Pasteurella multocida Toxin Facilitates Inositol Phosphate Formation by Bombesin through Tyrosine Phosphorylation of Gαq*

The intracellularly acting Pasteurella multocida toxin (PMT) is a potent mitogen that stimulates Gq-dependent formation of inositol trisphosphate. We show that PMT, a nontoxic mutant of PMT (PMTC1165S), and bombesin each stimulate time-dependent phosphorylation of Gαq at tyrosine 349. Although PMT and PMTC1165S each cause phosphorylation of Gαq, only the wild-type toxin activates Gq. Pretreatment of cells with wild-type or mutant PMT potentiated the formation of inositol phosphates stimulated by bombesin equally. These data show that PMT potentiates bombesin receptor signaling through tyrosine phosphorylation of Gq and distinguishes between the two proposed models of Gq activation, showing that tyrosine phosphorylation is not linked to receptor uncoupling.

The Pasteurella multocida toxin (PMT) 1 is a highly potent mitogen for mesenchymal cells, including Swiss 3T3 fibroblasts (1). Although the primary molecular targets of this intracellularly acting toxin have not been identified, a prominent role for heterotrimeric G proteins has been elucidated (2)(3)(4). The toxin affects several signal transduction pathways, resulting in increased inositol phosphate production, stimulation of protein kinase C activity, Ca 2ϩ mobilization, actin rearrangements, and increased protein tyrosine phosphorylation (5).
Heterotrimeric G proteins are guanine nucleotide-binding proteins that function as molecular switches that transduce signals from G protein-coupled receptors (GPCR) to effector proteins such as enzymes or ion channels (6 -8). The G␣ proteins are divided into four families: G␣ s , G␣ i/o , G␣ q , and G␣ 12 (8,9). The G␣ q class are widely expressed and regulate various effector proteins including phospholipase C␤ and Bruton's tyrosine kinase (10). Activation of GPCRs results in a conformational change in the G␣ subunit, favoring the exchange of bound GDP for GTP. GTP binding results in the dissociation of G␣-GTP and ␤␥ complexes, each of which can modulate effector proteins. The regulation of these processes in vivo has yet to be fully elucidated.
Recently it has been demonstrated that the ␣ subunit of G q is a target for tyrosine phosphorylation. Interestingly, phospho-rylation of G␣ q increased its ability to activate phospholipase C␤ in an in vitro model, suggesting that phosphorylation may modulate the activity of the G protein in vivo (11). Modulation of G␣ q phosphorylation using chemical inhibitors of tyrosine kinases or tyrosine phosphatases had a profound effect on the production of inositol trisphosphates (IP 3 ) in vivo (12,13). Furthermore, transient expression of a dominant active mutant of the Fyn tyrosine kinase elevated the phosphorylation of G␣ q but blocked receptor-stimulated IP 3 production (12). Taken together, these results implied that cellular kinases and phosphatases coordinately regulate the activity of G␣ q .
The experiments presented here were designed to determine whether PMT stimulated activation and tyrosine phosphorylation of G␣ q . We show that PMT is a potent stimulator of G␣ q tyrosine phosphorylation but that this phosphorylation step is not a prerequisite for G␣ q activation. Furthermore, using wildtype and mutant forms of PMT, we show that tyrosine phosphorylation of G␣ q can potentiate signaling through the G qcoupled bombesin receptor.

EXPERIMENTAL PROCEDURES
Cell culture reagents were obtained from Invitrogen. All of the primary antisera were obtained from Santa Cruz Biotechnology, Inc. Horseradish peroxidase-conjugated donkey anti-rabbit and anti-mouse IgG were from Sigma-Aldrich. [␥-35 S]GTP␥S (1 mCi/ml) was obtained from Amersham Biosciences. myo-[2-3 H]Inositol (1 mCi/ml) was obtained from New England Nuclear, Ltd. All other reagents were of the highest available grade from standard commercial sources. SYF Ϫ/Ϫ (CRL-2459) and YF Ϫ/Ϫ (CRL-2498) cells were purchased from the American Type Culture Collection. G␣ q/11 Ϫ/Ϫ double deficient fibroblasts were a generous gift from Professor Stefan Offermanns (Pharmakologisches Institut, Universitä t Heidelberg, Germany).
Preparation of Swiss 3T3 Cell Membranes-Quiescent cultures of Swiss 3T3 cells were rinsed twice with cold phosphate-buffered saline and scraped into phosphate-buffered saline containing 1 mM sodium orthovanadate and proteinase inhibitors. Following centrifugation (200 ϫ g, 10 min, 4°C), washed cell pastes were frozen at Ϫ70°C until required. The frozen cell pastes were thawed on ice and suspended in 5 ml of buffer A (10 mM Tris-HCl, 10 mM MgCl 2 , 0.1 mM EDTA, pH 7.5) containing 1 mM sodium orthovanadate and proteinase inhibitors. The cells were ruptured by 25 passes through a 23-gauge needle, and the resulting homogenate was centrifuged at 800 ϫ g for 10 min to remove * This work was supported by a Medical Research Council studentship (to M. R. B.) and a Wellcome Trust project grant (to A. J. L.). 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.
unbroken cells and nuclei. The supernatants were transferred to fresh tubes and centrifuged at 50,000 ϫ g for 10 min. The pellet was washed and suspended in 10 ml of buffer A containing inhibitors. After a second centrifugation step the membrane pellet was suspended in buffer A to a protein concentration of 3 mg/ml and stored at Ϫ70°C.
Determination of GDP-GTP Exchange on G q -Determination of GTP␥S binding was essentially as described previously (18). Briefly, membranes (30 g) were suspended in 100 l of assay buffer (50 mM Hepes, pH 7.4, 120 mM NaCl, 20 mM MgCl 2 , 2 mM KCl, 1 mM deoxycholate, 20 M GDP, 0.2% bovine serum albumin) and incubated for 10 min at 37°C. Following incubation, an equal volume of assay buffer containing either toxins (140 pM) or bombesin (40 nM) and 2 nM [ 35 S]GTP␥S (ϳ100,000 cpm/tube) was added, and incubation at 37°C was continued. The reactions were quenched by the addition of 1 ml of ice-cold assay buffer followed by centrifugation. The membrane pellets were solubilized in 50 l of solution containing 1.5% (v/v) Triton X-100, 0.2% (w/v) SDS and then diluted to 1 ml with immunoprecipitation buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 20 mM MgCl 2 , 2 mM KCl, 1 mM EDTA, 1% (v/v) Triton X-100, 0.1% (w/v) SDS, protease inhibitors). The solubilized lysate was incubated for 18 h at 4°C with rabbit anti-G q/11 (2 g) antibody coupled to 40 l of protein G-Sepharose. Immunoprecipitates were recovered by centrifugation and washed six times with immunoprecipitation buffer. The complexes were solubilized by boiling in 5% SDS, and the bound GTP␥S was quantitated using a Wallac BetaRack scintillation counter.
Purification of Wild-type and Mutant PMT-Recombinant PMT and inactive, mutant PMT C1165S were expressed and purified as described previously (19).
Expression of the Bombesin/GRP Receptor in SYF Ϫ/Ϫ and YF Ϫ/Ϫ Cells-Membranes were prepared from confluent cultures of cells as outlined above. The membrane proteins (50 g) were fractionated by SDS-PAGE, transferred to nitrocellulose, and Western blotted with a rabbit polyclonal antiserum to GRP receptor as described previously (20).
G Protein Mutation-Tyrosine residue 349 of G q was converted to phenylalanine using a QuikChange kit (Stratagene) in accordance with the manufacturer's instructions. The constructs were confirmed by restriction digest and full nucleotide sequencing using a Beckman Coulter CEQ 2000XL automated sequencer.
Transfection of G␣ q/11 Ϫ/Ϫ Cells-G␣ q/11 Ϫ/Ϫ cells (21) were grown in DMEM supplemented with 10% (v/v) heat-inactivated fetal bovine serum, L-glutamine (1 mM), and 1ϫ modified Eagle's medium nonessential amino acids in a 95% air and 5% CO 2 atmosphere at 37°C. The cells were plated at 5 ϫ 10 4 cells/well in 33-mm dishes, and after a 24-h growth period they were transfected using LipofectAMINE Plus TM (Invitrogen BV) according to the manufacturer's instructions. After 6 h the cells were washed twice with Opti-MEM I and then cultured in DMEM for a further 18 h. The cells were washed twice with DMEM/nutrient mixture F-12 3:1 (v/v) and then incubated in the same medium for a further 48 h to induce quiescence. A total of 2 g of pcDNA3 containing the relevant G␣ q construct was used to transfect each dish.
Statistical Analysis-Data for production of inositol phosphates in Fig. 5 are presented as the means and standard errors from seven independent experiments with each point repeated in triplicate. Statistical analysis was carried out using the STATA 7 program (STATA Corp.). The significance of differences between various treatments was analyzed using the Mann-Whitney U test.

RESULTS
PMT Stimulates Tyrosine Phosphorylation of G␣ q -Treatment of Swiss 3T3 cells with mitogenic concentrations of PMT potently stimulated the tyrosine phosphorylation of the ␣ subunit of G q (Fig. 1a). Increases in tyrosine phosphorylation were concentration-dependent, reaching a maximum at a PMT concentration of ϳ70 pM (Fig. 1a). The increase in phosphorylation occurred after a lag period of 1 h and peaked around 4 -6 h after exposure to PMT (Fig. 1b, upper panel). The lag period is consistent with the time required for binding and internalization of the toxin to occur. Mutation of cysteine 1165 to serine in PMT (PMT C1165S ) leads to a complete loss of mitogenic activity and toxicity without grossly affecting toxin structure (19). Surprisingly, PMT C1165S was found to stimulate an increase in G␣ q phosphorylation with kinetics matching those of the wild-type toxin (Fig. 1b, lower panel). Treatment of Swiss 3T3 cells with the GPCR agonist bombesin also stimulated phosphorylation of G␣ q (Fig. 1c). However, in contrast to PMT, phosphorylation of G␣ q stimulated by bombesin was highly transient, peaking at 1 min post exposure and returning to basal levels within 5 min. The increase in G␣ q phosphorylation stimulated by PMT or PMT C1165S , but not by bombesin, could be effectively blocked by preincubation of the cells with either methylamine or a PMT antiserum (Fig. 1, d and e). This strongly suggested that the induced phosphorylation was a specific effect that followed toxin internalization.
PMT Stimulates Phosphorylation of Tyrosine 349 -Umemori et al. (12) previously demonstrated that GPCR agonists promoted tyrosine phosphorylation of G␣ 11 at tyrosine residue 352. It was therefore important to identify the site of G␣ q/11 phosphorylation in response to stimulation with PMT. Murine G␣ q (1-353) was cloned into the eukaryotic expression vector pcDNA3 and the C-terminal tyrosine residue (349), analogous to G␣ 11 tyrosine 352, was changed to phenylalanine by sitedirected mutagenesis. The constructs were confirmed by restriction digestion and nucleotide sequencing. Embryonic fibroblasts deficient in both G␣ q and G␣ 11 (G␣ q/11 Ϫ/Ϫ cells) were transfected with either wild-type or mutant (Y349F) G␣ q , and the expression of the G␣ subunit was determined. In agreement with previous studies (28), an anti-G␣ q/11 antiserum directed against the C terminus of the G protein failed to detect the mutant Y349F G␣ q .
However, probing of the membranes with an antiserum directed against an internal sequence of G␣ q (115-133) revealed equivalent expression of both constructs (data not shown). Because the C-terminal antiserum fails to recognize mutant Y349F G␣ q , it was not possible to use this antiserum for immunoprecipitation. It was therefore decided to carry out G␣ q immunoprecipitation using an anti-phosphotyrosine antibody and probe blots with the antiserum directed against the internal sequence of G␣ q .
Wild-type or mutant (Y349F) G␣ q was expressed in G␣ q/11 Ϫ/Ϫ cells, and cells were stimulated either with a mixture of bradykinin and thrombin or with wild-type or mutant PMT. In cells expressing wild-type G␣ q treatment with GPCR agonists or toxins resulted in increased tyrosine phosphorylation of G␣ q . In contrast, mutant Y349F G␣ q was not immunoprecipitated using an antibody directed against phosphotyrosine in either untreated or treated cells (Fig. 2a). Parallel experiments revealed that the observed changes in phosphorylation were not due to major differences in the expression of either wild-type or mutant G␣ q (Fig. 2b). This indicated that G␣ q Tyr 349 represents the major site of tyrosine phosphorylation in response to either GPCR or toxin stimulation.
PMT, but Not PMT C1165S , Stimulates Activation of G q -It had been assumed that mutant PMT (PMT C1165S ) failed to activate pathways associated with wild-type toxin, although this had never been demonstrated. We decided to determine whether PMT and PMT C1165S each stimulated activation of G q . The initial steps of G protein activation in response to either bombesin or toxins were determined by analyzing binding of the GTP analog GTP␥S. Because PMT is predicted to act enzymatically, standard GTP␥S binding assays were carried out at 37°C for 1 h. The use of an in vitro assay system means prolonged incubation steps are not required because the toxin does not require cellular binding and internalization. PMT and bombesin each stimulated concentration-dependent binding of GTP␥S to Swiss 3T3 membrane fractions (Fig. 3a). PMT potently stimulated GTP␥S binding at concentrations (picomolar range) 100-1000-fold lower than the GPCR agonist bombesin. In contrast, PMT C1165S did not stimulate a significant increase in GTP␥S binding in parallel experiments (Fig. 3a)  variety of conditions (longer incubation periods or higher toxin concentrations; data not shown).
To determine whether G q was directly affected by both bombesin and PMT, we performed GTP␥S binding assays followed by immunoprecipitation with antisera against G␣ q . Membranes from Swiss 3T3 cells were stimulated with 70 pM PMT or PMT C1165S or 20 nM bombesin as indicated. Nonspecific binding was determined in the presence of excess GTP␥S and by using normal rabbit serum. Bombesin and PMT each stimulated direct GTP␥S binding to G q (Fig. 3b). The kinetics of GTP␥S binding differed between bombesin and PMT. As previously reported, bombesin stimulated a rapid increase in the levels of bound GTP␥S, which peaked 5-10 min after addition (22). By comparison, increases in GTP␥S binding stimulated by PMT occurred gradually, peaking 40 -50 min after addition. This result would support the concept that PMT has an enzymatic action, with GTP binding occurring downstream of a toxin-catalyzed event. PMT C1165S had no effect on GTP␥S binding to G␣ q (Fig. 3b). Thus despite stimulation of G␣ q tyrosine phosphorylation by PMT C1165S , this mutant failed to stimulate this key indicator of G q activation.
To further investigate the functional role of tyrosine phosphorylation in G protein activation, the production of inositol phosphates was determined. Activation of G protein-coupled receptors linked to members of the G␣ q subfamily stimulates phospholipase C␤-mediated hydrolysis of phosphatidylinositol 4,5-bisphosphate. As previously reported (1), bombesin and PMT were potent activators of this pathway. In contrast, we found PMT C1165S did not stimulate inositol phosphate production (Fig. 3c). A wide variety of experimental conditions (longer incubation periods or higher toxin concentrations) failed to elicit increases in inositol phosphates in response to PMT C1165S (data not shown). These findings demonstrated that PMT and bombesin but not PMT C1165S could activate G q and that tyrosine phosphorylation of G␣ q did not necessarily lead to its activation.

PMT Activation of G q Is Not Dependent on G q Tyrosine Phosphorylation-
The ability of PMT C1165S to promote tyrosine phosphorylation of G␣ q without leading to its activation conflicts with the model proposed by Umemori et al. (13) suggesting that tyrosine phosphorylation occurred downstream of GTP binding. We decided to investigate whether phosphorylation of G␣ q was required for activation of phospholipase C. Cultures of Swiss 3T3 cells were treated with the broad spectrum tyrosine kinase inhibitor genistein to block phosphorylation of G␣ q . Genistein blocked tyrosine phosphorylation stimulated by PMT in a concentration-dependent manner (Fig. 4a). The increased phosphorylation of G␣ q in response to either PMT or bombesin was completely blocked by prior exposure to 50 M genistein (Fig. 4b). Daidzein, an analog of genistein, which lacks tyrosine kinase inhibitory activity, had no effect on the phosphorylation of G␣ q (data not shown). Application of 20 nM bombesin or 70 pM PMT to Swiss 3T3 cells pretreated with the solvent Me 2 SO or daidzein resulted in the production of inositol phosphates.

FIG. 2. Identification of tyrosine 349 as the major phosphorylation site. Wild-type (G q WT) or mutant (G q YF) G␣ q was expressed in G␣ q/11
Ϫ/Ϫ double deficient fibroblasts. a, cells were stimulated with a mixture of 5 M bradykinin and 1 unit/ml thrombin for 2 min or with 70 pM wild-type or mutant PMT for 8 h. Following treatment, the cells were lysed and adjusted to 0.1 mg/ml protein prior to immunoprecipitation. G␣ q was immunoprecipitated from RIPA extracts with anti-Tyr(P) antibody and Western blotted with an anti-G␣ q antibody (internal sequence). b, whole cell lysates (80 g) from parallel cultures were subjected to Western blotting with an anti-G␣ q antibody (internal sequence). The experiment was repeated twice with similar results. However, at concentrations that blocked tyrosine phosphorylation, genistein only inhibited inositol phosphate production stimulated by bombesin and did not inhibit the stimulation of inositol phosphate production by PMT (Fig. 4c). This result demonstrated that phosphorylation of G q is not an absolute requirement for phospholipase C activation in vivo. Furthermore, these data strongly argue against PMT acting directly as a tyrosine kinase.
PMT Enhances Inositol Phosphate Production in Response to Bombesin-It was previously reported that pretreatment of Swiss 3T3 cells with subsaturating concentrations of PMT enhanced the production of IP 3 in response to neuropeptides but not platelet-derived growth factor (2). To further investigate this effect, we asked whether PMT C1165S could also enhance the production of IP 3 . As previously reported, PMT was able to facilitate the production of IP 3 in response to bombesin treatment (Fig. 5). Although treatment of Swiss 3T3 cells with PMT C1165S alone did not stimulate production of IP 3 , pretreat-ment of Swiss 3T3 cells with PMT C1165S significantly enhanced the production of IP 3 in response to bombesin (p Ͻ 0.01) (Fig.  5). The degree of potentiation stimulated by either wild-type or mutant PMT (approximately 140 and 142%, respectively, of additive values) was comparable (Table I) and thus argues that PMT facilitates the production of IP 3 in response to bombesin through phosphorylation of G␣ q tyrosine 349.
Role of Src Family Kinases in G q/11 Phosphorylation-The differential effects of wild-type and mutant PMT on the stimulation of phosphorylation and inositol phosphate production suggested that a cellular kinase phosphorylated G␣ q . Src family kinases (subsequently referred to as Src kinases) have been postulated to be possible regulators of this process (11)(12)(13). To clarify the role of Src kinases, we utilized the recently described Src/Yes/Fyn-deficient (SYF Ϫ/Ϫ ) cell line (23). We compared activity in these cells with SYF Ϫ/Ϫ cells rescued with a retroviral vector expressing murine c-Src (YF Ϫ/Ϫ cells). PMT potently stimulated DNA synthesis in both cell lines, indicating that the cells were responsive to toxin (Fig. 6, a and b). PMT C1165S did not stimulate DNA synthesis in either cell line under the same  experimental conditions. Interestingly, bombesin only stimulated DNA synthesis in the Src expressing control cells (YF Ϫ/Ϫ ). This suggested that Src kinases might be required for GPCR stimulated DNA synthesis. To ensure differences were not due to a lack of receptor expression, we analyzed membrane fractions for the presence of the bombesin/GRP receptor. Both cell lines expressed the receptor (Fig. 6c). The ability of PMT to induce phosphorylation of focal adhesion kinase (p125 FAK ) was investigated to check the phenotype of the cell lines. PMT stimulated tyrosine phosphorylation of p125 FAK in YF Ϫ/Ϫ cells but not in SYF Ϫ/Ϫ cells (Fig. 6d). Finally, we investigated the effect of toxins and bombesin on the phosphorylation of G␣ q in YF Ϫ/Ϫ and SYF Ϫ/Ϫ cells (Fig. 6, e and f, respectively). Treatment of quiescent YF Ϫ/Ϫ cells with PMT, PMT C1165S , or bombesin stimulated tyrosine phosphorylation of G␣ q . By comparison, no changes in tyrosine phosphorylation of G␣ q could be observed in quiescent SYF Ϫ/Ϫ cells, although there was a high basal level of tyrosine phosphorylation of G␣ q . Although it is not clear why G␣ q was highly phosphorylated in these cells, the data demonstrate that Src kinases mediate the phosphorylation of G␣ q . DISCUSSION Tyrosine phosphorylation of proteins can modulate their activity and/or promote interaction with other molecules (24). Several neuropeptides that regulate cell growth and differentiation induce tyrosine phosphorylation of G␣ q . Recent studies have indicated that G␣ q phosphorylation forms part of a novel cycle in which tyrosine kinases and phosphatases regulate G q activation (11)(12)(13). The aim of the present study was to further investigate the role of phosphorylation using PMT.
The results presented here show for the first time that PMT induces both dose-and time-dependent increases in tyrosine phosphorylation of G␣ q . A nontoxic mutant of PMT (PMT C1165S ) also stimulated phosphorylation, with kinetics matching those of wild-type PMT. Phosphorylation of G␣ q/11 in response to wild-type or mutant PMT occurs at G␣ q tyrosine residue 349. Previous studies indicated that this tyrosine residue is also phosphorylated in response to GPCR agonists, an event confirmed in these studies. Our data demonstrate that the induction of tyrosine phosphorylation by either wild-type or mutant PMT is a specific event following toxin internalization. Methylamine, an agent that increases endosomal and lysosomal pH (25) and therefore inhibits the entry and processing of many toxins, selectively blocked the induction of tyrosine phosphorylation by PMT. Similarly, the early addition of neutralizing antisera to PMT selectively blocked phosphorylation of G␣ q .
The addition of PMT to cell membranes induced an increase in binding of GTP␥S to G␣ q . This is the first description of an in vitro assay for PMT activity. PMT C1165S neither affected GDP/GTP exchange nor stimulated an increase in levels of inositol phosphates. Thus the stimulation of tyrosine phosphorylation by PMT C1165S does not lead to activation of G q . These data demonstrate that stimulation of G␣ q tyrosine phosphorylation can occur in the absence of G protein activation and confirm that phosphorylated G␣ q is not constitutively active in the basal GDP bound state (11).
Lui et al. (11) reported that in Rat-1 fibroblasts transformed with the v-src oncogene, inositol phosphate production stimulated by endothelin-1 was increased 6-fold, without changes in the number of receptors. This increased response was mediated through G q , which was phosphorylated on tyrosine residue(s). Moreover, when extracted G protein was reconstituted with exogenous phospholipase C, AlF 4 Ϫ -stimulated G q activity was significantly increased in extracts from v-src transformed cells. These data implied that phosphorylation of G q may have a regulatory role in vivo.
Subsequent studies by Umemori et al. (12,13) have further investigated the role of G q tyrosine phosphorylation in regulating G protein activity. These authors demonstrated that G q was phosphorylated upon ligand activation. This phosphorylation was suggested to prevent interaction of the GPCR with G q and was essential for the activation of G q by receptor stimulation (12). However, the data conflict with the previous report of Lui et al. (11), who clearly demonstrated that phosphorylated G q in v-src transformed cells could still interact with and be activated by the endothelin-1 receptor. Indeed, in such cells the production of IP 3 was potentiated by the phosphorylation of G q .
Similarly, overexpression of c-Src in mouse fibroblasts potentiates both agonist-induced signaling through ␤-adrenergic receptors and cAMP accumulation in response to cholera toxin (26,27). Analysis of the in vitro sites of phosphorylation catalyzed by c-Src identified residues Tyr 37 and Tyr 377 (27). Tyr 37 lies near the site of G␤␥ binding in the N terminus, whereas Tyr 377 is located in the extreme C terminus, within a region of G␣ s important for receptor interaction. Moreover, phosphorylation of G␣ s by immune-complexed c-Src resulted in enhanced rates of receptor-mediated GTP␥S binding and GTP hydrolysis (29). These data support the findings of Lui et al. (11) and suggest that tyrosine phosphorylation of G␣ subunits does not, in itself, prevent interaction of the G protein with the GPCR.
Most recently, Liu et al. (28) reported that an aromatic group is required for efficient information transfer from an agonist occupied receptor to G␣ 11 . However, their data established that tyrosine 352 could be substituted with phenylalanine or tryptophan without abolishing G protein activity. These findings, together with the data presented here, clearly demonstrate that tyrosine phosphorylation of G␣ q/11 is not a requirement for G protein activation in vivo.
To further address this problem we investigated the effects of wild-type and mutant PMT on the production of IP 3 in response to bombesin. As reported previously (2), wild-type PMT facilitated the production of IP 3 in response to bombesin. Interestingly, we have now shown that mutant PMT can also potentiate signaling via the bombesin receptor. Wild-type and mutant PMT each stimulated phosphorylation of G q , but importantly, only the wild-type toxin activated the G protein. These findings imply that phosphorylation of G q alone is not sufficient to uncouple the receptor from the G protein.
The mechanism through which tyrosine phosphorylation of G q is induced remains unclear. Our data support the suggestion that a cellular kinase is activated early in the G protein cycle to phosphorylate G q . Umemori et al. (13) proposed that a kinase is activated in response to GTP-GDP exchange on G q . However, the findings that mutant PMT can stimulate phosphorylation of G q in the absence of GTP-GDP exchange suggest that this model may be an oversimplification. To further understand the mechanisms through which phosphorylation of G␣ q is regulated, the putative kinase must be identified. Pre-vious work has indicated Src family kinases are critical mediators of this pathway. Our data using Src-deficient cell lines suggest that Src mediates phosphorylation of G q in response to PMT in vivo. However, because basal phosphorylation of G q is high in these cells, we cannot exclude the possibility that other kinases may also be involved.
In summary, these studies have conclusively demonstrated that PMT activates members of the G q family of heterotrimeric G proteins. Utilizing GPCR agonists and a mutant form of PMT, we have confirmed that tyrosine phosphorylation of G␣ q can be dissociated from G protein activation. Moreover, the finding that wild-type and mutant PMT can potentiate GPCR signaling highlights as yet undefined roles for tyrosine phosphorylation of G␣ q .