INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

POTASSIUM CHANNELS CONSIST of a diverse group of proteins with disparate structural features and controlling mechanisms. Large-conductance K⁺ channels activated by raised intracellular Ca²⁺ levels (BK_Ca channels) are a feature of many smooth muscle cell types, playing a central role in the control of cellular excitability. The presence of BK_Ca channels in nonpregnant human (24), pregnant human (1, 8), and pregnant rat (9), rabbit, or pig myometrium (22) has been clearly demonstrated. Furthermore, macroscopic Ca²⁺-sensitive outward K⁺ currents have been observed in rat myometrium at estrus (20) and during pregnancy (18).

The properties of BK_Ca channels of neurons (21), skeletal muscle (3), and smooth muscle (7, 15) share many features, including a high single channel conductance (200-250 pS), voltage-dependent activation, and a similar pharmacology. Thus intracellular Ba²⁺ blocks the BK_Ca channel of neurons (21), rabbit T-tubules (25), and smooth muscle (2, 15) at submillimolar concentrations, whereas tetraethylammonium ion (TEA) block of most BK_Ca channels is more potent when this quaternary ammonium ion is applied to the extracellular rather than the internal membrane surface (2, 26, 30). We reported previously (8) the presence of a Ca²⁺- and voltage-sensitive K⁺-selective ion channel in pregnant, human nonlabor myometrium that has the characteristics of the classical BK_Ca channel, in that activation of the channel is a function of membrane potential and intracellular Ca²⁺ concentration ([Ca²⁺]_i), and it is sensitive to block by internal Ba²⁺ but less so by TEA. In contrast, the electrophysiological properties of the channel most frequently recorded from human myometrium after the onset of labor are considerably different from those of the BK_Ca channel, in that the former lacks Ca²⁺ dependence, has a high open-state probability (P_o) in the absence of internal Ca²⁺, and exhibits little voltage dependence. In addition, this channel (which we have termed BK to indicate large conductance and Ca²⁺ independence) does not display the typical pharmacology of BK_Ca channels, in that TEA sensitivity is enhanced and Ba²⁺ sensitivity is reduced in inside-out patches (8). A large-conductance Ca²⁺-insensitive K⁺ channel with similar features has also been reported in bovine tracheal myocytes (7). Therefore, the aim of the present study was to investigate further the differences in sensitivity of BK_Ca and BK channels in human myometrium to relatively specific and nonspecific K⁺ channel blockers.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Tissue collection. Myometrial tissue was collected from women undergoing elective or emergency cesarean section under regional analgesia (spinal or epidural anesthesia for emergency cesarean sections, spinal only for elective cesarean sections) at full-term gestation (38-42 wk) after patient consent with Local Ethical Committee approval (Cambridge District Health Authority). Patients were defined as being in labor if they had regular uterine contractions at a frequency greater than one every 5 min. Biopsy tissue from the upper section of the lower uterine segment was obtained from the area immediately below the reflection of the visceral peritoneum from nonlabor and labor patients. The use of this site as a marker for tissue collection also ensured that the cervix was not mistaken for the myometrium. Furthermore, visual inspection of the biopsy confirmed the muscular nature of the tissue, and there appeared to be no obvious morphological differences in myometrium obtained from nonlabor or labor patients. Tissue was kept for up to 12 h in Ham's F-12 supplemented with 100 U/ml penicillin and 100 µg/ml streptomycin.

Cell isolation. Freshly isolated myometrial cells were obtained by enzymatic disaggregation of finely minced myometrium with 2 mg/ml collagenase (type IA, 300-400 U/mg; Sigma Chemical) in Hanks' buffered salt solution. The incubation with enzyme was performed at 37°C for 2 h followed by centrifugation (800-1,000 revolutions/min) in 60% Percoll for 10 min; then the pellicle was removed, washed, and spun in physiological solution composed of (mM) 135 NaCl, 5 KCl, 1 CaCl₂, 1 MgCl₂, and 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES, pH 7.2). Single cells were plated onto 35-mm petri dishes (Falcon) that had been pretreated with lectin; thus 1 mg/ml concanavalin A in physiological solution was added to the dish for 2 min; then the dish was rinsed with physiological solution before the cells were plated, and experiments were begun immediately. The pretreatment of plates with concanavalin A improved cell adhesion to the plastic substrate but had no noticeable effect on channel activity.

Electrophysiological recordings and data analysis. Inside-out and outside-out patches were utilized for this study. Patch pipettes with resistances of 8-12 M when filled with electrolyte were fabricated from borosilicate glass on a PB-7 puller (Narashige). Single channel recordings were made with an Axopatch 200 patch-clamp amplifier (Axon Instruments), collected on digital audiotapes through a digital audiotape recorder (model DTC 1000ES, Sony), and replayed for illustration onto a Gould RS 3200 chart recorder. Voltages were measured with respect to an AgCl reference electrode and are expressed according to the usual sign convention, i.e., inside negative.

(1)

(2)

Drugs and solutions. For inside-out patches the electrode contained in (mM) 140 KCl, 1 CaCl₂, 1 MgCl₂, and 10 HEPES (pH 7.2). The bath intracellular solution consisted of (mM) 140 KCl, 0.35-0.9 Ca²⁺, 1 potassium ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 1 MgCl₂, and 10 HEPES (pH 7.2). These solutions were reversed for outside-out patch recordings. The free Ca²⁺ concentrations in the bathing solution were calculated as described previously (8). Potassium-EGTA was held constant at 1 mM, and CaCl₂ was added as calculated. Thus 50 nM Ca²⁺ was obtained by the addition of 0.35 mM CaCl₂, 0.5 µM Ca²⁺ with 0.85 mM CaCl₂, and 0.8 µM Ca²⁺ with 0.9 mM CaCl₂. Physiological solution was used to bathe the cells before experiments.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Cell-attached recordings. Cell-attached recordings demonstrated a clear difference in channel activity, depending on whether the cells were obtained before or after the onset of established labor. Cell-attached recordings from labor uterine myocytes were characterized by an absence of channel activity (n = 7) or displayed spontaneous rhythmic activity (12 of 19 patches). For example, with physiological saline (5 mM K⁺) in the bath and 140 mM K⁺ in the electrode and no applied voltage, oscillations in membrane current on which bursts of action currents were superimposed could be recorded under voltage-clamp conditions (Fig. 1A, top trace). These persisted for the duration of the recording. When cell-attached patches were made from labor myometrium with physiological solution (135 Na⁺, 5 mM K⁺) in the bath and high (140 mM) K⁺ in the electrode, cyclical channel activity was frequently observed; i.e., the cell would undergo periods of intense activity involving several simultaneously active channels characterized by a high average P_o (0.62 ± 0.11, n = 5). This is illustrated in Fig. 1A (bottom trace) at an applied pipette potential of 80 mV, which corresponds to a membrane potential of +40 mV, assuming a mean myometrial cell resting membrane potential of 42 mV for labor cells recorded in the whole cell configuration (data not shown). In marked contrast, cell-attached patches of nonlabor myometrium rarely exhibited spontaneous activity (3 of 29 patches). Single BK_Ca channel activity in nonlabor myometrium was characterized by a low P_o (0.16 ± 0.04, n = 8) in the absence of any applied membrane voltage. However, as pipette potential was made more negative (i.e., depolarized in the cell-attached configuration), channel openings underwent an increase in P_o (0.36 ± 0.08, n = 5 at pipette potential of 80 mV) and conversely on hyperpolarization. Representative single BK_Ca channel activity from a cell-attached patch at a membrane potential of +40 mV (assuming a mean membrane potential of 47 mV for nonlabor cells, data not shown) is illustrated in Fig. 1B. The clearly distinct patterns of channel behavior recorded in cells obtained before and after the onset of labor imply distinct physiological controlling mechanisms operating in the intact cells.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   A: Labor myometrium. Top trace: cell-attached recordings showing spontaneous rhythmic activity with action currents superimposed recorded at rest (0 voltage). Electrode contained 140 mM K+; bathing solution was physiological solution. Bottom trace: cell-attached recording showing cyclical behavior of single K+ channels at a membrane potential of +40 mV. Membrane potential in cell-attached configuration was obtained by assuming a cell resting membrane potential of 42 mV and an applied potential of 80 mV. Open-state probability (Po) for this patch is 0.46. B: representative cell-attached recording from nonlabor myometrium at a membrane potential of +40 mV. Activity is low compared with labor myometrium. Membrane potential in cell-attached configuration was obtained by assuming a cell resting membrane potential of 47 mV and an applied potential of 80 mV. Channel Po is 0.13. Solutions as in A. Arrowheads, closed state of channel; inward current is shown as downward deflections.

Inside-out recordings. The identification of myometrial large-conductance K⁺ channels was verified by increasing the Ca²⁺ concentration from 50 nM to 0.5 µM at the intracellular surface of inside-out patches. Channels were identified as BK_Ca if this procedure resulted in marked enhancement of BK_Ca channel activity or as BK if channel activity remained high on lowering or raising Ca²⁺ concentration. Representative recordings of BK_Ca and BK channel activity are shown in Fig. 2. BK_Ca channels are characterized by little activity at 50 nM Ca²⁺, whereas at 0.5 µM Ca²⁺ there was a substantial increase in activity, and at both Ca²⁺ concentrations this channel type displayed marked voltage dependence, with depolarization causing an increase in activity (Fig. 2A). In contrast, BK channels are characterized by high levels of channel activity independent of membrane voltage (60 to +60 mV) or Ca²⁺ concentration (from 50 nM to 0.5 µM; Fig. 2B). In this study the single channel conductance of BK and BK_Ca channels in the inside-out configuration was 229 ± 13 (n = 7) and 219 ± 16 pS (n = 24), respectively (P > 0.05). These values are in close agreement (221 and 212 pS for BK and BK_Ca, respectively) with those reported previously (8). Furthermore, in accord with our previous study, only BK_Ca channels were detected in nonlabor (n = 34) and BK channels in labor (n = 17) myometrium. To verify the presence of the BK channel in both configurations, patches from labor myocytes were excised from the same cell; the inside-out patch always possessed Ca²⁺- and voltage-independent BK channel activity, whereas the outside-out patch always contained voltage-dependent BK-like channels.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Ca2+ dependence of human myometrial Ca2+-dependent (BKCa) and Ca2+-independent large-conductance K+ (BK) channels. Single channel currents were recorded from inside-out patches excised into symmetrical K+ conditions. A: BKCa channels (nonlabor) recorded in presence of 50 nM Ca2+ (left) and 0.5 µM free Ca2+ (right). Po were as follows: in 50 nM Ca2+, 0.09 at +40 mV, 0.02 at +20 mV, 0.001 at 20 mV, and 0.00 at 40 mV; in 0.5 µM Ca2+, 0.56 at +40 mV, 0.47 at +20 mV, 0.17 at 20 mV, and 0.14 at 40 mV. B: BK channels (labor) recorded in presence of 50 nM Ca2+ (left) and 0.5 µM Ca2+ (right). Po were as follows: in 50 nM Ca2+, 0.76 at +40 mV, 0.68 at +20 mV, 0.73 at 20 mV, and 0.81 at 40 mV; in 0.5 µM Ca2+, 0.67 at +40 mV, 0.58 at +20 mV, 0.71 at 20 mV, and 0.84 at 40 mV. Arrowheads, closed state of channel; inward current is shown as downward deflections.

Outside-out recordings. Recordings from outside-out patches obtained from nonlabor and labor myometrial cells did not display such marked differences between the channel types. In both cases, large-conductance K⁺ channels exhibited strong voltage dependence, such that depolarization resulted in an increase in BK (labor) and BK_Ca (nonlabor) channel activity compared with channel activity at negative membrane potentials (Fig. 3). In this patch configuration, BK and BK_Ca channels are characterized by unitary conductances of 224 ± 9.2 (n = 6) and 217 ± 8.7 pS (n = 9), respectively (Fig. 3B), and are not significantly different (P > 0.05) from the conductances reported from inside-out patch recordings (see above). However, although both channel types display voltage dependence, there was a marked difference in their sensitivity to Ca²⁺ and voltage. This is clearly seen in Fig. 3, A and C, where, at a [Ca²⁺]_i of 50 nM, BK channels recorded from labor tissue are more active at negative voltages, and the increase in activity on depolarization is much more marked than for the corresponding BK_Ca channels from nonlabor tissue. In addition, under this recording configuration, consistently higher numbers of channels were observed in patches obtained from labor myometrium. Analysis of NP_o at this concentration of Ca²⁺ indicated a significant difference between BK and BK_Ca channels at negative and positive membrane potentials. For example, at a membrane potential of 20 mV, mean values for NP_o were 0.31 ± 0.14 (n = 4) and 0.02 ± 0.001 (n = 6) for BK and BK_Ca channels, respectively (P < 0.05), and at +20 mV the corresponding NP_o values were significantly greater for BK channels (1.47 ± 0.34, n = 4) than for BK_Ca channels (0.16 ± 0.03, n = 6, P < 0.05).

View larger version (32K): [in this window] [in a new window]   Fig. 3.   BK and BKCa channel activity recorded from outside-out patches. A: representative records of BKCa (left) and BK (right) channel activity in presence of 50 nM Ca2+. Arrowheads, closed state of channel; inward current is shown as downward deflections. B: single channel current-voltage (I-V) relationships for BKCa () and BK () channels. Conductances of these channels determined from slope of line fitted to data by linear regression were 207 and 214 pS, respectively. C: channel activity (NPo)-V relationship. Activity increased dramatically after patch depolarization for BK channel () compared with BKCa channel ().

Effects of intracellular Ba²⁺. Application of Ba²⁺ (2-10 mM) to the intracellular aspect of inside-out membrane patches obtained from nonlabor myometrial cells (n = 7) resulted in blockade of BK_Ca channel activity at depolarized membrane potentials (Fig. 4A) characterized by the appearance of long-lived closed periods of several seconds duration. The block by Ba²⁺ is strongly voltage dependent, with only brief channel openings at positive potentials, whereas considerable channel activity is apparent on hyperpolarization, although this is accompanied by a slight, but insignificant, reduction in unitary current amplitude at all voltages in the presence of 10 mM Ba²⁺ (Fig. 4B). A plot of P_o vs. voltage in the absence and presence of 10 mM Ba²⁺ shows that Ba²⁺ caused an inversion of this relationship (Fig. 4C). Thus few channel openings occurred at +50 mV in the presence of Ba²⁺ [P_o = 0.02 ± 0.002 (n = 4) compared with P_o = 0.42 ± 0.33 in the absence of Ba²⁺], but P_o increased as the patch was hyperpolarized to 50 mV [P_o = 0.33 ± 0.27 (n = 4) compared with P_o = 0.04 ± 0.002 (n = 4) in the absence of Ba²⁺]. The effects of Ba²⁺ were only partially reversible on washing (data not shown). Although 2 mM Ba²⁺ caused a reduction in channel P_o, it did not result in the inversion of the P_o-voltage relationship (n = 3; data not shown). BK channels recorded from inside-out patches obtained from labor myometrium were much less sensitive to block by internal Ba²⁺. Neither single channel amplitude nor P_o of the BK channel was affected by 5 mM Ba²⁺ (n = 5); however, at 10 mM Ba²⁺, depression of unitary current amplitude (n = 3) was evident (Fig. 5A). This was observed only at depolarized potentials where inward rectification was induced, characterized by a reduction in single channel conductance (over the range 0 to +50 mV) from 233 pS in the absence of Ba²⁺ to 136 pS in the presence of 10 mM Ba²⁺ (Fig. 5B). In addition, 10 mM Ba²⁺ induced a flickery block of BK channel activity at depolarized potentials (Fig. 5A), and this was associated with a reduction in P_o, whereas there was no significant effect of 10 mM Ba²⁺ at negative membrane potential (Fig. 5C). The effects of external Ba²⁺ were not studied on either channel.

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Effects of intracellular Ba2+ on single BKCa currents from nonlabor myocytes. A: representative record of BKCa channel activity recorded from an inside-out patch in 50 nM Ca2+; channel activity increased as membrane potential was made positive. Po were as follows: 0.34 at +40 mV, 0.23 at +20 mV, 0.05 at 20 mV, and 0.02 at 40 mV. Ba2+ (5 mM) blocked BKCa channels in a voltage-dependent manner. Note long closed states at positive voltages in presence of Ba2+. Po were as follows: 0.03 at +40 mV, 0.06 at +20 mV, 0.14 at 20 mV, and 0.29 at 40 mV. Data shown without and with Ba2+ were recorded from the same patch. Arrowheads, closed state of channel; inward current is shown as downward deflections. B: single channel I-V relationship for BKCa channel of pregnant human uterine myocytes. Note reduction in single channel current with 10 mM Ba2+ () compared with control () recorded from the same patch. C: Po-V relationship in presence of 50 nM intracellular Ca2+ for BKCa with () and without 10 mM Ba2+ ().

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Effects of Ba2+ on BK channels from labor myocytes. A: representative records of BK channel activity from an inside-out patch in presence of 50 nM Ca2+ without (top) and with (bottom) 10 mM Ba2+. Po were as follows: 0.74 at +40 mV and 0.64 at 40 mV without Ba2+ (control) and 0.27 at +40 mV and 0.61 at 40 mV with Ba2+. Arrowheads, closed state of channel; inward current is shown as downward deflections. B: single channel I-V relationship for BK channel shows that 10 mM Ba2+ () decreased unitary current amplitude at positive but not negative voltages compared with control (). C: Po-V relationship for BK channel without Ba2+ () and with 5 mM () and 10 mM Ba2+ ().

4-AP sensitivity of BK and BK_Ca channels. In inside-out patch recordings from nonlabor myometrial cells, with 0.5 µM Ca²⁺ in the bathing solution to maintain a high level of channel activity, application of 0.1 mM 4-AP was without effect on BK_Ca current amplitude (data not shown) but slightly decreased P_o (n = 5; Fig. 6A). Raising the concentration of 4-AP to 1 mM resulted in a significant reduction in the P_o of the BK_Ca channel at all voltages examined (n = 4 of 7; Fig. 6A). In contrast, 4-AP (0.1-1 mM) applied directly to the intracellular aspect of patches excised from labor myometrial cells had no effect on the single channel current amplitude or P_o of BK channels (n = 4; Fig. 6B).

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Effects of 4-aminopyridine (4-AP) on BKCa and BK channels. A, top: representative records of BKCa channel activity from an inside-out patch at a membrane potential of +50 mV with 0.5 µM Ca2+. Top trace, control data (Po = 0.61); bottom trace, with 1 mM 4-AP (Po = 0.14). 4-AP (1 mM) had no effect on single channel amplitude. A, bottom: Po-V relationship with 0.1 () and 1 mM () 4-AP compared with control (). B, top: representative records of BK channel activity from an inside-out patch under conditions in A. 4-AP (1 mM) had no effect on BK channel current amplitude or Po in labor myometrium. Po were as follows: 0.87 and 0.84 for control and 4-AP, respectively. B, bottom: Po-V relationship with 1 mM 4-AP () compared with control (). Arrowheads, closed state of channel; inward current is shown as downward deflections.

Effect of intracellular TEA. Figure 7 illustrates the block of BK_Ca channels when TEA was applied to the cytoplasmic aspect of inside-out patches in the presence of 50 nM Ca²⁺ from nonlabor myometrium. This ion had little effect at concentrations between 1 and 5 mM on BK_Ca channels (n = 10; data not shown), but 10-100 mM TEA (n = 4) decreased unitary current amplitude of BK_Ca channels in a concentration-dependent manner (Fig. 7A). The block was voltage dependent and resulted in the appearance of inward rectification of the current-voltage relationship at voltages positive to +20 mV. There was no effect of TEA on single channel current amplitude at hyperpolarized voltages. A plot of i_TEA/i_C vs. [TEA] resulted in a K_d of 45.7 ± 2.4 mM (Fig. 7B). Analysis, using the Woodhull (29) method, of the reduction in current amplitude by TEA of the BK_Ca channel (Fig. 7C) resulted in values of 64.8 ± 3.6 mM (n = 3) for K_{d(0 mV)} and 0.27 ± 0.03 (n = 3) for , suggesting that the TEA binding site experiences 27% of the electric field and so is not strongly voltage dependent. An additional feature of TEA action on BK_Ca channels was observed at high (>20 mM) concentrations, where a noticeable increase in channel activity (Fig. 7A) was accompanied by an increase in the P_o (n = 4); however, this was not investigated further in this study.

View larger version (36K): [in this window] [in a new window]   Fig. 7.   Intracellular tetraethylammonium (TEA) sensitivity of BKCa channel currents. A: representative single channel current records of BKCa channel activity from an inside-out patch at a membrane potential of +40 mV in presence of 50 nM intracellular Ca2+. Actions of an ascending series of TEA concentrations are shown. Arrowheads, closed state of channel; inward current is shown as downward deflections. B: relationship between percent inhibition of single channel current amplitude (iTEA/ic, where iTEA and ic are single channel currents with and without TEA, respectively) and concentration of applied TEA ([TEA]). Values are means ± SE of 3-5 separate patches. Estimated Kd from this analysis is 46 mM. C: ln(i/iTEA  1)-V relationship at 50 mM intracellular TEA for 3 separate inside-out patches. Slope gives a value of 0.27 for location of blocking site within membrane electric field (), and y-intercept gives a value of 64.8 mM for Kd at 0 mV [Kd(0 mV)].

View larger version (31K): [in this window] [in a new window]   Fig. 8.   Intracellular TEA sensitivity of BK channel currents. A: representative single channel current records of BK channel activity from an inside-out patch at a membrane potential of +40 mV in presence of 50 nM intracellular Ca2+. Actions of an ascending series of TEA concentrations are shown. Arrowheads, closed state of channel; inward current is shown as downward deflections. B: relationship between percent inhibition of single channel current amplitude and concentration of applied TEA. Values are means ± SE from 3 separate patches; where no error bars are apparent, SE is within symbol. Estimated Kd from this analysis is 53 µM. C: ln(i/iTEA  1)-V relationship with 100 µM intracellular TEA. Slope gives a value of 0.013 for , and y-intercept gives a value of 163 µM for Kd(0 mV).

Effect of extracellular TEA. TEA inhibited current flow through BK_Ca (n = 9; Fig. 9A) and BK (n = 6; Fig. 9B) channels in outside-out patches from nonlabor and labor myometrial cells, respectively, at much lower concentrations (100 µM-5 mM) than those required to block BK_Ca channel activity from the internal membrane surface. The inhibition produced by extracellular TEA was flickery in nature and accompanied by increased channel noise in the open state and, therefore, typical of fast block, which is observed as a reduction in unitary current. The block was not strongly voltage dependent and was fully reversible. Figure 9C shows fractional inhibition of channel activity for BK_Ca and BK channels in the presence of TEA plotted as a function of the extracellular [TEA]. It is clear from these data that the TEA block of both channel types is identical. The estimated K_d values for BK_Ca and BK channels were 423.0 ± 65.1 and 395.3 ± 58.4 µM (P > 0.05), respectively, obtained by fitting the curves to Eq. 2.

View larger version (38K): [in this window] [in a new window]   Fig. 9.   Extracellular TEA sensitivity of BKCa and BK channel currents. Single channel current records are representative of outside-out patches obtained at a membrane potential of +30 mV and with 50 nM Ca2+ bathing intracellular aspect of membrane. A: recordings of BKCa channel activity without TEA (control, top trace; Po = 0.17), with 100 µM TEA (middle trace; Po = 0.15), and with 1 mM TEA (bottom trace; Po = 0.09). B: recordings of BK channel activity without TEA (control, top trace; Po = 0.41), with 100 µM TEA (middle trace; Po = 0.47), and with 1 mM TEA (bottom trace; Po = 0.36). Arrowheads, closed state of channel; inward current is shown as downward deflections. C: relationship between percent inhibition of single channel current amplitude and concentration of applied TEA for BKCa () and BK () channels. Values are means ± SE from 4 separate patches; where no error bars are apparent, SE is within symbol. Estimated Kd from this analysis are 423 and 395 µM for BKCa and BK channels, respectively.

Effect of ball peptide. A synthetic peptide of 20 amino acids found at the amino terminus of the Shaker B K⁺ channel causes block of BK_Ca channels when applied intracellularly to inside-out patches of rat brain (6) and pig coronary smooth muscle (23). In view of the observed differences in TEA and Ca²⁺ sensitivity between the two channel types described in this study, the effect of the synthetic 20-mer ball peptide was tested, using inside-out patches, on BK and BK_Ca channel activity. Application of 50 µM ball peptide inhibited the activity of single BK_Ca channels, and this was observed as a reduction in channel P_o (Table 1). An analysis of the channel dwell times demonstrated that the channel's closed state was best fitted by two exponentials. The mean closed times were significantly increased from 4.63 ± 1.41 to 9.98 ± 1.73 ms at +30 mV (n = 3, P < 0.05) and from 14.55 ± 3.31 to 24.65 ± 3.48 ms at 30 mV (P < 0.05). In some recordings of longer duration, closures of BK_Ca channels lasting several seconds were observed, which is in agreement with the findings of Toro et al. (23) and Foster et al. (6) for this peptide on smooth and skeletal muscle BK_Ca channels, respectively. A greater degree of blockade of BK_Ca channels by ball peptide was observed on depolarization than on hyperpolarization, confirming the voltage-dependent nature of the block. Ball peptide had no effect on the single channel amplitude, P_o, or closed times of the BK channel of labor myometrium (n = 2; data not shown).

Effect of ChTX. ChTX (50-100 nM) applied to the external aspect of outside-out patches (n = 9) from nonlabor and labor myometrial cells reduced BK_Ca and BK channel openings, respectively (Table 2), whereas single channel current amplitude remained essentially unaltered (data not shown). The block did not exhibit voltage dependence for either channel type. ChTX (100 nM) had no effect when applied to the intracellular aspect of an inside-out patch having BK (n = 2) or BK_Ca (n = 2) channels.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The observed changes in the Ca²⁺ and voltage sensitivity in addition to the altered sensitivity to internal Ba²⁺, TEA, ball peptide, and 4-AP, but not external TEA and ChTX, indicate modifications to the intracellular control mechanisms of human myometrial BK_Ca and BK channels during pregnancy and labor. The voltage-dependent reduction in BK_Ca channel P_o at positive membrane potentials by internal Ba²⁺ has been reported for smooth muscle (2, 15) and neuronal (21) BK_Ca channels. It has been suggested that Ba²⁺ blocks BK_Ca channels by a slow block mechanism (17), where Ba²⁺ entering the channel pore bind to a site located 80-95% of the way through the membrane (2, 17, 25). The inversion of the P_o-voltage relationship of the myometrial BK_Ca channel leads us to suggest that an additional mechanism may be operating in which Ba²⁺ may substitute for Ca²⁺ in promoting BK_Ca channel activity. However, previous studies suggest that Ba²⁺ cannot mimic Ca²⁺ in activating the channel (25), and this appears to be true for the BK channel, where a small reduction in unitary current amplitude was apparent.

The marked contrast in 4-AP sensitivity between BK_Ca (nonlabor tissue) and BK (labor) channels may indicate some variation in the structure of these large-conductance myometrial K⁺ channels in the region of the cytoplasmic aspect of S6, since studies on cloned Kv channels and mutants have led to the suggestion that 4-AP likely associates with regions of the putative cytoplasmic ends of transmembrane segments S5 and S6, with the intracellular portion of S6 providing the binding site (10). Alternatively, 4-AP no longer has access to its blocking site on the -subunit of the channel, thereby accounting for its lack of effect on the BK channel and suggesting a change in channel architecture associated with labor.

The K_d values for internal TEA inhibition of the BK_Ca channel are 20-70 mM compared with 100-300 µM for external TEA (13) and are mediated by clearly distinct external and internal TEA binding sites. The characteristics of external TEA block of both channels in human myometrium are similar to those characterized in vascular myocytes (12), chromaffin cells (30), and skeletal muscle (3, 26), in which the flickering associated with the open state represents rapid association and dissociation of TEA from its binding site.

The values obtained for block by internal TEA, K_{d(0 mV)} of 65 mM and  of 0.27 for the myometrial BK_Ca channel, are in close agreement with those for BK_Ca channels of cultured rat muscle [K_{d(0 mV)} = 60 mM,  = 0.26 (3)] and chromaffin cells [K_{d(0 mV)} = 24 mM,  = 0.10 (30)] and imply that TEA sensitivity is conserved among certain BK_Ca channels. The mechanism of internal TEA block of BK_Ca channels is consistent with the "quiet" fast block of these channels in rat skeletal muscle (26). In contrast, the K_d of 53 µM and K_{d(0 mV)} of 163 µM for internal TEA blockade of the BK channel of human labor myometrial cells are closer to values obtained for BK_Ca channels of cerebral artery smooth muscle (0.83 mM) (27), rat synaptosomes (0.80 mM) (5), and clonal anterior pituitary cells (0.08 mM) (28). Furthermore, the voltage-independent nature of the block of BK channels ( = 0.013) indicates that this TEA binding site is located outside the membrane electric field. The observed block of BK_Ca by ball peptide is in accordance with findings supporting the existence of discrete yet distinct receptor sites, which bind this peptide to produce short or long blocks (6, 23). However, the apparent enhanced TEA sensitivity and insensitivity to ball peptide of the BK channel points to the involvement of alternative mechanisms by which this channel senses intracellular cationic blockers.

In view of the clear pharmacological and physiological differences between the BK_Ca and BK channel described in the present study, we can only speculate as to the nature of this variability. It is possible that association and dissociation of regulatory proteins, e.g., -subunits (11), which can alter the Ca²⁺ sensitivity of the BK_Ca channel when expressed with the -subunit (16), maybe strong candidates in altering channel properties. Changes in phosphorylation states may be possible alternative mechanisms whereby BK_Ca channel behavior can be profoundly altered, as demonstrated in neurons (14) and cloned channels (4); indeed, in the myometrium, protein kinase A can inhibit or increase the activity of the BK_Ca channel, depending on whether the tissue is from a pregnant or a nonpregnant source (19).

In the third trimester of human pregnancy the maternal uterine physiology begins to change to prepare itself for childbirth. Inasmuch as myometrial contractions are Ca²⁺ mediated, in view of the extremely high sensitivity of the uterine BK_Ca channel to [Ca²⁺]_i (8), any rise in [Ca²⁺]_i would subdue excitability by the activation of the BK_Ca channel conductance. However, if the Ca²⁺ and voltage dependence of the BK_Ca channel were uncoupled (i.e., BK channel), the myometrium could still undergo relaxation but would be free from [Ca²⁺]_i changes, thereby effectively dissociating BK_Ca channel activation and [Ca²⁺]_i. A high level of BK channel activity would tend to hyperpolarize the cell, making it less excitable. During labor, however, the uterus is required to undergo periods of intense relaxation followed by powerful contractions. The likelihood that the BK channel by virtue of its large conductance and Ca²⁺ insensitivity may fulfill this role by providing a strong hyperpolarizing influence contributing to the relaxation phases cannot be excluded. This is an exciting phenomenon, since the BK channel does not appear to be directly controlled by changes in [Ca²⁺]_i once labor has commenced, as seen in cell-attached patches, but by some unknown intrinsic factors. The switch in myometrial K⁺ channel properties, closely associated with the progression to labor, is a dramatic and clear illustration of the dynamic nature of ion channel regulation in response to physiological adaptations.