Abstract
The large-conductance voltage-gated and calcium-dependent K+ (BK) channels are widely distributed and play important physiological roles. Commonly used BK channel inhibitors are peptide toxins that are isolated from scorpion venoms. A high-affinity, nonpeptide, synthesized BK channel blocker with selectivity against other ion channels has not been reported. We prepared several compounds from a published patent application (Doherty et al., 2004) and identified 1-[1-hexyl-6-(methyloxy)-1H-indazol-3-yl]-2-methyl-1-propanone (HMIMP) as a potent and selective BK channel blocker. The patch-clamp technique was used for characterizing the activity of HMIMP on recombinant human BK channels (α subunit, α+β1 and α+β4 subunits). HMIMP blocked all of these channels with an IC50 of ∼2 nM. The inhibitory effect of HMIMP was not voltage-dependent, nor did it require opening of BK channels. HMIMP also potently blocked BK channels in freshly isolated detrusor smooth muscle cells and vagal neurons. HMIMP (10 nM) reduced the open probability significantly without affecting single BK-channel current in inside-out patches. HMIMP did not change the time constant of open states but increased the time constants of the closed states. More importantly, HMIMP was highly selective for the BK channel. HMIMP had no effect on human NaV1.5 (1 μM), CaV3.2, L-type Ca2+, human ether-a-go-go-related gene potassium channel, KCNQ1+minK, transient outward K+ or voltage-dependent K+ channels (100 nM). HMIMP did not change the action potentials of ventricular myocytes, confirming its lack of effect on cardiac ion channels. In summary, HMIMP is a highly potent and selective BK channel blocker, which can serve as an important tool in the pharmacological study of the BK channel.
The large-conductance voltage-gated and calcium-dependent K+ (BK) channels are widely distributed in smooth muscle, neuron, and many other tissues, and they play an important role in many physiological events (for a recent review, see Ghatta et al., 2006). The BK channel contains four α subunits that form the channel pore (Atkinson et al., 1991) and different auxiliary β subunits that modulate the channel properties (Garcia-Calvo et al., 1994; Knaus et al., 1994a). Four mammalian β subunits have been identified: the β1 subunit increases the Ca2+/voltage sensitivity of the BK channel and slows its kinetics (McManus et al., 1995; Brenner et al., 2000b); the β2 (Wallner et al., 1999; Xia et al., 2003) and β3 (Brenner et al., 2000a; Xia et al., 2000) subunits cause BK channel inactivation; and the β4 subunit alters channel kinetics (Brenner et al., 2000a, 2005).
Several widely used BK channel inhibitors, such as iberiotoxin (Galvez et al., 1990), charybdotoxin (Miller et al., 1985), and slotoxin (Garcia-Valdes et al., 2001), are peptide toxins that are isolated from various scorpion venoms. In a recent study, two more BK channel blockers, BmBKTx1 (Xu et al., 2004) and BmP09 (Yao et al., 2005), were identified in scorpion toxins. Peptide toxins with different pharmacological properties are widely used to study BK channels. However, β subunits modulate the interaction between these peptide toxins and the α subunit, and the neuronal BK channels formed by α+β4 subunits are resistant to iberiotoxin, charybdotoxin, and slotoxin (Meera et al., 2000; Garcia-Valdes et al., 2001; Lippiat et al., 2003). Nonpeptide, alkaloid BK channel blockers, such as paxilline, penitrem A, verruculogen, and tetrandrine, are also useful pharmacological tools for examining BK channels (Knaus et al., 1994b; Wu et al., 2000; Tammaro et al., 2004). Paxilline blocks BK channels in a Ca2+-dependent manner, and it potently inhibits BK channels with low free Ca2+ concentrations (Sanchez and McManus, 1996). In a recent study, a small molecule, A-272651, has been reported to inhibit the BK channel with an IC50 of ∼5 μM, which is 1000-fold less potent than the peptide BK channel blockers (Shieh et al., 2007). High-affinity, synthesized small molecule BK channel blockers with good ion channel selectivity remain to be identified.
We prepared several compounds from a published patent application (Doherty et al., 2004) and identified a highly potent BK channel blocker, 1-[1-hexyl-6-(methyloxy)-1H-indazol-3-yl]-2-methyl-1-propanone (HMIMP), with single-digit nanomolar potency and good ion channel selectivity (>50-fold). HMIMP, as a highly potent and selective small molecule BK channel blocker, is a very useful tool for examining the pharmacology of BK channels.
Materials and Methods
Cell Culture. Chinese hamster ovary (CHO) cells were stably transfected to express different recombinant channels, including the human BK α subunit, BKα+β1 subunits, BKα+β4 subunits, human ether-a-go-go-related gene (hERG), or human KCNQ1+minK. Human embryonic kidney (HEK)293 cells were used to stably express the human NaV1.5 α subunit or CaV3.2 (α1H) channels. All cell lines were generated at GlaxoSmithKline (King of Prussia, PA) and maintained at 37°C with 5% CO2 in T-75 flasks or 6-well culture dishes in appropriate media.
Cell Isolation. The studies were carried out in accordance with the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, 1996) as adopted and promulgated by the U.S. National Institutes of Health and approved by the GlaxoSmithKline Animal Care and Use Committee. Single myocytes were isolated from the right ventricle of adult male New Zealand White rabbits using enzyme digestion as described previously (Rials et al., 1997). Myocytes were also isolated from the right ventricle of adult male guinea pigs using the same isolation protocol, with the exception that the enzyme solution perfusion was abbreviated to 5 min. Only rod-shaped, quiescent cells with obvious striations were used for the experiments. The procedure of detrusor smooth muscle cell isolation has been described in detail previously (Ghatta et al., 2007). To isolate single vagal neurons, mice of either sex were euthanized by carbon dioxide asphyxiation. Nodose jugular complex was rapidly removed bilaterally. They were incubated in enzyme solution, which was composed of collagenase type 1 (2 mg/ml) (Sigma-Aldrich, St. Louis, MO) and dispase II (2 mg/ml) (Roche, Indianapolis, IN) in Ca2+- and Mg2+-free Hank's balanced salt solution for 1 h at room temperature. Neurons were dissociated by trituration, washed by centrifugation (2 times at 1000g for 5 min), suspended in L-15 medium (Invitrogen, Carlsbad, CA) containing 10% fetal bovine serum, and transferred on to poly-l-lysine-coated coverslips (BD Biosciences Discovery Labware, Bedford, MA). Neurons adhered to coverslips and were maintained for 15 h before recording. Neurons were used up to 24 h after isolation, and there was no neurite outgrowth observed. To isolate arterial smooth muscle cells, adult male guinea pigs were sacrificed by sodium pentobarbital overdose. Small mesenteric arteries were removed from the animal. Fat and connective tissue were removed in a cold, nominal Ca2+-free solution containing 137 mM NaCl, 5 mM KH2PO4, 1 mM MgSO4, 10 mM glucose, 5 mM HEPES, 8 mM taurine, and 1 mg/ml bovine serum albumin, pH 7.4. The vessels were cut into tiny pieces and incubated in nominally Ca2+-free solution on ice for approximately 20 min. The tissue pieces were then transferred to an enzyme solution made by adding 50 μM CaCl2, 1.5 mg/ml collagenase type II (Worthington Biochemicals, Freehold, NJ), 1 mg/ml protease XXIV (Sigma-Aldrich), and 0.25 mg/ml trypsin inhibitor (Sigma-Aldrich) to nominally Ca2+-free solution. Tissues in the enzyme solution were stored on ice for 2 to 3 h and then incubated in 37°C water bath and bubbled with 100% O2. The enzyme solution was checked periodically for single smooth muscle cells. Cells were harvested in the supernatant by centrifugation. The digestion and collection procedure was repeated twice. Isolated cells were stored at 4°C in a solution composed of 80 mM potassium glutamate, 20 mM K2HPO4, 20 mM KCl, 5 mM MgCl2, 0.5 mM K2EGTA, 2 mM Na2ATP, 5 mM Na-pyruvate, 5 mM creatine, 20 mM taurine, 10 mM glycine, 10 mM glucose, and 5 mM HEPES. Cells were used for experiments within 6 h of isolation.
Current Recording. All currents were recorded at room temperature (∼23°C) (with the exception of NaV1.5 current measured at 13 ± 0.4°C) using an Axopatch 200B amplifier and Digidata 1322A digitizer (Molecular Devices, Sunnyvale, CA). Cells were placed in a small chamber (volume = 0.7 ml) and continuously perfused with an external solution (3–4 ml/min). Electrodes were pulled from thin wall glass (WPI, Sarasota, FL) using a P-97 horizontal puller (Shutter, Novato, CA) and fire polished with MF-830 microforge (Narishige, Long Island, NY). Electrode resistance ranged from 2 to 3 MΩ for whole-cell currents and 8 to 15 MΩ for inside-out patch recordings. Currents were elicited by different voltage protocols (described in text and figure legends) and acquired with pCLAMP 8 software (Molecular Devices). Single-channel currents were filtered at 2 kHz and digitized at 25 kHz.
Action Potential Recording. Action potentials were recorded from single guinea pig or rabbit ventricular myocytes at 36 ± 0.3°C, using the microelectrode technique at a stimulus frequency of 1 Hz. The microelectrode was made using a KOPF pipette puller (David Kopf Instruments, Tujunga, CA) and had a resistance of 25 to 40 MΩ when filled with 3 M KCl. Action potentials of single guinea pig detrusor smooth muscle cells were recorded at 35 ± 0.4°C, using the whole-cell current-clamp (1.6-ms superthreshold depolarizing current pulses given at 0.5 Hz).
Solutions. Unless otherwise stated, bath solution for current and action potential recording contained 140 mM NaCl, 4 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 10 mM glucose, and 10 mM HEPES, pH 7.4. Bath solution for NaV1.5 current contained 14 mM NaCl, 126 mM N-methyl-d-glucamine chloride, 4 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 10 mM glucose, and 10 mM HEPES, pH 7.4. Bath solution for L-type Ca2+ current contained 135 mM NaCl, 5 mM CsCl, 1 mM MgCl2, 2 mM CaCl2, 10 mM glucose, and 10 mM HEPES, pH 7.4. Bath solution for transient outward K+ current contained 140 mM N-methyl-d-glucamine chloride, 5 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 10 mM glucose, and 10 mM HEPES, 0.3 mM CdCl2, 0.1 mM BaCl2. Pipette solution, which contained 140 mM KCl, 5 mM EGTA, 1 mM MgCl2, 5 mM MgATP, 0.2 mM CaCl2, and 5 mM HEPES, pH 7.2 (6.5 nM free Ca2+), was used for recording detrusor action potentials and all the currents unless otherwise stated. A similar pipette solution containing 1.4 mM CaCl2 was used for neuronal BK current recording (61 nM free Ca2+). Pipette solution for L-type Ca2+ current contained 151 mM CsOH, 10 mM l-aspartic acid, 20 mM taurine, 20 mM tetraethylammonium chloride, 5 mM glucose, and 10 mM EGTA, and pH was adjusted to 7.5 with H3PO4. MgATP (5 mM) and guanosine triphosphate (0.4 mM sodium salt) were added to the pipette solution before use, and the final pH was 7.3. Internal solution for recording NaV1.5, CaV3.2, hERG, and KCNQ1+minK current contained 119 mM K-gluconate, 15 mM KCl, 5 mM EGTA, 5 mM K2ATP, 3.2 mM MgCl2, and 5 mM HEPES, pH 7.2. Single-channel bath solution contained 140 mM KCl, 1 mM MgCl2, 2 mM CaCl2 5 mM EGTA, and 5 mM HEPES, pH 7.2 (100 nM free Ca2+), and pipette solution contained 150 mM KCl, 1 mM MgCl2, 1 mM CaCl2, and 5 mM HEPES, pH 7.2. The free Ca2+ concentrations of solutions were calculated using Webmaxc (http://www.stanford.edu/~cpatton/webmaxcS.htm).
Chemicals. HMIMP (see Fig. 8 for structure) was synthesized at GlaxoSmithKline. HMIMP was dissolved in dimethyl sulfoxide to make 1 mM stock solution and stored at -20°C. The drug stock solution was diluted in external solutions to the desired concentrations and used within 3 h. The effects of HMIMP were determined when drug response reached steady-state after 3 to 5 min of perfusion with the compound.
Data Analysis. Whole-cell currents were analyzed using pClamp 8 (Molecular Devices). Single-channel currents were analyzed with Fetchan and PStat bundles in the pClamp 8 software. Events lists were created in Fetchan using a 50% threshold-crossing approach to determine open events, and no minimal-duration level was imposed when detecting transitions between closed and open states. The single-channel amplitudes were obtained by forming amplitude histograms of selected regions of recordings that had clear single open and closed current levels. Gaussian distributions were fitted to the amplitude histograms to determine the unitary current. The dwell time distributions of open and closed current levels were constructed from patches containing only one active channel. Open dwell time distribution histograms were fitted with a single exponential, and closed dwell time distribution histograms were fitted with a double exponential: where Afast and Aslow are the amplitude terms of the fast (τfast) and slow (τslow) time constants, respectively. A Simplex algorithm was used to perform nonlinear least-squared fits. Data were presented as the mean ± S.E. (n). Statistical significance (P < 0.05) was determined using one-way analysis of variance, one-population t test, or paired t test.
Results
We first studied the effects of HMIMP on the human BK α subunit stably expressed in CHO cells. Whole-cell currents were elicited by 200-ms depolarizing steps to different voltages (+30 –+140 mV in 10-mV increments) from a holding potential of 0 mV (Fig. 1A, inset), and the interpulse interval was 2 s. The characteristic, fast-activating BK currents were recorded under control conditions, and 30 nM HMIMP almost completely abolished the currents (Fig. 1A). We next examined the concentration-dependent inhibition of the human BK α subunit by HMIMP. BK current inhibition (percentage) was calculated using current amplitudes after drug effect reached steady-state against those before HMIMP treatment at various voltages. The IC50 value of each concentration-response curve was obtained by fitting the curve with the logistic equation. HMIMP blocked human BK α subunit with IC50 values of 1.7 ± 0.6, 2.1 ± 0.8, 2.3 ± 0.8, and 2.9 ± 0.9 nM (n = 5) at +80, +100, +120, and +140 mV, respectively (Fig. 1B). The differences in IC50 values were not statistically significant. We then examined whether the effect of HMIMP was use-dependent. BK current was recorded using the voltage-clamp protocol (as shown in the inset of Fig. 1A, but in the voltage range of +70 –+110 mV) in control solution (Fig. 1C, left). After perfusing the cell with 3 nM HMIMP for 5 min without stimulation, BK currents were recorded 12 times using the same voltage-clamp protocol, and the first and last recordings in drug were compared (Fig. 1C, middle and right). There was no additional block with the opening of BK channels.
Different β subunits of the BK channel conveyed differential sensitivity to iberiotoxin. Iberiotoxin can potently inhibit BK α, α+β1, and α+β3, but not α+β4 (Meera et al., 2000; Lippiat et al., 2003). Therefore, we examined whether different β subunits would modify the blocking effect of HMIMP on the BK α subunit. CHO cells stably expressing human BK α+β1 or α+β4 subunits were used. Whole-cell currents were elicited with the same protocol, and data were processed in the same way as described in Fig. 1. In the presence of β1 (the smooth muscle isoform) or β4 subunits (the neuronal isoform), BK channel activation was much slower compared with channels composed of the α subunit alone (comparing insets of Fig. 2 to Fig. 1A), as described elsewhere (Lippiat et al., 2003). The concentration responses measured at +100 mV showed that HMIMP inhibited BK channels with an IC50 of 2.1 ± 0.6 nM for α+β1 and 1.5 ± 0.4 nM for α+β4 (Fig. 2, A and B). The IC50 values were not significantly different from each other or from that of channels composed of the α subunit alone. HMIMP (30 nM) also blocked the BK current in freshly isolated rabbit detrusor smooth muscle cells (Fig. 2C) and mouse vagal neurons (Fig. 2D), suggesting that HMIMP can act on native BK channels of different species.
Single-channel current was recorded in inside-out patches of CHO cells stably expressing the human BK α subunit. HMIMP (10 nM) reduced single channel open probability from 0.120 ± 0.051 to 0.014 ± 0.004, an average 81 ± 4% reduction (n = 7), but had no significant effect on single-channel current, 18.1 ± 0.8 pA in control and 18.0 ± 0.8 pA (n = 7) in drug recorded at +80 mV (Fig. 3, A–C). The extent of BK α channel block by 10 nM HMIMP in inside-out patches was comparable with that observed in the whole-cell condition. HMIMP did not affect the time constant of open states (Fig. 3D), 7.8 ± 1.5 ms in control and 9.6 ± 1.6 ms (n = 7) in drug. HMIMP increased the fast and slow time constants of the closed state from 21.3 ± 3 and 194.7 ± 39.5 ms to 41.9.0 ± 7.1 and 485.3 ± 70.5 ms (n = 7), respectively, after drug treatment (P < 0.05) (Fig. 3E). However, the ratio of the amplitude for the fast component to the sum of the amplitude of both components [Afast/(Afast+Aslow)] was unchanged, 0.71 ± 0.07 versus 0.83 ± 0.04 (n = 7) for control and in the presence of 10 nM HMIMP, respectively.
The selectivity of HMIMP against other ion channels expressed in the cardiovascular system was examined. Na+ currents were elicited from HEK293 cells stably expressing human NaV1.5 channels by 30-ms depolarizing steps to different voltages (-60 –+30 mV in 10-mV increments) from a holding potential of -90 mV (Fig. 4A, inset), and the interpulse interval was 5 s. Na+ currents were recorded at a low temperature (13°C) and in a reduced extracellular Na+ concentration to ensure good voltage clamp. Peak current amplitudes at various voltages were normalized to the maximal peak current amplitude of the cell before HMIMP treatment to obtain normalized current-voltage relationship (I-V) curve. HMIMP at 1 μM had no significant effect on the I-V curve of NaV1.5 channels (Fig. 4A). T-type Ca2+ current was recorded from HEK293 cells expressing human CaV3.2 channels by using the same voltage-clamp protocol as that used for Na+ current, with the exception of longer depolarizing steps (160 ms). HMIMP (100 nM) did not affect T-type Ca2+ current (Fig. 4B). L-type Ca2+ current in guinea pig ventricular myocytes were elicited by 180-ms depolarizing voltage steps to different voltages (-30 –+50 mV in 10-mV increments) from a holding potential of -40 mV (Fig. 4C, inset) in the presence or absence of 100 nM HMIMP (Fig. 4C). The interpulse interval was 5 s. Currents were normalized to the maximal current amplitude of the cell before application of HMIMP to generate a normalized I-V curve (Fig. 4C). The L-type Ca2+ currents were not affected by 100 nM HMIMP.
The effect of HMIMP on hERG channels was investigated using the same protocols as described previously (Zeng et al., 2006). In brief, hERG currents in CHO cells were induced before and after 100 nM HMIMP application by 2-s depolarizing steps to different voltages (-70 –+60 mV in 10-mV increments) from a holding potential of -80 mV, followed by a 2-s hyperpolarizing step to -50 mV to measure tail currents (Fig. 5A). The interpulse interval was 10 s. To obtain normalized I-V curves, step current amplitudes measured as the mean current of the last 100 ms at various voltages were normalized to the maximal step current amplitude of the cell before HMIMP treatment. The bell-shape I-V curves revealed that HMIMP (100 nM) inhibited the hERG step currents only at +40 mV or above (Fig. 5B). However, 100 nM HMIMP had no effect on the voltage-dependent activation of hERG channels that were measured from the normalized peak tail currents (Fig. 5C).
Effects on KCNQ1+minK channels, which underlie the slowly activating delayed rectifier K+ current in cardiac tissue, were evaluated. Current from CHO cells stably expressing human KCNQ1+minK were excited by 2.8-s depolarizing voltage steps to different voltages (-30 –+60 mV in 10-mV increments) from a holding potential of -90 mV, followed by a 1-s hyperpolarizing step to -50 mV (Fig. 6A, inset). The interpulse interval was 6 s. To generate normalized I-V curves, step current amplitudes at various voltages were normalized to the maximal step current amplitude of the cell before HMIMP application. The result showed that HMIMP had no effect on the KCQN1+minK current at 100 nM (Fig. 6A). Likewise, the transient outward K+ current (Ito) of rabbit ventricular myocytes was examined with a pulse protocol (Fig. 6B, inset) with or without 100 nM HMIMP. Ito current was induced by 160-ms pulses to different voltages (-10 – +40 mV in 10-mV increments) from a holding potential of -80 mV. The interpulse interval was 5 s. Peak currents were normalized to the maximal current amplitude before HMIMP application to evaluate the effect of HMIMP on Ito. The result clearly demonstrated that 100 nM HMIMP did not affect Ito current (Fig. 6B). Besides cardiac ion channels, we also examined whether HMIMP had any effect on the voltage-gated K+ (KV) channel in vascular smooth muscle cells. KV currents were excited by 200-ms depolarizing voltage steps to different voltages (-40 –+40 mV in 10-mV increments) from a holding potential of -80 mV (Fig. 6C, inset) with or without 100 nM HMIMP. The interpulse interval was 5 s. Currents were normalized to the maximal current amplitude of the cell before HMIMP application to generate normalized I-V curves (Fig. 6C), showing no effect of HMIMP on KV current.
Finally, we examined the overall effect of 100 nM HMIMP on cardiac ion channels by measuring the action potential of ventricular myocytes freshly isolated from guinea pig or rabbit heart. Action potentials were recorded with microelectrodes as described previously (Rials et al., 1997). The action potential shape and duration were almost identical before or after application of 100 nM HMIMP in both guinea pig and rabbit ventricular myocytes (Fig. 7, top and middle panels). On the contrary, HMIMP (30 nM) dramatically prolonged the action potential and reduced after hyperpolarization of a detrusor smooth muscle cell (Fig. 7, bottom panel) as expected, because BK channels play a critical role in the repolarization of the action potential of detrusor (Heppner et al., 1997).
Discussion
In this study, we reported a highly potent and selective, synthesized small molecule BK channel blocker, HMIMP. It blocked recombinant human BK channels with an IC50 value of ∼2 nM, and the blocking potency was not affected by the membrane potential or coexpression of different BK β subunits. HMIMP also blocked native BK channels in smooth muscle cells and in vagal neurons, which are mostly comprised of α+β1 subunits and α+β4 subunits, respectively.
As a widely distributed ion channel, the BK channels contribute to cellular functions in many tissues and are involved in certain human diseases (Ghatta et al., 2006). Most widely used BK channel blockers are peptide toxins. HMIMP is a small molecule that has BK-blocking potency greater than or equal to the “gold standard”, iberiotoxin. Whereas both iberiotoxin and HMIMP have no effect on single BK channel current amplitude, they are distinct in many aspects. First, iberiotoxin blocks BK channels only from the extracellular side of the membrane, whereas HMIMP blocked BK channels from either side. Second, iberiotoxin produces long nonconducting silent periods, without affecting the gating kinetics of BK channels (Giangiacomo et al., 1992), whereas HMIMP decreased the BK α channel open probability primarily by increasing the duration of channel closure, stabilizing the channel in the closed state. Third, BK channel accessory β1 subunit attenuates the potency of iberiotoxin 10-fold (Lippiat et al., 2003), and β4 renders the channel insensitive to iberiotoxin (Meera et al., 2000), whereas neither β1 nor β4 affected the potency of HMIMP. Besides iberiotoxin, BK α+β4 channels are also resistant to charybdotoxin (Meera et al., 2000) and slotoxin (Garcia-Valdes et al., 2001). Our results indicate that HMIMP is the most potent BK α+β4 channel blocker known and will have particular value for studying the functions of neuronal BK channels that contain the β4 subunit.
The lack of impact of β subunits on the HMIMP-blocking effect on BK channels implies that HMIMP may directly bind to the BK α subunit, and the binding site is not interrupted by the association of β subunits. In addition, HMIMP blocked BK channels in a nonuse-dependent manner, which indicated that it was not an open channel blocker. Therefore, it is probable that HMIMP binds to the BK α subunit and blocks the channel allosterically, rather than directly occupying the channel pore.
Besides peptide BK channel blockers, alkaloids such as paxilline and tetrandrine also inhibit BK channels with an IC50 value of 1.9 nM for BK α subunit and 5 μM for BK channels of an endothelial cell line (Sanchez and McManus, 1996; Wu et al., 2000). However, paxilline potently inhibits BK channels only at low free Ca2+-concentrations (Sanchez and McManus, 1996). Tetrandrine also blocks voltage-dependent Ca2+ channels (Wang and Lemos, 1995). The only other known small molecule BK channel blocker, A-272651, has an IC50 value of ∼5 μM (Shieh et al., 2007), 2500 times less potent than HMIMP.
There seems to be no obvious similarity between the two-dimensional structures of HMIMP and paxilline (Fig. 8), although they have comparable IC50 values. As far as commonalities go, both structures have carbonyls as well as five, 6-ring systems (indazole/indole) and ether oxygens, but the spatial relationships between these two types of moieties are completely different when comparing HMIMP and paxilline. We cannot exclude the possibility of the existence of three-dimensional correspondences between these two molecules, but it's very difficult to justify without knowing whether they bind to the same location in the BK channel and the key structural domains of each molecule.
In addition to being highly potent, HMIMP was also a very specific BK channel blocker. At ≥50 times the IC50 value for BK channel inhibition, HMIMP had no effect on various cardiovascular ion channels including NaV1.5, CaV3.2, L-type Ca2+, KCNQ1+minK, transient outward K+, and KV channels. Although HMIMP slightly reduced hERG step current, the effect occurred only at membrane potentials ≥ +40 mV, which is nonphysiological for cardiac myocytes. The lack of effects of HMIMP on action potentials of ventricular myocytes substantiated its specificity.
In summary, HMIMP is a potent and highly selective BK channel blocker that can serve as a useful tool in the study of the BK channel function.
Acknowledgments
We thank Drs. Joseph Marino, Dennis Lee, and Steve Zhao at the Chemistry Department of Metabolic Pathway Center of Excellence in Drug Discovery for the synthesis of HMIMP, Dr. Srinivas Ghatta at the Biology Department of Respiratory and Inflammation Center of Excellence in Drug Discovery for isolation of neurons, and Dr. Eric Manas (Computational Chemistry Department of Metabolic Pathway Center of Excellence in Drug Discovery) for analyzing the structures of HMIMP and paxilline.
Footnotes
-
H.Z. and E.G. contributed equally to this work.
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.108.139733.
-
ABBREVIATIONS: BK channel, large-conductance voltage-gated and calcium-dependent K+ channel; A-272651, 2,4-dimethoxy-N-naphthalen-2-yl-benzamide; HMIMP, 1-[1-hexyl-6-(methyloxy)-1H-indazol-3-yl]-2-methyl-1-propanone; CHO, Chinese hamster ovary; hERG, human ether-a-go-go-related gene; HEK, human embryonic kidney; I-V, current-voltage relationship; Ito, transient outward K+ current; KV, voltage-gated K+.
-
↵1 Current affiliation: Merck Research Laboratories, West Point, Pennsylvania.
- Received April 3, 2008.
- Accepted June 26, 2008.
- The American Society for Pharmacology and Experimental Therapeutics