The bradykinin B2 receptor is a delayed early response gene for platelet-derived growth factor in arterial smooth muscle cells.

Bradykinin and platelet-derived growth factor (PDGF) are inflammatory mediators important in the response to vascular injury. Based upon the known effect of oncogenic Ras to increase bradykinin receptor expression and the ability of PDGF to stimulate Ras, we examined whether PDGF regulates bradykinin B2 receptor expression in cultured arterial smooth muscle cells. Treatment with PDGF (AB and BB, but not AA) produced a dose- and time-dependent increase in both mRNA (6-7-fold increase at 2-4 h) and cell surface receptors (2-4-fold at 6-12 h) for the B2 receptor. There was a 60-min delay between exposure to PDGF and the initial increase in B2 receptor mRNA. Transcriptional inhibitors, actinomycin D or 5,6-dichloro-1-β-D-ribofuranosylbenzimidazole, completely blocked the increase in B2 receptor mRNA when added up to 60 min after stimulation with PDGF. However, protein synthesis was not required, as treatment with cycloheximide did not block but rather superinduced the PDGF-induced increase in B2 receptor mRNA. Comparison with the immediate early response gene c-fos demonstrated that the increase in B2 receptor mRNA was similarly inhibited by the tyrosine kinase inhibitor, tyrphostin, as well as staurosporine. However, stimulation of c-fos was slightly more sensitive to genistein, while the B2 receptor mRNA was more sensitive to inhibition by the protein kinase C inhibitor, calphostin C. The increase in cell surface B2 receptors were functionally coupled to an increase in phosphoinositide-specific phospholipase C, and the effects of PDGF were selective as there was no increase in either angiotensin II- or arginine vasopressin-induced inositol phosphate formation or intracellular calcium release. Taken together, these results demonstrate that the B2 receptor is a delayed early response gene for PDGF in vascular smooth muscle cells.

Bradykinin is a potent vasoactive hormone released during activation of the intrinsic pathway of coagulation (1). In the presence of an intact endothelium, bradykinin acts on B 2 receptors present on the endothelium to produce vasodilatation and an increase in vascular permeability (2,3). However, when the endothelium is disrupted, bradykinin gains direct access to receptors present on the underlying vascular smooth muscle cells (4 -6). The biological role of these smooth muscle B 2 receptors is unclear. The available evidence suggests that these B 2 receptors activate a phosphoinositide-specific phospholipase C leading to increases in intracellular calcium and activation of protein kinase C similar to that described for the vasoconstrictor hormones, angiotensin II and arginine vasopressin (7). This suggests that the direct effects mediated by the B 2 receptor on arterial smooth muscle may antagonize the indirect endothelialdependent vasoactive effects of bradykinin. Understanding the functional role and regulation of B 2 receptor expression on vascular smooth muscle has important implications to elucidating the role of bradykinin in vascular injury and inflammation.
Previous studies have shown that cells transformed with the Ras oncogene have an increased number of cell surface bradykinin receptors and selectively enhanced bradykinin receptor coupling to phosphoinositide turnover (8 -11). Although initial studies suggested that the effects were somewhat selective for the Ras oncogene, more recent studies have demonstrated that transformation with other oncogenes also stimulates an increase in bradykinin receptor expression (10,11). Of particular interest, the greatest increase in bradykinin receptor expression was reported in fibroblasts transformed by the constitutively activated receptor tyrosine kinase oncogene, trk (11). These observations suggest that activation of receptor tyrosine kinases may regulate bradykinin receptor expression possibly by a Ras-dependent mechanism. Despite these long-standing observations, there is no information as to whether activation of receptor tyrosine kinases known to stimulate Ras leads to an increase in bradykinin receptor expression in nontransformed cells. To approach this question, we recently surveyed several growth factors known to bind receptor tyrosine kinases and activate Ras in vascular smooth muscle cells (6). The studies demonstrated that platelet-derived growth factor (PDGF), 1 and, to a lesser extent, epidermal growth factor increased bradykinin receptor expression in cultured arterial smooth muscle cells (6).
PDGF is a mitogenic and chemotactic factor which is important in the response to vascular inflammation and injury (12,13). When released from platelets, PDGF binds and dimerizes a receptor tyrosine kinase (reviewed in Refs. 14 and 15). Receptor dimerization and activation of the intrinsic tyrosine kinase promotes membrane association of a guanine nucleotide exchange factor (e.g. SOS) and activation of membrane-bound Ras (16 -19). Activation of Ras initiates several signaling cascades, the best characterized of which is the Raf-1/MEK/MAPK pathway leading to regulation of gene transcription and cellular proliferation (20,21 and reviewed in Refs. 22 and 23). Given the potential involvement of both bradykinin and PDGF in the response to vascular injury and inflammation, we further examined the effect of PDGF on bradykinin B 2 receptor expression in arterial smooth muscle cells. 2 The results demonstrated that the B 2 receptor gene is a delayed early response gene for PDGF in arterial smooth muscle cells.
Treatment with Tyrosine Kinase Inhibitors-In preliminary experiments, we observed that the activity of the tyrosine kinase inhibitors, genistein and tyrphostin B46, decreased rapidly with storage in solution (Ϫ20°C). Therefore, all experiments were performed using freshly dissolved kinase inhibitors. Immunoblotting with the PY20 antiphosphotyrosine antibody (Calbiochem) demonstrated that PDGF-induced tyrosine phosphorylation was blocked by either genistein or tyrphostin B-46 (100 M, data not shown). Inhibition of PDGF-stimulated tyrosine kinase activity was also confirmed by performing paired control studies examining the effect of the inhibitors on the PDGF-stimulated immediate early response gene, c-fos (see below).
Bradykinin Receptor Binding Assay-Cell surface binding was measured as described previously (25). Confluent cultures of arterial smooth muscle cells grown in 12-well panels were deprived of serum for 24 h before treatment with PDGF at the concentrations and times indicated under "Results." After treatment with PDGF, the cells were washed and allowed to equilibrate in binding buffer containing: 25 mM HEPES (pH 7.0), 140 mM NaCl, 2 mM MgCl, 1 mM phenanthroline, 30 M captopril (gift from Squibb), and 50 g/ml bacitracin at 4°C. To determine the effect of PDGF on bradykinin receptor binding, saturation binding experiments were performed using [ 3 H]bradykinin in concentrations varying from ϳ10 pM-15 nM (110 Ci/mmol) in 600 l of binding buffer per well. For experiments measuring the time course and dose response of various PDGF ligands, total binding was determined using 0.5 nM [ 3 H]bradykinin. Nonspecific binding was determined in the presence of 2 M unlabeled bradykinin. The binding was terminated after 16 h by aspirating the buffer and rapidly washing the cells with 3 ml of ice-cold binding buffer. The remaining cell-bound counts were solubilized with 1 ml of 1 N NaOH and quantitated by scintillation counting (Beckman LS 3801). The saturation binding data were analyzed by a nonlinear curve-fitting program ("LIGAND" developed by Peter Munson at the NIH) to determine the best model and model parameters (K 1,i and R 1,i ) which fit the observed nontransformed data (26).
Preparation of Antisense 32 P-Labeled Riboprobes-A plasmid (403-8) containing 1.9 kilobases of cDNA for the rat bradykinin B 2 receptor was graciously provided by Dr. Kurt Jarnagin (27). This plasmid was cut with NcoI and transcribed using T7 polymerase yielding a 515-bp riboprobe (459-bp protected length). To measure rat c-fos, a partial cDNA was cloned using RNA isolated from PDGF-treated vascular smooth muscle cells by reverse transcription followed by polymerase chain reaction with the primers 5Ј-CAGCCGACTCCTTCTCTCCAG-CATG and 5Ј-TCCAGTTTTTCCTTCTCTTTCAGTAGATTGG (28). The resulting 437-bp product was cloned into pCRII (Invitrogen), and the sequence was confirmed by the dideoxynucleotide chain termination method. The plasmid was cut with SalI and transcribed with T7 polymerase yielding a 334-bp riboprobe (278-bp protected length). The pTRI-␤-actin-mouse template containing cDNA for the mouse ␤-actin (Ambion) was cut with HaeIII and transcribed using T7 polymerase generating a 207-bp length riboprobe (170-bp protected length). In some experiments, the pTRI-␤-actin-rat template containing cDNA for the rat ␤-actin (Ambion, 125-bp protected length) was also used. Antisense 32 P-labeled riboprobes were prepared by mixing 800 ng of cut plasmid, 20 units of RNasin, 0.5 mM each rATP, rUTP, and rGTP, 8.8 M [␣-32 P]CTP (50 Ci/assay for either the B 2 receptor or c-fos riboprobe and 15 Ci/assay for the ␤-actin riboprobe) and 1 unit of T7 RNA polymerase in transcription buffer (in mM: Tris 40, pH 7.5, MgCl 2 6, spermidine 2, NaCl 10) at a final volume of 20 l. The reaction was incubated for 4 h at room temperature, and the DNA was digested with DNase (1 unit for 15 min at 37°C). The sample was then diluted with an equal volume of gel loading buffer (80% formamide, 0.1% xylene cyanol, 0.1% bromphenol blue, 2 mM EDTA) and the full-length 32 Plabeled probe separated by electrophoresis in a 8 M urea, 5% polyacrylamide gel. The probes were identified by autoradiography, then excised from the gel and eluted into 350 l of buffer (0.5 M NH 4 acetate, 1 mM EDTA, 0.2% SDS) overnight. A 2-l sample of the probe was removed for quantitation by liquid scintillation counting, and the remainder was used immediately for the ribonuclease protection assay.
RNA Extraction-Total RNA was extracted from cultured cells using a modification of published procedures (29). Confluent cultures of arterial smooth muscle cells grown in 6-well panels (9 cm 2 /well) were deprived of serum for 24 h prior to treating with the indicated agents. After the indicated treatment, the buffer was aspirated, and the cells were scraped into 690 l of ice-cold GITC buffer (5.8 M guanidinium isothiocyanate, 38.6 mM sodium citrate, pH 7.0, 0.76% sarkosyl, and 0.7% ␤-mercaptoethanol). RNA was extracted by adding sequentially 50 l of 2 M sodium acetate (pH 4.0), 500 l of ice-cold water-saturated phenol, and 150 l of chloroform/isoamyl alcohol with vortexing. After incubation on ice for 15 min, the sample was centrifuged at 10,000 ϫ g for 20 min at 4°C, and the upper aqueous phase was removed to a new 1.5-ml Eppendorf tube. For the ribonuclease protection assay, the 32 Plabeled riboprobes were added directly to the RNA at this time. The RNA was then precipitated by adding 750 l of ice-cold isopropyl alcohol, vortexing, and incubating at Ϫ20°C for at least 30 min before centrifuging at 10,000 ϫ g for 20 min at 4°C. The supernatant was discarded and the RNA pellet was used immediately for the ribonuclease protection assay.
RNase Protection Assay-The procedures are slightly modified from those provided with the RPA II kit (Ambion). Briefly, sample RNA and the 32 P-labeled riboprobes are combined and precipitated as described above. The resulting pellet was resuspended in 20 l of hybridization buffer (80% formamide, 10 mM sodium citrate, 300 mM sodium acetate, 1 mM EDTA, pH 6.4) with vortexing, heated at 90°C for 4 min, and then allowed to hybridize overnight at 42°C. Nonhybridized probe was hydrolyzed by incubating with a 1:300 dilution of RNase (combination of RNase A and RNase T1 in Ambion's digestion buffer) for 30 min at 37°C. RNase was inhibited, and the remaining protected RNA was precipitated with a proprietary buffer (300 l, Ambion) and incubated at Ϫ20°C. The supernatant was removed, and the pellet was resuspended in 8 l of gel loading buffer (80% formamide, 0.1% xylene cyanol, 0.1% bromphenol blue, 2 mM EDTA) with vigorous vortexing and heated to 90°C ϫ 5 min before being subjected to electrophoresis on a 5% acrylamide, 8 M urea gel. The bands were identified by autoradiography and quantitated by densitometry (PDI system). The results for the B 2 receptor and c-fos were normalized for the amount of ␤-actin measured simultaneously in each sample. Unless otherwise stated, the results are expressed as fractional increase over vehicle-treated control cells.
Calcium Imaging in Individual Cells-The effect of PDGF on bradykinin-stimulated calcium response was assessed by a video microscopic digital analysis system (Photon Technology International Inc., South Brunswick, NJ) as described previously (7). Nearly confluent vascular smooth muscle cells grown on 25-mm coverslips were loaded with the calcium-specific dye Fura 2 by incubating with 5 M Fura 2/AM (Molecular Probes, Eugene, OR) for 40 -45 min. After washing once with DMEM/bovine serum albumin, cells were reincubated in DMEM/bovine serum albumin for 30 -45 min at 37°C to allow complete hydrolysis of 2 B. S. Dixon and M. J. Dennis, unpublished data. Fura 2/AM to Fura 2 inside the cell. Fura 2-loaded cells displayed stable and bright fluorescence at 340, 360, and 380 nm excitation, while unloaded cells had no detectable autofluorescence. Intracellular calcium was measured at 37°C in HEPES-buffered (25 mM, pH 7.4) Eagle's salt solution containing 0.1% bovine serum albumin and no bicarbonate.
Excitation light wavelengths of 340, 360, and 380 nm were obtained using dual excitation monochromators and a 75-watt xenon arc lamp (7). Cellular images were recorded at 10-s intervals before and after addition of the hormones. Bradykinin, angiotensin II, or vasopressin were added as 2 ϫ concentrated solutions to yield the final concentration indicated under "Results." The fluorescent images obtained by excitation at 340 nm and 380 nm were corrected for background fluorescence and camera dark current. Ratio images were formed by dividing the corrected 340 nm image by the corrected 380 nm image on a pixel by pixel basis. The intensities of all the pixels within the entire cell were summed and converted to [Ca 2ϩ ] i using Equation 1.
where R represents the ratio of fluorescence at 340 nm to 380 nm in the absence of calcium (R min ) and in the presence of maximally saturating levels of calcium (R max ) and F 0 /F s represents the ratio of fluorescence at 380 nm in the presence of zero calcium (F 0 ) and saturating levels of calcium (F s ). The values used for R max ϭ 8.5, R min ϭ 0.45, F 0 /F s ϭ 6.8, and K D ϭ 224 have been determined experimentally as described previously (7) (30). Cells grown to confluence in 4-cm 2 wells were loaded with [ 3 H]myoinositol (5 Ci/well) for 48 h in MEM containing 2% bovine serum albumin. After loading, the cells were gently washed with physiologic saline solution (ϫ 2) to remove unincorporated [ 3 H]inositol and allowed to equilibrate at the assay temperature (30°C) for 15 min. Then the buffer was aspirated and the experiment was initiated by adding 1 ml of fresh buffer containing the indicated concentration of agonist for the indicated time before terminating the reaction by scraping the cells into 1 ml of ice-cold 20% trichloroacetic acid. The trichloroacetic acid-insoluble material is removed by centrifugation at 3000 ϫ g for 10 min, and the resulting supernatant is extracted with 10 volumes of diethyl ether. The sample is alkalinized by the addition of 2 ml of 50 mM Tris base and then applied to an 0.8 ϫ 2-cm column of Dowex AG 1X-8 DEAE-cellulose (200 -400 mesh), and the 3 H-labeled compounds were eluted sequentially with 40

Characterization of PDGF Regulation of Bradykinin Receptor
Binding-PDGF is a covalent dimer of two proteins, A and B, forming three possible PDGF ligands (AA, AB, and BB). When released from platelets, PDGF binds and dimerizes two cell surface tyrosine kinase receptors (␣ and ␤, reviewed in Refs. 16 and 17). Incubation of cultured arterial smooth muscle cells with PDGF AB produced a dose-and time-dependent increase in cell surface bradykinin receptors (Fig. 1). The increase in receptor binding occurred within 4 h and peaked at 6 -12 h after addition of PDGF (Fig. 1, upper panel). Subsequently, there was a gradual decrease in receptor binding which remained above control levels for over 48 h. Dose-response curves demonstrated that both the PDGF AB and BB isomers but not PDGF AA stimulated a dose-dependent increase in bradykinin receptor binding (Fig. 1, middle panel). This is consistent with the ability of PDGF AB and BB but not AA to stimulate mitogenesis in these cultured cells (data not shown). Association binding isotherms further demonstrated that treatment with PDGF (AB or BB) produced a 2-4-fold FIG. 1. Effect of PDGF on cell surface bradykinin receptor binding. Cultured mesenteric arterial smooth muscle cells were grown to confluence and incubated in defined serum-free media for 24 h before treating with PDGF as described below. After treatment, cell surface BK receptors were measured as described under "Experimental Procedures." In the upper panel, cells were incubated with 10 ng/ml PDGF AB for the indicated times before measuring BK receptor binding. The results represent the specific binding (cpm/well) expressed as mean Ϯ S.E. from 4 separate cultures. Similar results were found in 3 additional experiments with peak receptor binding occurring within 6 h. In the middle panel, cells were incubated with the indicated concentrations of PDGF AB (q), BB (å), or AA (f) for 16 h before measuring BK receptor binding. The results represent the specific binding (cpm/well) expressed as mean of at least 2 separate cultures at each point. The dose-response curve for PDGF AB has been confirmed in 3 additional experiments. In the lower panel, cells were incubated with either 25 ng/ml PDGF AB (f) or buffer (Ⅺ) for 16 h before measuring cell surface BK binding in the presence of varying concentrations of [ 3 H] BK. The resulting association binding isotherms were analyzed by LIGAND 4.1, and the results from three separate paired experiments are combined and presented as a Scatchard curve. The calculated displacement curve is also shown (inset). The data were best fit by a model with one independent binding site (mean Ϯ 95% confidence limits for control: increase in the total number of cell surface receptors (Fig. 1,  lower panel) associated with a small decrease in receptor affinity (K d ϭ 0.46 Ϯ 0.12 nM in control cells and 0.72 Ϯ 0.09 nM after PDGF). Cell surface bradykinin receptor binding from either control or PDGF-treated cells was completely displaced by incubation with the specific B 2 receptor antagonist, HOE140, indicating that the receptors are of the B 2 receptor subtype (data not shown). The increase in cell surface receptors required new protein synthesis since treatment with either cycloheximide or actinomycin D completely prevented the PDGF-induced increase in cell surface bradykinin receptor binding (Fig. 2). Hence, PDGF induced a rapid increase in the translation and expression of cell surface B 2 receptors in cultured arterial smooth muscle cells.
Effect of PDGF on mRNA for the B 2 Receptor-Messenger RNA for the B 2 receptor was quantitated using a ribonuclease protection assay. Treatment with PDGF induced an increase in B 2 receptor mRNA which peaked between 2 and 4 h and declined slowly thereafter (Fig. 3). In many experiments, a lag phase of approximately 60 min occurred before any significant increase in B 2 receptor mRNA was observed (Fig. 3, inset). Addition of the transcriptional inhibitors, actinomycin D (Fig.  4) or DRB (not shown), up to 60 min after addition of PDGF completely blocked synthesis of new B 2 receptor mRNA. To determine whether the lag phase in B 2 receptor mRNA induction was secondary to the synthesis of new proteins, we examined the effect of PDGF in the presence of cycloheximide. Treatment with cycloheximide, either 1 g/ml as shown in Fig. 5 or 10 g/ml (not shown), did not inhibit, but rather superinduced both basal and PDGF-stimulated B 2 receptor mRNA levels. At these concentrations of cycloheximide (1 and 10 g/ml), protein synthesis was inhibited over 90% and 95%, respectively. These results demonstrate that the B 2 receptor gene is a delayed early response gene for PDGF and are compatible with an effect of PDGF to increase transcription of the B 2 receptor gene. The effects of PDGF on B 2 receptor mRNA are selective since no increase in mRNA for the angiotensin II AT1 a receptor was observed (data not shown, but see Figs. 7 and 8 below and Ref. 31).
Mechanism of the PDGF-induced Increase in B 2 Receptor mRNA-Most intracellular second messengers generated by PDGF involve activation of the tyrosine kinase on the cytosolic domain of the receptor (13)(14)(15). To test whether the PDGFinduced increase in B 2 receptor mRNA involved activation of tyrosine kinase, we examined the effect of the tyrosine kinase inhibitors genistein and tyrphostin B-46 (32). Since PDGF stimulation of the immediate early response gene, c-fos, is known to involve activation of tyrosine kinase (33), we compared the effects of these inhibitors on the PDGF-stimulated increase in mRNA for both the B 2 receptor and c-fos. Preliminary experiments demonstrated that exposure to PDGF stimulated an increase in c-fos mRNA which peaked at 15-30 min and decayed rapidly thereafter (data not shown). Therefore, mRNA for c-fos was measured at 30 min while B 2 receptor mRNA was measured after 3 h of exposure to PDGF. The results are shown in Table I and Fig. 6. Tyrphostin and genistein significantly inhibited the PDGF-stimulated increase in both c-fos and B 2 receptor mRNA. Genistein inhibited the increase in c-fos more than the B 2 receptor; however, interpretation of this observation was complicated by the fact that genis-tein also significantly increased basal levels of c-fos mRNA (Table I). These observations are consistent with the known signaling mechanisms for the PDGF receptor and indicate that activation of tyrosine kinase mediates the increase in both c-fos and the B 2 receptor. To further determine whether activation of Ras was responsible for the increase in B 2 receptor mRNA, we examined the effects of two reported Ras inhibitors, lovastatin and perillyl alcohol. These agents inhibit the activation of Ras by two different mechanisms (34). Pretreatment with either 50 M lovastatin or 1 mM perillyl alcohol for 48 h potently inhibited PDGF-induced mitogenesis but did not significantly inhibit either farnesylation of Ras or the PDGF-induced increase in B 2 receptor mRNA (data not shown). Higher doses of either inhibitor were associated with significant morphological evidence of cellular toxicity.
Further insight into the mechanism of the PDGF-induced increase in B 2 receptor mRNA was sought by examining the effects of two additional protein kinase inhibitors, staurosporine and calphostin C (Table I and Fig. 6). The nonselective protein kinase inhibitor, staurosporine, significantly inhibited both the PDGF-stimulated c-fos and B 2 receptor mRNA. In contrast, the more selective protein kinase C inhibitor, calphostin C, produced only a slight inhibition (24%) of PDGF-induced B 2 receptor mRNA and had no effect on the PDGF-stimulated increase in c-fos. Overall, these results demonstrate a similar dependence on PDGF-stimulated kinase activation for both c-fos and the B 2 receptor. However, stimulation of c-fos was more sensitive to genistein while the increase in the B 2 receptor was more sensitive to inhibition by calphostin C.
Effect of PDGF on Bradykinin Activation of Phospholipase C-The B 2 receptor is known to activate phosphoinositidespecific phospholipase C in arterial smooth muscle cells (7). To determine whether the increase in cell surface B 2 receptors stimulated by PDGF are functionally coupled to activation of phosphoinositide-specific phospholipase C, we measured the effect of PDGF on bradykinin-stimulated inositol phosphate formation (Fig. 7) and intracellular calcium release (Fig. 8). As shown in Fig. 7 (upper panel), overnight treatment with PDGF produced at least a 2-fold increase in the maximal bradykininstimulated IP 2 and IP 3 formation. The EC 50 for bradykininstimulated IP 3 release appears to be shifted to the right which is consistent with the observed PDGF-stimulated decrease in receptor affinity (see above). However, analysis of the exact EC 50 is complicated because the time to maximal IP 3 formation varies with the concentration of bradykinin employed. Consistent with the effects on IP 3 formation, overnight treatment with PDGF also produced a marked increase in bradykinin-stimulated calcium release (Fig. 8, upper panel). As shown in Fig. 8  (upper panel), PDGF significantly enhanced the kinetics of FIG. 5. Effect of cycloheximide on the PDGF-stimulated increase in B 2 receptor mRNA. Confluent serum-deprived cultures were treated with either vehicle or cycloheximide (1 g/ml) for 60 min before adding either buffer or PDGF AB (12 ng/ml) for the indicated times. The upper panel shows representative autoradiographic data for vehicle-treated (A) and cycloheximide-treated (CHx, B) cells. The autoradiograms were exposed for 3 days with an intensifying screen to obtain the data shown for the B 2 receptor and 16 h without the screen for ␤-actin. Note that the control cells shown on the left side of B were not exposed to cycloheximide. In the lower panel, mRNA for the B 2 receptor was quantitated as described in Fig. 3, and the results are expressed as the fractional increase over the simultaneous vehicletreated control cells. Each point represents the mean Ϯ S.E. of 4 -5 separate cultures from 2 different experiments.

TABLE I Effect of protein kinase inhibitors on basal and PDGF-stimulated mRNA for the B 2 receptor and c-fos
Confluent serum-deprived cultures were treated with either vehicle (0.1% dimethylsulfoxide) or the indicated concentration of inhibitor for 30 min prior to adding additional media with or without PDGF AB (25 ng/ml). After incubation for an additional 30 min (c-fos) or 3 h (B 2 receptor), the cultures were washed and mRNA for the B 2 receptor, c-fos, and ␤-actin were measured as described under "Experimental Procedures." The results for the B 2 receptor and c-fos were normalized for differences in ␤-actin and expressed as fractional increase over vehicle-treated basal levels of mRNA (i.e. vehicle-treated basal mRNA for c-fos and B 2 receptor ϭ 1 Ϯ 0.2). Data for % of control ϭ (PDGF i Ϫ basal i ) ϫ 100/(PDGF c Ϫ basal c ), where the subscripts i and c represent inhibitor-treated and vehicle-treated, respectively. All data represent mean Ϯ S.E. of 8-12 independent determinations. bradykinin-stimulated calcium release and decreased the threshold concentration to less than 0.1 nM bradykinin. Overall, these results demonstrate that treatment with PDGF produced a similar increase in both the expression of cell surface B 2 receptors and maximal bradykinin-stimulated inositol phosphate generation suggesting the new receptors are functionally coupled to activation of phospholipase C. In addition, prolonged exposure to PDGF also appeared to further enhance the kinetics of bradykinin-stimulated intracellular calcium accumulation.

Inhibitor
To further examine whether the effects of PDGF were selective for bradykinin, we also examined the effect of PDGF on angiotensin II-and arginine vasopressin-stimulated inositol phosphate formation (Fig. 7) and calcium release (Fig. 8). As shown in the two lower panels of Figs. 7 and 8, angiotensin II produced only a small increase in inositol phosphate formation and calcium release while responses to vasopressin were similar to those produced by bradykinin in control cells. Chronic treatment with PDGF did modify the kinetics of hormonestimulated calcium release leading to a more rapid peak followed by a discrete second increase in intracellular calcium concentration (Fig. 8, lower two panels). However, in contrast to the effects on bradykinin, pretreatment with PDGF produced essentially no increase in either vasopressin-or angiotensin II-stimulated inositol phosphate formation or the total time-averaged increase in intracellular calcium concentration (Figs. 7 and 8, lower two panels). These results suggest that chronic exposure to PDGF has a general effect on the kinetics of hormone-stimulated intracellular calcium concentration. However, these effects of PDGF are not mediated by a generalized increase in receptor coupling to inositol phosphate formation. These results further demonstrate that PDGF stimulates a selective increase in bradykinin receptors which are directly coupled to increased inositol phosphate formation.

DISCUSSION
Bradykinin and PDGF are local mediators of the response to vascular injury and inflammation. PDGF is released by degranulation of platelets while bradykinin is released during activation of the contact activation pathway of coagulation. The present study now further demonstrates that the bradykinin B 2 receptor is also an early response gene for PDGF in cultured arterial smooth muscle cells. Similar to other early response genes induced by PDGF (35)(36)(37)(38)(39)(40), the increase in B 2 receptor mRNA did not require new protein synthesis and was superinduced by pretreatment with cycloheximide. After a short delay, PDGF-induced a 6 -10-fold increase in steady state mRNA which peaked between 2 and 4 h and lead to a 2-4-fold increase in cell surface B 2 receptors at 6 -12 h. Treatment with PDGF also increased bradykinin-stimulated inositol phosphate formation and markedly enhanced intracellular calcium release, suggesting that the increased expression of B 2 receptors are functionally coupled to activation of phosphoinositide-specific phospholipase C. Finally, the effects of PDGF were selective for the bradykinin receptor since PDGF did not increase mRNA for the AT 1a receptor (data not shown) or the activation of phosphoinositide-specific phospholipase C by either angiotensin II or arginine vasopressin. Consistent with these observations, previous investigators have also shown that PDGF decreased AT 1 receptor binding and mRNA in vascular smooth muscle cells (31). Overall, these results demonstrate that PDGF exerts a selective effect to enhance vascular responsiveness to cellular signals generated by bradykinin via the B 2 receptor on arterial smooth muscle cells.
The observation that the B 2 receptor is an early response gene for PDGF has important implications both for the cellular mechanisms which mediate the effect as well as the potential role of the B 2 receptor in the biological response to PDGF. Studies in fibroblast cell lines using differential screening have shown that short-term exposure to PDGF or serum induces a large but restricted population (approximately 0.1-1%) of total cellular genes (35,41,42). Kinetic studies have suggested that these early response genes can generally be divided into two groups exhibiting either an immediate (Յ30 min) or delayed (Ͼ30 min) induction (42). Immediate early response genes include the transcription factors c-fos (37-39), egr-1 (43), and the chemokine, KC (35,44), while delayed responses are typical for the transcription factor c-myc (36,39) and the chemokine, JE/MCP-1 (35,45 and see Ref. 40 for review). The PDGFinduced increase in B 2 receptor mRNA exhibited the slower kinetics observed for delayed early response genes. This was confirmed by the observation that addition of the transcriptional inhibitors, actinomycin D or DRB, 1 h after exposure to PDGF completely blocked the subsequent increase in B 2 receptor mRNA.
For most early response genes, the growth factor-induced increase in mRNA involves a combination of both enhanced gene transcription and increased mRNA stability (42). Several elements, particularly when located in the 3Ј-untranslated region of the gene have been shown to regulate mRNA stability (46). The best characterized of these are the AU-rich regions which are associated with destabilizing mRNA and increasing mRNA degradation (47)(48)(49). Factors which bind to these regions are candidates for regulation by signal transduction pathways such as protein kinase C (50,51). The rat B 2 receptor gene has two ATTTA motifs in its 3Ј-untranslated region which might be associated with regulating mRNA stability (52). However, in preliminary studies, we have found that the half-life for B 2 receptor mRNA is over 4 h and there was no increase in stability after treatment with PDGF. 2 Further studies are required, but overall these results suggest that the PDGF-induced increase in B 2 receptor mRNA is mediated primarily at the level of gene transcription.
Elucidation of the signaling pathways by which PDGF mediates gene transcription is still in its infancy. PDGF is a dimer of two proteins, A and B, forming three ligands (AA, AB, and BB) which bind and dimerize two PDGF receptors, ␣ and ␤ (13)(14)(15). PDGF-A ligand only binds the PDGF␣ receptor while the PDGF-B ligand can bind either receptor. The cytosolic domains of the two receptors are different, and differences in signal transduction pathways as well as gene transcription have also been observed (53)(54)(55)(56)(57). Nevertheless, both receptors have intrinsic tyrosine kinase activity and have been shown to activate Ras (19,57). In our studies, treatment with PDGF-AA had no effect on B 2 receptor expression, suggesting that the effects of PDGF are mediated via the PDGF␤ receptor. These results are compatible with the low level of expression of PDGF␣ receptors generally found on cultured rat vascular smooth muscle cells (55,58). However, PDGF␣ has been shown to stimulate some early response genes even when expressed at low levels in vascular smooth muscle cells (53,54). Hence, it will be important in future experiments to determine whether the inability of PDGF-AA to stimulate B 2 receptor expression involves differences in postreceptor signaling mechanisms between the two PDGF receptors.
Consistent with our observations for the B 2 receptor, most (but not all) PDGF-induced early response genes are blocked by inhibition of tyrosine kinase activation (33,59). Activation of tyrosine kinase stimulates a number of second messenger pathways which may regulate gene transcription (see Refs. 22, 23, and 60 for reviews). For the prototypical immediate early response gene c-fos, two pathways have been shown to be important: the Ras-mediated activation of mitogen-activated protein kinases which increase transcription via the serum response element and the STAT pathway which directly activates transcription via the sis-inducible element (61-63 and reviewed in Ref. 60). A role for Ras was suggested by earlier studies demonstrating increased expression of cell surface B 2 receptors in cells transformed with oncogenic Ras (8 -11). Analysis of the recently reported promoter region (Ϫ1225 to ϩ235) of the B 2 receptor gene (52) does not reveal any consensus serum response element or CArG box to bind the SRF and no STAT binding elements such as the sis-inducible element. However, there are several potential transcription factor binding sites which are known targets for Ras-activated signaling pathways. These include binding sites for Ets-related factors (i.e. Ets-1 and PEA) as well as several AP-1 sites. Moreover, the AP-1 sites may be functionally important as we observed that activation of protein kinase C with phorbol 12-myristate 13-acetate increases bradykinin receptor binding (6). However, maximal concentrations of phorbol 12-myristate 13-acetate induced only 30 -50% of the increase in B 2 receptor expression as that observed with PDGF. This suggests that additional signaling pathways are involved in mediating the effects of PDGF. Further studies will be needed to determine whether activation of Ras is involved in PDGF stimulation of the B 2 receptor gene.
An important feature of the B 2 receptor gene is its delayed response to PDGF. The kinetics of its response are quite different from the immediate early response genes such as c-fos and egr-1, which are regulated by Ras. The mechanisms which regulate delayed transcription of the immediate response genes are unclear. Stiles and co-workers have pursed these mechanisms for the monocyte chemotactic factor, JE/MCP-1, and have identified several novel, widely spaced elements in both the 5Ј-and 3Ј-flanking regions that are both required for PDGF-induced transcription of this gene (64,65). The 3Ј motif, called the 3Ј immediate response box (3Ј-IRB, TTTTGTA) was found to be present in the proximal 3Ј-untranslated region of all known immediate response genes (64). The immediate response box was found to be effective only when located in the 3Ј-flanking region of the gene; however, its activity was independent of its orientation as the sequence TACAAAA functions equally well (64). This latter sequence is also present in the 3Ј region of the B 2 receptor gene (ϩ1988 to ϩ1994). In addition, these investigators more recently identified two novel elements (sequences II and III) in the 5Ј region of the MCP-1 gene which bind a serine/threonine phosphoprotein and mediate the slow kinetics of induction (65). We examined the 5Ј region of the B 2 receptor gene and have identified two elements having 80% homology to these two elements (Ϫ858 to Ϫ843 of the B 2 receptor gene is homologous to sequence II and Ϫ803 to Ϫ789 of the B 2 receptor gene is homologous to sequence III). Further study will be required to determine whether these sequences are involved in the delayed response of the B 2 receptor gene to PDGF.
The observation that the B 2 receptor gene is an early response gene for PDGF suggests that it may have an important role in regulating vascular smooth muscle responses to PDGF during vascular injury. Since mitogenesis is an important biological response to PDGF, we have examined the effects of bradykinin on mitogenesis (6,7). Bradykinin alone had only a modest and variable effect on mitogenesis; however, we found that in arterial smooth muscle cells, bradykinin was a potent inhibitor of PDGF-induced mitogenesis. These potent inhibitory effects of bradykinin appeared to be mediated primarily by the B 1 receptor (6). The effects of the B 2 receptor on mitogenesis are not clear. However, based upon the observation that signal transduction via the B 2 receptor, like that of the angiotensin, AT 1 receptor, is primarily mediated by activation of a phosphoinositide-specific phospholipase C, its effects are likely to be mitogenic and hypertrophic (6,7). Other possible roles for up-regulation of the B 2 receptor during inflammation may be to regulate glucose transport and cell motility. The glucose transporter has been shown to be an early response gene for PDGF (66). Since bradykinin is known to stimulate glucose transport (67), the rapid PDGF-induced up-regulation of these two genes may allow increased glucose transport into metabolically active cells. Regulation of cell motility and chemotaxis is another important biological response mediated by PDGF during tissue injury (12). Recent studies in fibroblasts have demonstrated that bradykinin can regulate the actin cytoskeleton via activation of the small G protein, Cdc42 (68). This may suggest a role for up-regulation of B 2 receptors in modulating cell morphology and possibly the chemotactic response to PDGF.
Taken together, these observations suggest another interaction between platelets and the coagulation cascade during vascular injury. PDGF released from degranulating platelets may stimulate the rapid up-regulation of B 2 receptors on the underlying vascular smooth muscle cells. Although the exact biological role for these B 2 receptors is not currently known, based upon the similarities of the signal transduction pathways for the B 2 receptor and the angiotensin AT 1 receptor, we expect that the B 2 receptor may be proinflammatory (7). This hypothesis would be consistent with the recent observations that in the absence of an angiotensin-converting enzyme inhibitor, treatment with a specific B 2 receptor antagonist decreased neointimal formation after arterial injury (69). Further studies are in progress to better characterize the interactions between bradykinin and PDGF in the response to vascular injury.