Endogenous G protein-coupled receptor kinase 6 Regulates M3 muscarinic acetylcholine receptor phosphorylation and desensitization in human SH-SY5Y neuroblastoma cells.

We have previously shown that overexpression of G protein-coupled receptor kinase 6 (GRK6) enhanced the phosphorylation and desensitization of the endogenously expressed M(3) muscarinic acetylcholine (mACh) receptor in human SH-SY5Y neuroblastoma cells. In this study we have examined the potential role of endogenous GRK6 in the regulation of M(3) mACh receptor by blocking its action through the introduction of a kinase-dead, dominant-negative GRK6 ((K215R)GRK6). (K215R)GRK6 expression inhibited methacholine-stimulated M(3) mACh receptor phosphorylation by 50% compared with plasmid transfected control cells. Guanosine-5'-O-(3-[(35)S]thio)triphosphate binding and immunoprecipitation studies, conducted after agonist pretreatment (3 min), indicated that M(3) mACh receptor-G alpha(q/11) uncoupling was attenuated by 50% in cells expressing (K215R)GRK6 when compared with control cells. In contrast, expression of the related dominant-negative kinase (K215R)GRK5 had no effect on M(3) mACh receptor phosphorylation or uncoupling. Time course studies also showed that agonist-stimulated [(3)H]inositol phosphate accumulations were more sustained in cells expressing (K215R)GRK6 compared with control and (K215R)GRK5-expressing cells, whereas (K215R)GRK6 expression had no effect on the phospholipase C response to direct stimulation of G proteins with AlF(4)(-). The ability of (K215R)GRK6 to inhibit agonist-mediated M(3) mACh receptor phosphorylation and G protein uncoupling suggests that endogenous GRK6 mediates, at least in part, M(3) mACh receptor desensitization in the SH-SY5Y cell line.

The majority of G protein-coupled receptors desensitize when stimulated continuously or repetitively. This process can be initiated via phosphorylation of the receptor, which leads to receptor-G protein uncoupling (1)(2)(3) and reduced receptor signaling to downstream pathways. Second messenger-regulated kinases, such as PKC 1 (4) and cAMP-dependent protein kinase (5,6), and specific G protein-coupled receptor kinases (GRKs) (7,8) have been implicated in G protein-coupled receptor desensitization through phosphorylation of serine and threonine residues within the third intracellular loop or C-terminal tail of G protein-coupled receptors. Upon agonist stimulation, the human M 3 muscarinic acetylcholine (mACh) receptor is rapidly phosphorylated (9), and a number of different kinases have been implicated in this process, including PKC (10,11) and casein kinase 1␣ (CK1␣) (12). However, agonist-stimulated receptor phosphorylation by either CK1␣ or PKC does not appear to mediate rapid desensitization of the M 3 mACh receptor (10 -12).
More recently, we have examined the potential involvement of GRKs in the desensitization of the M 3 mACh receptor endogenously expressed in the SH-SY5Y cell line (13). This human neuroblastoma expresses GRKs 3 and 6, and overexpression of GRK3 and GRK6 leads to enhanced M 3 mACh receptor phosphorylation and reduced activation of phospholipase C (PLC). However, only GRK6 overexpression enhanced uncoupling of the receptor from G␣ q/11 assessed by a [ 35 S]GTP␥S binding/ immunoprecipitation protocol. In contrast, only GRK3 overexpression suppressed AlF 4 Ϫ activation of PLC (13). These data strongly support an action of GRK3 independent of phosphorylation, possibly via direct binding to activated G␣ q/11 and/or free G␤␥ subunits, but highlight a potential role for GRK6 via receptor phosphorylation and uncoupling of G␣ q/11 . Although GRK6 overexpression can be shown to enhance the M 3 mACh receptor desensitization process in SH-SY5Y cells, this experimental approach is unable to resolve whether endogenous GRK6 contributes to M 3 mACh receptor regulation. Therefore, in an attempt to block the actions of endogenous GRK6, we have introduced a kinase-dead, dominant-negative mutant form of GRK6, created by introducing a K215R point mutation into the ATP-binding domain, into SH-SY5Y cells to create cell lines stably expressing this construct. Using this approach we have shown that K215R GRK6 expression inhibits the phosphorylation and subsequent desensitization of the M 3 mACh receptor. Our data suggest that GRK6 plays a significant role in the desensitization of the endogenously expressed M 3 mACh receptor in human SH-SY5Y neuroblastoma cells.

EXPERIMENTAL PROCEDURES
Cell Culture and Creation of Stably Transfected Dominant-negative GRK Cell Lines-SH-SY5Y human neuroblastoma cells were cultured in minimal essential medium containing 5% fetal and 5% new born calf serum, penicillin (100 units/ml), streptomycin (100 g/ml), and fungizone (2.5 g/ml) (Invitrogen). All of the cells were maintained at 37°C in humidified conditions under 5% CO 2 . Wild-type SH-SY5Y cells were transfected with either pcDNA3 alone or human dominant-negative GRK5 or GRK6, both with a single point mutation at K215R, cloned into pcDNA3 at BamHI and XbaI, or EcoRI for GRKs 5 (14) and 6, respectively (both kindly provided by E. Kelly, University of Bristol, Bristol, UK), using FuGENE 6 according to the manufacturer's instructions. * This work was supported by Wellcome Trust Grant 062495. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
After 48 h, geneticin (300 g/ml) was added to the cells. The surviving colonies were selected and expanded into cell lines.
Western Blotting-The cells were lysed and subjected to electrophoretic separation exactly as described previously (15,16). The separated protein was transferred to nitrocellulose, and GRK expression was detected using anti-rabbit polyclonal IgG antibodies (1:1000 dilution) specific for GRK2, GRK3, GRK5, or GRK6 (Santa Cruz Biotechnology). G␣ q/11 expression was detected using an anti-rabbit polyclonal IgG (Santa Cruz Biotechnology). CK1␣ was detected using an antirabbit polyclonal as described previously (12). Protein expression was determined by the addition of ECL reagent (Amersham Biosciences), according to the manufacturer's instructions and exposure to Hyperfilm (Amersham Biosciences).
Determination GRK5 expression on the phosphorylation of endogenously expressed M 3 muscarinic receptors was assessed by the method of Tobin and Nahorski (9). Briefly, either plasmid control or cells expressing either K215R GRK6 or K215R GRK5 were seeded into 6-well culture plates. Confluent cells were loaded with [ 32 P]orthophosphate (5 Ci/ml; Amersham Biosciences) in phosphate-free Krebs buffer, pH 7.4, for 1 h prior to agonist challenge (methacholine, 100 M). After 3 min, the agonist was removed, and the cells were solubilized, immunoprecipitated, and electrophoretically resolved as described previously (13). Autoradiograms were documented and quantified using the GeneGenius system and software (Syngene, Cambridge, UK).
Casein Phosphorylation-Any potential kinase activity of dominantnegative GRK5 and GRK6 was assessed by measuring their ability to phosphorylate the substrate ␣-casein as described previously (13).

Measurement of Total [ 3 H]Inositol
Phosphate Accumulation-Either plasmid control or cells expressing K215R GRK5 or K215R GRK6 were seeded into 24-well plates at ϳ50% confluency. After 24 h, the cells were incubated in the presence of [ 3 H]inositol (1 Ci/ml) in geneticin-free medium for a further 24 h. Confluent cell monolayers were then washed twice with Krebs buffer (118.6 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO 4 , 1.2 mM KH 2 PO 4 , 4.2 mM NaHCO 3 , 10 mM HEPES, 11.7 mM glucose, and 1.3 mM CaCl 2 , pH 7.4) and incubated for 15 min at 37°C. LiCl (final concentration, 10 mM) was added to each well for 10 min prior to the addition of methacholine. Reaction termination, sample neutralization, and separation of the [ 3 H]inositol phosphate fraction (containing inositol mono-, bis-, and tris-phosphates) were performed as described previously (17).

Assessment of M 3 mACh receptor G␣ q/11 Coupling by [ 35 S]GTP␥S
Binding-Plasmid control or dominant-negative GRK-expressing cells were grown in 80-cm 2 cell culture flasks until confluent. The cells were then harvested in 10 mM HEPES, pH 7.4, 0.2% (v/v) EDTA, 0.9% (v/v) NaCl. After centrifugation the cells were resuspended in Krebs buffer, pH 7.4, at 37°C for 15 min prior to the addition of either vehicle or MCh (100 M). After 3 min, excess ice-cold Krebs buffer (50 ml) was added, and the cells were pelleted at 1000 ϫ g for 5 min. The cell pellets were then resuspended in 30 ml of 20 mM HEPES, pH 7.4, 10 mM EDTA and homogenized at maximum speed for 30 s using a PT210 Polytron. The resulting suspension was centrifuged at 20,000 ϫ g for 15 min. The pellet was resuspended in 30 ml of 20 mM HEPES, pH 7.4, 0.1 mM EDTA and centrifuged for a further 15 min at 20,000 ϫ g. The resulting pellet was resuspended at 1 mg/ml of protein and stored at Ϫ80°C until required.
[ 35 S]GTP␥S binding and immunoprecipitation were performed as described previously (18,19). Briefly, 50 g of membranes were added to 1 nM [ 35 S]GTP␥S (PerkinElmer Life Sciences), 1 M GDP, 100 M Ϯmethacholine in assay buffer (100 mM HEPES, 100 mM NaCl, 10 mM MgCl 2 , pH 7.4) and incubated for 2 min at 30°C. Nonspecific binding was determined by inclusion of 10 M GTP␥S. Termination, solubilization, and G␣ q/11 -specific immunoprecipitation were performed exactly as described previously (13). Desensitization was determined as a re-duction in [ 35 S]GTP␥S binding after pretreatment with methacholine and expressed as a percentage of the response found when compared with a nonpretreated matched control. In addition to desensitization experiments, M 3 mACh receptor activation of G␣ q/11 was assessed in nonpretreated cell membranes via generation of methacholine (300 nM to 1 mM) concentration-response curves.
Data Analysis-All concentration-response curves were fitted, and the EC 50 values were determined using nonlinear regression analysis (GraphPad Prism 3). All data were analyzed using one-or two-way analysis of variance (Microsoft Excel version 5). Significance was accepted when p Ͻ 0.05.

Creation of Stable Dominant-negative GRK Cell Lines-
Transfection of wild-type SH-SY5Y neuroblastoma cells with either K215R GRK6 or K215R GRK5 and selection with geneticin (300 g/ml) yielded several surviving colonies, which were expanded into cell lines. Two clones that expressed K215R GRK6 and that were matched with plasmid controls for receptor number based on [ 3 H]NMS binding were selected for further study (named D1 and D5; Fig. 1A). Estimation of the level of K215R GRK6 expression was determined by serial dilution of cell lysates and subsequent Western blotting, indicating ϳ30-fold greater levels than endogenous GRK6 expression (data not shown). In an attempt to examine the specificity of the effects of K215R GRK6, a single clone expressing the closely related K215R GRK5 (D2; Fig. 1B) was chosen for some experiments. The level of overexpression of K215R GRK5 was difficult to assess because there was no detectable endogenous GRK5 in SH-SY5Y cells. To determine whether K215R GRK6 lacked kinase activity (i.e. was kinase-dead), HEK293 cells were transiently transfected with empty vector, GFP-tagged wt-GRK6, or GFPtagged K215R GRK6. After 48 h the kinases were immunoprecipitated with either GFP polyclonal or wt-GRK6 antibodies. Kinase activity was assessed by addition of dephosphorylated  6). The cells were lysed and mixed with GFP polyclonal antibody (lanes 1, 2, and 4 -6) to immunoprecipitate GFP-tagged kinases, before the addition of [␥-32 P]ATP and dephosphorylated ␣-casein for 10 min at 37°C. GRKs were removed from the reaction mix by centrifugation, and the resultant supernatant was separated by SDS-PAGE. ␣-Casein phosphorylation was determined by autoradiography. Lane 3 shows the degree of ␣-casein phosphorylation seen after immunoprecipitation of endogenous wt-GRK6, by wt-GRK6 polyclonal antibody from HEK293 cells. The data are representative of three separate experiments.
␣-casein in the presence of [␥-32 P]ATP (as described under "Experimental Procedures"). No phosphorylation of ␣-casein was detected after transfection with empty vector, followed by immunoprecipitation with GFP antibody (Fig. 1C). As expected, enhanced ␣-casein phosphorylation was detected after transfection of GFP-tagged GRK6. Furthermore, a slight phosphorylation of ␣-casein was detected after immunoprecipitation of endogenously expressed GRK6 from nontransfected HEK293 cells. In contrast, ␣-casein phosphorylation was not evident in the presence of K215R GRK6.
Effects of K215R GRK5 or K215R GRK6 on Expression of Endogenous GRKs and CK1␣-In an attempt to determine whether overexpression of K215R GRK5 or K215R GRK6 altered the expression of other endogenously expressed kinases, we first blotted all the clones used in this study for GRKs 2, 3, 5, and 6, as well as CK1␣ (Fig. 2). Overexpression of K215R GRK5 had no effect on the expression of any of the kinases studied. Furthermore, K215R GRK6 overexpression had no effect on GRK2, GRK3, GRK5, or CK1␣ expression in any of the cell lines. However, it was not possible to assess whether K215R GRK6 altered the expression of endogenous GRK6 because both K215R GRK6 and wild-type GRK6 are not distinguishable with the antibody used (Fig. 2D).
Determination of M 3 Receptor Number-Whole cell binding studies were undertaken using [ 3 H]NMS to determine whether expression of dominant-negative constructs affected M 3 mACh receptor expression. The data obtained for B max and K D indicated that expression of either K215R GRK6 or K215R GRK5 had no significant effect on M 3 mACh receptor expression (Table I). Moreover, receptor expression levels did not alter with passage (data not shown).
M 3 mACh Receptor Phosphorylation-The effects of K215R GRK6 expression on methacholine-stimulated (100 M) M 3 mACh receptor phosphorylation was examined in clones D1 and D5. As shown in Fig. 3 (A and B), after 3 min of agonist exposure a robust phosphorylation was observed in the plasmid control cell lines (P1 and P3). In contrast, M 3 mACh receptor phosphorylation was reduced by ϳ50% in cells expressing K215R GRK6. Furthermore, expression of K215R GRK5 had no effect on agonist-stimulated M 3 mACh receptor phosphorylation (Fig. 3C). Densitometric analysis confirmed that K215R GRK6 had no effect on basal M 3 mACh receptor phosphorylation. However, K215R GRK6 significantly (p Ͻ 0.01) reduced agonist-stimulated M 3 mACh receptor phosphorylation by 50 and 43% for D1 compared with P3 and for D5 compared with P1 clones, respectively (Fig. 4). Furthermore, expression of K215R GRK6 did not affect PKC-mediated M 3 mACh receptor phosphorylation stimulated by phorbol 12,13-dibutyrate (PDBu) (1 M, 3 min; Fig. 3D).

Assessment of M 3 mACh Receptor Desensitization by [ 35 S]GTP␥S
Binding-To determine whether the inhibition of agonist-stimulated M 3 mACh receptor phosphorylation observed with K215R GRK6 expression had a subsequent effect on receptor function, we examined the direct interaction of the M 3 mACh receptor and G␣ q/11 at the point of receptor catalyzed GTP/GDP exchange. Thus, we have used agonist-stimulated [ 35 S]GTP␥S binding followed by immunoprecipitation of G␣ q/11 in a membrane preparation (see "Experimental Procedures").
First, we assessed whether expression of K215R GRK6 had any effects on acute M 3 mACh receptor-G␣ q/11 interaction. Concentration-response curves in membranes from nonpretreated K215R GRK6-expressing clones, undertaken at 2 min, the optimal time point for M 3 mACh receptor-G␣ q/11 activation (13, 19),   showed binding identical to that seen in controls (Fig. 5A). However, as described previously (11,13), pretreatment of intact cells with MCh (100 M) for 3 min leads to reduced agonist-stimulated [ 35 S]GTP␥S binding to immunoprecipitated G␣ q/11 in the subsequent membrane assay (Fig. 5B). Plasmid control cells (P1 and P3) showed 59 (D1 versus P3) and 46% (D5 versus P1) less M 3 mACh receptor-G␣ q/11 coupling when compared with nonpretreated controls. In contrast, the degree of receptor-G␣ q/11 uncoupling was inhibited by 50% with K215R GRK6 expression (Fig. 5B). Interestingly, despite agonist pretreatment, expression of K215R GRK5 had no effect on M 3 mACh receptor-G␣ q/11 uncoupling (Fig. 5B). These effects were not due to a loss of membrane-associated G␣ q/11 because total G␣ q/11 immunoreactivity in the membrane fraction was unaltered after 3 min of pretreatment (Fig. 5C).
Assessment of Phospholipase C Activity-To further study the effects of GRK6 on M 3 mACh receptor regulation, we examined the effects of K215R GRK6 on PLC activity using total [ 3 H]inositol phosphate accumulation. Time course studies indicated that clones expressing K215R GRK6 produced significantly (p Ͻ 0.01) greater [ 3 H]inositol phosphate accumulations than plasmid controls matched for receptor number (Fig. 6, A  and B). In addition, concentration-response curves to MCh produced after 3 min indicate that expression of K215R GRK6 enhanced [ 3 H]inositol phosphate accumulation when compared with plasmid control cells (Fig. 7). Interestingly, expression of the structurally similar dominant-negative kinase K215R GRK5 had no effect on [ 3 H]inositol phosphate accumulation (Fig. 6C)

M 3 mACh Receptor
Internalization-To assess whether GRK6 mediated M 3 mACh receptor internalization plasmid control, GRK6-overexpressing (clone 24) or K215R GRK6-expressing (D1) cells were exposed to MCh (100 M) for up to 1 h. Subsequent [ 3 H]NMS binding revealed that overexpression of GRK6 had no effect on M 3 mACh receptor internalization.
Furthermore, inhibition of endogenous GRK6 with K215R GRK6 failed to inhibit M 3 mACh receptor internalization (Fig. 8). DISCUSSION We have recently shown that recombinant GRK6 can enhance both M 3 mACh receptor phosphorylation and receptor/ G␣ q/11 uncoupling, subsequently increasing desensitization of the endogenously expressed M 3 mACh receptor in the SH-SY5Y neuroblastoma cell line (13). These data are suggestive of a role for endogenous GRK6 in the regulation of M 3 mACh receptor desensitization. Therefore, in an attempt to block the action of endogenously expressed GRK6 in SH-SY5Y cells, we have introduced a kinase-dead, dominant-negative mutant, K215R GRK6.
Our data show for the first time the potential role of endogenous GRK6 in the regulation of an endogenously expressed G␣ q/11 -coupled receptor. Despite the many studies indicating that GRKs are able to enhance the phosphorylation of overexpressed receptors (reviewed in Ref. 1), relatively few have examined the role of endogenous GRK6 activity in the regulation of endogenously expressed receptors. Introduction of antisense GRK6 oligonucleotides inhibited desensitization of the calcitonin gene-related peptide receptor stably expressed in HEK293 cells (20). However, several recent studies have reported the use of differing versions of dominant-negative GRK6 and have provided conflicting evidence as to their effectiveness in inhibiting endogenous GRK6. Lazari et al. (21) were able to inhibit endogenous GRK6 phosphorylation of follitrophin receptors in HEK293 cells by introducing a double point-mutated GRK6 (K215M/K216M). Zhou et al. (22) introduced a triple pointmutated (R215K/D484S/D485S) dominant-negative GRK6 that reduced agonist-stimulated thomboxane A 2 receptor phosphorylation when compared with equivalent overexpression of GRK6 (22). However, unlike the present data, where we show a 50% reduction in agonist-stimulated M 3 mACh receptor phosphorylation with expression of K215R GRK6, the action of endogenous GRK6 appeared not to be inhibited by this dominantnegative ( R215K/D484S/D485S GRK6) (22). In view of these findings we chose to determine the catalytic activity of our dominantnegative GRK6 using immunoprecipitation of GFP-tagged GRK6 or K215R GRK6 and ␣-casein as substrate. These studies indicated that unlike wt-GRK6, K215R GRK6 was devoid of ␣-casein phosphorylating activity.
The finding that certain GRKs can mediate inhibition of PLC activity via nonreceptor phosphorylation mechanisms (23,24) cautions against the use of dominant-negative GRKs. However, the ability of GRK2 and GRK3 to directly bind to activated, GTP-bound G␣ q/11 via their N-terminal RGS domain is not seen with GRK5 and GRK6 (23,24). In addition, direct stimulation of PLC via activation of G proteins using AlF 4 Ϫ , which is inhibited by GRK3 overexpression, was unaffected by either wt-GRK6 (13) or K215R GRK6. Nevertheless, to further examine the specificity of K215R GRK6 to inhibit M 3 mACh receptor desensitization, we introduced the dominant-negative version of GRK5 ( K215R GRK5). Interestingly, despite the close structural similarities between GRK5 and GRK6 (83.5% amino acid similarity for human forms of GRK5 and GRK6) (2), K215R GRK5 did not affect M 3 mACh receptor signaling. This might seem surprising because both GRK5 and GRK6 are membrane-associated and thought to be regulated in a similar manner, i.e. negatively by calmodulin (25) and positively through phosphatidylinositol 4,5-biphosphate binding (26). In addition, evidence from GRK5 knockout mice suggests that GRK5 regulates many muscarinic responses attributable to M 2 receptors in vivo (27). It is possible that the lack of effect of K215R GRK5 on M 3 mACh receptor signaling in SH-SY5Y cells may reflect a lack of wt-GRK5. Although we were unable to detect endogenous GRK5 by Western blotting, immunoprecipitation studies from plasmid control SH-SY5Y cells indicated the presence of a GRK5 immunoprecipitable protein able to phosphorylate ␣-casein (data not shown). However, if the M 3 mACh receptor was able to recruit GRK5, one might expect K215R GRK5 to block access of similar kinases to the receptor after agonist stimulation. The absence of this finding, combined with the inhibition of M 3 mACh receptor phosphorylation by K215R GRK6, suggests that the recruitment signals for GRK5 or GRK6 are likely to be different even for these closely related kinases. It is also of interest that phosphorylation of the M 3 mACh receptor by direct stimulation of PKC (by PDBu) was unaltered by the expression of K215R GRK6. All of these data imply that K215R GRK6-mediated inhibition of M 3 mACh phosphorylation is not due to a nonspecific physical association of receptor and kinase but requires the specific recruitment of GRK6 by the receptor after agonist activation.
Because GRK6 appears to regulate M 3 mACh receptor desensitization, it is interesting to speculate which phosphoac-ceptor sites GRK6 may phosphorylate. Despite the wealth of studies implicating GRK involvement in receptor desensitization, no consensus sequence has been agreed for GRK phosphorylation (3,28,29). Several studies have identified serine or threonine clusters as favored targets for GRK-mediated receptor phosphorylation (30,31). Wu et al. (32) have mapped the phosphorylation sites for GRK2 to two such regions ( 331 SSS 333 and 348 SASS 351 ) in the third intracellular loop of the M 3 mACh receptor. Recent evidence examining the phosphorylation profile of the B 2 bradykinin receptor has indicated that GRK (2, 3, 5 or 6) phosphorylation is limited to three distinct serines situated in the C-terminal tail (33). Closer examination of the individual serine moieties highlighted distinct phosphorylation patterns between GRKs, raising the possibility that different GRKs may regulate different receptor signaling mechanisms. These data suggest that GRK6 and GRK2 M 3 mACh receptor phosphorylation sites may not be mutually exclusive but that phosphorylation patterns may differ both temporally and positionally. However, it is also noteworthy that unlike for the bradykinin B 2 receptor, determination of M 3 mACh receptor phosphorylation patterns is potentially much more complex because of the presence of 36 serine and 16 threonine residues within the third intracellular loop alone. Nevertheless, we have previously shown that although GRK3 and GRK6 enhance M 3 mACh receptor phosphorylation, only GRK6 mediates uncoupling of the receptor from G␣ q/11 (13). Furthermore, inhibition of endogenous GRK6 with a 30-fold excess of K215R GRK6 only produced a 50% decrease in M 3 receptor phosphorylation and G␣ q/11 uncoupling. This finding could suggest either that increasing K215R GRK6 overexpression may further inhibit M 3 mACh receptor phosphorylation and G␣ q/11 uncoupling or more likely that other kinases such as GRK2 or 3 may contribute to M 3 mACh receptor desensitization in SH-SY5Y cells.
Many studies have reported that GRK overexpression not only leads to enhanced desensitization but also accelerated internalization of receptors (reviewed in Refs. 1 and 3). The present data indicate that although GRK6 is capable of desensitizing the M 3 mACh receptor, it appears not to play a significant role in receptor internalization. This finding may appear surprising, but for some receptors kinase-mediated phosphorylation does not always enhance receptor internalization. Indeed, in agreement with our data, Lazari et al. (21) found that although dominant-negative GRK2 or GRK6 are equally capable of inhibiting follitropin receptor phosphorylation, only inhibition of GRK2-mediated phosphorylation prevents receptor internalization. Moreover, our data suggest that the potential pattern of phosphoacceptor sites that mediate M 3 mACh desensitization and internalization may be different. The presence of GRK2, GRK3, CK1␣, and PKCs in SH-SY5Y cells and their ability to phosphorylate the M 3 mACh receptor (10 -13, 32) raise the possibility that one or more of these kinases may be responsible for regulating M 3 mACh receptor internalization. Indeed, recent preliminary studies have shown that dominant-negative CK1␣ inhibits M 1 receptor internalization when concurrently overexpressed in HEK293 cells (34). Furthermore, we have shown that overexpression of GRK3 in SH-SY5Y cells (13), and others have reported that GRK2 in Chinese hamster ovary cells (35) enhanced M 3 mACh receptor internalization. It will be interesting to examine which of these kinases plays a role in the internalization of the endogenously expressed M 3 mACh receptor in SH-SY5Y cells and whether this is crucial for recruitment of adaptor proteins and initiating alternative signaling cascades.
In conclusion we have examined the potential role of endogenously expressed GRK6 in the desensitization of M 3 mACh receptors in the human SH-SY5Y cell line. Introduction of a dominant-negative, catalytically inactive GRK6 ( K215R GRK6) inhibited both agonist-stimulated M 3 mACh receptor phosphorylation and receptor/G␣ q/11 uncoupling. Furthermore, K215R GRK6 expression partially reversed the time-related decay of agonist activation of PLC. In contrast, the closely related K215R GRK5 was unable to affect M 3 mACh signaling in any way. Despite the obvious effects of GRK6 on M 3 mACh receptor PLC-coupled signaling, manipulation of GRK6 activity had no effect on receptor internalization. Overall these data suggest that endogenous GRK6 regulates, at least in part, M 3 mACh receptor desensitization in human SH-SY5Y neuroblastoma cells.