INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

AGONIST ACTIVATION of smooth muscle initiates a complex sequence of excitation-contraction coupling events (36). These include coupling of cell surface receptor occupancy to the hydrolysis of membrane phosphatidylinositol 4,5-bisphosphate by a specific phospholipase C, with the resultant production of inositol trisphosphate (IP₃) (2), the release of intracellular Ca²⁺ stores (26), myosin light chain phosphorylation (20), and rapid force development (11). Additional mechanisms come into play to sustain contraction, despite the fact that the increases in IP₃, cytosolic free Ca²⁺, and myosin light chain phosphorylation are largely transient (2, 11, 26). The mechanisms responsible for force maintenance are less well understood than those responsible for force development but have been suggested to involve protein kinase C activation (19, 21) and/or an increase in Ca²⁺ sensitivity of the contractile proteins (5, 32).

Force maintenance has been shown to have a critical dependence on extracellular Ca²⁺ influx as well as a close coupling to membrane potential (17, 27, 37). In the absence of extracellular Ca²⁺, contractile responses to agonists cannot be maintained (9, 34). Agonist activation causes membrane depolarization, with subsequent activation of voltage-dependent Ca²⁺ influx (29), and also indirectly activates nonselective cation channels (33, 41).

Mechanisms contributing to the depolarization of the membrane potential by agonists have only recently been identified. Ca²⁺ released from the sarcoplasmic reticulum (SR) by agonists activates Ca²⁺-sensitive Cl channels (Cl_Ca) (41) and inhibits voltage-dependent K⁺ channels (K_v) (13). Both of these effects are thought to contribute to the membrane depolarization. There is direct evidence that membrane depolarization is sustained for the duration of agonist exposure on the basis of membrane potential measurements (27). Furthermore, there is indirect evidence based on the observation that organic Ca²⁺ channel blockers can inhibit sustained agonist-induced contractions in arterial smooth muscle (31), especially in resistance arteries (4) and in arteries from hypertensive subjects (1).

Considering the fact that SR Ca²⁺ release is transient (26) and Cl_Ca currents inactivate with sustained increases in cytosolic Ca²⁺ concentration ([Ca²⁺]) (43), the question arises as to how the membrane depolarization associated with agonist activation is sustained. Recently, it was demonstrated that L-type Ca²⁺ channels (Ca_L) exhibit measurable open probability under steady-state conditions at voltages between 40 and 20 mV in arterial smooth muscle (35). It has also been shown that a voltage "window" exists for Ca_L activation, which produces a similar voltage window of sustained, elevated cytoplasmic [Ca²⁺] (12). These results suggest that sustained Ca²⁺ influx through Ca_L associated with membrane depolarization may also provide a mechanism to inhibit K_v. It was the purpose of the experiments reported in this study to test the hypothesis that Ca²⁺ influx can inhibit K_v and provide a mechanism to produce sustained membrane depolarization during agonist activation, thereby contributing to force maintenance.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cells were isolated from rat small mesenteric artery branches using techniques previously described in detail (6, 8). Briefly, individual branches were cut open longitudinally and incubated at 37°C for 60 min in a Ca²⁺-free buffer. The segment was then cut into small pieces and incubated in buffer with collagenase (250 U/ml) and elastase (25 U/ml) for ~30 min without agitation at 37°C. The tissue was gently aspirated with a Pasteur pipette, and released cells were separated by filtration through 210-µm nylon mesh.

Electrophysiology. An aliquot of cells was placed in a perfusion chamber mounted on an inverted microscope (Nikon Diaphot), allowed to adhere to the glass bottom, then superfused with a HEPES-buffered solution. The external [Ca²⁺] was slowly increased to 2 mM to avoid damaging the cells. Cells were sealed to pipettes with gentle suction, and the membranes were ruptured by negative pressure at a holding potential of 40 mV. Membrane currents were recorded using whole cell or perforated patch-clamp techniques (18) with a voltage-clamp amplifier (model 8900; Dagan). Fire-polished micropipettes (2-3 M resistance) were fashioned from capillary tubing (Kwik-fil; WPI) using a micropipette puller (model P-80/PC; Sutter Instruments). Series resistance and stray capacitance compensation were employed using the amplifier circuitry. For perforated-patch conditions, junction potential was compensated before measurements. Current and voltage signals were converted from analog to digital form at a sampling rate of 2 kHz (Labmaster A/D board) and stored in a computer (Gateway 2000 486DX) for subsequent analysis.

Protocols. Whole cell currents were measured using both voltage-ramp and voltage-step protocols from holding potentials of 60 and 20 mV. With the ramp protocol, voltage was increased at 1 mV/ms from either holding potential to a maximum value of +60 mV. With the step protocol, 1-s voltage steps were applied from 60 to +60 mV from both values of holding potential in 10-mV increments at intervals of 15 s. For statistical analysis and comparisons, peak currents and currents averaged over the last 100 or 200 ms of each voltage-clamp step were determined. For most conditions, current responses were recorded before and during an intervention. Difference currents were determined between such conditions by subtracting digital current records obtained under control conditions from those obtained during an intervention at each value of test voltage. Current records were analyzed using pCLAMP software (version 5.5.1; Axon Instruments).

Chemicals and solutions. Collagenase was purchased from Worthington Biochemical, Freehold, NJ (type CLS3), and elastase was from ICN Nutritional Biochemicals, Cleveland, OH (hog pancreas). All other chemicals were obtained from Sigma Chemical, St. Louis, MO. Water (>18 M) was obtained from a Barnstead purification system with an organic final filter. The solution used for myocyte isolation contained (in mM) 140 NaCl, 5 KCl, 1 MgCl₂, 10 HEPES, and 10 dextrose at pH 7.4 (with NaOH). The external perfusion solution was the same, with 2 mM Ca²⁺ added. The pipette solution for whole cell studies contained (in mM) 140 KCl, 5 NaCl, 5 MgATP, 10 HEPES, and 0.2 or 10 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA) at pH 7.2 (with KOH) and had a resistance of 2.5-3.5 M. HEPES was increased with the 0.2 mM BAPTA internal solution to maintain osmolarity constant at ~300 mosM. The pipette solution for perforated-patch studies contained (in mM) 110 potassium gluconate, 30 KCl, 5 NaCl, 0.1 CaCl₂, and 10 HEPES with 150 µg/ml amphotericin at pH 7.2 (with KOH).

Statistical analysis. Statistical comparisons of membrane currents were performed using a two-way ANOVA with repeated measures for paired data using the STATWORKS application on a Power Macintosh computer. Probability values of <0.05 were considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the first series of experiments, the effects of 0.1 mM Cd²⁺ (to inhibit Ca_L) on K⁺ current (I_K) were determined. Figure 1 shows current responses to voltage ramps from each holding potential before and during the addition of 0.1 mM Cd²⁺ to the perfusate. When I_K was recorded from a holding potential of 60 mV (I_K60), Cd²⁺ increased net outward current at voltages just above the activation threshold (Fig. 1A). At more positive voltages up to +60 mV, however, net outward current was decreased with Cd²⁺. When recorded from a holding potential of 20 mV (I_K20), Cd²⁺ only decreased net outward current at voltages >0 mV (Fig. 1B). Difference currents, obtained by digital subtraction of current ramps before and during Cd²⁺, also shown in Fig. 1, confirm this description. We initially assumed that, under the conditions of these experiments, K⁺ currents dominate these responses. Whole cell K⁺ currents have been shown in many smooth muscle cells to be composed primarily of K_v and Ca²⁺-activated K⁺ channel (K_Ca) components (22, 29). Because K_v currents are expected to be inactivated (decreased availability) at a holding potential of 20 mV (22, 46), the I_K20 data should primarily represent the K_Ca component. The above I_K20 results suggest that inhibiting Ca_L with Cd²⁺ reduced K_Ca current. Because both K_v and K_Ca currents are expected to contribute to I_K60, and since Cd²⁺ inhibits I_K20 (i.e., K_Ca) only at voltages >0 mV, one interpretation of the I_K60 responses at voltages <0 mV is that K_v currents are augmented in the presence of Cd²⁺. The presence of both K_v and K_Ca components of I_K60 were observed in every cell studied.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Effects of external Cd2+ on current responses to voltage ramps. Current responses were recorded from holding potentials of 60 mV (A) and 20 mV (B). Arrows, beginning of voltage ramp. Each panel shows control ramp responses before () and during addition of 0.1 mM Cd2+ to external perfusion solution. Below these responses are difference currents obtained by digital subtraction of ramps obtained with Cd2+ minus those before Cd2+. Holding current before and after ramps are included as baseline references. Calibration bars in A apply to all currents and represent 200 pA and 100 ms, respectively.

Current responses to voltage ramps are only qualitative at best, since voltage-dependent activation and inactivation kinetics may obscure the details of currents recorded by this method. To study the effects of Cd²⁺ on I_K quantitatively, measurements of I_K were made in response to voltage steps from holding potentials of 60 and 20 mV with 2 mM external [Ca²⁺] and 0.2 mM pipette BAPTA in the absence then the presence of 0.1 mM Cd²⁺. Figure 2 shows representative results of whole cell currents recorded at two values of test potential (0 and +60 mV) from the two values of holding potential (60 and 20 mV). The addition of Cd²⁺ to the perfusion solution consistently increased I_K60 at a test voltage of 0 mV (Fig. 2A, top) but decreased I_K60 at a test voltage of 60 mV (Fig. 2C, top) compared with control conditions. Inspection of the current responses suggests that the kinetics of the current was also altered by Cd²⁺. Therefore, difference currents [I_K60(Cd)] were calculated between the two conditions. At 0 mV test voltage, I_K60(Cd) increased rapidly to a peak and subsequently declined to a steady level that was maintained for the duration of the voltage step (Fig. 2A, bottom). The peak value of I_K60(Cd) averaged 75 ± 13 pA, whereas the "steady" level (last 100 ms of the clamp step) averaged 31 ± 9 pA (n = 8). Although the majority of I_K60(Cd) recorded at the +60-mV test voltage was negative (smaller net outward current), there was an early positive phase for the difference current (arrow in Fig. 2C, bottom). The peak value of this early current averaged 115 ± 18 pA, whereas the steady level averaged 148 ± 52 pA (n = 8) at +60 mV. When recorded at a test voltage of 0 mV from a holding potential of 20 mV, the currents were very small (Fig. 2B, top) and the difference current [I_K20(Cd)] was essentially zero. When measured at a test potential of +60 mV (Fig. 2D, bottom), I_K20(Cd) was consistently decreased in the presence of Cd²⁺ and never exhibited an early positive peak as found in I_K60(Cd). The steady level of I_K20(Cd) averaged over the last 200 ms of the voltage-clamp step was 288 ± 48 pA (n = 8).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Effects of Cd2+ on K+ current (IK) measured with 2 mM external [Ca2+] and 0.2 mM pipette BAPTA. Currents were measured from holding potentials of 60 (A and C) or 20 mV (B and D). In each panel, currents before () and during addition of 0.1 mM Cd2+ to the perfusion solution are shown at test voltages of either 0 mV (A and B) or +60 mV (C and D). Difference currents obtained by digital subtraction of current with and without Cd2+ are shown in each panel. Calibration bars represent 60 pA (A and B) or 200 pA (C and D) and 100 ms (A-D).

To expand on these observations, measurements were repeated at several test potentials from both values of holding potential (i.e., I_K20 and I_K60). Examples of difference currents determined from currents measured before and during Cd²⁺ at various values of test potential are shown in Fig. 3. Over a test potential range from 20 to +20 mV (Fig. 3A), I_K60(Cd) rose to an initial peak that increased with test voltage then appeared to "inactivate" with time. This secondary decline increased with increasing test voltage and was also associated with an increase in current "noise." With further increases in test potential >20 mV, the initial peak became shorter in duration and was followed by a net negative outward current. I_K60(Cd) always exhibited an initial positive phase at all test potentials, but values during the sustained phase of the voltage-clamp step exhibited a transition from positive values at negative test potentials to negative values at positive test potentials. This behavior was different from that recorded from a holding potential of 20 mV (Fig. 3B). Values of I_K20(Cd) were zero at all voltages up to 0 mV (Fig. 3B). At test potentials >0 mV, values of I_K20(Cd) were only negative (decreased net outward current or increased net inward current) over the entire duration of the voltage-clamp step. I_K20(Cd) never demonstrated an early positive value.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Voltage dependence of difference currents (IK) associated with addition of Cd2+ to perfusate. A: IK determined from IK recorded with and without Cd2+ at test voltages from 20 to +60 mV as indicated from a holding potential of 60 mV. B: similar data recorded from a holding potential of 20 mV. Responses from a 60 mV holding potential develop a biphasic character with increasing test voltage (i.e., positive to negative values), whereas those from a 20 mV holding potential are only monophasic (negative). Calibration bars represent 400 pA and 100 ms, respectively.

Peak values and steady values of I_K60(Cd) averaged over the last 100 ms of the voltage-clamp step were determined at different test potentials, and average results are shown in Fig. 4A. Peak values of I_K60(Cd) "activated" at about 30 mV and increased with voltage up to +60 mV. Steady values of I_K60(Cd) also activated at about 30 mV, increased to a maximum value at about +10 mV, then decreased with increasing test potential, becoming negative at about +40 mV and higher. Steady values of I_K20(Cd) were determined over the last 200 ms of the voltage-clamp step, and results are summarized in Fig. 4B. Steady values of I_K20(Cd) were significantly different from zero only at voltages >0 mV, where they were always negative (smaller net outward current). Steady values of I_K20(Cd) were significantly larger (i.e., more negative) than steady values of I_K60(Cd) at all voltages >0 mV. These results are qualitatively similar to those presented in Fig. 1 for ramp current responses.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Summary of voltage dependence of difference currents measured in presence of Cd2+ from holding potential of 60 mV [IK60(Cd)] (A) and 20 mV [IK20(Cd)] (B). A: responses determined from peak value of IK60(Cd) () and value averaged over last 100 ms of voltage-clamp step (). B: value of IK20(Cd) averaged over last 200 ms of voltage-clamp step. Symbols are means and vertical lines are ±1 SE (n = 8).

The current responses at the two holding potentials have been interpreted above based on the assumption that only K_Ca currents contribute to I_K20, whereas both K_v and K_Ca contribute to I_K60. The validity of the conclusions concerning the effects of Cd²⁺ on I_K components rests in part on the validity of this assumption. Therefore, experiments were performed to determine the effects of blockers of K_v (4-aminopyridine) and K_Ca (iberiotoxin) on I_K20 and I_K60 to test this assumption. As shown in Fig. 5, 100 nM iberiotoxin completely inhibited I_K20, whereas it had a relatively small effect on I_K60 (Fig. 5B), including reducing the current noise. In the presence of iberiotoxin, prominent tail currents remain in I_K20, representing the recovery of K_v from inactivation following clamp steps to voltages negative to the 20-mV holding potential. When 1 mM 4-aminopyridine was added with iberiotoxin, I_K60 was reduced substantially, as was the tail current in I_K20 (Fig. 5C). Similar results were obtained in six cells. These results confirm the assumption that I_K20 represents primarily K_Ca currents, whereas I_K60 represents both K_v and K_Ca currents.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Pharmacology of IK recorded from mesenteric myocytes at 2 holding potentials. Families of IK were recorded from holding potentials of 60 (top traces) and 20 mV (bottom traces) under control conditions (A), after addition of 100 nM iberiotoxin to perfusate (B), and after addition of 1 mM 4-aminopyridine with iberiotoxin (C). Calibration bars below C represent 300 pA and 400 ms.

If the responses shown in Figs. 1-4 are the result of inhibiting Ca²⁺ influx on I_K components by Cd²⁺ as suggested, then it would be expected that decreasing Ca²⁺ influx by lowering external [Ca²⁺] should have similar effects. The effects of decreasing external [Ca²⁺] from 2 to 0.2 mM on I_K were determined to test this expectation. External [Mg²⁺] was simultaneously increased to 2.8 mM to maintain divalent ion concentration constant. The results are summarized in Fig. 6. At a test potential of 0 mV, decreasing external Ca²⁺ increased I_K60, and the difference current, I_K60(Ca), increased rapidly to a peak then declined slowly for the duration of the clamp step (Fig. 6A). The peak value of I_K60(Ca) at 0 mV test voltage averaged 120 ± 20 pA while the steady value averaged over the last 100 ms of the clamp step was 59 ± 19 pA (n = 9). As with Cd²⁺, I_K60(Ca) recorded at a +60 mV test voltage exhibited an early positive phase (arrow in Fig. 6C) followed by a sustained negative phase. The peak value of this early current phase averaged 187 ± 34 pA while the steady level averaged 296 ± 44 pA (n = 9). When recorded from a holding potential of 20 mV, however, I_K20 was only decreased when external Ca²⁺ was decreased, and only negative difference currents (smaller net outward currents) were observed over the entire duration of the voltage-clamp step. At a test potential of 0 mV, the difference currents were also essentially zero (Fig. 6B), but at +60 mV (Fig. 6D) I_K20(Ca) averaged 455 ± 78 pA. In contrast to I_K60(Ca), I_K20(Ca) never exhibited an early positive phase at a test potential of +60 mV (Fig. 6D). When the voltage dependence of the peak and late values of difference currents were determined (Fig. 6, E and F), they exhibited characteristics similar to those found with Cd²⁺ (Fig. 4) but with larger magnitudes.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Effects of decreasing external Ca2+ on IK. Currents were measured from holding potentials of 60 (A, C, and E) or 20 mV (B, D, and F). Measurements were made at 0.2 and 2 mM Ca2+ () in random order. External [Mg2+] was increased by 1.8 mM in the former to maintain total divalent ion concentration constant. Current traces show representative responses and difference currents from both holding potentials at test voltages of +0 (A and B) or +60 mV (C and D). Calibration bars represent 100 pA (A and B) or 300 pA (C and D) and 100 ms (A-D). E: average values of peak () and late () IK60. F: average values of late IK20 over voltage range of 60 to +60 mV. Symbols are means and vertical lines are ±1 SE (n = 9).

The positive difference currents produced by decreasing external Ca²⁺ could be due (in part) to a shift in the voltage dependence of I_K availability as a result of altered masking of surface charge even with the compensatory increase in external [Mg²⁺] (15). Figure 7 shows a comparison of the effects of external [Ca²⁺] on steady-state inactivation of I_K. There was no significant difference in the voltage dependence of this relationship when external [Ca²⁺] was reduced to 0.2 mM and external [Mg²⁺] increased to 2.8 mM. The addition of 0.1 mM external Cd²⁺ also had no significant effect on I_K availability (data not shown). It is unlikely, therefore, that a shift in the voltage dependence of I_K contributed to the effects of decreasing external Ca²⁺ on whole cell current described above.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Effects of external [Ca2+] on voltage dependence of IK availability recorded with 0.2 () and 2 mM Ca2+ () with 3 M total divalents (Ca2+ + Mg2+) in perfusion solution. Availability was determined using a 2-step protocol consisting of a 20-s conditioning voltage step (from 100 to 0 mV) followed by a 1-s step to a test voltage of 0 mV. Peak value of current recorded at each test potential was determined as a function of conditioning voltage, normalized to the maximum response range, and averaged (n = 9).

If these current responses are the result of an effect of Ca²⁺ influx on global intracellular [Ca²⁺], then I_K responses to Cd²⁺ should be decreased in cells dialyzed against a pipette solution containing 10 mM BAPTA to buffer intracellular [Ca²⁺] (45). As shown in Fig. 8, responses to the addition of 0.1 mM Cd²⁺ to the external perfusate with 10 mM pipette BAPTA were smaller but showed some similarities to those obtained with 0.2 mM pipette BAPTA (Fig. 4). Peak values of I_K60(Cd) measured with 10 mM pipette BAPTA exhibited a bell-shaped voltage dependence with a peak near 0 mV test voltage (Fig. 8A). I_K60(Cd) at 0 mV test voltage exhibited an early positive value (increased net outward current) that averaged 68 ± 11 pA (n = 7) and inactivated more completely toward baseline over the course of the 1-s voltage-clamp step (Fig. 8A). Values of I_K60(Cd) with 10 mM BAPTA measured at a +60-mV test voltage did not exhibit an early positive peak, but a significantly negative difference current developed during the sustained voltage-clamp step (Fig. 8C). The late value of I_K60(Cd) averaged 59 ± 23 pA at a +60-mV test voltage with 10 mM pipette BAPTA. Values of I_K20 recorded with 10 mM pipette BAPTA were generally smaller than those recorded with 0.2 mM pipette BAPTA as expected and were reduced by the addition of external Cd²⁺. These results suggest that, when cytosolic Ca²⁺ is dialyzed to low (nM) levels by 10 mM pipette BAPTA, blocking Ca²⁺ influx with external Cd²⁺ still increased I_K60 (smaller effect compared with 0.2 mM BAPTA), but to a smaller extent and over a smaller voltage range.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Effects of 10 mM pipette BAPTA on IK response to external Cd2+. Measurements were made before () and during addition of 0.1 mM Cd2+ to perfusate at holding potentials of 60 (A, C, and E) and 20 mV (B, D, and F). Current traces show representative responses at test voltages of +0 (A and B) or +60 mV (C and D). Calibration bars represent 100 (A and B) or 200 pA (C and D), and 100 ms. E: average values of peak () and late () IK60. F: average values of late IK20 over voltage range of 60 to +60 mV. Symbols are means and vertical lines are ±1 SE (n = 7).

The direct effects of Cd²⁺ on I_K were determined by recording responses in the absence of external Ca²⁺. These results are summarized in Fig. 9. I_K was again measured from holding potentials of 20 (I_K20) and 60 mV (I_K60) with 0 mM external Ca²⁺ (replaced by Mg²⁺) and 0.2 mM pipette BAPTA. Currents measured from a holding potential of 20 mV were significantly reduced at voltages above a test voltage of +40 mV by Cd²⁺ (Fig. 9F), but the effect was much smaller than noted under previous conditions. At test voltages above 20 mV, Cd²⁺ significantly reduced currents recorded from the 60-mV holding potential (Fig. 9E). Cd²⁺ never increased I_K under conditions of Ca²⁺-free perfusion.

View larger version (20K): [in this window] [in a new window]   Fig. 9.   Effects of external Cd2+ on IK during Ca2+-free perfusion. Measurements were made before () and during addition of 0.1 mM Cd2+ to perfusate at holding potentials of 60 (A, C, and E) and 20 mV (B, D, and F). Current traces show representative responses at test voltages of 0 (A and B) or +60 mV (C and D). Calibration bars represent 100 pA and 100 ms. E: average values of peak () and late () IK60. F: average values of late IK20 over voltage range of 60 to +60 mV. Symbols are means and vertical lines are ±1 SE (n = 5).

To directly test the interpretation of the previous results, experiments were performed after blocking K_Ca currents with 100 nM iberiotoxin. In addition, I_K was recorded using the perforated-patch configuration (amphotericin) to provide more physiological conditions. The effects of inhibiting Ca²⁺ influx with 0.1 mM Cd²⁺ under these conditions are summarized in Fig. 10. When measured in the presence of iberiotoxin, the increase in I_K60 (i.e., K_v) associated with the addition of Cd²⁺ now occurred over the entire voltage, where I_K60 was active (from 30 to +60 mV). The maximum increase in I_K60 averaged 100 ± 6 pA (n = 4) at a test voltage of +60 mV.

View larger version (15K): [in this window] [in a new window]   Fig. 10.   Effects of Cd2+ on IK recorded with perforated-patch methods in presence of 100 nM iberiotoxin. A: original IK60 records were measured at 0 (left) and +60 mV (right) before () and during () addition of 0.1 mM Cd2+ to perfusion solution. Difference currents () were determined by digital subtraction of those records and are shown below each pair. Calibration bars: 100 pA and 100 ms. B: peak values of difference currents were determined at each test voltage and are shown as means ± SE (n = 4). Solid curve shows a fit of the data with a Boltzmann function: peak value = 100 ± 6 pA; voltage at half-maximum current (V1/2) = 6.5 ± 2.9 mV; and slope factor = 10.4 ± 2.4 mV.

To further test the interpretation of the previous results, experiments were performed to determine the effects of increasing Ca²⁺ influx on I_K60. I_K was recorded with the perforated-patch configuration in the presence of 100 nM iberiotoxin to block K_Ca, and Ca_L current was increased by adding 1 µM BAY K 8644 to the perfusate. As shown in Fig. 11, I_K60 was decreased following the addition of BAY K 8644, and the effect was reversible. The decrease in I_K60 was significant at all voltages above 30 mV, where K_v was active, and had a maximum value of 97 ± 7 pA (n = 4) at a test voltage of +60 mV.

View larger version (31K): [in this window] [in a new window]   Fig. 11.   Effects of BAY K 8644 on IK recorded with perforated-patch methods in presence of 100 nM iberiotoxin. A: families of currents measured from 60 to +60 mV shown here were recorded (from top to bottom) 1) under control conditions, 2) after addition of iberiotoxin, 3) after addition of 1 µM BAY K 8644 in presence of iberiotoxin, and 4) after washing out BAY K 8644. Calibration bars: 100 pA and 100 ms. B: peak current values determined at each test voltage before (), during (), and after () addition of BAY K 8644 to perfusate. C: difference currents obtained by subtracting currents recorded after addition of BAY K 8644 from those recorded before. Solid curve shows a fit of the data with a Boltzmann function: peak value = 97 ± 7 pA; V1/2 = 12.9 ± 2.8 mV; and slope factor = 14.3 ± 1.1 mV. Data are shown as means ± SE (n = 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of these studies demonstrate that reducing Ca²⁺ influx either by 1) decreasing external Ca²⁺ or 2) adding external Cd²⁺ increased steady values of I_K60 at test voltages between about 40 and +10 mV but decreased I_K60 at test voltages above +30 mV and decreased I_K20 at test voltages above +10 mV. We have interpreted these results to indicate that decreasing Ca²⁺ influx relieves inhibition ("disinhibits") of K_v and decreases the activation of K_Ca in this tissue.

This interpretation of our experimental results based on whole cell current measurements relies on two assumptions: 1) the contribution of K_v and K_Ca currents to I_K20 and I_K60 and 2) the relative contribution of outward (primarily K⁺) and inward currents (primarily Cl and Ca²⁺) to the net current responses. Regarding the first assumption, I_K20 but not I_K60 is nearly completely inhibited by iberiotoxin, a selective inhibitor of K_Ca, suggesting that I_K20 is primarily K_Ca current. I_K60 is only partially inhibited by iberiotoxin but is further inhibited by 4-aminopyridine, suggesting that both K_v and K_Ca contribute to I_K60. This latter conclusion is consistent with the known steady-state availability of K_v in vascular myocytes, which have a voltage for 50% availability between 50 and 40 mV (22, 46), as well as with the known properties of K_Ca (10). Thus the first assumption appears to be valid.

Regarding the second assumption, the increase in difference currents associated with adding Cd²⁺ or reducing external [Ca²⁺] could be due to an increase in outward current, a decrease in inward current, or a combination of both. We have suggested that our results primarily represent changes in outward currents. However, several candidate inward currents could contribute to these responses, including Cl, Ca²⁺, and nonselective cation currents (29, 33, 41, 44). Under the conditions of our experiments, both Cl (symmetrical [Cl]) and nonselective cation currents would be expected to have reversal potentials near 0 mV (41, 42). The results shown in Figs. 2 and 6 demonstrate that reducing Ca²⁺ influx by either method increased whole cell current at a test potential of 0 mV (as well as others). It is unlikely that inhibition of either Cl or nonselective cation currents contributes significantly to the changes in I_K60 at 0 mV. Furthermore, if changes in inward currents did contribute significantly to these current responses, it would be expected that the holding current at 60 mV would be increased (i.e., more positive) following inhibition of Ca²⁺ influx, but it was not (Fig. 1). Furthermore, the difference currents shown in Figs. 3A and 5A remain positive over the entire 1-s voltage-clamp step. It is unlikely that Ca_L are directly responsible for this effect, since they are likely to be small under the conditions of these experiments and they inactivate rapidly to small steady-state values (35). No voltage-gated Na⁺ currents or T-type Ca²⁺ currents are present in this tissue (8, 24). Although a minor contribution from inward currents may occur at negative test voltages, it is likely that I_K60 and I_K20 primarily represent contributions from K_v and K_Ca.

On the basis of this discussion, we have interpreted the results of these experiments to indicate that decreasing Ca²⁺ influx in rat small mesenteric artery myocytes augments K_v currents and inhibits K_Ca currents. When recorded from a holding potential of 20 mV where the dominant current is carried by K_Ca, decreasing Ca²⁺ influx decreases I_K20 over the voltage range where K_Ca currents are active (i.e., >0 mV). When recorded from a 60 mV holding potential where both K_v and K_Ca currents contribute to I_K60, a mixed result is predicted. Over the voltage range where K_v current dominates (negative to +10 mV), the effect of decreasing Ca²⁺ influx is to increase I_K60. Over the voltage range where K_Ca current dominates (positive to +30 mV), the effect of decreasing Ca²⁺ influx is to decrease I_K60. The results obtained in the presence of iberiotoxin confirm this interpretation. When K_Ca currents are inhibited, decreasing Ca²⁺ influx (with Cd²⁺) increases I_K60 over the entire voltage range from 30 to +60 mV. Furthermore, increasing Ca²⁺ influx (with BAY K 8644) decreases I_K60, as would be predicted based on our interpretation of these results. These latter two sets of experiments provide strong evidence in support of our interpretation of the results obtained in the absence of K⁺ channel blockers.

In apparent conflict with this simple explanation was the observation that, at positive test potentials, an initial rapid, transient positive phase of the difference current was found in I_K60 (Figs. 2C and 6C) but not I_K20 (Fig. 2D and 6D). This initial peak of I_K60 was followed by a secondary decrease to negative net I_K60 values. This biphasic response can be explained on the basis of a faster rate of activation of K_v compared with K_Ca currents (30, 40). At positive test potentials, the effects of Ca²⁺ influx on K_v and K_Ca contribute algebraically to the overall response. The rapid, initial transient phase represents the "disinhibition" of the more rapidly activating K_v current, but the larger inhibition of the more slowly activating K_Ca current subsequently dominates the K_v contribution, and the difference current declines to its final net negative value. Consistent with this explanation, when difference currents between 20 and +60 mV are inspected, a progressive transition in I_K60 is seen over this voltage range (Fig. 3). Although K_Ca dominate steady-state responses at positive test potentials, the contribution of K_v disinhibition can be seen in the initial transient current phase (Fig. 3A). However, since K_v are not available at a holding potential of 20 mV, the inhibition of K_Ca by decreased Ca²⁺ influx completely explains the observed responses of I_K20 (Fig. 3B).

To test the interpretation of these results, we developed a simple model representation. We assumed that I_K could be represented by the sum of K_v and K_Ca components. The K_v component was represented by an activation function consisting of a single exponential function to the second power (n²) plus an inactivation function consisting of two exponential components plus a noninactivating component (40). The K_Ca component was represented by a similar activation function. Values for the amplitude and time constants of the various functions were determined for K_v and K_Ca components from I_K60 data obtained in the presence of iberiotoxin and from I_K20 data, respectively, using the curve-fitting functions of SigmaPlot. Figure 12 shows some of the results predicted by this model at a test voltage of +60 mV. First, we assumed that reducing Ca²⁺ influx increased the amplitude of K_v current by 30% and decreased the amplitude of K_Ca current by 50%. The net effect of these changes on I_K is shown in Fig. 12F. The model predicts an early, transient positive current response followed by a more slowly developing net negative current response. If only a change in K_Ca current is allowed to occur, the response in Fig. 12G is obtained, and there is no early positive peak, only a net negative current. This simple model accurately represents the results shown above (Figs. 2, 3, and 6) and gives support to the explanations provided. Thus the early net positive outward current peak in the composite whole cell I_K response is the result of the more rapid activation kinetics of K_v compared with K_Ca currents.

View larger version (17K): [in this window] [in a new window]   Fig. 12.   Model representation of effects of Ca2+ influx inhibition on IK. A: voltage-dependent K+ channel (Kv) activation represented by function [1  exp(t/0.01)]2. B: Kv inactivation represented by {0.25 + 0.25[exp(t/0.25)] + 0.5[exp(t/2)]}. C: Kv represented as product of activation and inactivation terms (A × B). D: Ca2+-activated K+ channel (KCa) activation represented by [1  exp(t/0.03)]2. E: total IK represented as sum of Kv and KCa components (C + D). F: IK difference current (IK before minus IK after) decreasing Ca2+ influx, assuming Kv current is increased by 30% and KCa current is decreased by 50%. G: IK difference current, assuming only a 50% inhibition of KCa current. Time constants are in seconds and were determined from experimental data recorded at a test voltage of +60 mV. IK component amplitudes were normalized to produce peak values of one. Calculations were performed using Excel (Microsoft, Redwood, WA) and plotted using SigmaPlot (Jandel, San Rafael, CA).

How does Ca²⁺ influx contribute to these effects on I_K components? The results obtained with 10 mM pipette BAPTA suggest that Ca_L current activated by voltage-clamp steps directly contributes to these responses. The primary evidence to support this conclusion comes from the voltage dependence of the peak value of I_K60(Cd) recorded under these conditions, which mirror the voltage dependence of Ca_L currents (7, 8). However, the voltage dependence of peak I_K60(Cd) or I_K60(Ca) recorded with 0.2 mM pipette BAPTA does not exhibit a similar voltage dependence. Instead of exhibiting a bell-shaped variation with voltage peaking near 0 mV, these responses extend to positive test potentials, where they achieve their largest values. This is particularly apparent in the results obtained in the presence of K_Ca blocked by iberiotoxin (Fig. 10). This suggests that the time-dependent Ca_L is not the primary determinant of the I_K responses to decreased Ca²⁺ influx under more normal conditions (0.2 mM pipette BAPTA). It is likely, therefore, that a finite open probability of Ca_L contributes significantly to global intracellular Ca²⁺ at a holding potential of 60 mV (35). This conclusion is supported by the observation that augmentation of Ca_L by BAY K 8644 inhibits K_v over the entire test voltage range up to +60 mV. The greater effect of time-dependent Ca_L in the presence of 10 mM BAPTA is likely due to the much larger current amplitude that is expected when global intracellular Ca²⁺ is reduced, resulting in the loss of Ca²⁺-induced inactivation of Ca_L (15)

We chose to use Cd²⁺ to inhibit Ca_L in these experiments, because in initial studies we found that 100 nM nisoldipine inhibited I_K60 with 0.2 mM external Ca²⁺ and 0.2 mM pipette BAPTA. Under these conditions, it is expected that only a small Ca_L would exist, so that the inhibition of I_K by nisoldipine probably represents a direct inhibitory effect on K_v currents. Data in the literature demonstrate that heterologously expressed channels of the Shaker family are inhibited by dihydropyridines (16). A similar effect of dihydropyridines has been reported in rabbit coronary artery myocytes (22).

The effect of Cd²⁺ on I_K60 in the voltage range from 20 to +10 mV was similar with 0.2 or 10 mM pipette BAPTA (Figs. 2, 4, and 8). With the higher pipette BAPTA it would be expected that global intracellular Ca²⁺ would be much lower but that the Ca²⁺ current would be larger because of a reduction in Ca²⁺-induced inactivation of Ca_L (3). These results could be interpreted to suggest that Ca_L and K_v may be colocalized in the plasma membrane in a region not readily accessible to BAPTA or in which free diffusion of Ca²⁺ may be restricted. The difference current recorded with 10 mM BAPTA exhibited a continuous decrease with time, consistent with slow diffusion and/or buffering of local Ca²⁺. Similar suggestions have been made concerning the colocalization of functional elements involved in smooth muscle excitation-contraction coupling, including SR Ca²⁺-release channels and plasma membrane K_Ca (28), plasma membrane Na⁺ pump and Na⁺/Ca²⁺ exchanger, and calsequestrin-containing regions of the SR (25).

The effects of Ca²⁺ influx on whole cell I_K are similar to those demonstrated for the I_K response to SR Ca²⁺ release mediated by either agonists or caffeine (13, 14). These investigators further demonstrated that Ca²⁺ added at micromolar levels to the cytoplasmic surface of inside-out patches inhibited the open probability of single-channel events by decreasing mean open time. Similar but more complicated results have been described for the effects of Ca²⁺ on inside-out patches of Xenopus oocytes expressing colonic smooth muscle K_v1.5 channels (39).

It has been known for some time that agonist activation of smooth muscle produces membrane depolarization (17, 27, 37). Ca²⁺ influx is then increased as a result of an increase in open probability of Ca_L associated with the membrane depolarization (29) and as a direct result of the effect of the agonist on Ca_L (23). Activation of Cl_Ca (33) and inhibition of K_v (13) by SR Ca²⁺ release have both been suggested to participate in this response. This has been suggested to be a positive-feedback mechanism to maintain Ca²⁺ influx and elevated cytosolic Ca²⁺ (13). However, it is known that agonists, in addition to promoting SR Ca²⁺ release, maintain the release channels in the open configuration, preventing subsequent SR Ca²⁺ accumulation and release, thereby allowing Ca²⁺ influx to sustain contraction (38). The results of the present experiments suggest that Ca²⁺ influx can also inhibit K_v, thereby sustaining the membrane depolarization and the increased open probability of Ca_L. Cl_Ca have been shown to inactivate during sustained increases in cytosolic Ca²⁺ (44) by a Ca²⁺/calmodulin kinase II-mediated process (43). This suggests that, while an action of SR Ca²⁺ release on Cl_Ca and K_v may be involved in the initial phase of force development, sustained membrane depolarization and Ca_L influx may be maintained by an action of Ca²⁺ influx on K_v, contributing to force maintenance by a positive-feedback mechanism.