The smooth muscle-type β₁ subunit potentiates activation by DiBAC₄(3) in recombinant BK channels

Abstract

Large-conductance Ca²⁺-activated (BK) channels, expressed in a variety of tissues, play a fundamental role in regulating and maintaining arterial tone. We recently demonstrated that the slow voltage indicator DiBAC₄(3) does not depend, as initially proposed, on the β₁ or β₄ subunits to activate native arterial smooth muscle BK channels. Using recombinant mslo BK channels, we now show that the β₁ subunit is not essential to this activation but exerts a large potentiating effect. DiBAC₄(3) promotes concentration-dependent activation of BK channels and slows deactivation kinetics, changes that are independent of Ca²⁺. K_d values for BK channel activation by DiBAC₄(3) in 0 mM Ca²⁺ are approximately 20 μM (α) and 5 μM (α+β₁), and G-V curves shift up to −40mV and −110 mV, respectively. β₁ to β₂ mutations R11A and C18E do not interfere with the potentiating effect of the subunit. Our findings should help refine the role of the β₁ subunit in cardiovascular pharmacology.

Introduction

BK channels, part of the voltage-activated potassium channel family, can be activated by membrane depolarization and/or intracellular Ca²⁺ increases. In smooth muscle as well as several other tissues, the opening of BK channels provides a negative feedback mechanism that opposes cell excitation events.^1-3 BK channel exist, in most tissues, as hetero-octomers formed by 2 types of subunits: α and β subunits. The regulatory β subunit, first described in smooth muscle, comprises 2 transmembrane domains (TM1-TM2) linked by an extracellular glycosylated loop.⁴ Four different subtypes of β subunits (β1 to β4) with distinctive tissue-specific expression patterns have been identified in mammals; these correspond to the products of 4 different genes, KCNMB1 to KCNMB4.^5-7 A recently described type of auxiliary subunit, named γ, belongs to the family of leucine-reach repeat-containing protein 26.⁸ We did not find evidence for this subunit in aortic smooth muscle cDNA.⁹ The α subunit comprises 7 transmembrane regions (S0-S6) and a large cytosolic region.¹⁰ This subunit and its multiple alternative splice variants are encoded by one gene, Slo/KCNMA1.^11-14

The transmembrane regions of the α subunit contain the pore gate domain (PGD) that expands between S5–S6 segments and the voltage-sensor domain (VSD) that, presumably, includes S0–S4 segments.^15,16 The large C-terminal cytoplasmic domain (CTD) comprises approximately two-thirds of the protein. CTD contains a tandem of 2 regions homologous to the regulator of K⁺ (RCK1 and RCK2) conductance domain; this forms the gating ring of the channel and involves Ca²⁺ binding sites.^17,18 The PGD, VGD, and CTD modules are proposed (Horrigan and Aldrich, HA model)^19-21 to interact allosterically to activate the channel, consistent with the modular nature of BK channels.^15,22 Specifically, structural rearrangements are transmitted to the PGD upon Ca²⁺ binding to the gating ring of the channel or activation of the voltage sensors, or both, thereby increasing the probability of opening the gate.^15,21,22 These allosteric components (Ca²⁺-dependent, voltage-dependent, or intrinsic gating) are potential effectors of BK channel openers.

Intensive research is being devoted to the identification of pharmacological BK channel openers.^23-28 In particular, a channel-opening compound with subunit specificity is relevant for cardiovascular physiology, where the β₁ subunit is predominant and acts as a molecular tuner critical to vasoregulation.²⁹ The importance of subunit specificity of channel activators extends to other fields, as well, because of the tissue-specific distribution of β subunits (i.e., β₄ in nervous system). The slow voltage indicator DiBAC₄(3) was found to have BK channel-specific effects. Moreover, these activating effects on recombinant BK channels were described to be specific only for the β₁ and β₄ subunit.³⁰ We recently demonstrated, however, that DiBAC₄(3) does not depend, as initially proposed, on the β₁ or β₄ subunits to activate native arterial smooth muscle BK channels.^9,30 In this addendum, we extended our studies to investigate the requirement of the smooth muscle-type β1 subunit to activate recombinant BK channels by DiBAC₄(3). We report that the β₁ subunit is not essential to this activation but exerts a large potentiating effect. These findings provide a foundation for the development of BK channel-mediated vasoregulators and for the dissection of intricate molecular mechanisms involved in BK channel modulation.

Results

DiBAC₄(3) can activate α alone BK channels and markedly activate α+β₁ BK channels

In our previous paper, we demonstrated that native single BK channels lacking β₁ (or β4) subunits were readily activated by DiBAC₄(3). Using the heterologous expression of recombinant BK channels in HEK cells, we analyzed the effect of DiBAC₄(3) on macroscopic BK currents from excised patches in the absence of Ca²⁺. In α alone BK channels, 10 μM DiBAC₄(3) produced an increase in the size of steady-state ionic currents and slowed the deactivation of the current when the channels were stepped back to −120 mV (Fig. 1A, left). For comparison, a steady-state current-voltage relationship was plotted in the absence and in the presence of 10 μM DiBAC₄(3); the dye shifted the threshold for current activation from 80 mV to 50 mV (Fig. 1B, left). The increase in current together with the slowing of current deactivation kinetics appear to be the most salient features for the effect of DiBAC₄(3) on recombinant BK channels. These 2 characteristic effects of DiBAC₄(3) were enhanced in the presence of the β₁ subunit. Currents from α+β₁ BK channels were recorded using solutions identical to those used for α alone and using a similar voltage protocol, except for a longer duration of pulses to allow the channels to reach steady-state. These currents showed typical slower kinetics, a signature of β₁ modulation of BK channels, and were enhanced ~3-fold, at steady-state, by the application of 10 μM DiBAC₄(3) (Fig. 1A, right). Indeed, DiBAC₄(3) displaced the threshold for current activation from 95 mV to 15 mV (Fig. 1B). The dramatic effect on deactivation kinetics can be observed from the traces.

Figure 1. DiBAC₄ (3) activates α alone and α+β₁ BK channels in 0 mM Ca²⁺. (A) Representative traces of macroscopic currents recorded in 0 mM Ca²⁺, elicited by the corresponding voltage protocol (top insets). Traces are plotted every 10 mV for clarity purposes. Macro patches were excised from HEK cells expressing α alone (left) or α+β₁ channels (right), in the absence (control) or presence of 10 µM DiBAC₄(3). Excised inside-out patches were bathed in symmetrical (140 mM) K⁺ solutions with 0 mM Ca²⁺. (B) Corresponding steady-state current vs. voltage. Symbols represent the average current from the last 15 ms of the activating step.

DiBAC₄(3) shifts the voltage-dependent activation of BK channels, and slows current deactivation, in a concentration-dependent manner

We performed a series of similar experiments to examine the concentration-dependent characteristics of DiBAC₄(3) effects (Fig. 2 and Table 1). Tail currents at −120 mV, normalized to the maximum tail current at 200 μM Ca²⁺, were used to obtain conductance vs. voltage (G-V) relationships. Grouped G-V relationships obtained from α alone channels (9 to 16 independent experiments) showed that increasing DiBAC₄(3) concentrations promoted a leftward shift in the G-V relationship in a concentration-dependent manner (Fig. 2A, left). Further, this effect was more pronounced in the presence of the β₁ subunit (Fig. 2A, right), investigated using α+β₁ channels (5 to 8 experiments).

Figure 2. Concentration-dependent effects of DiBAC₄(3) on α alone and α+β₁ BK channels. (A) Grouped G-V relationship at increasing concentrations of DiBAC₄(3). G-V curves were obtained from instantaneous tail currents at –120 mV measured 200 μs after the end of the preceding activating step. Currents were recorded as shown in Figure 1. Data points correspond to the average of 10–16 experiments for α alone channels, and from 5–8 experiments for α+β₁ channels. (B) Plot of V_1/2 vs. DiBAC₄(3) concentration. Symbols represent the averaged V_1/2 values shown in Table 1, and were obtained from individual Boltzmann fittings of the experiments used in (A). Data were fitted to a Hill equation (continuous line). (C) Plot of deactivation time constant vs. DiBAC₄(3) concentration. Symbols represent averaged deactivation time constant values obtained from single exponential fits of the current relaxation at –120 mV, after an activating step to 200 mV. Data were fitted to a Hill equation (continuous line). Insets show representative tail currents before (black) and after 10 μM DiBAC₄(3) (gray) (D) Voltage dependence of current deactivation time constants at large negative potentials. Data points were obtained from single exponential fits of deactivation kinetics at several voltages (from –65 to –190 mV) after an activating step to +90 mv (α alone channels) or to +150 mV (α+β₁ channels). Currents were recorded in the absence (control) or in the presence of 30 µM DiBAC₄ (3) for both, α alone or α+β₁ channels. Continuous lines represent single exponential fittings to the data. (E) Plot of activation time constant vs. DiBAC₄(3) concentration. Symbols represent averaged activation time constants obtained from single exponential fits of current activation at 200 mV from a holding potential of –120 mV. Data are expressed as mean ± s.e.

Table 1. DiBAC₄(3) concentration-dependent effects on BK channels.

	α				α+β1
DiBAC4 (3)	V1/2	Q	τActivation	τDeactivation	V1/2	Q	τActivation	τDeactivation
0 μM	167 ± 2	1.16 ± 0.04	2.53 ± 0.33	0.15 ± 0.01	180 ± 5	0.87 ± 0.14	21.80 ± 3.92	0.55 ± 0.09
0.12 μM	166 ± 1	1.09 ± 0.04	2.37 ± 0.34	0.17 ± 0.01	163 ± 8	0.82 ± 0.16	29.31 ± 7.38	0.76 ± 0.05
0.37 μM	163 ± 2	1.16 ± 0.08	3.08 ± 0.94	0.17 ± 0.01	158 ± 4	0.72 ± 0.10	25.76 ± 3.02	0.97 ± 0.16
1.1 μM	160 ± 3	1.13 ± 0.08	2.69 ± 0.51	0.18 ± 0.01	151 ± 8	0.68 ± 0.11	25.14 ± 4.30	1.26 ± 0.14
3.33 μM	156 ± 3	1.11 ± 0.07	3.08 ± 0.94	0.25 ± 0.02	124 ± 8	0.71 ± 0.07	23.70 ± 5.41	2.03 ± 0.06
10 μM	141 ± 4	1.01 ± 0.03	2.64 ± 0.64	0.45 ± 0.08	90 ± 11	0.83 ± 0.28	22.50 ± 5.33	5.02 ± 0.71
30 μM	123 ± 8	0.95 ± 0.03	2.81 ± 0.70	0.67 ± 0.10	69 ± 6	0.87 ± 0.17	17.42 ± 2.77	6.61 ± 0.92

V_1/2 is expressed in mV. Q is the number of elementary charges associated to channel activation. τ_Activation and τ_Deactivation are expressed in ms. n ≥ 5 independent experiments.

To accurately estimate the differential effect of DiBAC₄(3) on α vs. α + β1 BK channels, we fitted each individual experiment to a Boltzmann equation, and plotted the averaged V_1/2 as a function of DiBAC₄(3) concentration (Table 1). As shown in Figure 2B, V_1/2 decreased with increasing concentrations of DiBAC₄(3) in α alone channels, and to a greater extent in α +β₁ channels. Hill function fitting of the data suggested that the β₁ subunit promotes a 4-fold decrease in the K_d for DiBAC₄(3) from 22 μM, in α channels (with an N of Hill of 0.93), to 5 μM in α +β₁ channels (N of Hill 1.27). We also analyzed the concentration dependence of DiBAC₄(3) in the slowing of channel deactivation kinetics. Time constants obtained from mono-exponential fittings of current relaxation at −120 mV (from a 200 mV activation step) are plotted as a function of DiBAC₄(3) concentration in Figure 2C. The deactivation time constant increased in a concentration-dependent fashion. We performed a Hill function fitting of the data, and obtained a K_d value of 10 μM for α alone and 6.8 μM for α+β₁ BK channels; with Ns of Hill of 1.7 and 1.9, respectively.

We next examined the voltage-dependent characteristics of current relaxation kinetics after a depolarizing step in the presence and absence of 30 μM DiBAC₄(3). Time constants obtained from these currents are plotted as a function of voltage in Figure 2D. In α alone channels, DiBAC₄(3) produced an increase in the deactivation time constant with an apparent disruption of its voltage dependence. On the other hand, in α+β₁ channels the deactivation time constants were also increased in the presence of DiBAC₄(3) (approximately by a factor of 50), but the voltage dependence was unaltered. The slope of the exponential fitting remains similar.

We also evaluated activation kinetics at 200 mV. We were not able to detect any noticeable effect of DiBAC₄(3); time constants remained almost unchanged at increasing concentrations of DiBAC₄(3) (Fig. 2E).

BK channels at saturating Ca²⁺ concentrations can be further activated by DiBAC₄(3)

As DiBAC₄(3) activation appears to be independent of Ca²⁺-dependent activation of the channels, we examined whether DiBAC₄(3) can activate BK channels at saturating concentrations of Ca²⁺ when all high-affinity Ca²⁺ binding sites of the channel are likely to be occupied. Even under conditions of maximal activation by Ca²⁺ (200 μM), DiBAC₄(3) produced a further hyperpolarizing shift in the I-V relationship (Fig. 3, left). This effect was more pronounced in channels formed by α+β₁ subunits (Fig. 3, right).

Figure 3. DiBAC₄(3) activates α alone and α+β₁ BK channels in 200 μM Ca²⁺. Plots depict representative examples of the voltage dependence of BK channel activation in 200 μM Ca²⁺ in the absence and presence of 30 µM DiBAC₄(3). G-V data were obtained from tail currents at –120 mV, after 150 ms voltage steps (–200 to +20 mV for α alone channels; –200 to –70 mV for α+β₁ channels).

β₁ to β₂ mutations do not prevent the potentiating effect of the β₁ subunit on DiBAC₄(3) activation of BK channels

DiBAC₄(3) was initially described as a BK channel activator acting only when β₁ or β₄, subunits were part of the channel, with the β₂ subunit having no role—or even an inhibitory one—in the presence of DiBAC₄(3).³⁰ Additionally, specific β₁ to β₂ mutations were recently reported to dampen the differential effect of omega-3 fatty acids on channels containing β₁ vs. β₂ subunits.³¹ Our observations of the consistent enhancing influence of the β₁ subunit on the effects of DiBAC₄(3) prompted us to introduce these mutations into our experiments. We mutated β₁ subunits in the residues Arg11 or Cys18 to obtain β_R11A and β_C18E subunits. We co-expressed α subunits with either β_R11A or β_C18E. Differential effects were investigated by exposing the mutant channels to a DiBAC₄(3) concentration (1.1 μM) that is not high enough to activate α alone channels. DiBAC₄(3) activated BK currents in patches with both α+β_R11A and α+β_C18E channels (Fig. 4A andB). Further, application of 1.1 μM DiBAC₄(3) promoted a leftward shift in the current-voltage relationship and slowed the deactivation kinetics in both cases. The V_1/2 shifted from 178 ± 7 mV to 144 ± 10 mV for α + β_R11A channels (n = 6; P < 0.005) and from 162 ± 5 mV to 128 ± 4.8 mV for α + β_C18E channels (n = 6; P < 0.005) (Fig. 4C). This suggests that β₁ residues Arg11 and/or Cys18 are unlikely contributing to the differential effect of DiBAC₄(3) on α alone vs. α+β₁ channels.

Figure 4. β₁ to β₂ mutations do not impair the increase in BK current induced by DiBAC₄(3). (A) Representative traces of macroscopic current recordings elicited by the indicated voltage protocol (inset) from macro patches excised from HEK cells expressing α alone or together with mutated β₁ subunits, in the absence (control) or presence of 1.1 µM DiBAC₄(3). Data were recorded from inside-out patches in symmetrical (140 mM) K⁺ solutions with 0 mM Ca²⁺. (B) Corresponding steady-state current vs. voltage. Symbols represent the average current from the last 15 ms of the activating step. (C) Bar graph of average V_1/2 values in the absence and in the presence of 1.1 μM DiBAC₄(3) from HEK cells expressing α alone or together with WT or mutated β₁ subunits (mean ± s.e.; n ≥ 5; multiple comparisons performed with 2-way ANOVA; P < 0.05).

Discussion

In our recent single-channel study recording arterial smooth muscle BK channels we showed that DiBAC₄(3) is able to activate native BK channels in the absence of β₁ or β4 subunits.⁹ We now show that recombinant BK channels can also be activated by DiBAC₄(3) in the absence of auxiliary β subunits, suggesting a binding site for the dye is located in the α subunit and the β₁ subunit plays a modulatory role. This is not the first time an activator was originally described as β₁-specific but was later shown to activate α alone channels as well: DHS-I (dehydrosoyasaponin I)⁵ and tamoxifen³² also fit this description. Importantly, the substantial β₁ modulatory effects obtained with common recombinant BK channel variants may mislead the interpretation of activating effects as being produced by genuine β₁-specific ligands. Careful inspection of a wide range of concentration-dependent effects is advised for determining the stringency for β subunit requirements. The differential effect of DiBAC₄(3) in α vs. α+β₁ channels presented here might be at the root of the initially observed β₁/β₄ specificity.³⁰ A detailed inspection of the effect of DiBAC₄(3) on BK channels highlights a mechanism that involves a shift in the voltage-dependent activation of the channel together with a deceleration of the deactivation kinetics.

DiBAC₄(3) shifts voltage-dependent activation

In the virtual absence of Ca²⁺ we observed a concentration-dependent shift in the voltage-dependent gating of BK channels upon DiBAC₄(3) application. For α alone channels this shift is as much as 40 mV, with a small change in the associated charge observed at high concentrations of DiBAC₄(3) (> 10 μM). For α+β₁ channels, DiBAC₄(3) can produce a shift larger than 110 mV, without significant changes in slope values. These values are smaller than what we have observed in native channels. This disparity probably stems from the molecular heterogeneity of native BK channel α subunits,^13,33-36 and/or differences in post-translational modifications. Since DiBAC₄(3) also activates BK channels in saturating Ca²⁺ concentrations, the Ca²⁺-sensing domain of the channel appears not to be involved in the activating mechanism. Thus, DiBAC₄(3) is not acting as a Ca²⁺ surrogate for BK channels and can activate BK channels beyond the maximal effect of Ca²⁺. The G-V slopes were not greatly affected by DiBAC₄(3), suggesting that the voltage-sensing machinery of the channels might not participate; gating current measurements would be more appropriate to resolve this question.

DiBAC₄(3) slows deactivation kinetics

In the presence of DiBAC₄(3) we observed a consistent slowing of deactivation kinetics that was concentration-dependent. Indeed, (30 μM) DiBAC₄(3) prolongs the deactivation time constants more than 4-fold in both α and α+β₁channels, but does not affect activation kinetics. According to the HA model for BK channel gating, the changes induced by DiBAC₄(3) may result from a mechanism affecting the intrinsic closed-to-open (C-O) transitions of the channels. This process differs from the voltage-sensor movement but is still (weakly) voltage-dependent. Changes in relaxation kinetics are expected if DiBAC₄(3) interferes with the C-O conformational changes. The deceleration of deactivation kinetics suggests that DiBAC₄(3) may promote a destabilization of the closed state of the channel. If macroscopic deactivation kinetics are dictated by single-channel bursting behavior,³⁷ the observed prolongation of deactivation time constants with DiBAC₄(3) supports our previous observation in native single channels where DiBAC₄(3) shortened the channel’s interburst time in a concentration-dependent fashion (Scornik et al.,⁹ Figure 3C). Nevertheless, an intermediate Ca²⁺ concentration (3 μM) was used in that study, yielding a more complex gating scheme.

We also analyzed the voltage dependence of deactivation time constants at a single DiBAC₄(3) concentration. The intrinsic C-O transition is slightly voltage-dependent. Under control conditions, the voltage dependence of deactivation time constants can be described by an exponential function that has very similar slopes for both α and α+β₁ channels. The application of DiBAC₄(3) produces a prolongation of time constants. However, our analysis reveals a differential effect of DiBAC₄(3) for α vs. α+β₁ channels. At ­–190 to –70 mV, DiBAC₄(3) prolongs deactivation time constants of α alone channels more than 20-fold but disrupts the voltage dependence of channel closing by producing a shallower response to voltage. On the other hand, in the presence of the β₁ subunit, DiBAC₄(3) prolongs the deactivation time constants more than 50-fold, while preserving the voltage dependence associated to the process. The physical interpretation of this remains unclear; a wider voltage range is needed to firmly establish the differential effect.

The role of the β₁ subunit

The dye concentrations needed to half-maximally activate the channels are reduced by a factor of 4 in the presence of the β₁ subunit. This parameter (K_d) changes from around 20 μM to 5 μM when the β₁ subunit is present. The latter is very similar to the value obtained using native single channels in our previous work (3.4 μM in canine coronary BK channels), suggesting a similar affinity for the dye. The concentration-dependent effects of the dye on deactivation kinetics showed an expanded range of change (Table 1) in the presence of the β₁ subunit, although the K_d for DiBAC₄(3) in this series falls within a similar range (10 μM vs. 6.8 μM; α vs. α+β₁ channels, respectively). Collectively, these results support a potentiating role of the β₁ subunit in modifying G-V relationships and deactivation kinetics with DiBAC₄(3).

Point mutation analysis of β₁ molecular determinants

The β₁ subunit was recently reported to potentiate the activating effect of the omega-3 fatty acid DHA (docosahexaenoic acid) on BK channels.³¹ Two key residues, Arg11 in the N terminus and Cys18 in the TM1 region, are critical to produce this potentiation. We mutated those residues to understand the molecular mechanism by which the β₁ subunit enhances DiBAC₄(3) effects. DiBAC₄(3) and DHA share, in principle, β₁ and β₄ subunit potentiating effects and both lack a role for β₂ subunits. Our point mutation study, however, indicates that neither R11A nor C18E mutations reduce the β₁ potentiation of DiBAC₄(3) effects. In contrast, those mutations in β₁ reduce DHA potentiation of the BK channel current.³¹ The 2 β-subunit residues are proposed to be close enough to each other to establish a hydrogen bond or an electrostatic interaction. The 2 interacting residues would then interact directly with the pore helix S6 region or the S6-RCK1 linker of the α subunit to destabilize the closed state of the channel.³¹ Our analysis of DiBAC₄(3) argues for a mechanism that also involves destabilization of the closed state of the channel. However, β₁ residues, suspected to contribute to the destabilization of the closed state of the channel, do not appear to play a role during DiBAC₄(3) activation.

Notably, while DiBAC₄(3) conspicuously prolongs deactivation kinetics, DHA has only a slight effect on this process. This suggests that DiBAC₄(3) and DHA target different molecular mechanisms to activate the channel. Gessner et al.³⁸ recently suggested a classification for BK channel activators based on the HA model. They proposed activators could be grouped based on their capacity to modify, for example, the C-O equilibrium constant, and/or the resting-activated equilibrium constant of the voltage sensor (i.e., HA model parameters L, and J, respectively). It would be interesting to know if a unifying picture can emerge from this classification in relation to activators potentiated by β subunits, reconciling a classification based on mechanistic considerations with one based on auxiliary subunit (tissue) specificity. We did not perform an extended study to obtain HA model parameters. However, based on the effects presented here, DiBAC₄(3) may belong to L⁺ class activators, potentiated by the β₁ subunit, despite that specific β₁ residues do not influence its effects. Whether this feature constitutes a class of openers that can eventually be mapped to specific structural components of BK channels remains to be elucidated. Our findings shed light on specific molecular mechanisms of BK channel pharmacology that are key for the rational design of compounds of both research and therapeutic interest.

Materials and Methods

Heterologous expression of recombinant mouse (mslo) BK channels

Human Embrionic Kidney (HEK) tsa201 cells (Health Protection Agency Culture Collections) were grown in Dulbecco’s Modified Eagle Medium supplemented with 10% fetal bovine serum (GIBCO, Invitrogen) under standard cell culture conditions (5% CO₂ incubator at 37 °C). Mouse BK channels, mslo α cDNA (kcnma1, GenBank: MMU09383) subcloned into the mammalian expression vector pcDNA3 (Invitrogen, www.lifetechnologies.com), and mslo β₁ cDNA (kcnmb1, GenBank: EDL23733.1) subcloned into the mammalian expression vector pIRES2-EGFP (CLONTECH) were used for transfection. HEK cells were plated in 35 mm dishes and transfected 24 h later at 60–70% confluence, using the transfection reagent GenCellin (BioCellChallenge). In experiments assessing the α subunit alone, mslo was co-transfected with green fluorescent protein as a marker. In co-transfection experiments assessing mslo α+β₁ channels, the β₁ subunit (WT or mutated) was co-transfected at a ratio of 1:10, α to β₁, to ensure saturation of BK channels with β₁ subunits. The mouse β₁ subunit construct fluorescently labels cells with channel subunit expression. Four to six h after transfection, cells were trypsinized and re-plated on glass coverslips and maintained in culture for their use between 24–48 h post-transfection.

Site-directed mutagenesis

The construct pIRES2-EGFP-kcnmb1 was used as a template to engineer mutations p.R11A and p.C18E by site-directed mutagenesis using the QuikChange® Lightning Site-Directed Mutagenesis Kit (Stratagene) and the following primer pairs (mutation underlined): p.R11A forward, 5′GGTGATGGCC CAGAAGGCCG GAGAG 3′; reverse 5′GCTCGTGTCT CTCCGGCCTT CTGGGCC 3′. p.C18E forward, 5′GGAGAGACAC GAGCCCTCGA GCTGGG 3′, reverse 5′CCATTGCCAC TCCCAGCTCG AGGGCT 3′. The resultant constructs (kcnmb1_R11A and kcnmb1_C18E) were sequenced and prepared with Qiagen Maxi Prep kit (QIAGEN).

Electrophysiological recordings

Macroscopic BK currents were recorded from excised inside-out patches of transfected HEK cells. Pipette and initial bath (200 μM Ca²⁺) solution contained 140 mM KCl, 10 mM HEPES, 1 mM MgCl₂, and 0.2 mM CaCl₂ (pH 7.2). After patch excision an initial voltage protocol was routinely applied to establish the maximum BK current of the patch in the 200 μM Ca²⁺ bathing solution. The external recording solution was then extensively washed with a 0 mM Ca²⁺ solution containing 140 mM KCl, 10 mM HEPES, and 11 mM EGTA. Increasing concentrations of DiBAC₄(3), dissolved in the same 0 mM Ca²⁺ solution, were applied with a gravity driven perfusion system (Cell MicroControls). After each solution exchange, excisions were equilibrated for 2 min before recordings. Pipette resistance was between 1 and 3 MΩ. Data were acquired on an Axopatch 200B amplifier (Molecular Devices), filtered at 5 kHz, and sampled at 100 kHz by using a Digidata 1322A (Molecular Devices). All recordings were performed at room temperature. Leak correction was achieved using P/N leak subtraction protocols. DiBAC₄(3) was obtained from Invitrogen. DiBAC₄(3) was dissolved in dimethyl sulfoxide (DMSO) to a 20 mM stock solution, aliquoted, and stored at –20 °C until use. Unless stated otherwise, all chemicals were from Sigma-Aldrich.

Data analysis

To obtain G-V curves, we measured instantaneous tail currents at 200 μs after the end of the preceding activating pulse using Clampfit 10.2 (Molecular Devices). Instantaneous tail current values were then normalized to the maximum tail current amplitude obtained in 200 μM Ca²⁺. Data were imported into Origin, and each curve was individually fitted by using a built-in Boltzmann equation as a data descriptor. Individual V_1/2 and slope values were obtained from the fit. Charge (Q) associated with channel activation was estimated from the slope using RT/f = 25.3 mV. Activation and deactivation time courses were fit with a single exponential function using Clampfit and visually inspected for proper fit. For deactivation kinetics, a family of tail currents were elicited by step pulses at different voltages, preceded by an initial 250 ms activating step to 90 mV for α alone channels, and a 150 ms activating step to 150 mV for α+β₁ channels. Concentration-response curves were fitted using Origin built-in Hill function. Apparent dissociation constant (K_d) and N of Hill values were obtained from the fit.

Glossary

Abbreviations:

BK channel

large conductance calcium-activated potassium channel

DiBAC₄(3)

Bis (1,3-Dibutylbarbituric Acid) Trimethine Oxonol

Scornik FS, Bucciero RS, Wu Y, Selga E, Bosch Calero C, Brugada R, Pérez GJ. DiBAC₄(3) hits a “sweet spot” for the activation of arterial large-conductance Ca²⁺-activated potassium channels independently of the β₁-subunit. Am J Physiol Heart Circ Physiol. 2013;304:H1471–82. doi: 10.1152/ajpheart.00939.2012.

10.4161/chan.27212