Identification of Gα13 as One of the G-proteins That Couple to Human Platelet Thromboxane A2 Receptors*

Previous studies have shown that ligand or immunoaffinity chromatography can be used to purify the human platelet thromboxane A2 (TXA2) receptor-Gαq complex. The same principle of co-elution was used to identify another G-protein associated with platelet TXA2 receptors. It was found that in addition to Gαq, purification of TXA2 receptors by ligand (SQ31,491)-affinity chromatography resulted in the co-purification of a member of the G12 family. Using an antipeptide antibody specific for the human G13 α-subunit, this G-protein was identified as Gα13. In separate experiments, it was found that the TXA2 receptor agonist U46619 stimulated [35S]guanosine 5′-O-(3-thiotriphosphate) incorporation into G13 α-subunit. Further evidence for functional coupling of G13 to TXA2 receptors was provided in studies where solubilized platelet membranes were subjected to immunoaffinity chromatography using an antibody raised against native TXA2 receptor protein. It was found that U46619 induced a significant decrease in Gαq and Gα13 association with the receptor protein. These results indicate that both Gαq and Gα13 are functionally coupled to TXA2 receptors and dissociate upon agonist activation. Furthermore, this agonist effect was specifically blocked by pretreatment with the TXA2 receptor antagonist, BM13.505. Taken collectively, these data provide direct evidence that endogenous Gα13 is a TXA2 receptor-coupled G-protein, as: 1) its α-subunit can be co-purified with the receptor protein using both ligand and immunoaffinity chromatography, 2) TXA2 receptor activation stimulates GTPγS binding to Gα13, and 3) Gα13 affinity for the TXA2 receptor can be modulated by agonist-receptor activation.

Previous studies have shown that ligand or immunoaffinity chromatography can be used to purify the human platelet thromboxane A 2 (TXA 2 ) receptor-G␣ q complex. The same principle of co-elution was used to identify another G-protein associated with platelet TXA 2 receptors. It was found that in addition to G␣ q , purification of TXA 2 receptors by ligand (SQ31,491)-affinity chromatography resulted in the co-purification of a member of the G 12 family. Using an antipeptide antibody specific for the human G 13 ␣-subunit, this G-protein was identified as G␣ 13 . In separate experiments, it was found that the TXA 2 receptor agonist U46619 stimulated [ 35 13 ␣-subunit. Further evidence for functional coupling of G 13 to TXA 2 receptors was provided in studies where solubilized platelet membranes were subjected to immunoaffinity chromatography using an antibody raised against native TXA 2 receptor protein. It was found that U46619 induced a significant decrease in G␣ q and G␣ 13 association with the receptor protein.

S]guanosine 5-O-(3-thiotriphosphate) incorporation into G
These results indicate that both G␣ q and G␣ 13 are functionally coupled to TXA 2 receptors and dissociate upon agonist activation. Furthermore, this agonist effect was specifically blocked by pretreatment with the TXA 2 receptor antagonist, BM13.505. Taken collectively, these data provide direct evidence that endogenous G␣ 13 is a TXA 2 receptor-coupled G-protein, as: 1) its ␣-subunit can be co-purified with the receptor protein using both ligand and immunoaffinity chromatography, 2) TXA 2 receptor activation stimulates GTP␥S binding to G␣ 13 , and 3) G␣ 13 affinity for the TXA 2 receptor can be modulated by agonist-receptor activation.
Interaction of the prostaglandin endoperoxide analogue, TXA 2 1 (1, 2) with platelet receptors (3)(4)(5) has been shown to modulate not only hemostasis but also the development of thromboembolic diseases (6 -9). However, despite recent progress, the TXA 2 -mediated signal transduction pathway is not completely understood. In this regard, previous studies have shown that one mechanism by which TXA 2 receptors act is through stimulation of phospholipase C (PLC) leading to inositol 1,4,5-triphosphate (IP 3 ) production, and subsequent intracellular Ca 2ϩ mobilization (10 -14). Furthermore, separate studies have linked this stimulation of PLC activity to TXA 2 receptor signal transduction through the pertussis toxin-insensitive guanine nucleotide-binding protein (G-protein) G q (16,17). On the other hand, experiments conducted in our laboratory provided evidence for the existence of intraplatelet Ca 2ϩ mobilization, which is independent of IP 3 production (15). This finding raised the possibility that TXA 2 receptors may also couple to a G-protein family separate from G q . Additional evidence in support of this notion was provided by experiments showing that a C-terminal antibody which recognizes the ␣-subunit of G q and G 11 was not able to completely inhibit U46619-stimulated GTPase activity (16). Moreover, ligand and immunoaffinity chromatography purification of the TXA 2 receptor-G-protein complex allowed co-purification of G-proteins distinct from G q (17). Taken together, these results led to the hypothesis that TXA 2 receptors might couple to a G-protein(s) to stimulates platelet aggregation independently of the G q -PLC-IP 3 pathway.
Although this putative G-protein has not been identified, recent reports have provided indirect evidence that it may belong to the G 12 family (18,19). In one study, it was shown that activation of platelet TXA 2 receptors led to increased incorporation of the photo reactive GTP analogue [␣-32 P]GTP azidoanilide into both G 12 and G 13 ␣-subunits, which may suggest coupling of TXA 2 receptors to these ␣-subunits (20). On the other hand, as all the agonists tested (U46619, thrombin, ADP, and vasopressin) produced [␣-32 P]GTP azidoanilide incorporation, this labeling could also have been due to activation of a downstream signaling event or to cross-talk between these separate signal transduction pathways. In separate studies, it was shown that the affinity state of TXA 2 receptors transfected in COS-7 cells could be influenced by co-expression of G␣ 13 (21). Although this finding is consistent with the notion that TXA 2 receptors have the capacity to couple with G 13 , it is not clear whether such coupling occurs at physiological concentrations of receptor and/or G-protein. Consequently, two independent reports have provided indirect evidence that TXA 2 receptors may couple to a G␣ subunit in the G 12/13 family. Based on these considerations, in the present study we performed experiments to determine whether this phenomenon occurs in a native platelet preparation using endogenous concentrations of TXA 2 receptor and G␣ subunits. To this end, affinity purification of the receptor-G-protein complex was employed to measure direct physical association of TXA 2 receptors and G␣ 12/13 . This approach has been previously applied in our laboratory to the identification of G␣ q as one of the G-proteins associated with the platelet TXA 2 receptors (17). It was found that, in addition to G␣ q , purification of the TXA 2 receptors resulted in co-elution of G␣ 13 . Furthermore, agonist activation of TXA 2 receptors caused an increase in GTP␥S binding to G 13 ␣-subunit as well as dissociation of the receptor-G␣ 13 complex, providing evidence that G␣ 13 is indeed functionally coupled to platelet TXA 2 receptors. (Atlanta, GA); CHAPS, protein A-Sepharose CL-4B, GTP␥S, o-phenylenediamine, and rabbit preimmune IgG were from Sigma; Affi-Gel 102 and 4-chloro-1-naphthol (horseradish peroxidase color development reagent) were from Bio-Rad; and horseradish peroxidaseconjugated goat anti-rabbit IgG (HϩL), biotinylated goat anti-rabbit IgG (HϩL), and the Vectastain ABC kit were purchased from Vector Laboratories (Burlingame, CA).

Materials
Antibodies-A 9-amino acid peptide corresponding to residues 40 -48 of the human G␣ 13 (P 21 , Table I) (22), with a cysteine added at the N terminus to facilitate coupling to carrier protein was synthesized by Chiron Mimotropes (Raleigh, NC). The peptide was coupled to keyhole limpet hemocyanin using m-maleimidobenzoic acid N-hydroxysuccinimide ester and injected into White New Zealand Pasteurella multocida-free rabbits, according to previously described procedures (23). Rabbit polyclonal antibodies against the C-terminal region of G␣ q (G-QL, Table I) were produced as described previously (16). Antibodies were purified from rabbit serum by chromatography on protein A-Sepharose CL-4B, and the IgG fractions were labeled with carrier-free Na 125 I (Amersham Pharmacia Biotech) using the IODO-BEADS iodination reagent (Pierce). Rabbit polyclonal IgG raised against residues 2-21 of G␣ 12 (G-12) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit polyclonal IgG directed against the C-terminal region of G␣ 12 was a generous gift from J. Sylvio Gutkind (National Institutes of Health, Bethesda, MD) (24). Although this antibody has been suggested to be specific for G␣ 12 , C-terminal segments of G 12 and G 13 differ only by 4 amino acids (18,19). Consequently, this antibody may have limited cross-reactivity with G 13 and is named G-12/13. G␣ common , a rabbit polyclonal IgG recognizing G-protein ␣-subunits of the G s and G i families, was purchased from Calbiochem.
Membrane Preparation and Solubilization-Human platelet membranes were prepared from platelet concentrates and solubilized using 10 mM CHAPS as described previously (5). Typically, this method resulted in a 60 -70% solubilization of platelet membrane protein, yielding a final protein concentration of 2-3 mg/ml.
Ligand Affinity Chromatography Purification of the Thromboxane A 2 Receptor-G-protein Complex-TXA 2 receptors were purified as described previously (5). Briefly, the TXA 2 receptor antagonist SQ31,491 was immobilized to Affi-Gel 102, and CHAPS-solubilized membranes (4 mg of protein) in buffer A (20% glycerol, 500 mM KCl, 0.2 mM EGTA, and 0.5 mg/ml asolectin) were incubated with the matrix overnight. Unbound proteins were eluted as flow-through, and the column was washed with buffer D (20 mM Tris-HCl, 10 mM CHAPS, 20% glycerol, 500 mM KCl, 0.2 mM EGTA, 0.5 mg/ml asolectin, pH 7.4). TXA 2 receptors and receptor-associated proteins were then eluted with buffer D containing 50 mM TXA 2 receptor antagonist, BM13.177 (25,26). After elution of the first 1-ml fraction, the flow was stopped for 30 min and restarted to elute the subsequent 1-ml fractions. TXA 2 receptor binding activity as well as G␣ q immunoreactivity were found to be concentrated in the first fraction following the 30-min incubation (17). A modification of this method was used in order to allow further identification of the TXA 2 receptor-associated G-proteins (27). Specifically, after unbound proteins were washed with buffer D, 3 g/ml 125 I-G-QL IgG, 125 I-G-12 IgG, or 125 I-G-13 IgG (or the same protein concentration of 125 I-labeled preimmune IgG (PI IgG)) was added, and the reaction mixture was allowed to incubate for 1 h at 20°C. Unbound antibodies were washed with buffer D, and elution of TXA 2 receptors and receptor-associated proteins was performed as described above. The elution fractions were counted for 125 I activity and specific binding attributable to G␣ q , G␣ 12 , or G␣ 13 was defined as the difference between the counts eluted from the 125 I-antibody columns minus the counts eluted from the 125 I-PI IgG column.
Immunoaffinity Chromatography Purification of the Thromboxane A 2 Receptor-G-protein Complex-Solubilized platelet membranes were prepared as described (5), and the CHAPS concentration was adjusted to 2 mM. The preparation (4 mg of protein) was then incubated with an immunoaffinity matrix coupled to an anti-TXA 2 receptor antibody (TxAb) for 1 h at 20°C (28). 125 I-Labeled G-QL IgG, G-13 IgG, or PI IgG was added (final concentration 150 g/ml), and the reaction mixture was allowed to incubate for 5 min. The preparation was then incubated with vehicle or the TXA 2 agonist U46619 (100 nM) (29) for an additional 5 min. The matrix was loaded on a column and washed with buffer D to elute unbound proteins. The column was eluted with 100 mM glycine (pH 2.5), and the 3-ml elution fraction was counted for 125 I activity. Specific binding attributable to G␣ q or G␣ 13 was defined as the difference between the counts eluted from the 125 I-G-QL or 125 I-G-13 columns, respectively, minus the counts eluted from the 125 I-PI IgG column. Eluted counts were normalized to the amount of purified TXA 2 receptor protein, as measured by densitometric analysis of the immunoaffinity column elution fractions immunoblotted with TxAb. In experiments where the TXA 2 receptor antagonist BM13.505 (30) was used to block U46619 effects, BM13.505 (10 M) was incubated for 30 min before addition of U46619.
Assay of [ 35 S]GTP␥S Binding-Solubilized platelet membranes (4 mg of protein) were prepared as described (5) and incubated with 10 M GTP␥S (5 ϫ 10 6 cpm [ 35 S]GTP␥S) for 5 min at room temperature in the presence of 1 M GDP. The incubation was then allowed to proceed for an additional 15 min at room temperature in the presence or absence of 10 nM U46619. The preparation was added to 20 l of G-13 IgG, which had been preincubated with 55 l of a 10% (w/v) suspension of protein A-Sepharose. The immunoprecipitates were collected, washed to remove nonspecifically bound proteins and incubated with 1 mM G-13 peptide for 1 h at room temperature to specifically elute G 13 ␣-subunits. The fractions were then added to 5 ml of Econosafe TM (Research Product International, IL) and analyzed by scintillation spectrometry. Eluted counts were normalized to the amount of immunoprecipitated G 13 ␣-subunit, as measured by densitometric analysis of the elution fractions immunoblotted with G-13 IgG.
ELISA-Immulon 2 microtiter plates were coated with either 12.5 g of synthetic peptide or 125 g of solubilized platelet membranes. Following incubation for 1 h at room temperature, the plates were washed three times with modified Tyrode's buffer containing 0.1% bovine serum albumin, 5 mM dextrose, 1 mM CaCl 2 , 5 mM HEPES, pH 7.4, and then blocked by incubation for 1 h with 5% bovine serum albumin in the same buffer. Serial dilutions of antisera were applied to the wells and incubated for an additional 1 h at room temperature. The wells were washed three times with the modified Tyrode's buffer, and bound antibodies were detected by incubation for 1 h with goat anti-rabbit IgG (HϩL) conjugated to horseradish peroxidase. After extensive washing, the color reaction was developed by addition of 50 l of 0.4 mg/ml o-phenylenediamine, 0.012% H 2 O 2 in 80 mM citrate phosphate, pH 5. An equal volume of 2 N H 2 SO 4 was then added, and the presence of specific antibodies was measured by absorbance at 490 nm.
Polyacrylamide Gel Electrophoresis and Immunoblot Assay-The affinity column eluates were first concentrated using Millipore Ultrafree-MC filters. 20 -40 l of sample (30 -40 g of protein) was then subjected to SDS-PAGE according to the method of Laemmli (31) using 10% minigels, under nonreducing conditions, and the proteins were electrophoretically transferred onto nitrocellulose membranes according to the method of Towbin et al. (32). After transfer, the nitrocellulose membranes were blocked with 3% gelatin in Tris-buffered saline and incubated overnight at room temperature with the indicated dilution of G-QL, G-12/13, or G␣ common IgG. The blots were washed and treated with biotinylated goat anti-rabbit IgG (HϩL) as the secondary antibody. The immunoreactive proteins were detected with avidin and horseradish peroxidase, followed by 0.5 mg/ml 4-chloro-1-naphthol.

RESULTS
In order to purify TXA 2 receptor-associated G-proteins, solubilized platelet membranes were subjected to ligand affinity chromatography, and the column was eluted with the TXA 2 receptor antagonist BM13.177 (5,25,26). Elution fractions were then immunoblotted with G-protein ␣-subunit specific antibodies. As illustrated in Fig. 1, G-QL IgG, which recognizes G q and G 11 ␣-subunits, blotted a major protein band in the molecular mass region of 42 kDa and two minor bands at approximately 38 kDa in solubilized platelet membranes and the ligand column elution fraction. This pattern of primary labeling at 42 kDa and secondary labeling at lower molecular masses has been previously described in human platelets and other tissues and has been attributed to proteolytic fragments of G␣ q (16). Furthermore, blotting with G-12/13 IgG also revealed immunoreactivity for the ␣-subunit of G 12/13 . Thus, a single band in the molecular mass range of 43-44 kDa was observed both in solubilized platelet membranes and the ligand affinity column eluate (Fig. 1). On the other hand, an antibody against G␣ common revealed that members of the G␣ s and G␣ i families were only present in solubilized platelet membranes but not in the affinity column eluate (data not shown). These results indicate that the ligand chromatography-purified TXA 2 receptor-G-protein complex is selectively enriched in both G␣ q and a member(s) of the G␣ 12 family.
In order to determine whether TXA 2 receptors purified in complex with either G␣ 12 or G␣ 13 , the ligand affinity eluate was first probed using a new antibody specifically directed against G␣ 13 . This polyclonal antibody was generated by immunizing rabbits with a peptide sequence unique to an internal segment of the human G 13 ␣-subunit (G-13, Table I) (22). Evaluation of the sera and IgG revealed that specific antibodies against G␣ 13 were successfully raised. Specifically, the anti-G-13 serum was shown by ELISA to react in a concentrationdependent manner with its cognate peptide ( Fig. 2A). Moreover, evaluation of solubilized platelet membranes by ELISA revealed positive immunoreactivity against an endogenous platelet protein (Fig. 2B). In addition, immunoblotting of solubilized platelet membranes with G-13 serum and IgG demonstrated a predominant 43-44 kDa protein (Fig. 3, lanes 1 and  2, respectively), consistent with the molecular mass previously described for G 13 ␣-subunit (18,19,22). The blotting of this band could be prevented by preincubation of G-13 IgG with its cognate peptide (Fig. 3, lane 3). To utilize this antibody for quantitative evaluation of G␣ 13 , G-13 IgG was iodinated by standard procedures.
In the next experiments, solubilized platelet membranes were incubated with the ligand affinity matrix (5). The column was washed with buffer and then equilibrated with 125 I-G-13 IgG. The TXA 2 receptor-G-protein complex was next eluted by BM13.177 (25,26), and the elution fractions were quantitated for 125 I. As G␣ q is known to couple to TXA 2 receptors (16, 17), a positive control experiment was conducted using 125 I-G-QL IgG. Control experiments were also performed using 125 I-labeled preimmune IgG to determine nonspecific binding of both G-QL and G-13 IgG. Reported specific binding represents the difference between the counts eluted from the 125 I-anti-␣-subunit IgG column and the counts eluted from the 125 I-labeled preimmune IgG column. Using this procedure, it was found that TXA 2 receptors co-purified not only with G␣ q but also with G␣ 13 . Thus, Fig. 4 illustrates that the affinity column eluate contained 53 Ϯ 5% and 24 Ϯ 5% specific binding for G␣ q (solid bar) and G␣ 13 (open bar), respectively. These findings, therefore, provide evidence that both G␣ q and G␣ 13 are in direct physical association with endogenous platelet TXA 2 receptors.
The next series of experiments was performed to determine whether G␣ 12 also copurified as part of the TXA 2 receptor-Gprotein complex. These studies employed a specific antibody directed at a unique N-terminal segment of G␣ 12 (G-12, Table  I), which was also iodinated for quantitative purposes. Again, the TXA 2 receptor-G-protein complex was purified by ligand affinity chromatography and the amount of G␣ 12 present in the eluate was determined using 125 I-G-12 IgG. As above, specific binding was determined by parallel experiments using 125 Ilabeled preimmune IgG. It was found that, in contrast to G␣ 13 , G␣ 12 did not appear to co-purify with TXA 2 receptors (Fig. 4, hatched bar). In these experiments, it can be seen that the counts attributable to G␣ 12 are less than the counts representing nonspecific binding by preimmune IgG. Although this decrease is not significant, it can be explained on the basis that G-12 IgG is enriched in immunoglobulins against G␣ 12 and consequently contains a lesser percentage of nonspecific proteins than preimmune IgG. As nonspecific protein is presumably responsible for nonspecific binding observed with preimmune IgG, the difference between 125 I-G-12 IgG counts and 125 I-labeled preimmune IgG counts yields a negative number. The same phenomenon would also suggest that the specific binding observed for both G-QL and G-13 IgG (Fig. 4) is probably underestimated, as each of these IgG fractions contain less nonspecific proteins than their preimmune IgG controls. Furthermore, this consideration would indicate that the relative percentage of specific binding with G-QL and G-13 IgG may not necessarily represent the actual distribution of G␣ q and G␣ 13 within the TXA 2 receptor-G-protein complex. Taken together, the above results provide evidence that in addition to G␣ q , platelet TXA 2 receptors are coupled to endogenous G␣ 13 .
In the next series of experiments, the agonist U46619 was used to determine whether G␣ 13 is functionally coupled to TXA 2 receptors. In these studies, solubilized platelet membranes were incubated with [ 35 S]GTP␥S in the presence and absence of U46619 (10 nM) and subjected to immunoprecipitation with G-13 IgG. Immunoprecipitated G␣ 13 was eluted using 1 mM G-13 peptide, the elution fractions were counted for 35   and the counts were normalized for the amount of purified G 13 ␣-subunit. It can be seen in Fig. 5A that TXA 2 receptor activation led to a 44 Ϯ 18% (p ϭ 0.05) increase in [ 35 S]GTP␥S binding to G␣ 13 .
In separate experiments, an affinity column matrix coupled to an antibody raised against native TXA 2 receptor protein (TxAb) was used to purify the receptor-G-protein complex (16,28). Briefly, solubilized platelet membranes were incubated with the affinity matrix and the coupling of G-protein ␣-subunits to TXA 2 receptors was evaluated in the presence and absence of U46619 (29). Evidence has been previously provided that agonist activation of a seven-transmembrane receptor results in the dissociation of its coupled G-protein ␣-subunits (33). Based on these considerations, excess 125 I-labeled anti-Gprotein antibody was added to the column matrix prior to U46619 treatment in order to trap ␣-subunits that may be released upon receptor activation. The excess 125 I-labeled anti-␣-subunit antibody was then washed from the column matrix and the TXA 2 receptor-G-protein complex was eluted with glycine 100 mM, pH 2.5. Elution fractions were then counted for 125 I activity and normalized to the amounts of purified TXA 2 receptor protein. Again, specific binding was determined in parallel experiments using 125 I-labeled preimmune IgG. Consistent with the results obtained with ligand affinity chromatography, both G␣ q and G␣ 13 co-purified with platelet TXA 2 receptors. Specific binding attributable to G-QL IgG and G-13 IgG was 42 Ϯ 11% and 54 Ϯ 10%, respectively (Fig. 5B, open  bars). Furthermore, treatment with 100 nM U46619 (Fig. 5B, solid bars) caused a significant decrease in the amount of specifically eluted 125 I-labeled G-QL and G-13 IgG. The magnitude of the agonist-induced decrease in G-protein-receptor association was 44 Ϯ 15% and 32 Ϯ 1% for G␣ q and G␣ 13 , respectively. Finally, pretreatment with a TXA 2 receptor antagonist, BM13.505 (10 M) (30), completely inhibited U46619induced receptor-G-protein complex dissociation (Fig. 5B,  hatched bars). Taken together, these results demonstrate that in addition to G␣ q , human platelet TXA 2 receptors are functionally coupled to G␣ 13 .

DISCUSSION
The present study employed ligand affinity (5) and immunoaffinity chromatography techniques (28) to purify and identify G-proteins associated with human platelet TXA 2 receptors. These techniques have previously been used for the purification of TXA 2 receptor-G-protein complexes from solubilized platelet membranes (17). Using both ligand and immunoaffinity chromatography, it was found that in addition to G␣ q , a member of the G␣ 12 family of G-proteins co-purifies with platelet TXA 2 receptors. This G-protein was identified as G␣ 13 using an antibody raised against a unique internal sequence of human G 13 ␣-subunit (22). Additional studies demonstrated that this G 13 ␣-subunit was functionally coupled to TXA 2 receptors. Specifically, the TXA 2 receptor agonist U46619 stimulated [ 35 S]GTP␥S incorporation into G 13 ␣-subunit as well as caused dissociation of this subunit from TXA 2 receptors.
The G␣ 12 family of G-proteins defines the fourth and the  most recently discovered class of ␣-subunits (18,19). The members of this family share high sequence homology and are ubiquitous and immunodetectable in most membranes of various mammalian cells and tissues (22,34,35). However, despite intensive research in the past years, no definitive effector(s) has been assigned to either G␣ 12 or G␣ 13 (36,37). Both G␣ 12 and G␣ 13 are oncogenic, and expression of their mutationally activated forms stimulates cell proliferation and induces neoplastic transformation in NIH3T3 and Rat1 cells (24, 38 -40). Furthermore, GTPase-deficient mutants of G␣ 12 and G␣ 13 have been shown to stimulate Jun kinase/stress-activated protein kinase (JNK/SAPK) in NIH3T3, HEK293, and COS-1 cells (41,42). In addition, both activated ␣-subunits have been shown to stimulate stress fiber formation/focal adhesion assembly in Swiss 3T3 cells (43) and induce apoptosis when transfected in Chines hamster ovary or COS-7 cells (44,45). Finally, signal transduction through G␣ 12 and G␣ 13 appears to involve small molecular weight GTP-binding proteins such as RhoA, cdc42, and Ras (46 -48). However, even though there is similarity between G␣ 12 and G␣ 13 -associated pathways, evidence has been provided that both subunits seem to fulfill distinct cellular and biological functions. Specifically, G␣ 12 but not G␣ 13 has been shown to be involved in the transcriptional activation of the serum response element (47). On the other hand, G␣ 13 but not G␣ 12 is involved in the induction of inducible nitric-oxide synthase in MCT cells (49) and in lysophosphatidic acid-induced activation of Rho (50). Other studies showed that, whereas the guanine nucleotide exchange factor (GEF) for Rho, p155RhoGEF, was able to act as a GTPase-activating protein toward both G␣ 12 and G␣ 13 , only G␣ 13 bound to p155RhoGEF and stimulated its capacity to catalyze nucleotide exchange on Rho (51,52). In addition, disruption of the gene encoding G 13 ␣-subunit in mice impaired the ability of endothelial cells to develop into organized vascular system, resulting in intrauterine death and demonstrating a role for G␣ 13 in the regulation of cell movement and developmental angiogenesis (53).
Two potential effectors for G␣ 13 have been proposed that would be of interest in the signal transduction pathways associated with TXA 2 receptors in platelets. In this connection, G␣ 13 has been shown to stimulate the ubiquitously distributed Na/H exchanger isoform, NHE1 (46, 53-55). Moreover, substitution of C-terminal residues from ␣ z conferred on ␣ 13 the ability to respond to stimulation by the D 2 -dopamine receptor and to activate NHE1 in an agonist-dependent manner (54).
In platelets, regulation of Na/H exchange has been shown to modulate receptor-mediated phospholipase A 2 and phospholipase C activation as well as intracellular Ca 2ϩ mobilization (57,58). In regard to platelet TXA 2 receptors, it was found that U46619 caused an increase in intracellular pH, which was required for full U46619-induced Ca 2ϩ mobilization (59). Thus, coupling of TXA 2 receptors to NHE1 activity stimulation could be a possible mechanism by which G␣ 13 is involved in TXA 2mediated signal transduction in platelets.
In addition to its indirect effects on intracellular Ca 2ϩ via Na/H exchanger activity stimulation, G␣ 13 has also been implicated in the activation of L-type Ca 2ϩ channels (60,61). Specifically, in rat portal vein myocytes, the heterotrimer ␣ 13 ␤ 1 ␥ 3 couples to the angiotensin AT 1A receptors to increase cytoplasmic Ca 2ϩ concentration (60). Furthermore, it was found that the ␤␥ dimer released from ␣ 13 upon angiotensin AT 1A receptor activation was responsible for the activation of L-type Ca 2ϩ channels (61). Although extracellular Ca 2ϩ influx through L-type and non-L-type Ca 2ϩ channels has been associated with TXA 2 receptor-mediated contraction in rat aorta (62), no such channels have been identified on the platelet surface thus far.
In summary, the present data demonstrate that platelet TXA 2 receptors are functionally coupled to G␣ 13 . The physiological significance of the signal transduction pathway associated with such coupling requires further investigation.  13 was performed using a two-sample Student's t test. B, effect of U46619 treatment on G␣ q and G␣ 13 association with TXA 2 receptors. Solubilized platelet membranes were subjected to immunoaffinity chromatography purification as described under "Experimental Procedures," and the purified TXA 2 receptor-G-protein complex was probed with 125 I-labeled G-QL and G-13 IgG in the presence and absence of U46619 (100 nM). Results are expressed in counts per minute (cpm) normalized for the amount of purified TXA 2 receptor protein and represent the mean Ϯ S.E. of five separate experiments. Statistical analysis measuring the effect of U46619 on G␣ q and G␣ 13 association with TXA 2 receptors was performed using a two-sample Student's t test (*, p Ͻ 0.05; **, p Ͻ 0.005).