dependent PI hydrolysis by PKA is mediated via stimulatory phosphorylation of RGS4 and GRK2, leading to rapid inactivation of Gαq·GTP. PKG acts only via phosphorylation of RGS4.
- regulators of G protein signaling
- G protein-coupled receptor kinase 2
- phospholipase C
- cAMP-dependent protein kinase
- cGMP-dependent protein kinase
agonist-induced contraction of smooth muscle consists of a transient phase initiated by activation of PLC-β1 via Gαq or PLC-β3 via Gβγi, leading to generation of inositol 1,4,5-trisphosphate (IP3) and IP3-dependent Ca2+ release (18). The resultant increase in cytosolic Ca2+ [Ca2+]i activates Ca2+/calmodulin-dependent myosin light chain (MLC) kinase, leading to phosphorylation of MLC20, a prerequisite for smooth muscle contraction (11, 33). This Ca2+-dependent phase is followed by sustained contraction during which MLC20 phosphorylation is maintained by a Ca2+-independent MLC kinase and MLC phosphatase is inhibited via a RhoA-dependent cascade (18, 33).
Relaxation of contracted smooth muscle is mediated by cAMP- or cGMP-dependent protein kinase (PKA or PKG), which acts on several targets in the signaling cascades to induce dephosphorylation of MLC20 (1, 15, 32). Relaxation of initial contraction correlates with the decrease in [Ca2+]i, which could occur as the result of a decrease in IP3 generation and/or inhibition of IP3-induced Ca2+ release (16, 24). Only PKG is capable of phosphorylating IP3 receptor type I (IP3R-I) and inhibiting IP3-induced Ca2+ release, whereas both PKA and PKG are capable of inhibiting IP3 generation (12, 26, 27).
Several studies have examined whether inhibition of IP3 generation reflects inhibitory phosphorylation of PLC-β (2, 17, 37, 38). In various cell lines transfected with G protein subunits and PLC-β isozymes, phosphorylation of Ser26 and/or Ser1105 in PLC-β3 by PKA or PKG inhibited Gαq- or Gβγi-stimulated PLC-β3 activity (37, 38). However, in cell lines or native cells, PLC-β1 is not phosphorylated in vivo or in vitro by PKA or PKG (37). In smooth muscle cells, PLC-β3 is preferentially activated by Gβγi, whereas PLC-β1 is selectively activated by Gαq (20, 22, 25, 28). The mechanism by which PLC-β1 activity is inhibited by PKA or PKG could reflect a proximal step, such as inhibitory phosphorylation of Gαq or stimulatory phosphorylation of a RGS protein.
The strength and duration of Gα·GTP signaling are regulated by a family of GTPase-activating proteins known as regulators of G protein signaling (RGS) (8). The proteins contain a RGS domain, capable of binding to activated Gα subunits and accelerating their intrinsic GTPase activity. Smooth muscle cells of the gut express several RGS proteins, including RGS3, RGS4, RGS6, RGS8, RGS12, and RGS16 (Huang J and Murthy KS, unpublished studies). Activation of Gαq in these cells is regulated by RGS4 (10).
In the present study, we have shown that both PKA and PKG phosphorylated RGS4, enhancing its ability to bind Gαq, and accelerated the intrinsic GTPase activity of Gαq·GTP. Furthermore, PKA, but not PKG, phosphorylated G protein-coupled receptor kinase 2 (GRK2), which contains a RGS-like domain, enhancing its ability to bind Gαq and accelerate the intrinsic GTPase activity of Gαq·GTP. The rapid inactivation of Gαq via phosphorylation of GRK2 and/or RGS4 resulted in inhibition of PLC-β1 activity. The results provide firm evidence that PKA inhibits PLC-β1 activity by phosphorylating both RGS4 and GRK2, thereby enhancing their ability to accelerate deactivation of Gαq·GTP. In contrast, PKG inhibits PLC-β1 activity by phosphorylating RGS4 only.
MATERIALS AND METHODS
Gastric smooth muscle cell culture.
Smooth muscle cells were isolated from the circular muscle layer of rabbit stomach by sequential enzymatic digestion in 25 mM HEPES medium as described previously (19, 24, 28, 29). Briefly, muscle strips were incubated at 31°C for 20 min in HEPES medium with type II collagenase (0.1%) and soybean trypsin inhibitor (0.1%). The partly digested strips were washed, and muscle cells were allowed to disperse spontaneously for 30 min. The cells were harvested by filtration through 500-μm Nitex and then centrifuged twice at 350 g for 10 min. Dispersed smooth muscle cells were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum until they attained confluence and were then passaged once for use in various studies (28, 29). Experiments were performed in both freshly dispersed muscle cells and in cultured muscle cells at confluence in first passage.
RGS4, GRK2, and Gαq mutant constructs.
The following mutants were constructed: 1) RGS4(S52A), which lacks a PKA/PKG phosphorylation site, 2) Gαq(G188S), which binds GRK2 but not RGS protein (5–7), and 3) GRK2(S685A), which lacks a PKA phosphorylation site. These constructs were subcloned separately into the multiple cloning site (EcoRI) of the eukaryotic expression vector pEXV. Recombinant plasmid DNAs (2 μg each) were transiently transfected into cultured smooth muscle cells in first passage with the use of Lipofectamine Plus reagent for 48 h. The cells were cotransfected with 1 μg of pGreen Lantern-1 to monitor expression. Control cells were cotransfected with 2 μg of vector (pEXV) and 1 μg of pGreen Lantern-1 DNA. Transfection efficiency (∼75%) was monitored by the expression of green fluorescent protein using FITC filters (28, 29).
Assay of phosphoinositide hydrolysis.
Phosphoinositide (PI) hydrolysis was measured as total inositol phosphate formation with the use of anion exchange chromatography as described previously (20, 22, 28). Freshly dispersed and cultured smooth muscle cells were labeled with myo-2-[3H]inositol (0.5 μCi/ml) in inositol-free medium. Muscle cells (2 × 106 cells) were treated with isoproterenol or sodium nitroprusside (SNP) for 10 min and then with acetylcholine (ACh) for 60 s in 1 ml of 25 HEPES medium containing 115 mM NaCl, 5.8 mM KCl, 2.1 mM KH2PO4, 2 mM CaCl2, 0.6 mM MgCl2, and 14 mM glucose. The reaction was terminated by the addition of 940 μl of chloroform-methanol-HCl (50:100:1). After extraction with 340 μl of chloroform and 340 μl of H2O, the aqueous phase was applied to DOWEX AG-1 columns. [3H]inositol phosphates were then eluted, and the radioactivity was determined by liquid scintillation.
Protein phosphorylation (RGS4, GRK2, PLC-β1, PLC-β3, and Gαq).
Phosphorylation of specific proteins was determined from the amount of 32P incorporated into each protein after immunoprecipitation with specific antibody to RGS4, GRK2, PLC-β1, PLC-β3, or Gαq. Briefly, cultured muscle cells were incubated with [32P]orthophosphate for 24 h, whereas freshly dispersed cells were incubated with [32P]orthophosphate for 4 h (21). Samples (1 ml) were then incubated with isoproterenol or SNP for 10 min in the presence or absence of inhibitors for PKA (myristoylated PKI) and PKG [Rp-8-(4-chlorophenylthio)guanosine 3′,5′-cyclic monophosphate (Rp-cGMPS)] (21). Cell lysates were separated by centrifugation at 13,000 g for 10 min at 4°C, precleared with 40 μl of protein A-Sepharose, and incubated with RGS4, GRK2, Gαq, PLC-β1, or PLC-β3 antibody for 2 h at 4°C and with 40 μl of protein A-Sepharose for another 1 h. The immunoprecipitates were extracted with Laemmli sample buffer, boiled for 5 min, and separated by electrophoresis on SDS-PAGE. After transfer to polyvinylidene difluoride (PVDF) membranes, [32P]RGS4, [32P]GRK2, [32P]Gαq, [32P]PLC-β1, or [32P]PLC-β3 was visualized by autoradiography, and the amount of radioactivity in the band was measured using liquid scintillation. The results were expressed as counts per minute (cpm).
Immunoblot analysis of Gαq-bound RGS4 and GRK2 proteins.
Smooth muscle cells (3 × 106 cells/ml) treated with ACh in the presence of isoproterenol or SNP were lysed by incubation for 30 min at 4°C in 10 mM Tris (pH 7.5), 50 mM NaCl, 1% Triton X-100, and 60 mM octyl glucoside, and the lysates were centrifuged at 15,000 g for 30 min. The supernatant was precleared by incubation with 40 μl of protein A-Sepharose for 4 h and then incubated overnight with the antibody to RGS4 or GRK2. Protein A-Sepharose was then added and incubated for another 2 h, and the mixture was centrifuged at 13,000 g for 5 min. The immunoprecipitates were washed four times in lysis buffer and boiled in Laemmli buffer. Samples were separated by SDS-PAGE, transferred to PVDF membranes, and probed with antibody to Gαq subunit. After incubation with secondary antibody, the proteins were visualized. The intensity of the protein band on ECL film was determined using densitometry.
For translocation of RGS4 proteins, cells were treated with SNP or isoproterenol for different periods and the amount of RGS4 in cytosol and membrane fractions was quantified by Western blot analysis. Cell homogenates were prepared from freshly dispersed gastric smooth muscle cells in Tris·HCl (pH 7.5) medium containing 5 mM MgCl2, 2 mM EDTA, 250 mM sucrose, 1 mM DTT, 1 mM PMSF, 20 μg/ml of leupeptin, and 20 μg/ml aprotinin. The suspension was centrifuged at 100,000 g for 30 min at 4°C, and the supernatant was collected as the cytosolic fraction. Pellets were suspended in Tris·HCl buffer medium containing 1% Triton X-100 and 1% sodium cholate. The extract was centrifuged at 1,000 g for 15 min, and the supernatant was collected as membrane fraction. Cytosolic and membrane proteins were resolved by SDS-PAGE, electrophoretically transferred onto PVDF membrane, and probed with RGS4 antibody. The bands were identified using enhanced chemiluminescence.
Assay for GTPase activity.
GTPase activity was measured in smooth muscle membranes as described previously (30). Smooth muscle cells were incubated with ACh in the presence or absence of isoproterenol or SNP for various time periods. An equal amount of the membrane protein was incubated with 0.4 μCi of [32P]GTP for 30 min at 31°C. The reaction was terminated by the addition of 750 μl of 5% activated charcoal containing 20 mM phosphoric acid, and after centrifugation at 6,000 g for 15 min, the amount of radioactivity in the supernatant (500 μl) was counted by liquid scintillation. Nonspecific release of 32P in the absence of membrane protein was subtracted from the values. The results were expressed as counts per minute.
[3H]scopolamine binding to smooth muscle cells.
Binding of [3H]scopolamine to dispersed muscle cells was done as described previously (23). Muscle cells were suspended in HEPES medium containing 1% bovine serum albumin. Triplicate 0.5-ml (106 cell/ml) aliquots were incubated for 15 min with 1 nM [3H]scopolamine alone or in the presence of 10 μM ACh. Bound and free radioligands were separated by rapid filtration under reduced pressure through 5-μm polycarbonate Nucleophore filters. Nonspecific binding was calculated as the amount of radioactivity in the presence of 10 μM ACh. [3H]scopolamine binding was measured in the control cells and in cells pretreated for 10 min with isoproterenol or SNP.
Measurement of relaxation in dispersed smooth muscle cells.
Inhibition of ACh-induced contraction (i.e., relaxation) by SNP or isoproterenol was expressed as the decrease in maximal cell contraction induced by 0.1 μM ACh (24–26). The mean cell length of 50 muscle cells treated with ACh alone or in the presence of SNP (1 μM) or isoproterenol (1 μM) was measured using scanning micrometry and was compared with the length of untreated muscle cells (mean control cell length 102 ± 3 μM).
Myo-[3H]inositol and [32P]orthophosphate were obtained from NEN Life Sciences Products (Boston, MA). Polyclonal antibodies to Gαq, PLC-β1, PLC-β3, RGS4, and GRK2 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Western blotting and chromatography material were obtained from Bio-Rad Laboratories (Hercules, CA). Collagenase and soybean trypsin inhibitor were obtained from Worthington Biochemical (Freehold, NJ). All other reagents were obtained from Sigma.
Inhibition of ACh-stimulated PI hydrolysis by PKA and PKG.
Two relaxant agents were used to determine the ability of PKA and PKG to inhibit ACh-stimulated PI hydrolysis: SNP, which stimulates cGMP and selectively activates PKG at concentrations <1 μM, and isoproterenol, which stimulates cAMP and selectively activates PKA at concentrations <1 μM (21, 24). These agents were used in the presence or absence of selective inhibitors of PKA (myristoylated PKI14–22 amide) and PKG (Rp-cGMPS) to evaluate the involvement each kinase in inhibition of PI hydrolysis.
As shown previously, ACh acts via Gq-coupled m3 receptors to activate PLC-β1 and via Gi3-coupled m2 receptors to activate PLC-β3 (28). In the presence of the m2 receptor antagonist methoctramine, ACh caused a concentration-dependent increase in PI hydrolysis in freshly dispersed smooth muscle cells (Fig. 1). Pretreatment of the cells with isoproterenol or SNP inhibited ACh-stimulated PI hydrolysis: the inhibition by isoproterenol was selectively blocked by myristoylated PKI, whereas the inhibition by SNP was selectively blocked by Rp-cGMPS (Fig. 1). Inhibition of maximally stimulated PI hydrolysis by isoproterenol (EC50 25 nM) and SNP (EC50 60 nM) was concentration dependent. These results suggested that selective activation of PKA by isoproterenol and PKG by SNP inhibited PI hydrolysis. Inhibition of maximally stimulated PI hydrolysis by isoproterenol and SNP was time dependent: inhibition was significant within 2 min (61 ± 5% with SNP and 68 ± 5% with isoproterenol) and reached a peak within 5 min (78 ± 6% with SNP and 82 ± 4% with isoproterenol) that was sustained for 10 min.
To determine the level at which PKA or PKG exerted its effect, we measured m3 receptor binding, Gαq and PLC-β phosphorylation. Treatment of dispersed smooth muscle cells with isoproterenol (1 μM) or SNP (1 μM) had no effect on [3H]scopolamine binding (control specific binding: 1,835 ± 265 cpm; SNP: 1,769 ± 302 cpm; isoproterenol: 1,802 ± 274 cpm) and did not induce phosphorylation of Gαq or PLC-β1. Both agents, however, induced phosphorylation of PLC-β3 (data not shown) as reported previously (37, 38).
Phosphorylation of RGS4 by PKA and PKG and stimulation of RGS4:Gαq·GTP association.
Both isoproterenol and SNP phosphorylated RGS4 in freshly dispersed smooth muscle cells. Phosphorylation by isoproterenol was inhibited by myristoylated PKI but was not affected by Rp-cGMPS, whereas phosphorylation by SNP was inhibited by Rp-cGMPS but was not affected by myristoylated PKI (Fig. 2).
Isoproterenol and SNP also caused time-dependent translocation of RGS4 from the cytosol to the membrane fraction in the absence of ACh; the translocation was blocked by myristoylated PKI and Rp-cGMPS, respectively (Fig. 3). Translocation of RGS4 in the absence of ACh did not cause greater association of RGS4 and Gαq. In the presence of ACh, however, translocation of RGS4 to the membrane upon phosphorylation by PKA or PKG resulted in greater association of RGS4 with membrane-bound, activated Gαq·GTP (Fig. 4). The pattern demonstrated that RGS4 selectively bound to activated Gαq·GTP.
Augmentation of ACh-stimulated Gα·GTPase activity by PKA and PKG.
ACh stimulated GTPase activity in freshly dispersed smooth muscle cells in a time-dependent fashion. Pretreatment of the cells with isoproterenol or SNP significantly augmented ACh-stimulated GTPase activity (Fig. 5). The pattern reflected phosphorylation of RGS4 and its translocation to the membrane, resulting in increased association of RGS4 and Gαq·GTP and acceleration of Gαq·GTPase activity. The enhanced inactivation of Gαq resulted in decreased activation of PLC-β1.
Effect of RGS4 and Gαq mutants on ACh-stimulated Gαq:RGS4 association and GTPase and PLC-β1 activities.
The role of RGS4 in mediating the inhibitory effect of PKA and PKG on PLC-β1 activity was examined further in cultured smooth muscle cells expressing RGS4(S52A), which lacks the PKA/PKG phosphorylation site. ACh-stimulated PI hydrolysis is control cultured muscle cells (2,596 ± 362 cpm/mg protein above basal levels), and muscle cells expressing RGS4(S52A) (2,935 ± 321 cpm/mg protein) was not significantly different from PI hydrolysis in freshly dispersed smooth muscle cells (2,601 ± 402 cpm/mg protein). Expression of RGS4(S52A) blocked the increase in ACh-stimulated GTPase activity and the decrease in PI hydrolysis induced by PKA and PKG (Fig. 6, A and B). Blockade of PKG-dependent effects was complete, whereas blockade of PKA-dependent effects was only partial. The results implied that inhibition of PI hydrolysis by SNP was solely due to PKG-mediated phosphorylation of RGS4 and subsequent inactivation of Gαq, whereas inhibition of PI hydrolysis by isoproterenol was only partly due to PKA-mediated phosphorylation of RGS4 and subsequent inactivation of Gαq.
The increase in RGS4:Gαq·GTP association induced by PKA or PKG was blocked in cells expressing RGS4(S52A), implying that the increase in association was dependent on RGS4 phosphorylation (Fig. 7). Expression of Gαq(G188S), which does not bind RGS4 (5–7), prevented Gαq:RGS4 association (data not shown) and augmented ACh-stimulated PI hydrolysis by ∼50% (Fig. 8). In these cells, PKG did not inhibit PI hydrolysis, whereas PKA caused only partial inhibition (Fig. 8). The partial inhibition reflected the operation of an additional RGS4-independent mechanism (see below).
Selective phosphorylation of GRK2 by PKA: a role for the RGS domain of GRK2.
GRK2, a serine/threonine kinase that initiates homologous desensitization by phosphorylating ligand-occupied receptors, is a substrate for phosphorylation by PKA but not by PKG in smooth muscle cells (see below). In addition to a central catalytic domain, GRK2 contains a COOH-terminal pleckstrin homology domain that binds Gβγ and an NH2-terminal RGS domain that binds Gαq and can potentially inhibit PLC-β activity (3, 31, 34). Our group recently showed that phosphorylation of GRK2 by PKA augments GRK2-mediated desensitization of VIP (VPAC2) receptors (39). We postulated that selective phosphorylation of GRK2 by PKA stimulated the ability of the RGS domain in GRK2 to bind Gαq and enhance Gαq·GTPase activity.
Isoproterenol, but not SNP, induced GRK2 phosphorylation and increased GRK2:Gαq·GTP association in the presence of ACh to the same extent in freshly dispersed and cultured smooth muscle cells. Both GRK2 phosphorylation and GRK2:Gαq·GTP association were blocked in cultured muscle cells overexpressing GRK2(S685A) (Fig. 9). The mutated Ser685 site is one of three potential (RXS685) sites for phosphorylation by PKA in the COOH terminus of GRK2. Treatment of the cells with isoproterenol or ACh alone did not increase GRK2:Gαq·GTP association, implying that GRK2 was selectively bound to activated Gαq·GTP and the association was dependent on GRK2 phosphorylation by PKA (Fig. 9).
The increase in ACh-stimulated Gα·GTPase activity induced by isoproterenol was only partly inhibited in cells expressing GRK(S685A) (Fig. 10). The residual stimulation reflected the contribution of RGS4. The increase in ACh-stimulated Gα·GTPase activity induced by SNP was not affected in cells expressing GRK(S685A) (Fig. 10).
Inhibition of ACh-stimulated PI hydrolysis by isoproterenol was partly reversed in cells expressing GRK(S685A); inhibition of PI hydrolysis by SNP was not affected (Fig. 11). The residual inhibition by PKA in cells overexpressing GRK2(S685A) reflected the contribution of RGS4. In support of this notion, coexpression of GRK2(S685A) with RGS4(S52A) completely reversed inhibition of PI hydrolysis induced by isoproterenol (Fig. 12). Further support for this notion was obtained in cells coexpressing Gαq(G188S) and GRK2(S685A), where PKA had no effect on ACh-stimulated PI hydrolysis (Fig. 12). Expression of RGS4, GRK2, and Gαq mutants, singly and in combination, demonstrated that PKA inhibited PI hydrolysis by phosphorylating both RGS4 and GRK2 and enhancing their ability to accelerate inactivation of Gαq·GTP.
Inhibition of ACh-stimulated contraction by SNP or isoproterenol.
Treatment of smooth muscle cells with ACh caused a peak contraction within 30 s. Contraction was measured as the mean decrease in control cell length (control cell length: 102 ± 3 μm; mean decrease in cell length: 34 ± 3 μm). Pretreatment of muscle cells for 2 min with SNP and isoproterenol inhibited ACh-induced contraction by 75 ± 4 and 62 ± 3%, respectively. The inhibition of contraction induced by SNP was selectively blocked by Rp-cGMPS, whereas the inhibition of contraction induced by isoproterenol was selectively blocked by myristolylated PKI.
Inhibition of agonist-induced Ca2+ mobilization in smooth muscle by PKA and PKG reflects the net effect of three processes: a proximal process involving inhibition of IP3 generation by both kinases and two distal, PKG-specific processes involving phosphorylation of IP3R-I, leading to inhibition of IP3-dependent Ca2+ release and phosphorylation of the sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA2), leading to Ca2+ sequestration in sarcoplasmic stores (1, 12, 15, 16, 24, 26, 27, 35) . The mechanism for inhibition of IP3 generation by PKA in particular is not known but could involve phosphorylation of the receptor, G protein, a G protein regulator (i.e., RGS protein), or a PLC-β isozyme.
In smooth muscle of the gut, Gq-coupled receptor agonists preferentially activate PLC-β1 to stimulate PI hydrolysis and IP3 generation (18, 28). As shown previously (37, 38) and as confirmed in the present study, PLC-β1, unlike PLC-β3, is not a substrate for inhibitory phosphorylation by PKA or PKG. Involvement of a RGS protein is suggested by studies in which PKG inhibited PI hydrolysis in aortic smooth muscle cells expressing wild-type but not dominant negative RGS2 (9, 36). PKG phosphorylated RGS2 and inhibited PLC-β1 activity in aortic muscle cells from control but not RGS−/− mice (9, 36). In the present study we have shown that PKA phosphorylates RGS4 and GRK2, which together regulate Gαq activity in smooth muscle cells of the gut, whereas PKG selectively phosphorylates RGS4. The stimulatory phosphorylation of RGS4 and GRK2 accelerates deactivation of Gαq, resulting in inhibition of PLC-β1 activity.
RGS proteins bind to switch regions in the activated Gα subunit and increase its intrinsic GTPase activity (8). The ability of RGS4 and GRK2 to bind Gαq·GTP and enhance GTPase activity was augmented upon phosphorylation of both RGS4 and GRK2 by PKA or of RGS4 only by PKG. The rapid inactivation of Gαq·GTP precluded further activation of PLC-β1. Several lines of evidence support this notion. First, PKG and PKA induced phosphorylation of RGS4, causing its translocation to the plasma membrane and increasing its association with activated Gαq, enhanced ACh-stimulated GTPase activity, and inhibited ACh-stimulated PI hydrolysis. All these effects were dependent on phosphorylation of RGS4 in the case of PKG but only partially dependent on RGS4 in the case of PKA. Expression of phosphorylation-deficient RGS4(S52A) abolished RGS4:Gαq association and blocked completely (for PKG) or partly (for PKA) the increase in GTPase activity and the inhibition of PI hydrolysis. Unlike PKG, PKA induced phosphorylation of GRK2, leading to an increase in GRK2:Gαq association that contributed to the increase in GTPase activity and the decrease in PI hydrolysis; coexpression of phosphorylation-deficient RGS4(S52A) and GRK2(S685A) completely blocked the inhibition of PI hydrolysis induced by PKA.
Expression of a Gαq(G188S), which binds GRK2 but not RGS4 (5, 6), shed further light on the roles of RGS4 and GRK2 in mediating inhibition of PI hydrolysis by PKA and PKG. In the absence of RGS4 binding, ACh-stimulated PI hydrolysis was not inhibited by PKG, and was only partly inhibited by PKA; the partial inhibition by PKA reflected the effect of GRK2, since coexpression of Gαq(G188S) and GRK2(S685A) completely blocked the inhibition induced by PKA. The pattern provided further evidence for the participation of RGS4 and GRK2 in inhibition of PI hydrolysis by PKA. Although earlier studies in cell lines had shown that GRK2 bound to activated forms of Gαq, Gα11, and Gα14 and inhibited Gα-mediated PI hydrolysis (4), the present study is the first to demonstrate that phosphorylation of GRK2 by PKA augments GRK2 binding to Gαq·GTP and enhances the intrinsic GTPase activity of Gαq, resulting in inhibition of PI hydrolysis; furthermore, the effect was specific to PKA.
It is worth noting that ACh stimulated PI hydrolysis and GTPase activity and induced association of Gαq with RGS4 or GRK2 to the same extent in freshly dispersed and cultured gastric smooth muscle cells in first passage. Furthermore, isoproterenol and SNP inhibited PI hydrolysis, increased ACh-stimulated GTPase activity and association of Gαq with GRK2 and/or RGS4 to the same extent in both cell preparations. These similarities provide further confirmation of the functional state of cultured smooth muscle cells.
Relaxation of smooth muscle was mediated by PKA and PKG, and they act on several targets to mediate inhibition of initial and sustained contraction, resulting in MLC20 dephosphorylation and relaxation (18). Relaxation of initial contraction correlated with the decrease in cytosolic Ca2+, which could be achieved by inhibiting IP3 generation and/or by decreasing Ca2+ release from or stimulating Ca2+ uptake into the sarcoplasmic stores. Our previous studies (24, 26, 27) have shown that both PKA and PKG inhibit agonist-stimulated Ca2+ release and that only PKG phosphorylates IP3 receptors and inhibits Ca2+ release. The present studies identify a mechanism for inhibition of IP3 generation by both PKA and PKG via phosphorylation of RGS4 and GRK2 and rapid inactivation of Gαq. The importance of this mechanism is in its location at the inception of the signaling cascade that leads to Ca2+ release and MLC20 phosphorylation during initial contraction. Inhibition of IP3 formation and Ca2+ release by PKA and PKG is relevant only to the initial contraction mediated by Ca2+/calmodulin-dependent activation of MLC kinase and MLC20 phosphorylation (18, 28). Sustained contraction is largely Ca2+ independent and reflects inhibition of MLC phosphatase via RhoA-dependent pathways involving phosphorylation of regulatory proteins MYPT1 and CPI-17 that mediate inhibition of MLC phosphatase (28). Relaxation of sustained contraction is mediated by both PKA and PKG via phosphorylation of RhoA at serine 188 and inhibition of RhoA activity (29).
In summary, we have identified a mechanism by which PKA and PKG inhibit agonist-induced PI hydrolysis in smooth muscle cells. Both PKA and PKG phosphorylate RGS4, stimulating its binding to Gαq·GTP and enhancing its ability to accelerate the intrinsic GTPase activity of Gαq. PKA, but not PKG, also phosphorylates GRK2, leading to a similar enhancement in GTPase activity. The rapid inactivation of Gαq·GTP prevents further activation of PLC-β1 and should lead to inhibition of IP3 generation, IP3-dependent Ca2+ release, and initial muscle contraction (Fig. 13). This appears to be the sole mechanism for relaxation mediated by PKA. Relaxation mediated by PKG involves inhibition of PI hydrolysis and IP3-dependent Ca2+ release, as well as inhibitory phosphorylation of IP3 receptors.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-28300.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society