Phenotypic alteration of a human BK (hSlo) channel by hSlobeta subunit coexpression: changes in blocker sensitivity, activation/relaxation and inactivation kinetics, and protein kinase A modulation

Abstract

A human homolog of the large-conductance calcium-activated potassium (BK) channel beta subunit (hSlobeta) was cloned, and its effects on a human BK channel (hSlo) phenotype are reported. Coexpression of hSlo and hSlobeta, in both oocytes and human embryonic kidney 293 cells, resulted in increased Ca2+ sensitivity, marked slowing of BK channel activation and relaxation, and significant reduction in slow inactivation. In addition, coexpression changed the pharmacology of the BK channel phenotype: hSlo-mediated currents in oocytes were more sensitive to the peptide toxin iberiotoxin than were hSlo + hSlobeta currents, and the potency of blockade by the alkaloid BK blocker tetrandrine was much greater on hSlo + hSlobeta- mediated currents compared with hSlo currents alone. No significant differences in the response to charybdotoxin or the BK channel opener NS1619 were observed. Modulation of BK channel activity by phosphorylation was also affected by the presence of the hSlobeta subunit. Application of cAMP-dependent protein kinase increased P(OPEN) of hSlo channels, but decreased P(OPEN)of most hSlo + hSlobeta channels. Taken together, these altered characteristics may explain some of the wide diversity of BK channel phenotypes observed in native tissues.

Fig. 1.

Predicted primary sequence of a cloned human BK channel β subunit (hSloβ) and its alignment with a bovine β subunit. The identical amino acids are boxedtogether, resulting in 85% identity between these sequences with the putative M2 transmembrane domain being completely conserved. The putative transmembrane domains M1 and M2 are marked within the consensus sequence by the blackboxes.

Fig. 2.

Activation records for hSlo andhSlo + hSloβ BK channels in oocytes and HEK 293 cells. Activation of hSlo + hSloβ BK currents was significantly slower than currents mediated by hSloexpression in Xenopus oocytes (A) and HEK 293 cells (B; whole-cell patch recordings, C; ensemble average currents from excised membrane patches, D; excised patch recordings, single traces). Currents were normalized for presentation. Peak current values in A, 6.4 μA for hSlo and 4.6 μA for hSlo + hSloβ; currents shown are in response to a depolarizing step from −60 to 80 mV. In B, 17.1 nA for hSloand 11.7 nA for hSlo + hSloβ; currents shown resulted from voltage steps from −60 to 100 mV. In C, 114 pA for hSlo and 355 pA for hSlo + hSloβ. Horizontal axis scale: 20 msec for A andB, 40 msec for C. The currents in Cand D resulted from voltage steps from −60 to 40 mV. The pulse duration for the multichannel recordings in D is 200 msec; note that the patches contain different (and large) numbers of channels; initial amplitude (i.) is 110 pA. No significant difference was observed in the amplitudes of hSlo orhSlo + hSloβ currents in either expression system after comparable incubation periods; likewise, no difference was observed in the single-channel slope conductance of channels coded by these constructs when currents from patches with a small number of channels were recorded (hSlo = 286 ± 12 pS,n = 5; hSlo + hSloβ = 289 ± 6 pS, n = 4). Note that deactivation ofhSlo + hSloβ currents was also slower, as indicated in A and visible in C and D(see Table 1).

Fig. 3.

Slow current inactivation reduction in oocytes expressing hSlo + hSloβ relative to those expressing hSlo. A, In a paired-pulse paradigm, to examine the rate of recovery from inactivation, a 1 sec voltage step (−60 mV hold to 100 mV) was used as the “conditioning” stimulus to produce a significant level of inactivation (time 0, end of the conditioning pulse), followed at an increasing interval by a single identical “test” voltage step. Initial inactivation was measured by comparing the early peak current of the conditioning voltage step with the residual current at the end of the first step. Peak current amplitudes of the subsequent step in each paired episode were used to measure recovery. The conditioning step produced significantly greater levels of inactivation of hSlo currents compared with expressing hSlo + hSloβ currents (ttest; p < 0.001), and recovery was significantly slower (repeated measures ANOVA, F = 15.9, p = 0.005). Full recovery was not achieved by 7.5 sec after the conditioning pulse. With longer single voltage steps (10 sec), a significant increase in the time course of inactivation was observed (hSlo τ₁ = 309.7 ± 36.2 msec, n = 5; hSlo + hSloβ τ₁ = 702.0 ± 88.2 msec, n = 5; p = 0.003, two-tailed t test; τ₂ values not reported because of contamination by slowly activating native current). B, Examples of current inactivation and its recovery resulting from the paired-pulse paradigm (8 sweeps) in oocytes expressing hSlo or hSlo + hSloβ. The level of inactivation was independent of current amplitude (expression level) within the normal limits encountered in this study. There was no inactivation of native Ca²⁺-activated Cl⁻current, and currents represent IbTX-sensitive current components, after subtraction of residual currents in supramaximal IbTX.

Fig. 4.

Coexpression of hSlo + hSloβ increased the sensitivity of BK channels to intracellular Ca²⁺. A, Conductance (g/g_max)/voltage (g/V) plots generated for hSlo (i.) and hSlo + hSloβ (ii.) in 28.2 μmCa²⁺ (a) and 0.79 μm Ca²⁺ (b); recordings obtained from inside-out excised patches from transiently transfected HEK 293 cells in response to voltage ramps (−100 to 100 mV, 4 sec duration; data from a minimum of 25 ramps per patch). Note that at both concentrations of intracellular Ca²⁺, the g/V relationship forhSlo + hSloβ is shifted to the left, indicating increased Ca²⁺ sensitivity. Curves were generated with a standard Boltzmann relationship in whichg/g_max = (1 +exp[(V_1/2 − V_m)/K])⁻¹.B, The half-maximal activation values (V_1/2) for three concentrations of intracellular Ca²⁺ are plotted forhSlo and hSlo + hSloβ;V_1/2 values were consistently lower (as plotted) for hSlo coexpressed with the hSloβ subunit.

Fig. 5.

IBTX and ChTX pharmacology. A, Application of the BK channel-blocking peptide IbTX to oocytes expressing hSlo or hSlo + hSloβ revealed that coexpression with the hSloβ subunit resulted in a nearly 10-fold decrease in the sensitivity to IbTX blockade;n = 5–10 oocytes/IbTX concentration. Maximal effect was defined as the effect produced by incubation in 100–250 nm IbTX; maximal levels of suppression did not differ significantly between hSlo and hSlo + hSloβ. B, Application of the peptidyl blocker ChTX did not reveal a significantly different profile of blockade forhSlo and hSlo + hSloβ. Maximal effect was defined as the response to 250–500 nmChTX, and maximal levels of effect did not differ betweenhSlo and hSlo + hSloβ.

Fig. 6.

Tetrandrine pharmacology in oocytes and HEK 293 cells. A, Suppression of hSlo-mediated whole-cell membrane currents in HEK 293 cells by the alkaloid type II BK blocker tetrandrine was reduced relative to the degree of block observed in cells transfected with hSlo + hSloβ;n = 5–9 cells/tetrandrine concentration. The fitted curves suggest multiple affinity interactions of tetrandrine with both constructs; absolute maximal levels of tetrandrine block could not be determined because of limited solubility. The traces are examples of the relative effects of 2.5 μmtetrandrine on whole-cell HEK 293 BK membrane currents. Note that tetrandrine slowed activation of both types of current, but this effect was most pronounced with a greater level of block produced in cells expressing hSlo + hSloβ. B, A pronounced increase in the latency to current peak was observed only in oocytes expressing hSlo and hSlo + hSloβ when exposed to tetrandrine. Tracesdepict the effect of 50 μm tetrandrine (b.) on BK currents in oocytes expressing hSlo(top traces; i.), or hSlo + hSloβ (bottom traces; ii.), relative to control values (a.). The dotted trace in the bottom setis the control hSlo trace for comparison. Tetrandrine had no effect on the current activation in the hSlo oocyte, but it significantly delayed the peak of the current in the oocyte expressing hSlo + hSloβ (EC₅₀ = 8.5 μm). The currents are the response to voltage steps from a holding potential of −60 to 100 mV, and only the initial portions of the normalized currents are depicted.

Fig. 7.

Effects of the application of the BK channel opener NS1619. At a range of NS1619 concentrations, oocytes injected with either hSlo or hSlo + hSloβ cRNA showed similar levels of activation of whole-cell BK currents. Measured currents represent the response to a voltage step from −60 to 140 mV. A minimum of five oocytes were used for each concentration of NS1619.

Fig. 8.

Modulation of hSlo and hSlo + hSloβ channels by exposure to PKA. A, Application of the catalytic subunit of PKA produced an increase inNP_open when applied to the cytosolic side of an inside-out membrane patch excised from an HEK 293 cell transiently expressing hSlo. B, PKA reducedNP_open in a patch excised from an HEK 293 cell transiently coexpressing hSlo + hSloβ. The mean increase in NP_open (averaged over the entire 240 sec control period vs the 240 sec PKA application) forhSlo patches was 97.1 ± 35.9% (SEM all) above control values (n = 6; increases seen in all 6 patches); the mean decrease in NP_open recorded fromhSlo + hSloβ patches was −19.0 ± 15.6% (n = 8; 5 decreased, 1 no change, 2 increased, all data included in average; the difference was significant, p < 0.03, two-tailed t test). Partial recovery after removal of PKA was observed in some patches during a 240 sec wash period (data not shown).