Many excitable cells contain at least two different voltage-dependent Ca channels (L- and T-type). The cardiac, slow, L-type Ca channel is further modulated by cyclic AMP-dependent phosphorylation, which increases the probability of it being open, and is readily blocked by Ca channel blockers including dihydropyridines and phenylalkylamines. The tritiated congeners of these blockers bind in vitro to sites which have the same pharmacological characteristics as those observed in vivo, that is, stereospecific and allosteric interaction between distinct sites. The dihydropyridine-binding site purified from skeletal muscle t-tubules contains three peptides of relative molecular mass (Mr) 142,000 (142K), 56K and 31K. The cAMP kinase incorporates one mol phosphate per mol of the 142K peptide and binding of (+)PN-200/110, a potent Ca antagonist, is allosterically affected by D-cis-diltiazem and verapamil. The purified dihydropyridine-receptor complex has also been incorporated into phospholipid bilayer membranes. Here, we show for the first time that the complex can be reconstituted to form a functional 20-pS Ca channel that retains the principal regulatory, biochemical and pharmacological properties of membrane-bound L-type Ca channels.

The dihydropyridine receptor was purified from rabbit skeletal muscle microsomes in the presence of [3H]nitrendipine plus diltiazem or [3H](+)PN 200-1 10 to an apparent density of 1.5 -2 nmol binding sites/mg protein. Sodium dodecyl sulfate gel electrophoresis in the absence of reducing agents yielded three peptide bands of 142, 56 and 30 kDa in a relative ratio of 11 : 1 : 1.3, whereas in the presence of 40 mM dithiothreitol bands of 142, 122, 56, 31, 26 and 22 kDa were obtained in a relative ratio of 5.5:2.2: 1 :0.9: 14:0.09. This gel pattern was observed regardless of whether the receptor was purified as a complex with nitrendipine plus diltiazem or with (+)PN 200-1 10. CAMP-dependent protein kinase phosphorylated preferentially the 142-kDa band up to a stoichiometry of 0.82 f 0.07 (15) mol phosphate/mol peptide. The 56-kDa band was phosphorylated only in substoichiometric amounts. These results suggest that the purified dihydropyridine receptor retains the basic properties of the membranebound receptor and contains separate sites for at least dihydropyridines and phenylalkylamines.Most excitable cells contain two types of voltage-regulated calcium channels, namely a transient (t) and a long-lasting (1) channel [l -41. The opening of the cardiac muscle 1-type depends on membrane depolarization to less then -35 mV and is facilitated by CAMP-dependent phosphorylation of the channel [5-91. Both types of calcium channels are present in skeletal muscle [4]. The skeletal muscle 1-type channel shares many electrophysiological properties with the cardiac muscle 1-type channel, although these channels are not identical [lo]. Apparently, the skeletal muscle 1-type channel is affected by CAMP-dependent phosphorylation [ll]. The conductance of -1 10,isopropy1-4-(2,1,3-benzoxadia-dine-3-carboxylate; PhMeS02F, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; WGA, wheat germ agglutinin; I&, concentration required to inhibit 50% of a maximal effect. the 1-type but not the t-type calcium channel is blocked by a heterogenous group of organic compounds which include the dihydropyridines nitrendipine and PN 200-1 10, the phenylalkylamines verapamil and desmethoxyverapamil and the benzothiazepine diltiazem. The tritiated congeners of these compounds bind to specific sites in skeletal and cardiac muscle membranes 112 -191. The high-affinity binding sites for these compounds interact allosterically with each other, suggesting that the membrane-receptor complex contains distinct sites for dihydropyridines, phenylalkylamines and presumably diltiazem [12-15, 17, 181. The high-affinity sites have been considered to represent the voltage-operated calcium channel, although until recently a direct proof has not been presented for this suggestion.Purification of the 1-type calcium channel has not been achieved from cardiac muscle since this muscle contains a very low density of this channel [17, 181. In contrast, t-tubular membranes of skeletal muscle have a high density of binding sites for dihydropyridines [13,15] and have b...

The distribution of the bovine cardiac binding sites for the organic calcium-channel blockers was studied. Crude microsomal membranes were separated into three fractions, which contained mainly membranes derived from sarcolemma, 'junctional' sarcoplasmic reticulum containing transversal tubuli, and free sarcoplasmic reticulum. The high-affinity binding site for the dihydropyridines, determined in the presence of nitrobenzylthioinosine, was enriched 12-fold and 17-fold in sarcolemma and junctional sarcoplasmic reticulum. The binding sites for the phenylalkylamines, determined with [3H]verapamil or [3H](-)desmethoxyverapamil, were enriched 1.5-3.4-fold in sarcolemma and junctional sarcoplasmic reticulum but 6-10-fold in free sarcoplasmic reticulum. The phenylalkylamine-binding site, present in free sarcoplasmic reticulum, was partially destroyed by chymotrypsin or phospholipase A2 and C treatment. Specific binding was proportional to the concentration of the added membrane protein. The binding of (-)desmethoxyverapamil was half-maximally inhibited by 6.5 mM calcium chloride and was optimal in the presence of 5 mM EGTA. In three out of five preparations (-)desmethoxyverapamil bound to a single site with an apparent Kd value of 191 +/- 42.8 nM and a density of 34.5 +/- 7.7 pmol/mg protein. In two out of five preparations an additional high-affinity site (Kd approximately 0.67 nM) was detected. The low-affinity site bound other phenylalkylamines, but stereospecific binding of phenylalkylamines was not observed. Binding of phenylalkylamines to the low-affinity site was inhibited by some but not all calmodulin 'antagonists'. Furthermore dihydropyridines did not affect the binding of (--)desmethoxyverapamil suggesting that the low-affinity site differs considerably from the high-affinity sarcolemmal site. These results suggest that free sarcoplasmic reticulum contains a binding site for phenylalkylamines at a relative high density, which is not related to the high-affinity site present in the voltage-dependent calcium channel.

Nonionic and ionic detergents were used to solubilize the bovine cardiac sarcolemmal binding sites for nimodipine and (-)desmethoxyverapamil in the absence of added ligand. Only Chaps, digitonin and sucrose monolauryl ester were able to solubilize the binding sites in a form that bound radioligands. About 45% of each of the membrane-bound high-affinity site was solubilized by 0.4% Chaps (w/v) in the presence of 48% (w/v) glycerol. The solubilized binding sites were destroyed by trypsin or by a 10-min incubation at 50°C. Calcium stimulated nimodipine binding slightly at 0.3 mM and inhibited (-)desmethoxyverapamil binding completely with an ICs0 of 1.2 mM. Nimodipine binding was reduced by 20% in the presence of EGTA. The solubilized receptors sedimented in sucrose density gradients with an apparent szo,w of 21 S . An identical sedimentation value was obtained for the cardiac sarcolemmal and skeletal transverse tubulus receptor which were prelabeled with nitrendipine and solubilized by digitonin.Solubilization reduced the affinity of nimodipine for its high-affinity site slightly from 0.35 nM to 1.2 nM and that for its low-affinity site from 33 nM to 130 nM. Solubilization did not affect significantly the specific density of these sites. Binding of nimodipine to the low-affinity site was completely abolished by 0.1 pM nitrobenzylthioinosine. After solubilization only the high-affinity site for (-)desmethoxyverapamil could be measured with tenfold reduced affinity (Kd = 15.3 nM) but unchanged specific density. Binding to the solubilized high-affinity site for nimodipine and (-)desmethoxyverapamil was stereospecific and showed a similar rank order as the particulate binding sites. Binding of nimodipine was inhibited allosterically by phenylalkylamines. Similarly, (+)PN200-110 inhibited allosterically (-)desmethoxyverapamil binding. d-cis-Diltiazem stimulated nimodipine binding at 20°C 1.2-fold, reduced the dissociation rate from 0.018 min-' to 0.0083 min-' and had no effect on the association rate (0.173 min-' . nM-'). The Kd calculated from the rate constants was 0.1 nM and in close agreement with the value of 0.49 nM measured under equilibrium conditions in the presence of nitrobenzylthioinosine. In contrast, desmethoxyverapamil increased the dissociation rate of nimodipine to 0.03 min-'. The association and dissociation rate constants for (-)desmethoxyverapamil were 0.024 min-' . nM -and 0.025 min-', respectively. d-cis-Diltiazem inhibited (-)desmethoxyverapamil binding in an noncompetitive manner with an ICs0 value of 9 pM and stimulated 3.4-fold the dissociation rate of (-)desmethoxyverdpdmil, suggesting that the solubilized membranes contained a third site specific for diltiazem.These results suggest that the cardiac sarcolemmal binding sites for nimodipine, (-)desmethoxyverdpamil and d-cis-diltiazem are solubilized in a form which is kinetically indistinguishable from the membrane-bound binding sites. They suggest further that these binding sites are localized either on distinct domains of the same protein or...