Enhanced Activity of a Large Conductance, Calcium-sensitive K+ Channel in the Presence of Src Tyrosine Kinase*

Large conductance, calcium-sensitive K+ channels (BKCa channels) contribute to the control of membrane potential in a variety of tissues, including smooth muscle, where they act as the target effector for intracellular “calcium sparks” and the endothelium-derived vasodilator nitric oxide. Various signal transduction pathways, including protein phosphorylation can regulate the activity of BKCa channels, along with many other membrane ion channels. In our study, we have examined the regulation of BKCa channels by the cellularSrc gene product (cSrc), a soluble tyrosine kinase that has been implicated in the regulation of both voltage- and ligand-gated ion channels. Using a heterologous expression system, we observed that co-expression of murine BKCa channel and the human cSrc tyrosine kinase in HEK 293 cells led to a calcium-sensitive enhancement of BKCa channel activity in excised membrane patches. In contrast, co-expression with a catalytically inactive cSrc mutant produced no change in BKCa channel activity, demonstrating the requirement for a functional cSrc molecule. Furthermore, we observed that BKCa channels underwent direct tyrosine phosphorylation in cells co-transfected with BKCa channels and active cSrc but not in cells co-transfected with the kinase inactive form of the enzyme. A single Tyr to Phe substitution in the C-terminal half of the channel largely prevented this observed phosphorylation. Given that cSrc may become activated by receptor tyrosine kinases or G-protein-coupled receptors, these findings suggest that cSrc-dependent tyrosine phosphorylation of BKCa channels in situ may represent a novel regulatory mechanism for altering membrane potential and calcium entry.

In the large family of voltage-gated K ϩ channels, large conductance, calcium-sensitive potassium (maxi-K or BK Ca ) 1 channels represent a unique class whose gating depends primarily on membrane voltage but which can be shifted in the negative direction by intracellular free calcium. A direct physiologic consequence of this behavior is that BK Ca channels act as "coincidence detectors" and regulate, in a feedback fashion, cellular processes stimulated by close temporal changes in membrane potential and intracellular calcium. That BK Ca channels indeed play such a role is evidenced by the fact that blocking these channels increases the degree of myogenic tone observed in arterial smooth muscle (1)(2)(3) and enhances the presynaptic calcium-dependent release of neurotransmitter at neuromuscular junctions (4,5).
Given their potential to influence cellular processes, it is not surprising that BK Ca channels are also targets of cellular signaling pathways, including phosphorylation and dephosphorylation reactions (6 -11), heterotrimeric GTP-binding proteins (12,13), and the endothelium-derived vasodilator nitric oxide (14). To date, however, many of the molecular aspects of these regulatory events remain poorly understood.
Of these various cellular pathways, protein phosphorylation remains as one of the most common forms of intracellular signaling and is critically involved in every aspect of eukaryotic cell function from muscle contraction to gene transcription to cell division. Adding to this richness was the discovery in the late 1970s that tyrosine residues may also be phosphorylated, and since then, tyrosine phosphorylation has emerged as a critical process, particularly in the regulation of cellular growth and development (15). However, more recent observations have pointed to an additional role for tyrosine phosphorylation in the control of cellular excitability, via the modulation of both voltagedependent and ligand-operated ion channels. Nonreceptor tyrosine kinases, Src or pp60 c-Src in particular, have been shown to regulate a variety of membrane ion channels, including fast type A ␥-aminobutyric acid (GABA) receptors (16,17), glutamate receptors (18), nicotinic acetylcholine receptors (19), voltage-gated calcium channels (20), and potassium channels (21)(22)(23)(24). These effects range from enhancement to inhibition of channel activity, and in some cases, a Src tyrosine kinase has been shown to be physically associated with the ion channel complex (18,22). These findings thus expand our picture of ion channel regulation to include tyrosine along with serine/threonine phosphorylation as critical regulatory events in the control of membrane excitability.
In the current study, we have examined the effects of the human Src tyrosine kinase on the functional behavior of a murine BK Ca channel using a transient co-expression strategy.
Our results indicate that the cSrc tyrosine kinase can enhance BK Ca channel activity in a calcium-dependent manner and that BK Ca channels undergo tyrosine phosphorylation as a consequence of co-expression. Akhand et al. (25) have recently reported that nitric oxide can directly activate cSrc via thionitrosylation; these results taken together with our own would suggest additional mechanisms by which cSrc may be involved in the regulation of BK Ca channels. Src-dependent tyrosine phosphorylation may thus represent a generalized cellular mechanism by which BK Ca channel activity can be regulated in response to diverse cellular stimuli.

EXPERIMENTAL PROCEDURES
Reagents and Chemicals-LipofectAMINE and high glucose-containing Dulbecco's modified Eagle's medium cell culture medium were purchased from Life Technologies, Inc. DNA modifying enzymes were obtained from New England Biolabs. The anti-phosphotyrosine mouse monoclonal antibodies 4G10 and PY20 were purchased from Upstate Biotechnology Inc. and Transduction Labs, respectively. The anti-Src monoclonal antibody 327 was purchased from Calbiochem. The anti-mSlo rabbit polyclonal antibody and horseradish peroxidase-linked goat anti-mouse and goat anti-rabbit secondary antibodies were obtained from Chemicon International. The SuperSignal chemiluminescence detection reagent was purchased from Pierce. Chemicals used in the preparation of solutions for electrophysiolgical recordings were from Sigma-Aldrich.
Construction and Transfection of cDNA Plasmids-The cDNA encoding the mouse brain mSlo ␣ subunit (26) was obtained from Dr. L. Pallanck (University of Wisconsin), and a ϳ3.7-kilobase fragment was subcloned into the SV-40 promoter-based mammalian expression plasmid SR␣ (27) as follows. The EcoRI site of SR␣ and the BamHI site at the 5Ј end of the mSlo cDNA were blunted with Klenow fragment and a NotI linker was ligated to these ends. The XbaI site at the 3Ј end of the mSlo cDNA was then blunted with Klenow fragment, and the insert cDNA was ligated between the NotI and EcoRV restriction sites in the polylinker region of the plasmid. The cDNA encoding the wild-type green fluorescent protein (28) was subcloned in the SR␣ plasmid between the PstI and EcoRI sites. A full-length cDNA encoding human Src tyrosine kinase (Dr. D. Fujita, University of Calgary) was subcloned into SR␣ using the EcoRI site. Site-directed mutagenesis of both the mSlo ␣ subunit and cSrc cDNAs was carried out using the Transformer mutagenesis kit (CLONTECH Laboratories). An epitope-tagged form of mSlo was also prepared by insertion of the seven-amino acid sequence EEFMPME (29,30) at position 1121 near the C terminus; the sequence of this modified mSlo ␣ subunit reads 1115 HSIPSTAEEFMPMENR-PNR. The presence of this tag did not alter the intrinsic voltage-or calcium-dependent gating of these modified channels following expression (data not shown). The enzymatically "dead" form of cSrc was prepared by a Lys to Met mutation at position 298 in the catalytic domain of the kinase (31).
HEK 293 cells (32) were obtained from Dr. M. Calos (Department of Genetics, Stanford University) and were maintained at 37°C in a 5% CO 2 incubator in Dulbecco's modified Eagle's medium containing Lglutamine, 4.5 g/liter D-glucose and 10% (v/v) characterized fetal bovine serum (Hyclone Laboratories, Logan, UT). Transient transfection of cells at 50 -80% confluency was carried out in 35-mm tissue culture dishes using the lipofection technique. Briefly, 6 -8 l of Lipo-fectAMINE was mixed together with 1.2-1.5 g of plasmid cDNA in 1 ml of serum-free Dulbecco's modified Eagle's medium and placed on cells for 5-6 h at 37°C in a humidified incubator containing 5% CO 2 . DNA-containing medium was then aspirated and replaced with serumcontaining medium. The following day, cells were detached using 0.025% trypsin/0.5 mM EDTA in phosphate-buffered saline and replated onto sterile glass coverslips and 10-cm culture dishes. Electrophysiological recordings were typically performed on days 2-4 following transfection. For Western blotting and enzymatic assays, cells were harvested on day 3 or 4 following transfection.
Electrophysiology-Macroscopic currents were recorded at 35 Ϯ 0.5°C from excised inside/out membrane patches of HEK 293 cells using an Axopatch 200B patch clamp amplifier and pClamp 6.03 software. BK Ca channel currents from native mSlo or the epitope-tagged mSlo (see above) were activated by voltage clamp pulses delivered from a holding potential of 0 mV to membrane potentials ranging from Ϫ180 to 240 mV; tail currents were recorded at ϩ50, Ϫ80, or Ϫ120 mV. Current traces were filtered at 2-5 kHz (4-pole Bessel filter) and acquired on a Dell Dimension XPS computer at a sampling frequency of 8 -10 kHz using a Digidata 1200 analogue/digital interface. Recording micropipettes were pulled from thin walled borosilicate glass capillaries (inner diameter, 1.2 mm; outer diameter, 1.5 mm; World Precision Instruments, Sarasota, FL) using a Sutter P-89 horizontal electrode puller. Micropipettes were filled with a solution containing 5 mM KCl, 140 mM KOH, 1 mM MgCl 2 , 1 mM CaCl 2 , and 10 mM HEPES, with a pH adjusted to 7.3 with methanesulfonic acid and had tip resistances of 2-4 M⍀. The bath solution contained 5 mM KCl, 140 mM KOH, 1 mM MgCl 2 , 2 mM EGTA or N-(2-hydroxyethyl)EDTA, and 10 mM HEPES with a pH adjusted to 7.2 with methanesulfonic acid; variable amounts of a 0.1 M CaCl 2 solution were added to give the desired free calcium concentrations. The level of free calcium in each solution was confirmed using a calcium electrode (Orion model 93-20) with calibration standards (WPI, Sarasota, FL) ranging from pCa 8 to 2. The recording chamber (volume, ϳ0.3 ml) was perfused at a constant rate of 1-1.5 ml/min, using a set of manually controlled solenoid valves to switch between various solutions.
Transfected HEK 293 cells plated on coverslips were placed in a temperature-controlled recording chamber on the stage of a Nikon TE 300 Eclipse inverted microscope. Individual cells expressing BK Ca channels were then identified visually by co-expression of the marker protein green fluorescent protein under epifluorescence using 480 nm excitation and 510 nm emission filters. Based on the number of green fluorescent cells, transfection efficiency was judged to be 20 -30% in any given experiment.
In Vitro Phosphorylation Assay-Frozen pellets of transfected cells were thawed on ice and then resuspended in 1.0 ml of ice-cold RIPA buffer containing 50 mM Tris HCl, pH 7.5, 150 mM NaCl, 1 mM Na 3 VO 4 , 10 mM NaF, 1 mM EDTA, 0.2 mM EGTA, 0.5% (v/v) Nonidet P-40, 0.1% (w/v) SDS, 0.3% (w/v) sodium deoxycholate, and 2 g/ml each aprotinin and leupeptin. The crude cell lysates were kept on ice for ϳ10 min and then centrifuged at 15,000 rpm for 5 min at 4°C to remove insoluble materials. Following transfer to clean microcentrifuge tubes, the supernatants were precleared by incubation with 20 l of protein A/G-agarose (Santa Cruz Biotech) on a rotator for 1 h at 4°C. The agarose beads were pelleted by centrifugation at 15,000 rpm for 5 min, and the supernatants were again transferred to clean tubes and then incubated for 2 h at 4°C with 4 -5 g/tube of the mouse anti-Src monoclonal antibody 327. Following addition of 20 l of protein A/G-agarose per tube, samples were again rotated at 4°C for an additional hour and then centrifuged for 2 min at 10,000 rpm. The pelleted materials were washed twice by gentle resuspension in 1 ml of RIPA buffer and then centrifuged as above. Following the second wash, the pelleted materials were washed in 1 ml of ice-cold buffer containing 50 mM HEPES, pH 7.4, 100 mM NaCl, 1 mM dithiothreitol, 0.1% (v/v) Nonidet P-40, and 10% (v/v) glycerol. Liquid from each sample was then aspirated, and the immunoprecipitates were assayed immediately for tyrosine kinase activity.
cSrc-dependent tyrosine kinase activity was measured at 30°C in a final assay volume of 50 l containing 50 mM HEPES, pH 7.8, 150 mM NaCl, 5 mM MgCl 2 , 1 mM dithiothreitol, 0.2 mM Na 3 VO 4 , 4 mg/ml p-nitro-phenylphosphate, and 100 M cdc2 6 -20 peptide substrate (Upstate Biotechnology, Inc.). The reaction was started by addition of [␥-32 P]ATP (final concentration, 10 M; 2000 -3000 cpm/pmol) and stopped at various times by adding a 40-l aliquot from each reaction mixture to tubes containing 25 l of 50% (v/v) acetic acid. A 60-l aliquot from each tube was then spotted onto a P81 paper disc (Whatman), followed by washes for 5-10 min in 3ϫ 500 ml volumes of 0.5% (v/v) phosphoric acid. P81 paper discs were then briefly rinsed in acetone and dried. Radioactivity bound to the discs was measured by Cerenkov counting in a Beckman model LS6000SC liquid scintillation counter.
Western Blotting-Transfected cells were detached on day 3 by brief incubation with sterile phosphate-buffered saline containing 0.05% trypsin and 0.5 mM EDTA centrifuged in 15-ml culture tubes at ϳ100 ϫ g for 5 min and stored at Ϫ80°C as intact cell pellets. These pellets were resuspended in 0.5-1 ml of ice-cold lysis buffer containing 1ϫ TBS, 1% (v/v) Triton X-100, 1 mM EGTA, 2 mM EDTA, 10 mM NaF, 1 mM Na 3 VO 4 , 1 mM phenylmethylsulfonyl fluoride, and 5 g/ml each of leupeptin and aprotinin and then sonicated for 5-10 s to shear the genomic DNA. Lysates were mixed with Laemmli sample buffer containing 0.5% (v/v) ␤-mercaptoethanol and incubated for 20 -30 min at 70°C, and the proteins then separated by SDS-polyacrylamide gel electrophoresis. The resolved proteins were electrotransferred to nitrocellulose membrane at 4°C in a buffer containing 25 mM Tris, 192 mM glycine, 0.1% (w/v) SDS, and 20% (v/v) methanol either for ϳ2 h at 80 -90 V or overnight at 35 V. Membranes were first dried in a fume hood to fix proteins and then briefly rinsed in a buffer containing 20 mM Tris HCl, pH 7.4, 150 mM NaCl, and 0.1% (v/v) Tween-20 (TTBS). Membranes were incubated at room temperature for 20 -30 min in TTBS containing 5% (w/v) skim milk powder to block nonspecific binding of antibodies and then rinsed three times for 5 min in TTBS. Incubation of membranes with primary antibodies was carried out in TTBS containing 1% (w/v) skim milk powder for 1-2 h at room temperature, followed by three to five 10-min washes with TTBS alone. Membranes were then incubated for ϳ1 h with the appropriate secondary antibody also diluted in TTBS/1% (w/v) skim milk powder, followed by three to five 5-min washes with TTBS. After the final wash, blots were immediately developed by applying the SuperSignal chemiluminescence reagent for 2-3 min and then exposing the blots to x-ray film (Hyperfilm; Amersham Pharmacia Biotech).

RESULTS
In HEK 293 cells transiently transfected with cDNAs encoding green flourescent protein and a mouse brain mSlo ␣ subunit (26), the pore-forming subunit of a large conductance, calcium-sensitive K ϩ channel, voltage clamp steps ranging from Ϫ180 to 240 mV gave rise to large, outwardly rectifying macroscopic currents in excised inside/out membrane patches (Fig. 1). In contrast, membrane patches from cells transfected with green fluorescent protein alone displayed only negligible currents in response to the same voltage clamp steps (data not shown). Increasing the cytoplasmic free Ca 2ϩ from 0 to ϳ120 M was observed to have several effects on the macroscopic currents: 1) a shift in the voltage for half-maximal activation of current to more negative membrane potentials, 2) an increase in the maximal current amplitude, and 3) an increase in the speed of current activation (Fig. 1, A-E). Normalized conductance-voltage (G-V) relations derived from tail current measurements (Fig. 1F) document the shifts to the left in channel gating along the voltage axis with increasing free intracellular calcium. These observations are thus similar to those recently reported by others (33)(34)(35)(36)(37)(38)(39) for heterologous expression of mammalian Slo genes encoding BK Ca channels.
Several investigators have already reported that BK Ca channel activity can be modulated by G-protein and phosphorylation/dephosphorylation pathways (6 -10), suggesting that these channels are important cellular targets for various regulatory mechanisms. More recently, much interest has been generated by observations that cellular tyrosine kinases, in particular cSrc, are able to modulate the activity of both ligand-and voltage-gated ion channels (16 -20), including K ϩ channels (21-24). Such findings may now have broader implications, given that many G-protein-coupled receptors, which can regu-late cellular excitability, are also now recognized to activate various tyrosine kinases as part of their signal transduction cascades (40,41). To determine whether BK Ca channel activity could also be modulated by tyrosine phosphorylation, we utilized our transient transfection strategy to co-express the BK Ca channel with cDNAs encoding active and catalytically inactive forms of the human cSrc tyrosine kinase (42). Site-directed mutagenesis was used to generate the inactive form of the enzyme by a Lys to Met substitution in the catalytic domain (see "Experimental Procedures"). The protein expression and tyrosine kinase activities of cSrc transfected HEK 293 cells were then confirmed using two complementary approaches. Fig. 2A shows a Western blot of total cell lysates probed with an antibody (4G10) to detect proteins containing phosphorylated tyrosine residues. In cells transfected with either the BK Ca channel alone (first lane, BK alone) or BK Ca channel plus the inactive form of cSrc (second lane, BK ϩ dead Src), similar levels of phosphotyrosine-containing proteins were observed. The major phosphoprotein with a molecular mass of ϳ60 kDa in the second lane likely represents the inactive cSrc, which still undergoes tyrosine phosphorylation in the intact cell. However, co-transfection of cDNAs encoding the BK Ca channel and active cSrc produced a large increase in the level of phosphotyrosine containing proteins ( Fig. 2A, third lane), indicating functional expression of cSrc tyrosine kinase activity in situ.
To directly confirm that both the active and inactive forms of the kinase were indeed expressed following transient transfection, a second Western blot of the same samples shown in Fig.  2A was performed, using a monoclonal antibody against the cSrc protein. In HEK 293 transfected with the mSlo cDNA alone, no signal was detected (Fig. 2B, first lane). However, strong ϳ60-kDa immunoreactive bands were readily observed in cells co-transfected with cDNAs for mSlo and either the active or inactive forms of cSrc (Fig. 2B, second and third  lanes). This result thus directly confirms the expression of cSrc protein under the latter two transfection conditions. In the absence of calcium (A), voltage clamp steps ranged from Ϫ30 to ϩ240 mV; tail currents were measured at ϩ50 mV. For 0.9 and 4 M free calcium (B and C), steps ranged from Ϫ90 to ϩ180 mV, and tail currents were measured at Ϫ80 mV. For 12 and 120 M free calcium (D and E), steps varied from Ϫ180 to ϩ90 mV, with tail currents measured at Ϫ120 mV. Refer to the diagrams of voltage clamp protocols shown below the panels. Scale bars shown in A apply to all current traces.
To provide further evidence that cSrc was most likely responsible for the large increase in cellular phosphoproteins observed following co-transfection ( Fig. 2A, third lane), we directly measured the expressed cSrc tyrosine kinase activity by first immunoprecipitating cSrc and then performing an in vitro phosphorylation assay. In cells transfected with mSlo alone or mSlo plus the inactive form of cSrc, very little protein kinasedependent phosphorylation of the selective cSrc peptide substrate cdc2 6 -20 (43) was observed (Fig. 2C). However, the immunoprecipitate from cells co-transfected with mSlo and active cSrc demonstrated a robust, time-dependent phosphorylation of the substrate under the same assay conditions. This observation thus strongly suggests that the increased level of phosphotyrosine containing proteins detected in cSrc transfected cells ( Fig. 2A, third lane) is due primarily to the activity of the cSrc tyrosine kinase itself. Fig. 3 shows macroscopic BK Ca channel currents recorded in an excised membrane patch from a cell co-transfected with cDNAs for mSlo and active cSrc (panels A-D) or mSlo and dead cSrc (panels E-H). In the absence of free cytoplasmic calcium (i.e. 2 mM EGTA in the bath), BK Ca channel currents resembled those observed with expression of the BK Ca channel alone (data not shown; refer to Fig. 1A). However, upon increasing free calcium from 0.9 to 120 M, we observed that BK Ca channel gating was shifted to the left to a much greater degree in the presence of cSrc (Fig. 3, A-D) compared with currents from mSlo channels either in the absence of cSrc co-expression (Fig.  1, B-E) or co-expressed with the inactive form of the enzyme (Fig. 3, E-H). In contrast, expression of just cSrc alone had no significant effect on the level of endogenous membrane currents observed under the same recording conditions (Fig. 3J). Thus, at levels of free cytosolic calcium Ն4 M, the open probability (Popen) of BK Ca channels is greater at any given voltage in the presence of cSrc compared with channels expressed alone (Fig. 1) or in the presence of dead cSrc (Fig. 3, E-H). Most

FIG. 2. Immunoblot analyses and in vitro kinase assay of HEK 293 cells transiently transfected with cDNAs encoding the human cSrc tyrosine kinase, an inactive form of Src and BK Ca channels.
To verify both the expression and catalytic activity of recombinant human Src tyrosine kinase expressed in HEK 293 cells, Western blots of cells transfected with cDNA for BK Ca ␣ subunits alone or together with cDNAs for either catalytically inactive (dead) Src or wild-type cSrc were probed with the anti-phosphotyrosine monoclonal antibody, 4G10 (A), or the anti-Src monoclonal antibody, 327 (B). Positions of molecular mass markers (in kDa) are shown on the right. C shows in vitro phosphorylation of the selective Src tyrosine kinase peptide substrate cdc2 6 -20 derived from the cSrc substrate, p34 cdc2 . Src tyrosine kinase activity was immunoprecipitated from HEK 293 cells transfected with either BK Ca ␣ subunits alone or together with either dead or active Src. The time course of cdc2 6 -20 peptide phosphorylation by immunoprecipitated kinase activity is plotted for all three transfection conditions; these conditions are indicated by symbols as follows: E, BK Ca channels expressed alone; Ⅺ, BK Ca channels co-expressed with the catalytically inactive cSrc mutant; ‚, BK Ca channels co-expressed with wild-type cSrc. Background substrate phosphorylation (i.e. no immunoprecipitate added) has been subtracted from all three conditions. In the presence of 0.9 and 4 M free calcium (A, B, E, and F), steps ranged from Ϫ90 to ϩ180 mV; tail currents were measured at Ϫ80 mV. For 12 and 120 M free calcium (C, D, G, and H), steps varied from Ϫ180 to ϩ90 mV, with tail currents measured at Ϫ120 mV. Scale bars shown in J apply to all current traces.
interesting, raising cytoplasmic calcium to 120 M (Fig. 3E) leads to very strong activation (i.e. Popen near 1) of BK Ca channels over the entire range of membrane potentials from Ϫ180 to 90 mV. By comparison, in the absence of cSrc expression, BK Ca channels did not reach their half-maximal open probability until a membrane potential of ϳϪ70 mV at the same level of free calcium (Fig. 1F). These observations demonstrate that BK Ca channel gating can be enhanced by coexpression with the human cSrc tyrosine kinase. The observed lack of effect of the inactive form of cSrc on the voltage dependence of gating and the ability of free calcium to shift gating to the left along the voltage axis suggest that the enhancement of BK Ca channel activity observed following co-transfection of mSlo and cSrc is dependent upon cellular tyrosine kinase activity and not simply the presence of the cSrc protein itself.
If the observed enhancement of BK Ca channel activity following co-expression of mSlo and active cSrc were due to tyrosine phosphorylation of the channel protein itself, as suggested by the maintained activity in excised membrane patches, it may be possible to demonstrate this directly. To test this possibility, HEK 293 cells were first transfected with mSlo alone, mSlo together with active cSrc or mSlo plus inactive cSrc. Following protein expression, cells were lysed and phosphotyrosine-containing proteins were immunoprecipitated from each dish of cells. The immunoprecipitates were then separated by SDS-polyacrylamide gel electrophoresis, and a Western blot was performed. Using a polyclonal antibody against the BK Ca channel ␣ subunit, we detected a major ϳ125-kDa band in the immunoprecipitate from cells transfected with mSlo and cSrc, but no immunoreactivity was observed under either of the other two conditions (Fig. 4A). This finding thus provides direct evidence that the BK Ca channel ␣ subunit undergoes direct tyrosine phosphorylation in the presence of active cSrc. To exclude the possibility that this observation may be due to variable expression of the BK Ca channel itself, aliquots of the initial total cellular lysates were probed for expression of BK Ca channel protein. Fig. 4B shows that under all three transfection conditions, similar amounts of immunoreactive BK Ca channel were detected using the same antibody. Taken together, these results support our initial hypothesis that in the presence of active cSrc tyrosine kinase, the BK Ca channel ␣ subunit undergoes direct tyrosine phosphorylation. This phosphorylation may thus be responsible for the sustained enhance-ment of BK Ca channel activity observed in excised membrane patches in the presence of micromolar concentrations of free calcium.
In an attempt to identify the site(s) of tyrosine phosphorylation, site-directed mutagenesis was used to make Tyr to Phe substitutions at three potential cSrc phosphorylation sites (42,44) within the BK Ca channel ␣ subunit. Following single substitutions at Tyr 766 (GSIEY 766 LKRE), Tyr 935 (DTELY 935 LT-QP), or Tyr 1027 (DGGCY 1027 GDLF), mutant BK Ca channels were co-expressed in the presence of cSrc and then examined for their level of tyrosine phosphorylation. Using the same experimental strategy as described for Fig. 4, we observed that Tyr to Phe substitutions at positions 935 and 1027 had no effect on cSrc-dependent phosphorylation of BK Ca channels (Fig. 5A). However, substitution of Tyr 766 dramatically decreased the level of in situ BK Ca channel tyrosine phosphorylation in the presence of cSrc. A Western blot of the total cell lysates (Fig.  5B) demonstrates that following transfection, the expression levels of both the native and mutant channels are very similar, indicating that this observed difference in the degree of tyrosine phosphorylation is not due to differential expression of the BK Ca channel constructs. Fig. 6 shows BK Ca channel currents recorded in an excised inside-out membrane patch from a cell co-expressed with cSrc and BK Ca channels containing the Tyr 766 to Phe substitution. In response to increasing concentrations of free cytosolic calcium, mutant BK Ca channel activity appeared to resemble that from native channels in the absence of cSrc, with no clear enhancement observed. This finding suggests that phosphorylation of Tyr 766 is likely responsible for the enhanced gating of BK Ca channels observed in the presence of cSrc. The effects of co-expression of cSrc or inactive cSrc on the gating of native and mutant BK Ca channels are summarized in a plot of the half-maximal voltages of activation (V1 ⁄2 values) versus the free cytosolic calcium concentrations (Fig. 7). Slope values derived from single Boltzmann fits of G-V curves did not vary significantly for wild-type and mutant BK Ca channels in the absence and presence of cSrc (range, 20.4 Ϯ 3.8 at 120 M free Ca to 28.3 Ϯ 5.1 at 4 M Ca). This plot quantifies the calcium-induced shifts to the left of channel gating for each of the expression conditions described above. FIG. 4. BK Ca channels undergo tyrosine phosphorylation in the presence of cSrc. BK Ca channels were expressed in HEK 293 cells either alone (BK alone) or co-expressed together with either catalytically inactive Src (BKϩ dead Src) or wild-type cSrc (BKϩcSrc). Following expression, cells from each group were lysed, and phosphotyrosinecontaining proteins were immunoprecipitated (IP) using the monoclonal antibodies 4G10 and PY20 as described. A shows a Western blot of immunoprecicipitated proteins that have been probed with a BK Ca channel-specific antibody. B shows a Western blot of the total cellular lysates from each group probed using the same BK Ca channel antibody.
FIG. 5. Mutation of Tyr 766 prevents phosphorylation of BK Ca channels co-expressed with wild-type cSrc. Site-directed mutagenesis was used to make Tyr to Phe substitutions at positions 766, 935, and 1027 in the BK Ca channel ␣ subunit. Wild-type and mutated channels were then individually co-expressed with cSrc, and immunoprecipitation (IP) of phosphotyrosine-containing proteins was performed as described. A shows a Western blot of immunoprecipitated proteins probed with the anti-BK Ca channel antibody. Mutant BK Ca channels are denoted as Y766F, Y935F, and Y1027F above the corresponding lanes. B shows the levels of expression of the native and mutant BK Ca channel proteins in the total cellular lysates from each of the transfection conditions shown in A.

DISCUSSION
The regulation of membrane ion channel activity by signal transduction pathways has long been considered an important mechanism for the alteration of cellular excitability and physiologic function. Examples of such events include: 1) the ␤-adrenergic regulation of cardiac L-type calcium channels (45,46) and epithelial CFTR chloride channels (47,48), 2) the regulation of inwardly rectifying K ϩ channels and neuronal calcium channels by G-protein ␤␥ subunits (49,50), and 3) the enhance-ment of neuronal N-methyl-D-aspartate and 2-amino-3-hydroxy-5-methyl-4 isoxalone propionic acid (AMPA) receptors by protein phosphorylation and dephosphorylation (51,52). The results of our study now demonstrate for the first time that the pore-forming ␣ subunit of the mammalian BK Ca channel can undergo direct tyrosine phosphorylation in the presence of the human cSrc tyrosine kinase (Fig. 4) and that this phosphorylation correlates with an enhancement of channel gating (see Figs. 3 and 6). This phosphorylation was entirely dependent upon cSrc tyrosine kinase activity and was not observed in the absence of cSrc or with co-expression of an inactive form of the enzyme. The simplest interpretation of these results is that cSrc itself is directly responsible for this tyrosine phosphorylation of the BK Ca channel protein and the subsequent enhancement of calcium-sensitive gating. However, we can not exclude the possibility that the observed phosphorylation of BK Ca channels is mediated by a different tyrosine kinase that is regulated by cSrc, such as Fak or Pyk2 (53). Alternatively, it is also possible that the observed enhancement of BK Ca channel activity may result from a direct protein-protein interaction between the BK Ca channel and a SH2 domain-containing protein, such as cSrc, for example, that is promoted by tyrosine phosphorylation and maintained in excised inside-out membrane patches.
An important aspect of our findings is that this enhancement was not observed under all experimental conditions, but rather, it occurred in a calcium-sensitive manner. In the presence of 2 mM EGTA with no added free calcium in the bath solution, BK Ca channels demonstrate largely voltage-dependent gating, with conductance-voltage relations shifted strongly to the right along the voltage axis (refer to Fig. 1, A and F). Under these same recording conditions (i.e. 2 mM EGTA only), co-transfection of BK Ca channels with either the active or inactive forms of cSrc had no effect on channel gating properties (Fig. 7). However, in the presence of mSlo plus active cSrc, an enhancement of BK Ca channel activity was observed as the level of free cytosolic free calcium was subsequently raised from 0.9 to 120 M (Fig. 3, B-E). This observation indicates that the enhancement of BK Ca channel gating by cSrc co-expression occurs in a calcium-sensitive manner. Therefore, the magnitude of enhancement by tyrosine phosphorylation on channel gating appears to be dependent upon calcium binding to the channel.
What are the possible physiologic consequences of cSrc-dependent tyrosine phosphorylation of BK Ca channels in situ? In vascular smooth muscle cells and neurons, along with other cell types that utilize BK Ca channels to dampen cellular activity, enhancing BK Ca channel gating by tyrosine phosphorylation would tend to further decrease excitability by turning off calcium influx through voltage-dependent calcium channels. In blood vessels and the nervous system, this would lead to vasodilation and reduced neurotransmitter release, for example. However, in many inexcitable cells, such hematopoietic and endothelial cells, which can produce a sustained calcium influx in response to a stimulus, an enhancement of BK Ca channel gating by tyrosine phosphorylation would be predicted to support the rise in intracellular calcium by causing greater membrane hyperpolarization and increasing the driving force for calcium entry. Therefore, tyrosine phosphorylation of BK Ca channels may lead to either an increase or decrease of cellular activity, depending upon the presence of other concomitant ionic fluxes. Alternatively, tyrosine phosphorylation may have a more important role in the density and/or distribution of BK Ca channels at the cell surface or the binding of regulatory/ signaling molecules, such as protein kinases and/or phosphatases, to BK Ca channels. Ultimately, examination of these pre-  dictions in native cells will help determine the physiologic significance of BK Ca channel tyrosine phosphorylation.