The effects of verapamil and diltiazem on N-, P- and Q-type calcium channels mediating dopamine release in rat striatum

Abstract

1. The putative inhibitory effects of verapamil and diltiazem on neuronal non-L-type Ca²⁺ channels were studied by investigating their effects on either K⁺- or veratridine-evoked [³H]-dopamine ([³H]-DA) release in rat striatal slices. Involvement of N-, P- and Q-type channels was identified by sensitivity of [³H]-DA release to ω-conotoxin GVIA (ω-CTx-GVIA), ω-agatoxin IVA (ω-Aga-IVA) and ω-conotoxin MVIIC (ω-CTx-MVIIC), respectively.

2. KCl (50 mM)-evoked [³H]-DA release was abolished in the absence of Ca²⁺, and was insensitive to dihydropyridines (up to 30 μM). It was significantly blocked by ω-CTx-GVIA (1 μM), ω-Aga-IVA (30 nM) and was confirmed to be abolished by ω-CTx-MVIIC (3 μM), indicating involvement of N-, P- and Q-type channel subtypes.

3. Verapamil and diltiazem inhibited K⁺-evoked [³H]-DA release in a concentration-dependent manner. The inhibitory effects of verapamil or diltiazem (each 30 μM) were fully additive to the effect of ω-CTx-GVIA (1 μM), whereas co-application with ω-Aga-IVA (30 nM) produced similar effects to those of ω-Aga-IVA alone.

4. As shown previously, veratridine-evoked [³H]-DA release in Ca²⁺ containing medium exclusively involves Q-type Ca²⁺ channels. Here, diltiazem (30 μM) did not inhibit veratridine-evoked [³H]-DA release, whereas verapamil (30 μM) partially inhibited it, indicating possible involvement of Q-type channels in verapamil-induced inhibition. However, verapamil (30 μM) inhibited this release even in the absence of extracellular Ca²⁺, suggesting that Na⁺ rather than Q-type Ca²⁺ channels are involved.

5. Taken together, our results suggest that verapamil can block P- and at higher concentrations possibly N- and Q-type Ca²⁺ channels linked to [³H]-DA release, whereas diltiazem appears to block P-type Ca²⁺ channels only.

Introduction

Synaptic transmission is dependent upon the entry of Ca²⁺ through presynaptic voltage-activated Ca²⁺ channels (VACCs). In peripheral and central mammalian neurons, one low (T-type) and at least six high VACCs (L-, N-, O-, P-, Q-, and R-type) have been identified on the basis of their electrophysiological and pharmacological properties (Olivera et al., 1994; Varadi et al., 1995; Randall & Tsien, 1995). Striatal dopamine (DA) transmission is mediated mainly by N-type Ca²⁺ channels which are blocked selectively by ω-conotoxin-GVIA (ω-CTx-GVIA) (Tsien et al., 1988), whereas L-type Ca²⁺ channels are not involved (Harvey et al., 1996; Prince et al., 1996; Dobrev et al., 1998). Recently, it has been demonstrated that P-type Ca²⁺-channels which are blocked by ω-agatoxin-IVA (ω-Aga-IVA) (Randall & Tsien, 1995) and Q-type Ca²⁺ channels (Wheeler et al., 1994) which are blocked by ω-conotoxin-MVIIC (ω-CTx-MVIIC) also modulate dopamine synaptic transmission (Turner et al., 1993; Kimura et al., 1995; Soliakov & Wonnacott, 1996; Harvey et al., 1996; Dobrev & Andreas, 1997; Okada et al., 1998). The role of R- and O-type Ca²⁺ channels in synaptic transmissions is still unclear, because selective ligands of these channels have not yet been discovered (Olivera et al., 1994; Randall & Tsien, 1995).

Agents that selectively modulate VACCs are valuable tools in basic research. For example, low concentrations (⩽1 μM) of dihydropyridines (DHPs), inhibit Ca²⁺ currents in a frequency- and voltage-dependent manner due to selective blockade of L-type VACCs. However, higher concentrations of these drugs (⩾30 μM) are no longer selective, but also block other ion channels as for instance Na⁺ channels (Yatani & Brown, 1985). Together with the peptide neurotoxins that block other specific classes of VACCs (Olivera et al., 1994), the DHPs have been used to delineate the contributions of calcium channel subtypes to specific biological functions; however the Ca²⁺ channel blockers of phenylalkylamine and benzothiazepine structure have not been sufficiently investigated.

Although phenylalkylamines and benzothiazepines in low concentrations are also considered to be selective for L-type Ca²⁺ channels, there is now evidence that they may also affect non-L-type VACCs at higher concentrations (for review see McDonald et al., 1994). Recently, Ishibashi et al. (1995) demonstrated that micromolar concentrations of verapamil and diltiazem block P-type Ca²⁺ currents in freshly dissociated cerebellar Purkinje neurons and Diochot et al. (1995) showed that these drugs also have inhibitory effects on other types of neuronal non-L-type Ca²⁺ currents (see also Cai et al., 1997). Little data is currently available concerning the effects of verapamil and diltiazem on dihydropyridine-resistant Ca²⁺ channels involved in striatal dopamine transmission. There is evidence from one recent study that verapamil strongly inhibits K⁺-evoked [³H]-DA release from striatal synaptosomes in a concentration-dependent manner (Prince et al., 1996). Since K⁺-evoked [³H]-DA release is known to be predominantly mediated by N- and P-type but not by L-type Ca²⁺ channels, and also to be tetrodotoxin-resistant (Turner et al., 1993; Prince et al., 1996), its block by verapamil and diltiazem would suggest involvement of non-L-type Ca²⁺ channels.

Here, we investigated the putative inhibitory effects of verapamil and diltiazem on neuronal non-L-type VACCs. Since direct electrophysiological investigation of the Ca²⁺ currents of most nerve terminals in central neurons is generally limited by their small size (see Dunlap et al., 1995), we chose to study the effects of verapamil and diltiazem on K⁺-evoked [³H]-DA release from rat striatal slices in order to establish the contribution of N- and P-type Ca²⁺ channels to [³H]-DA release. In addition, we also investigated the drugs' effects on veratridine-evoked [³H]-DA release because this release mechanism involves predominantly Q-type VACCs (Dobrev et al., 1998). Thus, the involvement of N-, P- and Q-type channels was identified pharmacologically by sensitivity of evoked [³H]-DA release to ω-CTx-GVIA, ω-Aga-IVA and ω-CTx-MVIIC, respectively (see Dobrev & Andreas, 1997; Dobrev et al., 1998). Some of the results reported here have appeared previously in abstract form (Dobrev & Andreas, 1998).

Methods

All studies complied with the German home office regulations governing the care and use of laboratory animals.

[³H]-DA loading of rat striatal slices

The release experiments were performed as previously described (Dobrev & Andreas, 1997; Dobrev et al., 1998). In brief, the rats were anaesthetized with CO₂, and after decapitation the brain was immersed in chilled (4°C) with atmospheric oxygen equilibrated Krebs-phosphate buffer (KPB) of the following composition (in mM): NaCl 118, KCl 4.7, CaCl₂ 1.8, MgSO₄ 7H₂O 1.2, Na₂HPO₄2H₂O 15.9, ascorbic acid 0.6 and glucose 5.6 (pH 7.4). Subsequently 200-μm-thick coronal slices of the striatum were prepared and pre-incubated for 20 min at room temperature (22–24°C). The slices were then incubated with 0.1 μM [³H]-DA for additional 30 min at 37°C. After loading and washing of the slices, 10 μM nomifensine was applied to KPB to block [³H]-DA re-uptake, 10 μM pargyline to prevent its metabolism and 1.3 mM Na₂-EDTA to prevent auto-oxidation of [³H]-DA. The slices (∼7 mg of tissue wet weight) were then transferred into a superfusion chamber (37°C) and were superfused with KPB using a peristaltic pump (flow rat 0.6 ml min⁻¹, Desaga PLG-Peristaltic pump, Heidelberg, Germany).

Release experiments

Base-line release became stable within 20 min of wash-out, 30 or 40 successive 1 min fractions were collected. The release of [³H]-DA from the slices was stimulated either once with 25 μM veratridine (S₁) or twice at fraction 7 (S₁) and 27 (S₂) (i.e. 27 and 47 min after the onset of superfusion) with 50 mM KCl (NaCl was replaced by KCl to maintain an isosmotic condition, see inset in Figure 1). Unless otherwise stated, the test drugs were applied to the medium 12 min before S₂ in the experiments with K⁺ stimulation, and 6 min before starting fraction collection in those with veratridine stimulation. At the end of the experiment, the slices were sonicated and after adding of scintillation cocktail to the probes, total radioactivity (³H) in the collected fractions and in 500 μl aliquots of the homogenized slice preparation was determined by liquid scintillation-spectrometry (LKB 1219 Rackbeta scintillation counter).

Figure 1.

Effects of tetrodotoxin, Ca²⁺ and ω-CTx-MVIIC on K⁺-evoked [³H]-DA release. After 20 min washout period, 1 min fractions of superfusate were collected, and [³H]-DA release was evoked twice (S₁, S₂, see inset) for 2 min with 50 mM KCl containing buffer. Tetrodotoxin was added to the superfusion medium or CaCl₂ was replaced by equimolar MgCl₂, 12 min before S₂, ω-CTx-MVIIC was applied 10 min before S₂. Values are means±s.e.mean (bars) of the S₂/S₁ ratios in per cent of control, n indicates number of experiments. Statistical analysis was performed on untransformed data (S₂/S₁) by one-way ANOVA and LSD multiple comparison test. ^*P<0.05 compared with the controls.

Data presentation and statistics

The stimulation-evoked [³H]-DA release was calculated from the total release during the 2 min stimulation period and the subsequent 9 min (in the case of KCl) or 15 min (in the case of veratridine), corrected for basal release during this time. The evoked release during S₁ and S₂ was then expressed as per cent of the total radioactivity present in the slices just before the respective stimulation (S₁ and S₂%). To quantify drug effects on [³H]-DA release evoked by K⁺ stimulation, the ratio S₂/S₁ was calculated and expressed as per cent of that in the control. When stimulation was performed with veratridine, S₁ values obtained with the test drugs were expressed as per cent of those in the control group. If drug effects were expressed in per cent of the mean S₂/S₁ ratio or S₁ values of the corresponding control, the law of error propagation was considered for calculating the final s.e. mean values.

Data are presented as means±s.e.mean of the S₂/S₁ ratios or S₁ values for the indicated number of experiments. In experiments on inhibition of release by various drugs, all data were statistically analysed as S₂/S₁ ratios or S₁ values before transformation of data into percentage of control release.

Statistical significance was evaluated by one-way analysis of variance (ANOVA), followed by LSD multiple comparisons test when P<0.05 by using the computer program SPSS for Windows (SPSS software GmbH, version 6.0.1, München, Germany).

Drugs

[³H]-Dopamine (3,4-[7,8-³H]-dihydroxyphenylethylamine; specific activity 21.5 Ci  mmol⁻¹) was from American Radiolabeled Chemicals Inc., St. Louis, MO, U.S.A. Diltiazem, nifedipine, nitrendipine, nicardipine, (±)-Bay  K  8644, nomifensine maleate, pargyline hydrochloride were from RBI, Natick, MA, U.S.A. L-ascorbic acid, (±)-verapamil, veratridine, tetrodotoxin, ω-conotoxin-GVIA, ω-agatoxin-IVA, and ω-conotoxin-MVIIC were from Sigma, Deisenhofen, Germany. Nominally Ca²⁺ free solution was prepared by replacing CaCl₂ with equimolar MgCl₂ (1.8 mM).

Results

Pharmacological classification of VACCs involved in K⁺-evoked [³H]-DA release

Under the conditions of the present study, stimulation of the slices with 50 mM K⁺ for 2 min (S₁) caused a [³H]-DA release of 38.3 ± 1.8% (n=24) of total radioactivity in excess of basal release. When two 2 min pulses of KCl (50 mM) were applied 20 min apart, the S₂/S₁ ratio in the controls averaged 0.82±0.02 (n=24) (see inset in Figure 1). The nominal absence of Ca²⁺ with Ca²⁺ replaced by equimolar MgCl₂ completely abolished K⁺-evoked [³H]-DA release. In contrast, the Na⁺ channel blocker TTX (1 μM) did not affect K⁺-evoked [³H]-DA release (Figure 1).

Moreover, the non-selective N-, P/Q-type Ca²⁺ channel blocker ω-CTx-MVIIC (3 μM), reduced K⁺-evoked [³H]-DA release to 12% of control, this value being similar to the inhibition of release to 7% of control in the absence of external Ca²⁺ (Figure 1, difference not statistically different). Finally, K⁺ evoked [³H]-DA release was not sensitive either to nifedipine (10–30 μM, Figure 2), nitrendipine (30 μM), nicardipine (30 μM) or to the activator of L-type VACCs (±)-Bay K 8644 (1 μM) (Table 1).

Figure 2.

Concentration-response curves for the inhibition produced by nifedipine, verapamil and diltiazem on K⁺-evoked [³H]-DA release. The drugs were added 12 min before S₂ and were kept at the indicated final concentrations until the end of the experiments. The maximum inhibition was to 8.9±1.3% and 49.4±7.7% of control, for 100 μM verapamil and 100 μM diltiazem, respectively. Each point represents the mean S₂/S₁±s.e.mean for the given concentration from 4–7 independent experiments, with error bars when bigger than symbols. The control S₂/S₁ ratios were 0.79±0.02 (n=8), 0.79±0.04 (n=7) and 0.81±0.03 (n=6) for nifedipine, verapamil and diltiazem, respectively, and are defined as 100%. Statistical analysis was performed on untransformed data (S₂/S₁) by one-way ANOVA and LSD multiple comparison test. ^*P<0.05 compared with the respective controls.

Table 1.

Effects of L-type Ca²⁺ channel blockers on K⁺-evoked [³H]-DA release from rat striatal slices

In contrast to the lack of effects to the selective L-type VACC blockers of the DHP group (±)-verapamil and diltiazem inhibited K⁺-evoked [³H]-DA release in a concentration-dependent manner (Figure 2), with profound quantitative differences between verapamil and diltiazem at 100 μM (see below).

Verapamil- and diltiazem-induced inhibition of multiple classes of VACCs

In order to investigate whether (±)-verapamil and diltiazem were acting on N- or P-type VACCs in our preparation, we tested the effects of co-application of selective neurotoxins and (±)-verapamil or diltiazem. The selective N-type Ca²⁺ channel blocker ω-CTx-GVIA (1 μM) inhibited [³H]-DA release to 78.5±11.6% (n=4) and the selective P-type Ca²⁺ channel blocker ω-Aga-IVA (30 nM) to 40.5±3.9% (n=4) of control, indicating involvement of both channel subtypes (Figure 3). Applicating of 30 μM (±)-verapamil alone produced an inhibition of K⁺-evoked [³H]-DA release to 55.7±6.5% (n=7) of control (compare Figure 2), and simultaneous application of 30 μM (±)-verapamil and 1 μM ω-CTx-GVIA decreased the release to 35.4±1.6% (n=4) of control (Figure 3). In contrast, 30 μM (±)-verapamil could not enhance inhibition of release caused by P-type VACC blocker ω-Aga-IVA (30 nM) alone (Figure 3). The inhibitory effects of 30 μM diltiazem and 1 μM ω-CTx-GVIA were also fully additive, whereas the inhibition produced by co-application with 30 nM ω-Aga-IVA was the same as with ω-Aga-IVA alone (Figure 3).

Figure 3.

Inhibitory effects of ω-CTx-GVIA alone and in combination with verapamil or diltiazem and respective effects of ω-Aga-IVA on K⁺-evoked [³H]-DA release. Verapamil (or diltiazem) and ω-CTx-GVIA were simultaneously applied 12 min before S₂, whereas verapamil (or diltiazem) and ω-Aga-IVA were applied sequentially. In these experiments verapamil or diltiazem were added 12 min before S₂, 6 min later the superfusion buffer was switched to medium containing both verapamil or diltiazem and ω-Aga-IVA. All drugs were then present until the end of the experiments. The control S₂/S₁ ratio was 0.82±0.02 (n=29), and is considered as 100%. Each column represents the mean±s.e.mean (bars) values from n independent experiments. Statistical analysis was performed on untransformed data (S₂/S₁) by one-way ANOVA and LSD multiple comparison test. All drug effects are significantly different from controls. ^*P<0.05 compared with ω-CTx-GVIA alone.

Effects of verapamil and diltiazem on veratridine-evoked [³H]-DA release

From the almost complete inhibition of K⁺-evoked [³H]-DA release by 100 μM (±)-verapamil (see Figure 2) we hypothesized that Q-type Ca²⁺ channels might be involved as well. This was tested by employing veratridine-evoked [³H]-DA release, of which we have shown prevously that Ca²⁺-dependent release is mediated solely by Q-type VACCs (Dobrev et al., 1998).

Like DHPs at the concentration of 30 μM, diltiazem did not inhibit veratridine-evoked [³H]-DA release (Table 2), whereas 30 μM (±)-verapamil significantly inhibited veratridine-evoked [³H]-DA release. With 100 μM diltiazem applied the release was inhibited and with 100 μM (±)-verapamil it was blocked further (Table 2). In order to determine whether the inhibition of release by 30 μM (±)-verapamil was exclusively mediated through blocking of Q-type VACCs, we investigated the effect of this drug on the component of veratridine-evoked [³H]-DA release which was independent of external Ca²⁺. (±)-Verapamil (30 μM) partially inhibited the veratridine-evoked [³H]-DA release in the absence of external Ca²⁺ (Table 2).

Table 2.

Effects of L-type Ca²⁺ channel blockers on veratridine-evoked [³H]-DA release from rat striatal slices

Discussion

We provide evidence that in rat striatal slices verapamil inhibits P- and at higher concentrations possibly also N- and Q-type VACCs, whereas comparable concentrations of diltiazem block P-type Ca²⁺ channels only. In our experimental models of K⁺- or veratridine-evoked [³H]-DA release, L-type VACCs were not involved, since DHPs (up to 30 μM) were without effect. The latter conclusion is in agreement with a large body of evidence demonstrating that neurotransmitter release in central neurons is insensitive to modulations of L-type VACCs (Miller, 1987; Takahashi & Momiyama, 1993). Further evidence against the involvement of L-type channels comes from the lack of stimulatory effect of Bay K 8644, as this activator of L-type VACCs would have to enhance DA release if these channel subtypes played a significant role in the release process (Herdon & Nahorski, 1989).

In contrast to dihydropyridines, the L-type Ca²⁺ channel blockers (±)-verapamil (1–100 μM) and diltiazem (10–100 μM) led to a concentration-dependent inhibition of K⁺-evoked [³H]-DA release (Figure 2). Similar inhibitory effects of verapamil and diltiazem have been demonstrated in other neurotransmitter release systems such as K⁺-evoked release of endogenous glutamate from cerebellar slices (Barnes & Davies, 1988), of [³H]-D-aspartate from hippocampal slices (Mangano et al., 1991), and of [³H]-GABA from cortical synaptosomes (Carvalho et al., 1986) as well as in striatal dopamine release measured by microdialysis in freely moving rats (Kato et al., 1992). Since we have demonstrated that block of N-, P- and possibly Q-type VACCs entirely abolishes K⁺-evoked [³H]-DA release, it is difficult to envision a scheme that could also accommodate inhibition of release by (±)-verapamil and diltiazem via block of L-type VACCs. The finding that DHPs were ineffective but (±)-verapamil and diltiazem inhibited K⁺-evoked [³H]-DA release suggests the involvement of presynaptic non-L-type Ca²⁺ channels or other ion channels as for instance Na⁺ channels. Although (±)-verapamil and diltiazem selectively block L-type VACCs in concentrations up to 1 μM, they may have additional effects on neuronal Na⁺ channels at higher concentrations (Bustamante, 1985; Yatani & Brown, 1985; Yokoo et al., 1998). However, the finding that the K⁺-evoked [³H]-DA release was not modified by tetrodotoxin in the first place, excludes involvement of Na⁺ channel block by (±)-verapamil and diltiazem in this release system. Therefore, (±)-verapamil and diltiazem must block non-L-type VACCs. In other neuronal systems, both verapamil and diltiazem block P-type Ca²⁺ currents (rat cerebellar Purkinje neurons, Ishibashi et al., 1995) or inhibit N- and P/Q-type Ca²⁺ currents (mouse sensory neurons, rat motoneurons: Diochot et al., 1995). Even recombinant Ca²⁺ channels of P/Q- and R-type were inhibited by verapamil and diltiazem (Cai et al., 1997) and many of those effects occurred in the same concentration range as reported here for inhibition of dopamine release. Therefore, it is tempting to explain our findings by interactions of (±)-verapamil and diltiazem with striatal N-, P- or Q-type Ca²⁺ channels.

These possibilities were tested by employing ω-CTx-GVIA at 1 μM (saturating concentration for N-type VACCs, see Olivera et al., 1994) and ω-Aga-IVA at 30 nM (saturating concentration for P-type VACCs, see Mintz et al., 1992). Each of the two peptides inhibited a fraction of K⁺-evoked [³H]-DA release which was completely blocked by the non-selective N-, P-, Q-type VACC blocker ω-CTx-MVIIC (3 μM), suggesting also involvement of Q-type VACCs.

The inhibiting effects of 1 μM ω-CTx-GVIA and of 30 μM (±)-verapamil on K⁺-evoked [³H]-DA release were additive, indicating that both compounds acted at different channel subtypes. Since no additional (±)-verapamil-induced inhibition of release was observed on top of 30 nM ω-Aga-IVA, we conclude that (±)-verapamil acts on P-type Ca²⁺ channels in striatal neurons. Similar conclusions hold true for diltiazem. Interaction of diltiazem with P-type Ca²⁺ channels is indirectly supported by the lack of effect of the drug in a neurotransmitter release model which does not involve any P-type VACCs in the first place, i.e. the veratridine-evoked [³H]-DA release (Dobrev et al., 1998).

Increasing the (±)-verapamil concentration to 100 μM completely abolished the K⁺-evoked [³H]-DA release, indicating that at this high concentration the drug may also affect N- and Q-type VACCs in addition to P-type. Although we did not test the additive effects of 100 μM diltiazem and 30 nM ω-Aga-IVA, 100 μM diltiazem inhibited K⁺-evoked [³H]-DA release to the same degree as ω-Aga-IVA (compare Figures 2 and 3), suggesting no significant interaction of diltiazem with N- or Q-type VACCs.

Veratridine-evoked [³H]-DA release was used as an additional probe for involvement of Q-type VACCs (Dobrev et al., 1998). In the present study, the application of 30 μM diltiazem had no effect on veratridine-evoked [³H]-DA release, excluding interaction with Q-type VACCs at this concentration. In contrast, 30 μM (±)-verapamil and very high concentrations of either (±)-verapamil or diltiazem (100 μM) inhibited veratridine-evoked [³H]-DA release, suggesting possible interaction with Q-type VACCs (see Table 2). In order to test whether low concentrations of (±)-verapamil (30 μM) interact with Q-type VACCs, veratridine-evoked [³H]-DA release in the absence of external Ca²⁺ was measured. (±)-Verapamil (30 μM) reduced the release even in the absence of external Ca²⁺ and this inhibitory effect of (±)-verapamil could be explained by an antagonistic effect on VACCs only, if we assume that in the absence of external Ca²⁺, Na⁺ enters into the neurons through VACCs. However, we have recently demonstrated that the Ca²⁺-independent component of veratridine-evoked [³H]-DA release is abolished by tetrodotoxin (Dobrev et al., 1998). Moreover, the (±)-verapamil- and diltiazem-induced inhibition of release at 100 μM was much greater both than that which could be obtained with ω-CTx-MVIIC (∼35%, see Dobrev et al., 1998) and than the Ca²⁺-dependent component of release (Table 2). Inhibitory effects of L-type channel blockers on Na⁺ channels when used at high concentrations were reported in various cell types (Callanan & Keenan, 1984; Yatani & Brown, 1985; Yokoo et al., 1998; for review see McDonald et al., 1994). Thus, although we cannot exclude an interaction of (±)-verapamil and diltiazem with Q-type VACCs, it is reasonable to suggest that in the case of veratridine-evoked [³H]-DA release they inhibit the release via block of Na⁺ channels.

Finally, because of its strong lipophilicity, (±)-verapamil–especially at high concentrations–could impair channel function indirectly by perturbation of membrane fluidity. At this time, we cannot exclude such an effect of (±)-verapamil from our results. If non-specific membrane perturbation was the operative mechanism, verapamil would also be expected to inhibit the release of other neurotransmitters. However, the verapamil effects reported in other systems were highly variable including inhibition (Barnes & Davies, 1988; Mangano et al., 1991; Prince et al., 1996), enhancement (Ebstein & Daly, 1982) or even no effect at all (Bentue-Ferrer et al., 1993). In addition, verapamil did not inhibit ionomycin-evoked [³H]-D-aspartate release from hippocampal slices (Mangano et al., 1991).

In summary, we have demonstrated that at low μM concentrations (⩽30 μM), (±)-verapamil and diltiazem block P-type VACCs in rat striatum. The inhibition of P-type Ca²⁺ channels by these drugs occurs at concentrations which overlap with those required for complete block of cardiac L-type Ca²⁺ channels (Diochot et al., 1995). At concentrations >30 μM, (±)-verapamil, but not diltiazem, also appears to interact with N- and Q-type Ca²⁺ channels. The significance of these findings awaits further study, but has practical consequences, since P-type VACCs are found in many brain regions besides the striatum and in peripheral neurons. Therefore, (±)-verapamil and diltiazem are expected to modify the function of those neuronal cells through interaction with P-type VACCs. On the other hand, at higher concentrations both drugs also appear to have inhibitory effects on Na⁺ channels, so that lack of specificity of these drugs should be taken into account when probing for presence of L-type channels in central neurons. Finally, these effects may contribute to their therapeutic efficacy.

Abbreviations

ω-Aga-IVA

ω-agatoxin IVA

ANOVA

analysis of variance

DA

dopamine

ω-CTx-GVIA

ω-conotoxin GVIA

ω-CTx-MVIIC

ω-conotoxin MVIIC

KPB

Krebs-phosphate buffer

LSD-test

least significant difference test

VACCs

voltage-activated calcium channels