Modulation of Voltage-dependent Ca 2 1 Channels in Rabbit Colonic Smooth Muscle Cells by c-Src and Focal Adhesion Kinase*

There is emerging evidence indicating that smooth muscle contraction and Ca 2 1 influx through voltage-dependent L-type Ca 2 1 channels are regulated by tyrosine kinases; however, the specific kinases involved are largely un-known. In rabbit colonic muscularis mucosae cells, tyro-sine-phosphorylated proteins of ; 60 and 125 kDa were observed in immunoblots using an anti-phosphotyrosine antibody and were identified as c-Src and focal adhesion kinase (FAK) by immunoblotting with specific antibodies. FAK co-immunoprecipitated with c-Src, and the phosphorylation of the c-Src z FAK complex was markedly enhanced by platelet-derived growth factor (PDGF) BB. The presence of activated c-Src in unstimulated cells was identified in cell lysates by immunoblotting with an antibody recogniz-ing the autophosphorylated site (P416Y). In whole-cell patch-clamp studies, intracellular dialysis of a Src substrate peptide and anti-c-Src and anti-FAK antibodies suppressed Ca 2 1 currents by 60, 62, and 43%, respectively. In contrast, intracellular dialysis of an anti-mouse IgG or anti-Kv1.5 antibody did not inhibit Ca 2 1 currents. Co-dialysis of anti-c-Src and anti-FAK antibodies inhibited Ca 2 1 currents (63%) equivalent to dialysis with the anti-c-Src antibody alone. PDGF-BB enhanced Ca 2 1 currents by 43%, which Immunoprecipitations and Western Blotting— Following a 5-min treatment with or without 50 ng/ml PDGF-BB, cell suspensions were centrifuged, and cell pellets were lysed in a Triton X-100 lysis buffer (50 m M Tris-HCl (pH 8.0), 150 m M NaCl, 0.02% sodium azide, 1 m M phen-ylmethylsulfonyl fluoride, 1 m M aprotinin, 1 m M leupeptin, 1 m M pepsta- tin A, and 1% Triton X-100). Lysates were centrifuged at 12,000 rpm at 4 °C for 10 min, and protein contents of the supernatants were deter- mined. For immunoprecipitation study, equal amounts of lysate protein (500 m g) were incubated overnight at 4 °C with a monoclonal anti-c-Src antibody. Immune complexes were recovered with protein A-Sepharose beads. The beads were washed, resuspended in sample buffer (50 m M Tris-HCl (pH 6.0), 5% b -mercaptoethanol, 2% SDS, 0.1% bromphenol blue, and 10% glycerol), and boiled at 100 °C for 5 min. Protein samples were separated by 8% SDS-polyacrylamide gel electrophoresis. After electrophoresis, proteins were transferred to nitrocellulose membranes. Membranes were blocked using 2% nonfat dried milk in phosphate-buffered saline (pH 7.2) and incubated overnight with an anti-Tyr(P) (PY20), monoclonal anti-FAK, anti-pp60 c- src , or anti-c-Src (P416Y) antibody (1 m g/ml each) at 4 °C. Immunoreactive bands were visualized by chemiluminescence. The L-type Ca 2 1 channel polyclonal antibody raised against residues 818–835 of the a 1C subunit was used to detect phosphorylation and interaction of the Ca 2 1 channel with c-Src and FAK. Colonic

The influx of extracellular Ca 2ϩ is a prerequisite for many cellular functions including cell proliferation and motility. In gastrointestinal smooth muscle, the upstroke of action potential is principally mediated by Ca 2ϩ influx through voltage-dependent L-type Ca 2ϩ channels and is responsible for initiation of contraction. A variety of neurotransmitters and hormones modulate Ca 2ϩ channel activity through protein phosphorylation (1). Modulation of Ca 2ϩ channel activity by serine/threonine kinases such as cAMP-dependent protein kinase and protein kinase C has been well established (2,3), and phosphorylation sites have been identified on the ␣ subunit of L-type Ca 2ϩ channels in vascular smooth muscle (4). In addition to their roles in growth and differentiation, accumulating evidence suggests that tyrosine kinases are involved in the regulation of smooth muscle contraction. For instance, activation of both G protein-coupled receptors and growth factor receptors leads to smooth muscle contraction that is accompanied by tyrosine phosphorylation of a number of proteins (5,6). Moreover, smooth muscle contraction can also be inhibited by structurally unrelated tyrosine kinase inhibitors (7). The fact that Ca 2ϩ currents are attenuated by tyrosine kinase inhibitors and enhanced by growth factors in smooth muscle cells (8) points toward a novel mechanism for tyrosine kinases in the regulation of smooth muscle function.
One of the earlier signaling events associated with activation of G protein-coupled receptors, particularly G i protein-coupled receptors, and receptor tyrosine kinases involves activation of c-Src (9 -11). Downstream signaling events of c-Src include the formation of complexes among Shc, Grb2, and Sos and activation of the Ras and mitogen-activated protein (MAP) 1 kinase cascade (12). Focal adhesion kinase (FAK) is a potential substrate for c-Src, and once phosphorylated, it may provide a docking site for SH2 domains of the adaptor proteins such as Grb2. In addition, FAK may facilitate activation of c-Src by displacement of the inhibitory C-terminal tyrosine phosphorylation (13). Interestingly, activation of G protein-coupled receptors and receptor tyrosine kinases increases the activities of c-Src and FAK in smooth muscle (10,14,15). The inhibition of basal Ca 2ϩ currents by tyrosine kinase inhibitors suggests that there may be constitutively activated tyrosine kinase(s) that up-regulates Ca 2ϩ channel activity. A possible candidate is c-Src because of its high levels in smooth muscle (16). Recent studies in vascular smooth muscle cells indicate that c-Src may be involved in the regulation of Ca 2ϩ channels based on the finding that intracellular dialysis of c-Src enhances Ca 2ϩ currents (17). However, it is not known whether other downstream signaling molecules may be involved in the regulation of smooth muscle Ca 2ϩ channels.
In this study, we have examined the kinase activity of c-Src and its association with FAK in rabbit colonic smooth muscle cells. We have also evaluated the roles of c-Src and FAK as well as their downstream components Ras and MAP kinase in the regulation of basal Ca 2ϩ channel activity and their involvement in platelet-derived growth factor (PDGF)-induced enhancement of Ca 2ϩ currents. Our results demonstrate that basal Ca 2ϩ currents are modulated by c-Src and FAK, but not Ras/MAP kinase, in differentiated smooth muscle cells. Furthermore, c-Src and FAK are involved in the functional cou-pling of PDGF receptors and Ca 2ϩ current enhancement, which is consistent with an increased phosphorylation of these two kinases leading to the tyrosine phosphorylation of the Ca 2ϩ channel.

EXPERIMENTAL PROCEDURES
Electrophysiological Recordings-Single smooth muscle cells were freshly dispersed from rabbit colonic muscularis mucosae as described previously (8). Ca 2ϩ currents were recorded using the whole-cell configuration of the patch-clamp technique (18). All experiments were performed at room temperature (ϳ25°C) using an Axopatch 200A patch-clamp amplifier (Axon Instruments, Inc., Foster City, CA). Patch pipettes were pulled from thin-walled borosilicate glass, and the resistance was 3-5 megohms when filled with internal solution. The internal solution consisted of 100 mM cesium aspartate, 30 mM CsCl, 2 mM MgCl 2 , 5 mM HEPES, 5 mM EGTA, 5 mM ATP, and 0.1 mM GTP (pH 7.2 with CsOH). The cells were continuously perfused with HEPES-buffered physiological salt solution (135 mM NaCl, 5.4 mM KCl, 0.33 mM NaH 2 PO 4 , 5 mM HEPES, 1 mM MgCl 2 , 2 mM BaCl 2 , and 5.5 mM glucose (pH 7.4 with NaOH)). Antibodies were applied directly to the cells by means of diffusional exchange during standard whole-cell patch-clamp recording, and current recordings were initiated 4 min after rupture of the membrane to allow adequate intracellular dialysis. Pulse generation and data acquisition were performed with a PC computer (Deskpro 486/33M, Compaq, Houston, TX) with pclamp6.0 software (Axon Instruments, Inc.). Currents were filtered at 1 kHz and normalized with respect to cell capacitance. Series resistance did not exceed 5 megohms and was not compensated. The average cell capacitance was 64.6 Ϯ 0.9 pF (n ϭ 128). Currents in the absence and presence of antibody dialysis were obtained on the same day within the same population of cells.
Immunoprecipitations and Western Blotting-Following a 5-min treatment with or without 50 ng/ml PDGF-BB, cell suspensions were centrifuged, and cell pellets were lysed in a Triton X-100 lysis buffer (50 mM Tris-HCl (pH 8.0), 150 mM NaCl, 0.02% sodium azide, 1 mM phenylmethylsulfonyl fluoride, 1 M aprotinin, 1 M leupeptin, 1 M pepstatin A, and 1% Triton X-100). Lysates were centrifuged at 12,000 rpm at 4°C for 10 min, and protein contents of the supernatants were determined. For immunoprecipitation study, equal amounts of lysate protein (500 g) were incubated overnight at 4°C with a monoclonal anti-c-Src antibody. Immune complexes were recovered with protein A-Sepharose beads. The beads were washed, resuspended in sample buffer (50 mM Tris-HCl (pH 6.0), 5% ␤-mercaptoethanol, 2% SDS, 0.1% bromphenol blue, and 10% glycerol), and boiled at 100°C for 5 min. Protein samples were separated by 8% SDS-polyacrylamide gel electrophoresis. After electrophoresis, proteins were transferred to nitrocellulose membranes. Membranes were blocked using 2% nonfat dried milk in phosphatebuffered saline (pH 7.2) and incubated overnight with an anti-Tyr(P) (PY20), monoclonal anti-FAK, anti-pp60 c-src , or anti-c-Src (P416Y) antibody (1 g/ml each) at 4°C. Immunoreactive bands were visualized by chemiluminescence. The L-type Ca 2ϩ channel polyclonal antibody raised against residues 818 -835 of the ␣ 1C subunit was used to detect phosphorylation and interaction of the Ca 2ϩ channel with c-Src and FAK. Colonic smooth muscle was homogenized in 300 mM KCl/phosphate-buffered saline with 1% digitonin. For immunoprecipitation, 1 mg of total protein was precleared with protein A-Sepharose beads for 30 min at 4°C.
Statistical Analysis-Data are expressed as means Ϯ S.E. Data analysis were performed using Student's t test, and differences with p Ͻ 0.05 were considered significant.

Presence of c-Src and FAK in Rabbit Colonic Smooth Muscle
Cells-Tyrosine-phosphorylated proteins were identified in freshly isolated smooth muscle cells of rabbit colonic muscularis mucosae by immunoblotting with the anti-phosphotyrosine antibody (PY20). Fig. 1A shows several tyrosine-phosphorylated proteins in unstimulated cells, including bands at ϳ60 and 125 kDa. These two bands were identified as c-Src and FAK, respectively, by stripping the anti-phosphotyrosine blot and reprobing with specific antibodies (data not shown). The phosphorylation of both c-Src and FAK was enhanced following treatment of cells with PDGF-BB (50 ng/ml) for 5 min (Fig. 1A). Immunoprecipitation of the cell lysates with an anti-c-Src antibody followed by immunoblotting with PY20 showed that a tyrosine-phosphorylated 125-kDa protein co-precipitated with c-Src in unstimulated cells, and the association of the phosphorylated protein with c-Src was enhanced following PDGF-BB treatment (Fig. 1B).
To further demonstrate that c-Src and FAK co-precipitated, cell lysates were immunoprecipitated with the anti-c-Src antibody and blotted with an anti-FAK antibody. Fig. 2A shows that there was a significant association of c-Src with FAK in unstimulated cells, which was enhanced following PDGF-BB treatment. The presence of a c-Src⅐FAK complex in unstimulated cells suggests that a constitutively activated c-Src may be present under resting conditions. This was confirmed by immunoblotting with an anti-c-Src antibody (P416Y) (Fig. 2B) that recognizes c-Src that is phosphorylated at Tyr 416 and correlates with activated c-Src (19). This observation is consistent with the previous findings of a high degree of activated c-Src in resting smooth muscle (16) and the ability of tyrosine kinase inhibitors to attenuate basal Ca 2ϩ channel activity (8).
In addition, activated c-Src was also enhanced by PDGF-BB (Fig. 2B).
Inhibition of Ca 2ϩ Currents by Src Substrate Peptide and Anti-c-Src and Anti-FAK Antibodies-To demonstrate whether c-Src and FAK may be involved in the regulation of Ca 2ϩ currents, we examined the effects of intracellular dialysis of a c-Src substrate peptide and anti-c-Src and anti-FAK antibodies on Ca 2ϩ currents in single smooth muscle cells using the patchclamp technique. Ca 2ϩ currents were recorded from a holding potential of Ϫ50 mV using Ba 2ϩ (2 mM) as the charge carrier and normalized with respect to cell capacitance (8). The synthetic Src substrate peptide at high concentrations can result in a significant inhibition of Src kinase activity (20) and has FIG. 1. Anti-phosphotyrosine immunoblot of whole-cell lysates from rabbit colonic smooth muscle. The Western blots show tyrosine-phosphorylated proteins in whole-cell lysates (A) and following immunoprecipitation (I.p.) with the anti-c-Src antibody (B). Lysates were prepared from unstimulated cells (Con) and after treatment for 5 min with PDGF-BB (50 ng/ml). Proteins corresponding to ϳ60 and 125 kDa were identified as c-Src and FAK by stripping and reprobing with specific antibodies. Tyrosyl-phosphorylated proteins were enhanced following treatment with PDGF. B shows that phosphorylated protein corresponding to FAK immunoprecipitates with c-Src, and this association is enhanced following PDGF-BB treatment. The equal staining of mouse IgG heavy chain indicates equal loading of proteins.
previously been shown to inhibit Ca 2ϩ currents in vascular smooth muscle cells (17). Intracellular dialysis of the Src substrate peptide Cdc2-(6 -20)-NH 2 (60 M), derived from p34 cdc2 (21), suppressed the Ca 2ϩ currents from Ϫ6.19 Ϯ 0.65 (n ϭ 6) to Ϫ2.52 Ϯ 0.45 (n ϭ 8) pA/pF (p Ͻ 0.001), corresponding to a decrease of 59.3% at the test potential of ϩ10 mV (Fig. 3, A and  B). Neither the threshold nor the reversal potential was altered by the Src substrate peptide. Because the Src substrate peptide does not discriminate between members of the Src family, we studied the effects of a monoclonal c-Src antibody (anti-c-Src antibody) on Ca 2ϩ currents to determine the involvement of c-Src in the modulation of Ca 2ϩ channel activity. In these sets of experiments, the control Ca 2ϩ current at the test potential of ϩ10 mV was Ϫ6.88 Ϯ 0.70 pA/pF (n ϭ 6), whereas it was reduced to Ϫ2.64 Ϯ 0.54 pA/pF (n ϭ 10; p Ͻ 0.0003) by intracellular application of the anti-c-Src antibody (10 g/ml), representing an inhibition of 61.6% (Fig. 4, A and C). As illustrated in Fig. 4B, inhibition began soon after the onset of the dialysis process and reached its maximum within 4 min. The steadystate inactivation kinetics of the Ca 2ϩ currents were not altered by the anti-c-Src antibody (data not shown). To determine the specificity of the anti-c-Src antibody on Ca 2ϩ currents, we examined the effects of intracellular dialysis of an anti-mouse IgG antibody (10 g/ml) and an anti-Kv1.5 antibody (20 g/ml). As shown in Fig. 5, the amplitudes of Ca 2ϩ currents were not altered by either antibody.
Since FAK co-immunoprecipitates with c-Src in colonic smooth muscle cells, we examined whether FAK could modulate Ca 2ϩ channel activity. Cells were dialyzed with an anti-FAK antibody raised against residues 748 -1053 of human FAK, which includes the region required for c-Src-induced Grb2 association with FAK (22). Fig. 6A shows that dialysis of the anti-FAK antibody (7.3 g/ml) resulted in a decrease of 43.1% in the peak amplitude of Ca 2ϩ currents (from Ϫ7.31 Ϯ 1.31 (n ϭ 6) to Ϫ4.16 Ϯ 0.43 (n ϭ 6) pA/pF (p Ͻ 0.05)). To test whether the action of c-Src and FAK on Ca 2ϩ currents was additive, the cells were co-dialyzed with anti-c-Src and anti-FAK antibodies. Application of both antibodies to the cells reduced the peak currents by 62.9% (from Ϫ8.91 Ϯ 0.70 (n ϭ 4) to Ϫ3.31 Ϯ 0.29 (n ϭ 4) pA/pF (p Ͻ 0.05)) (Fig. 6B). This inhibition is comparable to that obtained with the anti-c-Src antibody alone, suggesting that c-Src and FAK may modulate Ca 2ϩ currents through a complex of c-Src and FAK, rather than acting separately.

PDGF-induced Enhancement of Ca 2ϩ Currents Is Mediated by c-Src and FAK-Previously
, we have shown that Ca 2ϩ currents are enhanced by epidermal growth factor (8). In the present study, Ca 2ϩ currents were also enhanced by PDGF-BB (Fig. 7A). Perfusion of PDGF-BB (50 ng/ml) resulted in a 42.8 Ϯ 4.2% (n ϭ 11; p Ͻ 0.001) increase in peak currents. The onset of this enhancement was rapid and reached a maximum at ϳ2 min. It is well known that c-Src and FAK are involved in the regulation of PDGF-mediated responses (23). We therefore investigated the potential roles of c-Src and FAK in PDGF-BBmediated enhancement of Ca 2ϩ currents. Similar to the observation described above, dialysis of the cells with the anti-c-Src antibody (10 g/ml), but not the anti-mouse IgG antibody (10 mg/ml), resulted in a reduction of the basal Ca 2ϩ currents. PDGF-BB-mediated increases in Ca 2ϩ currents were abolished by the anti-c-Src antibody, whereas they remained unchanged in the presence of the anti-mouse IgG antibody (Fig. 7B). Moreover, dialysis with the anti-FAK antibody also suppressed the basal Ca 2ϩ currents and blocked PDGF-BB-mediated enhancement of Ca 2ϩ currents (Fig. 7B).
Effects of Anti-Ras Antibody and PD 098059 on Ca 2ϩ Currents-Since activation of c-Src and FAK stimulates the Ras/ MAP kinase cascade (12,13), and the anti-FAK antibody used in this study might block the binding of Grb2 to FAK, we investigated the roles of the Ras/MAP kinase pathway in the regulation of Ca 2ϩ currents using an anti-Ras antibody and the MEK inhibitor PD 098059 (24). As shown in Fig. 7, intracellular dialysis of the anti-Ras antibody (5 g/ml) did not depress the basal Ca 2ϩ current or PDGF-BB-stimulated Ca 2ϩ currents. In the presence of the anti-Ras antibody, PDGF-BB enhanced Ca 2ϩ currents by 39.5 Ϯ 6.7% (n ϭ 5). Furthermore, perfusing the cells with 30 M PD 098059 also did not inhibit basal Ca 2ϩ currents or PDGF-mediated enhancement (37 Ϯ 8.1%, n ϭ 5). Interaction of c-Src with ␣ 1C Ca 2ϩ Channel-To determine whether c-Src or FAK directly associates with the Ca 2ϩ channel, rabbit colonic tissue was treated with PDGF-BB and homogenized in 1% digitonin. The samples were immunoprecipitated with cardiac anti-␣ 1C , anti-Tyr(P) (PY20), anti-Src, and anti-FAK antibodies and immunoblotted with the anti-␣ 1C antibody. Fig. 8 shows the presence of the ␣ 1C L-type Ca 2ϩ channel (first lane), corresponding to a band of ϳ210 kDa. Bands of similar size were also observed following immunoprecipitation with anti-phosphotyrosine (PY20) (second lane) and anti-c-Src (third lane) antibodies and immunoblotting with the anti-␣ 1C antibody, suggesting that the Ca 2ϩ channel is phosphorylated by tyrosine kinases and associates with c-Src kinase. However, the anti-FAK antibody failed to immunoprecipitate the ␣ 1C subunit (data not shown).

DISCUSSION
In this study, we have examined the roles of the non-receptor tyrosine kinases c-Src and FAK in the regulation of Ca 2ϩ channel activity in differentiated smooth muscle cells of rabbit colon. The results demonstrate the presence of activated c-Src, which interacts with FAK to form a c-Src⅐FAK complex, in unstimulated smooth muscle cells. The phosphorylation of the c-Src⅐FAK complex is enhanced by PDGF-BB. Furthermore, we provide evidence that Ca 2ϩ channels are constitutively modulated by c-Src, and enhanced phosphorylation of the c-Src⅐FAK complex by PDGF-BB correlates with enhanced Ca 2ϩ currents. This study also shows that c-Src directly associates with the ␣ subunit of the Ca 2ϩ channels in smooth muscle. The modulation of Ca 2ϩ channel activity by c-Src and FAK implicates an important role for these two kinases in regulating excitability of smooth muscle cells.
A significant finding of this study is that PDGF-BB-induced enhancement of Ca 2ϩ channel activity was blocked by anti-c-Src and anti-FAK antibodies. These data provide the first evidence for a direct involvement of c-Src and FAK in the regulation of Ca 2ϩ currents in differentiated cells by PDGF. The activities of c-Src and FAK are elevated by PDGF (9, 14), and c-Src is required for PDGF mitogenic signaling (25,26). We confirmed these findings in colonic smooth muscle cells by FIG. 5. Effects of anti-mouse IgG and anti-Kv1.5 antibodies on Ca 2؉ currents. Cells were dialyzed with the anti-mouse IgG (Ig G-Ab; 10 g/ml) or anti-Kv1.5 (Kv1.5-Ab; 20 g/ml) antibody, and currents were obtained by a 600-ms depolarization step from a holding potential of Ϫ50 mV and normalized with respect to cell capacitance. Each point represents the mean Ϯ S.E. obtained from five to eight cells. Currents in the absence of intracellular antibodies were obtained on the same day within the same population of freshly dispersed cells (Control). showing that phosphorylation of these two kinases is enhanced by PDGF-BB, which is consistent with the observations in the electrophysiological study. Two binding sites for c-Src have been identified in the ␤-type PDGF receptors, Tyr 579 and Tyr 581 (23). However, an absence of both SH2 and SH3 domains in FAK excludes the possibility of a direct interaction of FAK with receptor tyrosine kinases (27), and c-Src is suggested as an intermediate between receptor tyrosine kinases and FAK (28). Our finding that PDGF-BB enhances the association of c-Src with FAK supports such a role for c-Src.
The formation of a c-Src⅐FAK complex is crucial for cell adhesion and integrin signaling (12) and may also be required for modulating Ca 2ϩ channels. While FAK does not directly associate with the Ca 2ϩ channel, its involvement is of particular relevance to smooth muscle contraction since one of the potential targets for FAK are the cytoskeletal proteins talin and paxillin, which are phosphorylated following activation of G protein-coupled receptors and receptor tyrosine kinases (6,14,15,29). The membrane-associated dense plaques of smooth muscle are structurally similar to the focal adhesion sites of cultured cells in that both contain the cytoskeletal proteins. FAK autophosphorylation at Tyr 397 promotes the binding of FAK to the SH2 domain of Src family kinases (27,30,31). Subsequent phosphorylation of FAK in the kinase domain at Control currents were recorded 2 min prior to perfusion with PDGF-BB (50 ng/ml). The peak amplitude of currents was normalized with respect to cell capacitance (n ϭ 11). B, bar graph showing the effects of the anti-mouse IgG (IgG-Ab), anti-c-Src (c-Src-Ab), anti-FAK (FAK-Ab), anti-Ras (Ras-Ab) antibodies and the MEK inhibitor PD098059 on PDGF-BB-induced enhancement of Ca 2ϩ currents. Cells were dialyzed with the anti-mouse IgG (10 g/ml), anti-c-Src (10 g/ml), anti-FAK (7.3 g/ml), or anti-Ras (5 g/ml) antibody, and Ca 2ϩ currents were recorded before and 2 min after bath application of PDGF-BB (50 ng/ml) and normalized with respect to cell capacitance. The MEK inhibitor PD 098059 (30 M) was applied to the bath, and Ca 2ϩ currents were recorded before and after application of PDGF-BB. Data are expressed as means Ϯ S.E. obtained from five to six cells. *, p Ͻ 0.05 versus cells not treated with PDGF-BB.
FIG. 8. L-type Ca 2؉ channel (␣ 1C ) immunoblot of rabbit colonic smooth muscle. Cell lysates were prepared as described under "Experimental Procedures"; immunoprecipitated (IP) with the anti-␣ 1C (2 g), anti-phosphotyrosine (PY20), or anti-c-Src (c-Src) antibody; and immunoblotted with the anti-␣ 1C antibody. The tissues were pretreated with PDGF-BB for 5 min prior to homogenization. The ␣ subunit immunoprecipitated with PY20 and c-Src antibodies, indicating tyrosine phosphorylation and association with c-Src kinase following PDGF treatment.
Tyr 576 and Tyr 577 enhances the activity of FAK (32). By forming such a complex, c-Src activity is also up-regulated (13) and leads to modulation of the smooth muscle Ca 2ϩ channel. A recent study in HEK 293 cells also suggests that formation of the c-Src⅐FAK complex is essential for coupling FAK to the Ras signaling pathway (33). In rat glomerular mesangial cells, PDGF was shown to enhance voltage-independent Ca 2ϩ channels through Ras (34). The failure of the anti-Ras antibody and the MEK inhibitor PD 098059 to suppress basal Ca 2ϩ currents or to alter PDGF-induced enhancement of Ca 2ϩ currents suggests that the Ras/MAP kinase cascade, the downstream signaling components of c-Src and FAK, is not involved in the regulation of voltage-dependent Ca 2ϩ channels.
In transfected HEK 293 cells, v-Src associates with the human delayed rectifier-type K ϩ channel Kv1.5, and tyrosine phosphorylation of the channels is accompanied by an inhibition of K ϩ currents (35). Furthermore, the association of c-Src with N-methyl-D-aspartate channels has been observed in rat central neurons (36). Our studies show similar modulation of smooth muscle L-type Ca 2ϩ currents by c-Src. The L-type Ca 2ϩ channel is tyrosine-phosphorylated and associates with c-Src following PDGF treatment (Fig. 8). The full cDNA sequence of the L-type Ca 2ϩ channel (␣ 1C subunit) from rat vascular smooth muscle reveals a potential phosphorylation site of tyrosine kinases, which is located at residues 1869 -1876 (RL-SEEVEY) of an alternately spliced region (residues 1854 -1921) (4). It is known that the SH3 domain interacts with proline-rich sequences (RPLPXXP) in the target protein (37), and RPLPRYIP, a sequence similar to the SH3 domain-binding motif, is present in the rat aorta Ca 2ϩ channel (4).