The auxiliary subunit gamma 1 of the skeletal muscle L-type Ca2+ channel is an endogenous Ca2+ antagonist

Abstract

Ca2+ channels play crucial roles in cellular signal transduction and are important targets of pharmacological agents. They are also associated with auxiliary subunits exhibiting functions that are still incompletely resolved. Skeletal muscle L-type Ca2+ channels (dihydropyridine receptors, DHPRs) are specialized for the remote voltage control of type 1 ryanodine receptors (RyR1) to release stored Ca2+. The skeletal muscle-specific gamma subunit of the DHPR (gamma 1) down-modulates availability by altering its steady state voltage dependence. The effect resembles the action of certain Ca2+ antagonistic drugs that are thought to stabilize inactivated states of the DHPR. In the present study we investigated the cross influence of gamma 1 and Ca2+ antagonists by using wild-type (gamma+/+) and gamma 1 knockout (gamma-/-) mice. We studied voltage-dependent gating of both L-type Ca2+ current and Ca2+ release and the allosteric modulation of drug binding. We found that 10 microM diltiazem, a benzothiazepine drug, more than compensated for the reduction in high-affinity binding of the dihydropyridine agent isradipine caused by gamma 1 elimination; 5 muM devapamil [(-)D888], a phenylalkylamine Ca2+ antagonist, approximately reversed the right-shifted voltage dependence of availability and the accelerated recovery kinetics of Ca2+ current and Ca2+ release. Moreover, the presence of gamma 1 altered the effect of D888 on availability and strongly enhanced its impact on recovery kinetics demonstrating that gamma 1 and the drug do not act independently of each other. We propose that the gamma 1 subunit of the DHPR functions as an endogenous Ca2+ antagonist whose task may be to minimize Ca2+ entry and Ca2+ release under stress-induced conditions favoring plasmalemma depolarization.

Fig. 1.

Voltage-dependent activation of Ca²⁺ current and Ca²⁺ release. (A) Successive rectangular voltage steps applied at 60-s intervals from a holding potential of −80 mV to the indicated voltage levels. (B) Voltage-activated L-type Ca²⁺ inward currents showing typical slow-onset kinetics. (C) Fura-2 fluorescence ratio signals recorded simultaneously with the signals in B. (D) Calculated flux of Ca²⁺ underlying the signals in C before and after correction for putative SR depletion (see Materials and Methods). Traces for pulse voltages between 0 and +50 mV were corrected individually. For pulse voltages of −10 mV and smaller the average value of the individually determined baseline SR contents was used for the correction (see ref. 13). (E) Current–voltage relation derived from the recordings in B. (F) Ca²⁺ conductance (circles) and SR Ca²⁺ permeability (diamonds) during the plateau of the depletion-corrected signals in D as functions of pulse voltage. Representative data were obtained from a γ−/− muscle fiber.

Fig. 2.

Voltage-dependent activation independent of D888. Normalized activation curves for Ca²⁺ release permeability and conductance as in Fig. 1F for 0 μm (circles) and 10 μM D888 (diamonds). V_0.5 and k values (in millivolts) for control and drug application are as follows: (A) Permeability, γ−/−, −11.23 ± 1.90; 8.2 ± 0.52 (6) and −13.26 ± 2.92; 8.50 ± 0.7 (3). (B) Conductance, γ−/−, 3.2 ± 0.95; 6.3 ± 0.73 (6) and 1.22 ± 0.97; 5.62 ± 0.3 (3). (C) Permeability, γ+/+, −8.46 ± 1.40; 7.80 ± 0.93 (3) and −8.24 ± 1.70; 8.10 ± 0.5 (8). (D) Conductance, γ+/+, 3.71 ± 3.20; 5.44 ± 0.61 (3) and 4.81 ± 0.84; 5.83 ± 0.4 (6). The corresponding absolute values at +50 mV for peak permeability or conductance are: (A) 5.73 ± 0.98%ms⁻¹ and 3.55 ± 0.20%ms⁻¹. (B) 180.34 ± 38.17 S F⁻¹ and 133.59 ± 14.13 S F⁻¹. (C) 5.39 ± 1.51%ms⁻¹ and 3.82 ± 0.45%ms⁻¹. (D) 109.09 ± 29.66 S F⁻¹ and 92.13 ± 8.20 S F⁻¹. The continuous lines were calculated as described in Materials and Methods by using the means of the best fit parameters.

Fig. 3.

Effect of D888 on voltage-dependent inactivation. (A) Experimental protocol to determine the voltage dependence of inactivation. (B and C) Calculated Ca²⁺ release fluxes and L-type Ca²⁺ inward currents, respectively, at different voltages in the absence of D888. (D and E) Release fluxes and inward currents, respectively, in the presence of 5 μM D888. (F and G) Normalized steady-state voltage dependence of inactivation of peak Ca²⁺ release and Ca²⁺ inward current, respectively, at 0 μm (circles) and 5 μM D888 (triangles). Both fibers were from γ−/− mice.

Fig. 4.

Changes of steady-state availability caused by D888 in γ−/− and γ+/+ fibers. Mean values of normalized steady-state inactivation at different voltages obtained with the experimental protocol of Fig. 3 for peak Ca²⁺ release flux (A and C) and L-type Ca²⁺ current (B and D), respectively. (A and B) γ−/− fibers; (C and D) γ+/+ fibers. Numbers of experiments for 0 μm (circles), 5μM (triangles), and 10 μM D888 (diamonds) were 13, 10, and 9 in Fig. 5A, 13, 9, and 9 in B, 11, 6 and 13 in C, and 11, 7, and 13 in D, respectively. Note that bars indicating SEM are often smaller than the symbols. (E and F) Parameters V_0.5 and k of voltage-dependent inactivation and their alteration by D888. Closed symbols, γ+/+; open symbols, γ−/−. Asterisks indicate significant differences.

Fig. 5.

Alterations of the time course of recovery from depolarization-induced inactivation caused by D888 in γ−/− and γ+/+ fibers. (A) Experimental protocol for studying recovery. (B and C) Representative recordings (γ+/+) showing recovery at −80 mV of Ca²⁺ release and Ca²⁺ current, respectively. (D and E) Mean time constants of recovery of peak Ca²⁺ release flux and Ca²⁺ current (end of pulse), respectively, in the absence and in the presence of 5 and 10 μM D888. Asterisks indicate significant differences between γ+/+ (filled columns) and in γ−/− (open columns) for identical drug concentration. The values at 5 μM D888 in γ−/− were not significantly different from the controls in γ+/+ (#, P = 0.56 and P = 0.28 for release and current, respectively). Error bars indicate SEM. Numbers of experiments are indicated in the figure.

Fig. 6.

Interaction of γ₁-subunit with allosteric modulation of isradipine binding by diltiazem. (A and B) Equilibrium isotherms of [³H]isradipine binding to skeletal muscle microsomes from γ+/+ (filled symbols) and γ−/− animals (open symbols) in the absence (A) and presence (B) of d-cis-diltiazem (10 μM) at 37°C. Least-squares fitting of canonical binding curves to the data resulted in the following values for B_max (pmol/mg). Control, γ+/+: 1.71 ± 0.12 (4); control, γ−/−: 0.72 ± 0.09 (4); diltiazem, γ+/+: 3.28 ± 0.15 (4); diltiazem, γ−/−: 2.97 ± 0.25 (4). Except for the difference between the latter two values, all pairwise differences were highly significant (P < 0.001). K_d values (nM) were as follows: 2.08 ± 0.37, 2.93 ± 0.79, 2.25 ± 0.26, and 2.54 ± 0.50. They showed no significant differences. (C and D) Effect of increasing concentrations of d-cis-diltiazem on the amount of [³H]isradipine (0.5 nM) bound to microsomes from γ+/+ (filled symbols) and γ−/− animals (open symbols) at 37°C (C) and 21°C (D). The figure shows representative data from individual experiments that were repeated at least twice.