INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

POTASSIUM CHANNELS play an important role in the control of contractility in vascular smooth muscle by influencing the level of membrane potential (24). It is well recognized that steady-state Ca²⁺ influx is required to maintain myogenic tone and sustained contractile force of vascular smooth muscle; depolarizations positive to approximately 50 mV evoke L-type Ca²⁺ channel activity, increase intracellular Ca²⁺ concentration, and induce contraction. In contrast, hyperpolarization of membrane potential results in relaxation due to voltage-dependent decline in the open probability of L-type Ca²⁺ channels. However, the ionic basis of membrane potential in vascular smooth muscle is poorly characterized. It appears to be due to a dynamic balance between steady-state inward current, carried by L-type Ca²⁺, Cl, and/or nonselective cation channels, and outward current due to at least four different types of K⁺ channels, including delayed rectifier K⁺ channel (K_DR), Ca²⁺-sensitive large-conductance K⁺ channel (BK_Ca), ATP-sensitive K⁺ channel (K_ATP), and/or inward rectifier K⁺ channels, depending on the tissue source and physiological condition (24).

Vasoactive agonists are believed to influence smooth muscle tone, at least in part, by altering the activity of ion channels and changing the level of membrane potential. For example, vasodilators that induce relaxation by elevating intracellular cAMP concentration in vascular myocytes cause hyperpolarization of membrane potential by increasing K⁺ channel open probability (24). Depending on the source of the vascular tissue, the hyperpolarization and relaxation induced by vasodilators acting via cAMP has been shown to be suppressed by selective block of either K_DR channels with 4-aminopyridine (4-AP) (13, 30), BK_Ca channels with iberiotoxin (IbTX) or charybdotoxin (16, 30), or K_ATP channels with glibenclamide (24).

Direct evidence for the regulation of macroscopic and/or unitary currents carried by K_DR, BK_Ca, and K_ATP channels of vascular smooth muscle cells by vasoactive ligands and/or second messengers was obtained using the patch-clamp method. For example, the activity of K_ATP and BK_Ca channels was shown to be enhanced by vasodilators via cAMP and cAMP-dependent protein kinase (PKA) (24) and to be inhibited by vasoconstrictors via activation of protein kinase C (24). We previously demonstrated that the slowly inactivating, 4-AP-sensitive, delayed rectifier K⁺ current (I_KDR) of myocytes from rabbit portal vein and coronary artery is also regulated by alterations in cAMP due to stimulation of -adrenoceptors or adenylyl cyclase (1, 11). An increase in I_KDR was observed after treatment with isoproterenol (1 µM) or forskolin (1 µM). We concluded that a phosphotransferase reaction mediated by PKA was involved because dialysis with the specific peptide inhibitor of PKA, PKI (10 µM), was found to prevent the increase in I_KDR due to -adrenoceptor stimulation (1). More recently, a role for 4-AP-sensitive K⁺ channels has been identified in endothelium-dependent relaxation of rabbit middle cerebral and pulmonary arteries in response to prostacyclin (via adenylyl cyclase, cAMP, and PKA) (13) and nitric oxide (NO) (via guanylyl cyclase, cGMP, and protein kinase G) (37, 38) release, respectively. The conductance of the K_DR channels affected by NO was found to be 45 pS (37).

The identity of the specific 4-AP-sensitive K_DR channel(s) influenced by -adrenoceptor and/or activation of PKA in vascular myocytes has not been determined. This is significant because 1) whole cell I_KDR of smooth muscle cells is likely composed of more than one component based on an analysis of activation and inactivation kinetics, as well as sensitivity to 4-AP and tetraethylammonium ion (8); 2) several voltage-gated K⁺ channels with different slope conductances have been observed in membrane patches of smooth muscle cells; and 3) 4-AP-sensitive, voltage-gated K⁺ current of canine gastrointestinal myocytes is apparently due to the expression of Kv1.2 and Kv1.5 channel -subunit proteins, but vascular myocytes of large-conduit arteries and veins, for example, portal vein, in this species only express Kv1.5 based on Northern analysis (19, 22). However, mRNAs encoding other Kv channels, including Kv1.1, Kv1.2, Kv1.4, Kv1.5, and Kv2.1, were detected in Northern blots of rat aorta (29), suggesting the possibility of tissue- and/or species-dependent variations in the composition of I_KDR of vascular myocytes. Small-conductance 5- to 20-pS K_DR channels are present in airway, portal vein, coronary artery, and colonic myocytes (5, 7, 21, 35), but larger channels of 30-50 pS and 70-95 pS have also been observed (3, 6, 17, 20, 21, 26). The smaller conductance K_DR channels of airway, coronary arterial, and colonic myocytes inactivate during steps to positive potentials with similar kinetics to the inactivating component of whole cell I_KDR (7, 21, 35). In contrast, it is unknown whether the larger conductance channels inactivate.

Accordingly, in the present study we employed the patch-clamp technique to identify the K_DR channel(s) of rabbit portal vein myocytes that contribute to the increase in macroscopic I_KDR after activation of -adrenoceptors or PKA. Isoproterenol and purified catalytic subunit of PKA in the presence of ATP were applied, and K_DR channel activity was monitored at a physiologically relevant membrane potential (40 mV) in cell-attached (C-A) and inside-out (I-O) patches, respectively. A preliminary account of these data was previously published (2).

	METHODS

Top Abstract Introduction Methods Results Discussion References

Cell isolation. Rabbits were housed and killed with an overdose of Nembutal (pentobarbital sodium) according to standards of the Canadian Council for Animal Care and a protocol approved by a local animal care committee of the Medical Research Council of Canada. Portal vein myocytes were isolated as previously described by Aiello et al. (1). Myocytes were placed in a 300-µl constant-flow bath containing solution at room temperature (20-22°C). Complete turnover of the bath contents was accomplished in 5-10 s by a multibarrel perfusion device obtained from Norscan Instruments (Winnipeg, Canada). Only cells of a spindle-shaped, optically refractive, and relaxed nature were used in this study.

Patch clamp. C-A and I-O patch-clamp methods were employed in this study (18) to record K⁺ channel activity under symmetrical KCl recording conditions. For the C-A patches, Sylgard-coated pipettes contained (in mM) 140 KCl, 1 CaCl₂, 1 MgCl₂, 5.5 glucose, and 10 HEPES (pH 7.4 with KOH), and IbTX (100 nM) was added to block the activity of K_Ca channels. The bath solution contained (in mM) 140 KCl, 1 MgCl₂, 5.5 glucose, and 10 HEPES (pH 7.4 with KOH) and was nominally Ca²⁺ free (i.e., no added Ca²⁺; free Ca²⁺ ~1 µM). For I-O membrane patch experiments, the pipette and bath solutions were identical except that the bath solution contained 5 mM Na₂ATP and had a pH of 7.2. Pipettes were prepared from capillary glass (7052 glass, Richland Glass) with a Sutter P-87 puller (Sutter Instruments) and MF-83 microforge (Narashige Scientific Instrument Laboratory). Recordings were performed using an Axopatch 200A amplifier (Axon Instruments). Pipette potential and capacitance were nulled, and a 10- to 15-G seal was formed with the cell membrane. Voltage-clamp protocols were applied using pCLAMP 6.0 software (Axon Instruments). Data were filtered at 1-2 kHz by an on-board eight-pole Bessel filter before digitization (3-10 kHz) with a Digidata 1200 analog-to-digital convertor (Axon Instruments) and stored to hard disk in a 486 PC clone. Data were displayed and analyzed using pCLAMP (Axon Instruments) and Origin software (MicroCal Software).

Drugs and chemicals. Isoproterenol and 4-AP were obtained from Sigma Chemical, and IbTX was from Research Biochemicals. Stock solutions of isoproterenol in bath solution were prepared fresh each day and diluted to a final concentration of 1 µM immediately before use. The pH of 4-AP-containing solutions was maintained at 7.2 or 7.4 with HCl. IbTX was kept as a frozen stock solution, thawed, and diluted to 100 nM in pipette solution immediately before use. Constitutively active catalytic subunit of PKA was purified to electrophoretic homogeneity according to Demaille et al. (12). Aliquots of stock solution were diluted to a final concentration of 50 nM in bath solution immediately before use in I-O membrane patch experiments.

Statistics. Data were compared by paired or unpaired Student's t-test. A level of P < 0.05 was considered to be significantly different.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Two predominant types of unitary K⁺ currents were identified in C-A and I-O membrane patch-clamp recordings of rabbit portal vein myocytes (Fig. 1), similar to the previous observations of Beech and Bolton (5). These included currents due to large-conductance BK_Ca channels and small-conductance 15-pS K_DR channels (Fig. 1), which were blocked by inclusion of IbTX (100 nM) in the pipette and 4-AP (1-5 mM) in the bath solution, respectively. Two additional channel types were also observed: an intermediate-conductance channel of ~35 pS (34.4 ± 3.5 pS) and a large-conductance channel of ~95 pS (93.6 ± 4.5 pS) slope conductance (Fig. 1). BK_Ca channels were observed in every patch (in the absence of IbTX), whereas the 15-pS K_DR channels were present in 70% (16 of 23) and the 35- and 90-pS channels were present in 70% (16 of 23) and 17% (4 of 23), respectively, of the membrane patches studied. Similar observations were made for C-A and I-O membrane patches; however, the activity of the 15-pS K_DR channels exhibited rundown after ~5 min of patch excision, such that all K_DR channel activity in the patches disappeared within 10 min, as was previously reported by others (7, 35). All subsequent experiments concerning the properties of the 15-pS K_DR channels and their regulation by isoproterenol and PKA were conducted with IbTX (100 nM) in the patch pipettes to suppress the activity of BK_Ca channels. Additionally, only those membrane patches containing no or only minimal 35- and 95-pS activity were selected for determining the effects of isoproterenol and PKA on 15-pS channel NP_o.

View larger version (37K): [in this window] [in a new window]   Fig. 1.   Four types of K+ channel of portal vein myocytes based on slope conductance. Five representative sweeps from a single cell-attached membrane patch during 4.2-s steps from 50 to +40 mV in which activity of 230-pS large-conductance Ca2+-sensitive K+ channel (BKCa), 15-pS delayed rectifier K+ channel (KDR), as well as 35- and 95-pS K+ channels was observed. Symmetrical KCl recording conditions were used to obtain these and all subsequent recordings. Horizontal dashed lines in this and subsequent figures indicate baseline level of each trace. Vertical dashed line in this and some subsequent figures indicates onset and end of each voltage step.

Figure 2 shows representative recordings of K_DR channel activity in an I-O membrane patch evoked by command steps to several voltages, as well as average unitary current-voltage relationships for the channels in this recording configuration. The K_DR channels were voltage gated; no channel activity was recorded at 80 mV (data not shown), but positive to approximately 60 mV, channel activity was evident at all voltages tested (Fig. 2). The calculated slope conductance for the small-conductance K_DR channels in five I-O membrane patches in which the actual transmembrane K⁺ gradient was known was 15.3 ± 0.6 pS. This value was not different from the slope conductance of the channels in five C-A membrane patches (14.8 ± 0.6 pS). Two C-A membrane patches were also studied using 5.4 mM K⁺ in the pipette solution. In this case, the slope conductance was reduced to ~8 pS at 0 mV, and the reversal potential of the unitary currents was negative to approximately 60 mV (the most negative potential at which single-channel activity was observed) (data not shown). These values are consistent with those previously reported for the slope conductance of slowly inactivating vascular K_DR channels of myocytes from rabbit portal vein (5 pS in asymmetrical KCl; Ref. 5), rabbit coronary artery (7 and 15 pS, respectively, in asymmetrical and symmetrical KCl; Ref. 35), canine trachea (12.7 pS in asymmetrical KCl; Ref. 7), and colonic myocytes (20 pS in symmetrical KCl; Ref. 21).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Current-voltage relationship for KDR channel activity in an inside-out membrane patch. A: representative recordings of single KDR channel activity during 9-s depolarizations to a range of membrane potentials between 30 and +60 mV after a 2-s step to 60 mV from a holding potential of 0 mV in presence of iberiotoxin (100 nM) added to pipette to suppress BKCa activity. Note transitions of a 35-pS K+ channel in recordings at +30 and +40 mV (arrows). B: average unitary current-voltage relationship for channel delayed rectifier K+ currents in 5 inside-out membrane patches. Each data point is mean ± SE. Average slope conductance was 15.3 ± 0.6 pS.

Figures 3 and 4 show representative recordings of K_DR channel activity obtained during single steps to +40 mV from 50 mV, as well as ensemble currents based on more than 30 sweeps containing only the small-conductance channels. Ensemble currents activated with a time constant of 10.7 ± 3.02 ms in four myocytes (based on 58-77 sweeps) (Fig. 3). In two myocytes, membrane patches were held for a sufficient period to estimate the kinetics of inactivation of ensemble current during 4.2-s command pulses (Fig. 4). The decay of ensemble current in both myocytes based on 32 and 42 sweeps was best fitted by a double exponential function; the respective fast and slow time constants for the myocytes were 0.34 and 2.96 and 0.3 and 2.3 s, respectively. These kinetics of activation and inactivation are consistent with those previously reported for the small-conductance K_DR channels of airway and coronary artery myocytes (7, 35), as well as the inactivating component of macroscopic 4-AP-sensitive I_KDR of several types of smooth muscle cells, including portal vein and coronary artery (2, 5, 28). We did not study the properties of the 35- and 95-pS channels in detail; however, neither showed evidence of inactivation during steps to +40 mV (data not shown).

View larger version (27K): [in this window] [in a new window]   Fig. 3.   Single-current sweeps and ensemble-average current for a representative cell-attached patch containing two 15-pS KDR channels. A: 12 representative traces recorded during a voltage step to +40 mV from 50 mV. B: ensemble average of 85 sweeps with initial activation of current shown in lower expanded trace. Smooth curve through expanded ensemble recording was best fit to equation I = A(1  et/)2, where A is amplitude and  is time constant.

View larger version (47K): [in this window] [in a new window]   Fig. 4.   Single-current sweeps and ensemble-average current of a cell-attached membrane patch containing 4 inactivating KDR channels. A: 15 representative traces of KDR channel activity during a 4.2-s step to +40 mV from 50 mV. B: ensemble average of 42 sweeps. Smooth curve represents best fit to a double exponential function with form I = A2e[(t  K)/2] + A1e[(t  K)/1] + C with indicated time constants ().

Figure 5 illustrates the sensitivity of the 15 pS K_DR channels to 4-AP. A double-pulse protocol was employed to assess channel block at 40 and +40 mV. The representative traces and amplitude histograms of Fig. 5 show that there was complete inhibition of channel activity at 40 mV after 3-min exposure to 4-AP (1 mM). However, the inhibition of channel activity was less effective at +40 mV, with channel open probability (determined from amplitude histograms) falling by 86 ± 6% (n = 6) in the presence of 4-AP. This less effective inhibition of K_DR channel activity by 4-AP at positive voltages is consistent with the voltage-dependent unblock of whole cell I_KDR of coronary arterial myocytes at positive potentials reported by Remillard and Leblanc (28).

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Inhibition of KDR channel activity by 4-aminopyridine (4-AP). A: 5 representative recordings of KDR channel activity of a cell-attached membrane patch during a double-step protocol to 40 and +40 mV from 60 mV before and after treatment with 1 mM 4-AP. Note complete inhibition of channel transitions at 40 mV and partial suppression of channel activity at +40 mV. B: amplitude histograms for channel activity at 40 and +40 mV in patch shown in A. Each histogram derives from 10 sweeps before and 10 sweeps after 3-min exposure to 4-AP. Channel open probability (NPo) is indicated for each voltage and condition (determined from amplitude histogram).

We previously demonstrated that I_KDR of rabbit portal vein myocytes was enhanced by -adrenoceptor activation or treatment with forskolin (1). Accordingly, we first determined whether the activity of the 4-AP-sensitive 15-pS K_DR channels of C-A membrane patches was affected by isoproterenol added to the bath solution. To facilitate the long-duration recordings (>20 min) required in these experiments, we employed a two-step voltage-clamp protocol applied from a holding potential of 0 mV, which we found enhanced the stability of the seal between the pipette and membrane patch. Hyperpolarizing steps of 2-s duration to either 80 or 60 mV were then applied every 20 s to permit recovery from inactivation followed by a 9-s test step to 40 mV to evoke steady-state channel activity. The voltage of this test step of 40 mV was chosen because it is within the physiological range of membrane potential in rabbit portal vein and other vascular myocytes (24). Figure 6 shows six representative recordings obtained during steps to 40 mV and corresponding amplitude histograms from 30 steps in the absence (control) and after 5.5 min in the presence of isoproterenol. Isoproterenol was applied to 10 patches; in 8 patches that showed evidence of 15-pS K_DR channel, 55.6 ± 6.2% of the 9-s recordings at 40 mV failed to show evidence of channel activity under control conditions and NP_o for the channels was low at 0.014 ± 0.005 (determined from idealized traces or amplitude histograms). In contrast, null sweeps were only observed 19.6 ± 6.8% of the time in the presence of isoproterenol, and K_DR channel activity increased in each of these eight patches; on average, NP_o increased by approximately threefold to 0.041 ± 0.02 (P < 0.05 by paired Student's t-test). No 15-pS channel activity was apparent in the two other patches before or after the application of isoproterenol. Because it was not known if the 15-pS channel was present in these membrane patches, the data were not included in the above analysis of NP_o (however, inclusion of these patches in the analysis did not affect the result: NP_o still increased from 0.012 ± 0.004 to 0.033 ± 0.008; n = 10, P < 0.05 by paired Student's t-test). Additionally, treatment with isoproterenol did not induce activity of any additional channels in the patches tested, and the activity of the intermediate, 35-pS channel present under control conditions in one patch was unaffected (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Representative effect of isoproterenol on KDR channel activity. A: 6 representative records obtained from a cell-attached membrane patch under symmetrical KCl recording conditions during 9-s depolarizing steps to 40 mV after a 2-s step to 80 mV from a holding potential of 0 mV under control conditions with iberiotoxin (100 nM) added to pipette solution to suppress BKCa channel activity. Six sweeps are shown on right for same patch after 5.5-min treatment with isoproterenol (1 µM). Note the following: 1) records are inverted to show openings of K+ channels at 40 mV in outward direction and 2) increased KDR channel activity and fewer null traces during -adrenoceptor stimulation. B: amplitude histograms for 30 sweeps during control (left) and isoproterenol (right) treatments in experiment of A. Values for NPo determined from idealized traces.

We previously demonstrated that the increase in I_KDR of rabbit portal vein myocytes due to isoproterenol treatment was suppressed by pretreatment with the peptide inhibitor of PKA, PKI (1). This implies that the activity of vascular K_DR channels may be regulated by phosphorylation by PKA. To determine the effect of PKA on the activity of 15-pS K_DR channels, the intracellular surface of I-O membrane patches was first exposed to a bath solution containing 5 mM Mg-ATP before switching to a second solution containing 5 mM Mg-ATP and 50 nM purified catalytic subunit of PKA. Figure 7 shows five representative recordings and amplitude histograms (based on 30 sweeps) of K_DR channel activity in the absence and presence of PKA during a voltage-clamp protocol similar to that described for C-A membrane patches. PKA was applied to 13 membrane patches; 11 membrane patches exhibited 15-pS K_DR channel activity during the experiments, and in 2 patches, no activity was observed before or after PKA. In the absence of PKA, the activity of K_DR channels in the 11 I-O membrane patches was low, and null traces were observed 84.6 ± 6.5% of the time. On average, NP_o was 0.011 ± 0.003 (n = 11; determined from idealized traces or amplitude histograms), and in 2 of the 11 patches, no transitions were observed under control conditions (but channel activity was observed during exposure to the kinase). This level of K_DR channel activity was slightly less than that observed for the channels in C-A membrane patches. As indicated above, we found that channel activity began to disappear after ~5 min due to rundown after membrane patch excision, so this slightly lower NP_o for the I-O membrane patches likely reflects this loss of channel activity during the late control sweeps in some experiments.

View larger version (34K): [in this window] [in a new window]   Fig. 7.   Representative effect of catalytic subunit of PKA in presence of ATP on KDR channel activity. A: 5 representative sequential records obtained from an inside-out membrane patch under symmetrical KCl recording conditions during 9-s depolarizations to 40 mV after a 2-s step to 60 mV from a holding potential of 0 mV under control conditions (including 5 mM ATP in bath solution) with iberiotoxin (100 nM) added to pipette solution to suppress BKCa channel activity. Same patch is shown after 6.8-min treatment with PKA (50 nM) in continued presence of ATP. Note the following: 1) that traces have been inverted to show openings of KDR channels at 40 mV in outward direction, 2) increased KDR channel activity and lack of null traces during PKA treatment, and 3) lack of change in activity of 35-pS intermediate conductance K+ channel (transitions present in sweeps 3 and 2 and 4 in A and B, respectively, as indicated by arrows). Double arrow indicates residual BKCa transition. B: amplitude histograms for control (left) and PKA + ATP (right) treatments in experiment of A based on 30 sweeps in each condition. Values for NPo were determined from idealized traces.

Application of PKA (Fig. 7) in the presence of ATP caused a marked increase in the activity of K_DR channels in all patches showing 15-pS channel activity under control conditions and induced 15-pS K_DR channel activity in two patches that did not demonstrate activity in the absence of the kinase. On average, PKA increased NP_o of the channels by ~10-fold to 0.138 ± 0.03, which was significantly greater than that recorded in the absence of the kinase (n = 11; P < 0.05 by paired Student's t-test). PKA was applied to two additional patches that did not demonstrate any channel activity before or after treatment with the kinase. These data were not included in the above analysis of NP_o (their inclusion did not affect the outcome: NP_o still increased from 0.010 ± 0.003 to 0.117 ± 0.028; n = 13, P < 0.05 by paired Student's t-test). Additionally, PKA treatment did not induce any new unitary currents, and the representative data of Fig. 7 indicate that the activity of the 35-pS channel and residual BK_Ca channel transitions not completely suppressed by IbTX recorded at 40 mV were unaffected by exposure to the kinase in the presence of ATP.

Two sets of control experiments were conducted to rule out the possibility that changes in K_DR channel activity occurred as a result of nonspecific effects of treatment with the catalytic subunit of PKA. First, when PKA was applied at an identical concentration of 50 nM but in the absence of ATP from the bath solution, neither a change in K_DR channel activity nor the restoration of channel activity was observed in six patches (data not shown). Second, patches were treated with boiled PKA in the presence of ATP, and no effect on K_DR channel activity was observed (n = 3; data not shown). These data imply that the effect of treatment with PKA in the presence of ATP was due to a phosphotransferase reaction leading to an increase in channel open probability. The activity of K_DR channels induced by PKA treatment was also sensitive to block by 4-AP (5 mM) in the bath (data not shown). These experiments were conducted in two ways: I-O patches were exposed to 4-AP (5 mM) after PKA in the presence of ATP had enhanced the activity of the K_DR channels, or patches containing K_DR channels were treated with 4-AP before the application of PKA. K_DR channel activity was not observed before treatment with PKA and ATP, and subsequent exposure to 4-AP in the continued presence of the kinase caused an inhibition of the PKA-induced K_DR channel activity in three membrane patches. No K_DR channel activity was apparent in the presence of 5 mM 4-AP, as indicated above for 1 mM, and subsequent exposure to the kinase in the presence of the channel blocker did not produce any change in the activity of 15-pS K_DR channels. Also, in both sets of 4-AP experiments, no new unitary currents were observed during PKA treatment, and neither the 35- and 95-pS channels nor residual BK_Ca channel activity showed any change.

To determine the mechanism by which the catalytic subunit of PKA and isoproterenol increased K_DR channel activity, the distributions of open and closed times before and after treatment were compared. However, only one patch in each condition did not show evidence of more than one K_DR channel during more than 20 min of recording. For this reason, our analysis was necessarily limited to these two patches. In the presence of PKA, mean open time was unchanged from 1.66 ± 0.04 to 1.67 ± 0.12 ms and mean closed time decreased from 461.50 ± 48.90 to 69.58 ± 2.12 ms in the presence of PKA (330 and 7,234 events before and after treatment, respectively). Figure 8 shows open- and closed-time histograms for control and PKA treatments. Open-time (1-ms bin width; unresolved openings of <1 ms were discarded) and closed-time (100-ms bin width) histograms were best fitted by single and double exponential decay functions, respectively. The values of the fast and slow closed-time constants declined from 73 and 1,025 ms before to 35 and 278 ms after PKA, respectively. Additionally, treatment with PKA decreased the probability of the channels entering the long-lived compared with the short-lived closed state: the fractional time spent in the longer closed state declined from 0.40 to 0.16, whereas that for the short-lived closed state increased from 0.6 to 0.84. The open-time constant was 0.51 ms under control conditions and increased slightly to 0.85 ms in the presence of PKA. Treatment with isoproterenol had a similar effect on K_DR channel activity; mean open times were 1.49 ± 0.04 and 1.40 ± 0.03 ms, and mean closed times were 431.83 ± 8.71 and 23.99 ± 3.80 ms in the absence and presence of -adrenoceptor activation, respectively. Closed-time constants declined from 51.04 and 1,746.30 to 29.59 and 520.59 ms, respectively, in the presence of isoproterenol, but the single open-time constant was unchanged (1.02 and 0.99 ms, respectively). These data suggest that the change in vascular K_DR channel activity due to phosphorylation by PKA in response to -adrenoceptor stimulation may be due to a decreased time spent in a long closed state.

View larger version (29K): [in this window] [in a new window]   Fig. 8.   Closed- and open-time distributions for an inside-out membrane patch before (control) and after exposure to catalytic subunit of PKA (50 nM; 5 mM ATP). Closed- and open-time histograms were prepared from an idealized events list and best fitted with double (I = A1et/1 + A2et/2) and single (I = Aet/) exponential functions, respectively. Time constants of fit () are indicated.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

This study demonstrates the novel finding that the activity of slowly inactivating 15-pS 4-AP-sensitive K_DR channels of rabbit vascular myocytes is regulated by -adrenoceptor occupancy and phosphorylation by the catalytic subunit of PKA. K_DR channel open probability at 40 mV in C-A and I-O patches was increased by ~3- and 10-fold, respectively, during treatment with isoproterenol and the purified catalytic subunit of PKA in the presence of ATP. Application of the kinase in the absence of ATP, or treatment with boiled kinase in the presence of ATP, failed to affect the activity of the K_DR channels. The enhanced K_DR channel activity after exposure to isoproterenol and PKA identified in this study provides the single-channel basis for previous observations of 4-AP-sensitive hyperpolarization and/or enhanced I_KDR of vascular myocytes by treatments that elevate intracellular cAMP and PKA activity (1, 15, 30).

Voltage-gated K⁺ currents due to K_DR channel activation have been observed in all smooth muscle cells studied to date and include 4-AP-sensitive and -insensitive components (8, 21, 24). For example, macroscopic and/or unitary I_KDR have been characterized in myocytes from portal vein (1, 5), renal artery (17), pulmonary artery (3, 26), coronary artery (22, 35), and mesenteric artery (33), as well as nonvascular myocytes from trachea (7) and colon (8). The findings of these studies indicate that vascular I_KDR is very likely due to the activity of more than one type of voltage-gated K⁺ channel, similar to the situation reported for canine colonic myocytes, in which at least two different K_DR channels contribute to macroscopic current (8, 21). This view is based on the range of values reported for 1) single-channel conductance, ranging from 5 to 20 pS for the small-conductance K_DR channels of portal vein and coronary arteries, as well as airway and gastrointestinal myocytes (5, 7, 21, 35), to between 30 and 90 pS for the larger conductance K_DR channels of coronary, pulmonary, and renal arteries as well as gastrointestinal muscle (3, 6, 17, 20, 21, 26); 2) sensitivity to block by 4-AP, with values for half-maximal inhibition of I_KDR ranging from 0.2 to 1.5 mM, as well as the presence of 4-AP-insensitive components; and 3) voltage dependence of inactivation, with values for half-maximal availability of I_KDR varying between 70 and 20 mV (24). We previously found that macroscopic I_KDR of portal vein myocytes was enhanced by exposure to isoproterenol or forskolin and that a peptide inhibitor of PKA prevented the response to isoproterenol (1). Given the reported variability in the properties of vascular K_DR channels and the potential contribution of more than one population of channels to macroscopic I_KDR, the identity of the K_DR channel(s) affected by -adrenoceptor occupancy and PKA activation was therefore not directly evident. The findings of the present study indicate for the first time that a small-conductance, 15-pS inactivating K_DR channel contributes to the increase in I_KDR of vascular myocytes at negative membrane potentials during treatment with -adrenoceptor agonist and activation of PKA.

At least two 4-AP-sensitive macroscopic K⁺ currents are present in the rabbit portal vein. These include a transient K⁺ current, I_KTO, with rapid activation and inactivation kinetics, as well as more slowly inactivating I_KDR (1, 4, 5). In the present experiments, the conductance of the 4-AP-sensitive K_DR channels exhibiting regulation by PKA was 15 pS under symmetrical KCl recording conditions. This small conductance is consistent with the observations of Beech and Bolton (5): channels of 5-8 pS (at 0 mV) were observed in outside-out membrane patches of rabbit portal vein myocytes with an asymmetrical K⁺ gradient (134/6 mM) across the membrane. These channels exhibited slow inactivation during 6-s test pulses from 70 to 0 mV and were suggested to contribute to I_KDR. In the present study, the time constants of activation (~10 ms) and inactivation (~0.3 and 2.5 s) at +40 mV of the 15-pS K_DR channels were similar to those previously described for 4-AP-sensitive macroscopic I_KDR, e.g., 10 ms for activation and 0.25 and 3 s for inactivation (2, 28). We do not believe that the 15-pS channels described in this paper contribute to macroscopic 4-AP-sensitive I_KTO for the following reasons: 1) inactivation of the 15-pS channels was slow, in contrast to the fast inactivation of I_KTO, and 2) I_KTO of portal vein myocytes was shown to be completely inactivated positive to approximately 65 mV (half-maximal potential of 80 mV; Ref. 4). Our experiments demonstrated steady-state activity of the 15-pS channels at 40 mV.

In the present study, parallel experiments were conducted with isoproterenol and purified catalytic subunit of PKA using C-A and I-O patches, respectively, to identify the K_DR channel underlying the increase in I_KDR in rabbit portal vein myocytes associated with elevated intracellular cAMP (1). We previously demonstrated that -adrenoceptor occupancy, via a signal transduction mechanism involving adenylyl cyclase, cAMP, and PKA, enhances I_KDR of rabbit portal vein myocytes (1). Increased I_KDR end-pulse and tail current amplitude was observed during treatment with isoproterenol (1-10 µM) or forskolin (1 µM). The increase in current due to -adrenoceptor activation was reversed upon exposure to 4-AP (5 mM) or propranolol (10 µM) and was prevented by pretreatment with a specific peptide inhibitor of PKA (1). An increase of approximately threefold in NP_o from 0.014 ± 0.005 to 0.041 ± 0.02 for 15-pS K_DR channels of C-A membrane patches was observed in the presence of bath-applied isoproterenol (1 µM) in the present study. This is consistent with the magnitude of change in I_KDR of isolated rabbit portal vein myocytes during -adrenoceptor activation with 1 µM isoproterenol (1). The ability of isoproterenol to alter the channel activity of C-A membrane patches when added outside the pipette is consistent with the presence of a diffusible second messenger, cAMP, in the signal transduction pathway rather than a direct G protein activation of K_DR channels. To further test this mechanism, we also applied the catalytic subunit of PKA in the presence of ATP to I-O membrane patches containing K_DR channels. An increase in K_DR channel open probability at 40 mV of ~10-fold was observed only in the presence of the active kinase and ATP; application of boiled kinase or either ATP or the kinase alone did not influence the activity of K_DR channels. The 10-fold increase in NP_o from 0.011 ± 0.003 to 0.138 ± 0.03 during PKA and ATP treatment was substantially greater than the threefold change in activity after isoproterenol application in the C-A recording mode. Two differences between these recording methods could account for the enhanced activation of K_DR channels in I-O membrane patches. First, the concentration of PKA (50 nM) at the cytoplasmic surface of the channels may have been greater in the I-O membrane patches than in the intact myocytes during C-A membrane patch recordings. Second, a phosphatase activity could have been lost from the I-O membrane patches after excision. Application of PKA and ATP in the absence of competing dephosphorylation of the channels by the phosphatase would be expected to lead to a greater change in open probability. However, we favor the first explanation based on the presence of channel rundown in the I-O membrane patch recording configuration.

In addition to the inactivating 15-pS K_DR channels studied in detail here, the membrane patches of rabbit portal vein myocytes were also found to contain channels of ~35 and 95 pS. Preliminary experiments to define the identity of these 35- and 95-pS channels have shown them to be insensitive to 1 mM 4-AP (T. Malcolm and W. C. Cole, unpublished data). This finding is in contrast to the results of the present study, which show the 15-pS channels to be inhibited by this blocker at the same concentration. Moreover, treatment with isoproterenol or PKA did not affect the activity of these 35- and 95-pS channels at 40 mV, and none of these agents were found to induce the appearance of openings of novel channels not apparent under control conditions. This is consistent with our previous findings that noninactivating residual currents recorded in the presence of 4-AP were insensitive to isoproterenol and forskolin (1). For these reasons, we attribute the increase in whole cell I_KDR in portal vein myocytes to a change in the activity of the 15-pS inactivating channels alone.

The effects of isoproterenol and PKA on the activity of single vascular K_DR channels demonstrated in this study are consistent with previous findings from whole cell patch-clamp experiments employing isolated myocytes of rat portal vein (15), as well as whole cell and single-channel experiments employing isolated canine colonic smooth muscle cells (21, 32). Isoproterenol (14), vasoactive intestinal polypeptide (32), and the catalytic subunit of PKA (21) were shown to cause a 4-AP-sensitive hyperpolarization, to increase I_KDR amplitude, and/or to enhance the open probability of a 20-pS small-conductance K_DR channel activity of canine colonic smooth muscle cells. Interestingly, PKA was also shown to enhance the activity of BK_Ca channels and a larger conductance, 90-pS K_DR channel of canine colonic myocytes at positive membrane potentials (9, 21). However, only the phosphorylation of the 20-pS K_DR channels was concluded to be physiologically relevant; blockade of BK_Ca channels with tetraethylammonium ion or charybdotoxin failed to antagonize the hyperpolarization induced by isoproterenol in intact tissues (14), but the response was effectively antagonized by 4-AP (32). Moreover, the catalytic subunit of PKA increased the open probability of only the 20-pS K_DR channels at the physiologically relevant transmembrane potential of 60 mV (21). Interestingly, Koh et al. (21) attributed the PKA-induced change in activity of the 20-pS K_DR channels to an increase in time spent in a long-lived open state. In the present study, only one open state was apparent for the single 15-pS channels present in two patches, and the change in activity due to isoproterenol or PKA was associated with a decreased time spent in a long closed state. The reason for this apparent difference in behavior between the respective 20- and 15-pS K_DR channels of gastrointestinal and vascular myocytes is not clear at this time. It may be related to a difference in molecular identity of the I_KDR or the properties of the Kv1.5 channels that are important components of the current in both preparations (10, 19, 25); canine colonic myocytes express both Kv1.2 and Kv1.5, but canine portal vein myocytes express only Kv1.5 based on Northern analysis (19, 25). It is unknown whether Kv1.2 is similarly absent from the rabbit portal vein. We recently cloned and sequenced Kv1.5 from this tissue and demonstrated the presence of the channel in freshly dispersed portal vein myocytes by immunocytolocalization (10). Rabbit portal vein Kv1.5 exhibits only 85% identity to the canine Kv1.5 channel and, significantly, one (⁵⁶²RRGS⁵⁶⁵) of two potential PKA phosphorylation consensus sites (⁵³⁹RKAS⁵⁴²) in the intracellular COOH terminus of the rabbit portal vein Kv1.5 channel is not present in the canine channel (25). It is possible that one or both of these consensus sites in the rabbit portal vein Kv1.5 channel is responsible for the increased activity of the native K_DR channels observed in the present study. However, we cannot discount the possible involvement of a modulatory -subunit associated with the channel (23).

Smooth muscle relaxation due to elevated intracellular cAMP and activation of PKA, as occurs during -adrenoceptor occupancy, results from several different phosphoprotein-dependent mechanisms (36). Considerable evidence indicates the involvement of altered Ca²⁺ sequestration, extrusion, and influx due to PKA-mediated phosphotransferase reactions (36). With regard to changes in Ca²⁺ influx, increased K⁺ channel activity leading to hyperpolarization of membrane potential and reduced voltage-gated Ca²⁺ channel activity is now also recognized to play an important role in vasodilation due to elevated cAMP and PKA activation (24). At least three different K⁺ channels appear to contribute to the vasodilation in different vessels because of elevated cAMP based on 1) reduced relaxation with selective blockade of either BK_Ca, K_ATP, or K_DR channels (13, 16, 27, 30) and 2) activation of BK_Ca channels by isoproterenol (34) or the stable analog of prostacyclin, iloprost (31), and K_ATP channels with calcitonin gene-related peptide and adenosine via A₂ receptors (24). The present study provides the first evidence that the slowly inactivating, 4-AP-sensitive, 15-pS K_DR channels of vascular myocytes are also affected by PKA, consistent with a contribution to the vasodilatory mechanism activated, for example, during -adrenoceptor and prostanoid receptor stimulation by -agonists and prostacyclin, respectively.