Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Three chemical classes of Ca⁺⁺ antagonists such as dihydropyridines (e.g., nifedipine, nitrendipine), phenylalkylamines (e.g., verapamil) and benzothiazepines (e.g., diltiazem) are known to bind to distinct binding sites on alpha-1 subunit of the L-type Ca⁺⁺ channel and to exert reciprocal allosteric effect on each other's binding sites (McDonald et al., 1994).

The potentiation of the specific binding of dihydropyridines by diltiazem is well known, and such an allosteric effect has often been used as a proof for the selectivity to the benzothiazepine binding site. Some of the 1,5-benzothiazepine derivatives, however, have been reported to partially inhibit the binding of dihydropyridine, which is interpreted as the negative allosteric modulation by the derivatives (Narita et al., 1990; Striessnig et al., 1990; Ikeda et al., 1991). Thus the degree and direction of the allosteric effect (positive or negative) on the dihydropyridine site appear to be varied among the 1,5-benzothiazepine derivatives.

The receptor domain on the alpha-1 subunit of the L-type Ca⁺⁺ channel for dihydropyridines, as well as phenylalkylamines, has been determined by high-affinity photoaffinity ligands selective to the respective sites. Because of lack of a high-affinity ligand, the benzothiazepine site has not been investigated as extensively as the other two binding sites. Another powerful approach for determining the critical site for Ca⁺⁺ antagonists has been electrophysiological experiments with membrane-impermeable quaternary derivatives (McDonald et al., 1994) or recombinant Ca⁺⁺ channels (Grabner et al., 1996; Hering, et al., 1996). The affinity for the quaternary diltiazem has been reported to be less than 1/40 of that for diltiazem; thus, the extremely high concentration (more than 10⁴ M) required for testing its effects might cause a secondary effect and complicate the results (Adachi-Akahane et al., 1993). On the other hand, the recombinant Ca⁺⁺ channels expressed in Xenopus laevis oocytes appear to be less sensitive to Ca⁺⁺ antagonists than the native channels; thus, again, the extremely high concentration of Ca⁺⁺ antagonists might attenuate their selectivity to the respective binding sites. Thus a high-affinity ligand for the benzothiazepine site will be a useful tool for investigating the benzothiazepine binding site.

DTZ323 (fig. 1) is a novel d-cis-1,5-benzothiazepine derivative in which the chemical structure of diltiazem is fully conserved. This compound has been reported to block voltage-dependent L-type calcium channel currents selectively in single guinea pig ventricular myocytes (Nagao et al., 1994).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Chemical structure of 1,5-benzothiazepine derivatives used in this study.

To elucidate the possibility of DTZ323 as a high-affinity ligand for the benzothiazepine site, we examined DTZ323 for the affinity and selectivity to the 1,5-benzothiazepine binding site of rabbit skeletal muscle membranes with radiolabeled dihydropyridine ((+)-[³H]PN200-110), phenylalkylamine (()-[³H]D888) and benzothiazepine (d-cis-[³H]diltiazem). Clentiazem, a diltiazem derivative, has been reported to have higher affinity for the benzothiazepine site than diltiazem (Zobrist and Mecca, 1990; Suzuki et al., 1991). Therefore we here compared the binding affinity of DTZ323 with that of clentiazem as well as diltiazem.

This report shows the high affinity and the selectivity of DTZ323 to the diltiazem binding site of skeletal muscle L-type Ca⁺⁺ channels.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Membrane preparation. Membranes were isolated from rabbit skeletal muscle according to the methods of Naito et al. (1989) and Ikeda et al. (1991). After cervical dislocation, the back muscles were quickly removed from male New Zealand white rabbits (about 1 kg), minced with scissors and homogenized with a Potter-Elvehjem glass homogenizer with a loosely fitting Teflon pestle and with a Polytron (setting 7-8, 10 s × 3) in about 5 volumes of ice-cold buffer A (20 mM NaHCO₃, 0.2 mM PMSF, 1 mM IAA, 1 µM pepstatin A). The homogenate was filtered through two layers of cheesecloth and centrifuged at 1,500 × g for 15 min. The resulting supernatant was filtered through two layers of cheesecloth and centrifuged at 45,000 × g for 20 min. The pellet was resuspended in ice-cold buffer B (50 mM Tris-HCl, 0.2 mM PMSF, 1 mM IAA, 1 µM pepstatin A, pH 7.4 at 4°C) with a glass-Teflon homogenizer and sedimented as before. The resulting pellet was washed with buffer B by centrifugation at 45,000 × g for 20 min. The pellet was finally resuspended in buffer B to yield a protein concentration of approximately 10 mg/ml, frozen in liquid nitrogen and stored at 70°C until use. To obtain a good yield of membrane preparation, the first pellet (1,500 × g 15 min) was resuspended, filtered and sedimented as described above. The resulting supernatant was then centrifuged at 45,000 × g for 20 min and the resulting pellet was resuspended and sedimented in buffer B. The pellet obtained was finally mixed with the last original pellet. All procedures were conducted at 4°C.

Binding assay. Radioligand binding studies with d-cis-[³H]diltiazem, (+)-[³H]PN200-110 and ()-[³H]desmethoxyverapamil (()-[³H]D888) were carried out at 37°C in 50 mM Tris-HCl (pH 7.4) as described previously (Glossmann and Ferry, 1985; Naito et al., 1989; Ikeda et al., 1991). Radioligand concentration, protein concentration, total assay volume and incubation time of competition binding experiments were 30 nM, 0.06 mg/assay, 0.3 ml and 60 min for d-cis-[³H]diltiazem; 1 nM, 0.01 mg/assay, 0.5 ml and 30 min for (+)-[³H]PN200-110; and 3 nM, 0.03 mg/assay, 0.5 ml and 30 min for ()-[³H]D888, respectively. Nonspecific binding was measured in the presence of 100 µM d-cis-diltiazem for d-cis-[³H]diltiazem binding, 1 µM nicardipine for (+)-[³H]PN200-110 binding and 30 µM (±)verapamil for ()-[³H]D888 binding. In equilibrium binding experiments, 0.1 to 10 nM (+)-[³H]PN200-110 and 1 to 100 nM ()-[³H]D888 were used. At the end of the incubation period, samples were diluted with 5 ml of ice-cold washing buffer (50 mM Tris-HCl, pH 7.4) and immediately filtered through GF/C filters which had been presoaked in 0.5% polyethyleneimine. Filters were washed three times with 5 ml of ice-cold buffer. Brandel cell harvester (Biomedical Research & Development Laboratories, Inc. Gaithersburg, MD) was used for the filtration procedure. In case of d-cis-[³H]diltiazem binding, 100 µM diltiazem was added to the washing buffer, and the filters were presoaked in 0.5% polyethyleneimine and 20 µM diltiazem for more than 2 hr at room temperature to eliminate undesirable binding (Balwierczak et al., 1987). Radioactivity on the filter was measured by liquid scintillation counting. All experiments were performed in duplicate and data were presented as mean ± S.E. of four experiments.

Data analysis. Data were analyzed with nonlinear least-squares programs. The SP123 program made by Dr. H. Ono (University of Tokyo) was used for Scatchard analysis of saturation binding data and pseudo Scatchard analysis of competition binding data to obtain values for K_d and B_max. The LBS program by Dr. A. Seo (Hiroshima University) was used to obtain values for the IC₅₀ and the slope factor from competition binding data. Values of the inhibition constant (K_i) were calculated from IC₅₀ values by use of the following equation: K_i = IC₅₀/(1 + L/K_d), where L is the concentration of the radioligand (Cheng and Prusoff, 1973). The dissociation rate constant (K₁) was determined from a first-order plot of ln B_t/B₀ versus time, where B_t is the ligand-receptor complex at time t and B₀ is that at time zero.

Drugs. d-cis-[³H]diltiazem (80.3-86.1 Ci/mmol) was purchased from New England Nuclear (Boston, MA). (+)-[³H]PN200-110 (79-86 Ci/mmol) and ()-[³H]D888 (85 Ci/mmol) were from Amersham (Buckinghamshire, U.K.). DTZ323, d-cis-diltiazem and clentiazem were kindly supplied by Tanabe Seiyaku Co. (Saitama, Japan). Nicardipine, pepstatin A, PMSF and IAA were purchased from Sigma (St. Louis, MO). (±)-Verapamil was purchased from Nacalai Tesque (Kyoto, Japan).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of DTZ323 on d-cis-[³H]diltiazem specific binding. We examined the effect of DTZ323 on the benzothiazepine binding site. Figure 2 shows the effect of DTZ323, diltiazem, clentiazem and verapamil on d-cis-[³H]diltiazem binding to benzothiazepine site on rabbit skeletal muscle membranes. DTZ323, as well as diltiazem and clentiazem, exhibited concentration-dependent and complete inhibition of d-cis-[³H]diltiazem binding to the membrane with slope factor close to unity, indicating competitive inhibition at benzothiazepine site. IC₅₀ values in table 1 indicates that DTZ323 was 48 times more potent than diltiazem and 9 times more potent than clentiazem. Pseudo Scatchard analysis of inhibition data of d-cis-[³H]diltiazem binding to the membranes revealed a single class of binding site. Calculated K_d and B_max values were 315 ± 36 nM and 4797 ± 336 fmol/mg protein, respectively. K_i values derived from the K_d were 6.6 ± 0.6 nM for DTZ323, 314 ± 25 nM for diltiazem and 61 ± 2 nM for clentiazem. Phenylalkylamine, (±)-verapamil, also showed complete inhibition of d-cis-[³H]diltiazem binding with a slope factor close to unity.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Inhibitory effects of drugs on d-cis-[3H]diltiazem binding to rabbit skeletal muscle membranes. Membranes were incubated at 37°C for 60 min with 30 nM d-cis-[3H]diltiazem in the presence of various concentrations of DTZ323 (), diltiazem (), clentiazem () and verapamil (). Nonspecific binding was measured in the presence of 100 µM diltiazem. Each point represents the mean ± S.E. of four experiments.

Effect of DTZ323 on (+)-[³H]PN200-110 specific binding. We examined whether DTZ323 affected the specific binding of the dihydropyridine ligand. Figure 3 shows that DTZ323 partially inhibited the (+)-[³H]PN200-110 binding. Maximal inhibition and EC₅₀ value were 83 ± 1% and 21 ± 2 nM, respectively. This effect was different from those of diltiazem and clentiazem which showed stimulation of (+)-[³H]PN200-110 binding up to 201 ± 6% and 174 ± 3%, respectively. EC₅₀ values were 280 ± 73 nM for diltiazem and 60 ± 7 nM for clentiazem. Phenylalkylamine, verapamil, partially inhibited (+)-[³H]PN200-110 binding to 54 ± 3% with an EC₅₀ value of 160 ± 28 nM. Dihydropyridine, nicardipine, inhibited (+)-[³H]PN200-110 binding completely in a concentration-dependent manner with a slope factor close to unity, which indicated competitive inhibition (IC₅₀ = 6.9 ± 0.2 nM, slope factor = 0.98 ± 0.01). Scatchard analysis of (+)-[³H]PN200-110 equilibrium binding to the membrane showed single class of binding site. K_d and B_max values were 0.38 ± 0.01 nM and 4303 ± 401 fmol/mg protein, respectively. K_i value derived from the K_d value was 1.9 ± 0.1 nM for nicardipine.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Effects of drugs on (+)-[3H]PN200-110 binding to rabbit skeletal membranes. Membranes were incubated at 37°C for 30 min with 1 nM (+)-[3H]PN200-110 in the presence of various concentration of DTZ323 (), diltiazem (), clentiazem (), verapamil () and nicardipine (). Nonspecific binding was measured in the presence of 1 µM nicardipine. Each point represents the mean ± S.E. of four experiments.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Effects of drugs on the binding of (+)-[3H]PN200-110 at two fixed concentrations of 1.5 nM and 15 nM. Membranes were incubated at 37°C for 30 min with either 1.5 nM or 15 nM (+)-[3H]PN200-110 in the presence of various concentrations of DTZ323 (), diltiazem () and nicardipine (). Nonspecific binding was measured in the presence of 1 µM nicardipine. Each point represents the mean ± S.E. of three experiments (1.5 nM (+)-[3H]PN200-110) or four experiments (15 nM (+)-[3H]PN200-110).

Effect of DTZ323 on ()-[³H]D888 specific binding. We then examined the effect of DTZ323 on the phenylalkylamine site. (±)-Verapamil, DTZ323, clentiazem and diltiazem showed complete inhibition of the specific binding of the phenylalkylamine ligand, ()-[³H]D888, with a slope factor close to unity (fig. 5). Relative potency of unlabeled ligands (table 3) was similar to the result from d-cis-[³H]diltiazem binding experiment as described above. Scatchard analysis of ()-[³H]D888 equilibrium binding to the membrane showed a single class of binding site. k_d and B_max values were 5.9 ± 0.9 nM and 5232 ± 59 fmol/mg, respectively. K_i value derived from the K_d value was 48 ± 3 nM for verapamil.

View larger version (18K): [in this window] [in a new window]   Fig. 5.   Inhibitory effects of drugs on ()-[3H]D-888 binding to rabbit skeletal muscle membranes. Membranes were incubated at 37°C for 30 min with 3 nM ()-[3H]D-888 in the presence of various concentrations of DTZ323 (), diltiazem (), clentiazem () and verapamil (). Nonspecific binding was measured in the presence of 30 µM verapamil. Each point represents the mean ± S.E. of four experiments.

View this table: [in this window] [in a new window]   TABLE 3 IC50 values and slope factors for inhibition of ()-[3H]D-888 binding to rabbit skeletal muscle membranes Membranes were incubated at 37°C for 30 min with 3 nM ()-[3H]D-888 in the presence of various concentrations of DTZ323, diltiazem, clentiazem and verapamil. Nonspecific binding was measured in the presence of 30 µM verapamil. Each point represents the mean ± S.E. of four experiments.

Effects of DTZ323 on dissociation kinetics of d-cis-[³H]diltiazem. After the binding of d-cis-[³H]diltiazem had reached the equilibrium at 2°C, dissociation of d-cis-[³H]diltiazem from the d-cis-[³H]diltiazem-receptor complex was initiated by the addition of 30 µM unlabeled diltiazem. This reaction displayed first order kinetics as shown in figure 6. Concomitant addition of 30 µM verapamil increased the dissociation rate constant. On the other hand, 1 µM DTZ323 had no effect on the dissociation rate constant.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Effects of DTZ323 and verapamil on the dissociation kinetics of d-cis-[3H]diltiazem. The rate of dissociation was measured at 2°C. After equilibrium of the binding of d-cis-[3H]diltiazem to membranes had been reached, the dissociation of the ligand-receptor complex was initiated by addition of 30 µM unlabeled diltiazem, and the remaining bound [3H]diltiazem was measured by filtration at the respective times. Effects of 1 µM DTZ323 and 30 µM verapamil on the rate of dissociation of d-cis-[3H]diltiazem were examined by concomitant addition of drugs with 30 µM diltiazem. Dissociation rate constants obtained by linear regression analysis for control (), DTZ323 () and verapamil () were K1 = 0.012 min1 (r = 1.00), K1 = 0.012 min1 (r = 0.99) and K1 = 0.023 min1 (r = 0.99), respectively.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Searching for a high-affinity ligand for the benzothiazepine binding site within the L-type calcium channel, a novel 1,5-benzothiazepine derivative, DTZ323 was characterized by radioligand-binding experiments in terms of the affinity and selectivity to the binding sites for the three classical Ca⁺⁺ antagonists (dihydropyridine, benzothiazepine and phenylalkylamine sites). Binding sites for new calcium antagonists other than those for the classical three classes have been reported (Striessnig et al., 1988; King et al., 1989; Schmid et al., 1989; Staudinger et al., 1991). At present, however, the information may still be insufficient to define such new sites and details remain to be elucidated (Spedding and Paoletti, 1992). Therefore, in the present study, we examined the specificity of DTZ323 to the binding sites for the three classical calcium antagonists.

In this study, DTZ323 modulated the binding of the radiolabeled ligands specific to the respective binding sites for the three chemical classes of calcium antagonists in rabbit skeletal muscle membranes in a manner similar to the compounds which have been reported to compete with diltiazem for the benzothiazepine binding site (King et al., 1988; Striessnig et al., 1990; Ikeda et al., 1991).

The results from the competition experiment suggest that DTZ323 interacts directly at either the benzothiazepine or phenylalkylamine site of the L-type Ca⁺⁺ channel. According to the simplified allosteric models by Ehlert (1988) and Tomlinson and Hnatowich (1988), the inhibition of the specific binding of a radioligand by an allosteric modulator is likely to be observed as apparently competitive inhibition when the magnitude of negative heterotropic cooperativity is very large and the concentration of the radioligand is much smaller than its K_d value. Such possible negative allosteric interaction with either phenylalkylamine or benzothiazepine site may complicate the interpretation of the results.

To clarify the selectivity of DTZ323 to the benzothiazepine and the phenylalkylamine site, we examined the effects of DTZ323 on dissociation kinetics of d-cis-[³H]diltiazem. Garcia et al. (1986) and Balwierczak et al. (1987) reported that verapamil increases the dissociation rate of d-cis-[³H]diltiazem through an allosteric interaction between benzothiazepine and phenylalkylamine binding sites. Changes in rates of ligand dissociation in the presence of a test compound is most likely ascribed to an allosteric interaction when no multiple interacting sites for a ligand exists on the receptor complex. Compounds that affect receptor binding as an allosteric effector are expected to alter ligand dissociation rates, whereas those that compete directly with ligands for the same binding site are expected not to affect dissociation rates. In our study, pseudo Scatchard analyses revealed that d-cis-[³H]diltiazem binds to a single class of noncooperative binding sites. DTZ323 (1 µM) exhibited no effect on the dissociation rate of d-cis-[³H]diltiazem, whereas verapamil increased the dissociation rate. Such results with DTZ323 were similar to those obtained with other compounds, which have been reported to bind competitively to the benzothiazepine site [tetrandrine (King et al., 1988); azidodiltiazem (Striessnig et al., 1990); l-cis-, d-trans-, and l-trans-diltiazem (Ikeda et al., 1991)].

In radioligand-binding experiments with Ca⁺⁺ antagonists, kinetic analysis of effects of the test compound on the binding of each ligand, based on the current allosteric model of calcium antagonist binding sites, has been considered to provide a determinant for classifying the binding site of a novel ligand. However, the selectivity between the benzothiazepine and the phenylalkylamine sites has not been satisfactorily determined in radioligand-binding experiments with membrane preparations. Prinz and Striessnig (1993) proposed a multipoint attachment model in binding experiments and concluded that the increase in the dissociation rate produced by a drug cannot be taken as an evidence for the existence of the multiple-binding domain. We also confirmed the acceleration of the dissociation of d-cis-[³H]diltiazem by the high concentration of benzothiazepine ligands (data not shown). Further investigation may be required for clarifying the difference in the binding characteristics between benzothiazepine- and phenylalkylamine-binding sites.

Whether binding sites for phenylalkylamines and benzothiazepines are composed of distinct or identical receptor sites within the L-type Ca⁺⁺ channels has been questioned. Several reports have provided contradictory evidence. In studies investigating skeletal muscle T-tubule membranes, Goll et al. (1984) and Glossmann et al. (1984) reported the noncompetitive interaction between diltiazem and verapamil binding sites. On the other hand, Galizzi et al. (1986) reported that phenylalkylamine and benzothiazepine sites may be identical because these compounds showed the reciprocal inhibition of the binding in apparently competitive manner by increasing K_d values in Scatchard analysis. In studies with cardiac muscle membranes, Garcia et al. (1986) reported that the two binding sites are distinct because the dissociation rate of diltiazem is markedly increased in the presence of verapamil. Balwierczak et al. (1987) also provided similar results by examining the effects of benzothiazepines and phenylalkylamines on dissociation kinetics of diltiazem. King et al. (1988) and Felix et al. (1992) concluded that the binding sites for diltiazem and phenylalkylamines are distinct by use of tetrandrine analogs which are selective to benzothiazepine site.

In electrophysiological experiments with isolated guinea pig ventricular myocytes, DTZ323 and its membrane-impermeable quaternary analog (DTZ417) inhibited the L-type Ca⁺⁺ channel currents preferentially from the extracellular side of the membrane in contrast to D890, a quaternary phenylalkylamine, which acts from the intracellular side (Kurokawa et al., in press). These results indicate that the specific binding site for DTZ323 is distinct from that for phenylalkylamines.

The diltiazem derivatives, which have been reported to compete with diltiazem for the benzothiazepine binding site, can be classified into two groups based on the way they interact with the dihydropyridine binding site at 37°C. d-cis-Diltiazem and clentiazem stimulate dihydropyridine binding (Ferry and Glossmann, 1982; Suzuki et al., 1991). On the other hand, other derivatives such as azidodiltiazem, azidobutyryl-diltiazem, azidobenzoyl-diltiazem, l-cis-diltiazem, d-trans-diltiazem and l-trans-diltiazem have been reported to inhibit the dihydropyridine binding (Striessnig et al., 1990; Narita et al., 1990; Ikeda et al., 1991). DTZ323 partially inhibited the (+)-[³H]PN200-110 binding (fig. 2). The results show that DTZ323 belongs to the group which exerts inhibitory effect on the [³H]dihydropyridine binding at 37°C.

In addition, nonbenzothiazepine compounds, such as KB944, MDL12330A and BM20.1140, have been reported not to interact competitively with diltiazem at the benzothiazepine binding site but to stimulate the binding of dihydropyridine radioligands at 37°C (Holck et al., 1984; Lee et al., 1985; Staudinger et al., 1991). Together with the heterogeneity of the effects of the diltiazem derivatives on the (+)-[³H]PN200-110 binding as mentioned above, the potency to stimulate the dihydropyridine binding may not be a sufficient criteria for classifying a new Ca⁺⁺ antagonist ligand as a specific ligand for the benzothiazepine site.

In competition experiment (fig. 4), the maximal effects of DTZ323 and diltiazem were reduced when the concentration of (+)-[³H]PN200-110 was extremely high (30 times larger than its K_d value), which was in contrast to the competitive ligand such as nicardipine which completely inhibited the (+)-[³H]PN200-110 binding. These results indicate that both DTZ323 and diltiazem allosterically modulate the dihydropyridine binding site.

We investigated the manner in which DTZ323 inhibits the specific binding of (+)-[³H]PN200-110, and found that DTZ323 reduces the affinity for (+)-[³H]PN200-110 in a concentration-dependent manner with slight decrease of the density of the binding sites (table 2). This was in contrast to diltiazem which has been shown to increase the affinity for (+)-[³H]PN200-110 with slight increase in B_max (Ikeda et al., 1991; Kanda, et al.,1997) at 37°C.

Among 1,5-benzothiazepine derivatives, azidobutyryl-diltiazem and azidobenzoyl-diltiazem have been reported to have binding affinities close to that of diltiazem (2°C, Naito et al., 1989). But the affinity of l-cis-, d-trans- and l-trans-diltiazem (2°C or 37°C) and azidodiltiazem (2°C) have been reported to be less than 1/10 of that of diltiazem in radioligand-binding experiments (Ikeda et al., 1991; Striessnig et al., 1990). Nonbenzothiazepine compounds such as tetrandrine bind to the 1,5-benzothiazepine site with an affinity three times higher than that of diltiazem at 37°C (King et al., 1988), and SQ32,910 7, a benzazepine analog, is 7 times more potent than diltiazem at 2°C (Hering et al., 1993). Hence, the potency of DTZ323, which is 48 times more potent than diltiazem, is the highest of all the diltiazem derivatives ever studied.

Our results demonstrate that DTZ323 modulates the specific binding of the radiolabeled dihydropyridine, phenylalkylamine and 1,5-benzothiazepine Ca⁺⁺ antagonists to their distinct binding sites on the alpha-1 subunit of skeletal muscle L-type Ca⁺⁺ channels. The manner of modulation was consistent with those of the compounds which are known to compete with diltiazem for the 1,5-benzothiazepine site. Thus we conclude that DTZ323 appears to be the most potent ligand for the benzothiazepine site of all the diltiazem derivatives ever studied. Further pharmacological investigation with radiolabeled DTZ323 is to be carried out. DTZ323 is expected to give a breakthrough for characterizing the benzothiazepine binding sites on the L-type calcium channel.