Effect of the Brugada syndrome mutation A39V on calmodulin regulation of Cav1.2 channels

Abstract

Background: The L-type calcium channel Cav1.2 is important for brain and heart function. The ubiquitous calcium sensing protein calmodulin (CaM) regulates calcium dependent gating of Cav1.2 channels by reducing calcium influx, a process known as calcium-dependent inactivation (CDI). Dissecting the calcium-dependence of CaM in this process has benefited greatly from the use of mutant CaM molecules which are unable to bind calcium to their low affinity (N-lobe) and high affinity (C-lobe) binding sites. Unlike CDI, it is unknown whether CaM can modulate the activation gating of Cav1.2 channels.

Results: We examined a Cav1.2 point mutant in the N-terminus region of the channel (A39V) that has been previously linked to Brugada syndrome. Using mutant CaM constructs in which the N- and/or C-lobe calcium binding sites were ablated, we were able to show that this Brugada syndrome mutation disrupts N-lobe CDI of the channel. In the course of these experiments, we discovered that all mutant CaM molecules were able to alter the kinetics of channel activation even in the absence of calcium for WT-Cav1.2, but not A39V-Cav1.2 channels. Moreover, CaM mutants differentially shifted the voltage-dependence of activation for WT and A39V-Cav1.2 channels to hyperpolarized potentials. Our data therefore suggest that structural changes in CaM that arise directly from site directed mutagenesis of calcium binding domains alter activation gating of Cav1.2 channels independently of their effects on calcium binding, and that the N-terminus of the channel contributes to this CaM dependent process.

Conclusions: Our data indicate that caution must be exercised when interpreting the effects of CaM mutants on ion channel gating.

Figure 1

The Brugada syndrome mutation A39V disrupts N-lobe CDI of Cav1.2 channels. A) Representative Ba²⁺ (black) and Ca²⁺ (red) traces of WT and A39V-Cav1.2 channels (B) expressed with Cavβ2a/Cavα2δ in the presence of low calcium buffering (0.5 mM EGTA) and CaM₃₄. Note that the peak of the Ba²⁺ trace is normalized to that in the presence of Ca²⁺. For reference the WT-Cav1.2 Ca²⁺ trace is displayed in grey in (B). The plots shown below the current traces reflect average CDI (f₃₀₀) which is quantified by the fraction of current remaining after 300 ms (r300) in calcium, and is then subtracted from the fraction of current remaining in barium at the same time point. The f₃₀₀ value at 10 mV (arrows) is significantly less for A39V, than WT-Cav1.2 (# p ≤ 0.04 by student’s t-test). The inset bar graph displays additional f₃₀₀ values over a potential range from -10 mV to +30 mV. A significant difference is observed in the f₃₀₀ values between WT-Cav1.2 and A39V-Cav1.2 at −10, 0 and +10 mV (p ≤ 0.05 by student’s t-test), but not at +20 (p = 0.28 by student’s t-test) or +30 mV (p = 0.26 by student’s t-test). C) Under high calcium buffering (10 mM BAPTA) conditions WT-Cav1.2 channels expressed as in (A) no longer exhibit N-lobe CDI. D) A39V-Cav1.2 does not show significant N-lobe CDI compared to WT-Cav1.2 in high calcium buffering at +10 mV (arrows). For reference the WT-Cav1.2 Ca²⁺ trace (C) is displayed in grey.

Figure 2

The Brugada syndrome mutation A39V does not affect C-lobe CDI of Cav1.2 channels. A) Representative Ba²⁺ (black) and Ca²⁺ (red) traces of WT-Cav1.2 channels expressed with Cavβ2a/Cavα2δ and CaM₁₂ in the presence of high calcium buffering (10 mM BAPTA). Note that the peak of the Ba²⁺ trace is normalized to that in the presence of Ca²⁺ and that average CDI (f₃₀₀) of WT-Cav1.2 is displayed below in the graph. B) A39V-Cav1.2 channels expressed with Cavβ2a/Cavα2δ and CaM₁₂ in the presence of high calcium buffering (10 mM BAPTA). Inset graphs show average CDI (f₃₀₀) which is not statistically different (p ≤ 0.55 by student’s t-test) between the two channel types.

Figure 3

The Brugada mutation A39V does not alter N-terminal binding to CaM. A) CaM sepharose pull-down experiments of N_1-EX, A39V-N_1-EX and N₁ GFP fusion proteins in 0.5 mM Ca²⁺, or 5 mM EGTA (B) run on SDS-PAGE with corresponding lysates (C) and blotted for GFP. Black lines mark where the gel picture was cut and irrelevant samples removed. These experiments were performed twice each.

Figure 4

CaM lobe mutants differentially shift the voltage dependence of activation for WT and A39V-Cav1.2 channels. A) Current voltage relationships for WT-Cav1.2 channels expressed transiently in tsA-201 cells with Cavβ2a/Cavα2δ and recorded in barium with one of four CaM conditions: CaM_WT, CaM₁₂, CaM₃₄, or CaM₁₂₃₄. All experiments were recorded with high calcium buffering intracellularly (10 mM BAPTA). B) Current–voltage relationships for A39V-Cav1.2 channels expressed as in (A) with one of four CaM conditions: CaM_WT, CaM₁₂, CaM₃₄, or CaM₁₂₃₄. C) A bar graph displaying the half activation potentials for Cav1.2 channels recorded in barium. WT-Cav1.2 channels recorded with any CaM mutant have a significant leftward shift in the voltage-dependence of activation in barium compared to CaM_WT (*p ≤ 0.05 by one-way ANOVA). D) A bar graph displaying the voltage dependence of activation for A39V-Cav1.2 channels recorded in barium. A39V-Cav1.2 channels recorded with CaM₁₂ have a significant leftward shift in the voltage-dependence of activation in barium compared to CaM_WT (*p ≤ 0.05 by one-way ANOVA), and CaM1234 (# p ≤ 0.05 by one-way ANOVA).

Figure 5

Calmodulin lobe mutants differentially affect the kinetics of activation for WT and A39V-Cav1.2 channels in the absence of calcium. A) Plot illustrating the time to maximum activation (Tau) at various voltages for WT-Cav1.2 channels expressed transiently in tsA-201 cells with Cavβ2a/Cavα2δ and recorded in barium with one of CaM_WT, CaM₁₂, CaM₃₄ or CaM₁₂₃₄ and with 10 mM BAPTA intracellular. Voltages for which all mutant CaMs differ significantly from the CaM_WT condition (* p ≤ 0.05 by one-way ANOVA), and where CaM₃₄ and CaM₁₂₃₄ differ from CaM_WT condition (# p ≤ 0.05 by one-way ANOVA). B) Plot for the kinetics of activation of A39V-Cav1.2 channels expressed as in (A). Voltages for which CaM₁₂₃₄ differs significantly from the CaM_WT condition ($ p ≤ 0.05 by one-way ANOVA), and where CaM₁₂ and CaM₁₂₃₄ differ from CaM_WT condition (& p ≤ 0.05 by one-way ANOVA). C) Sample traces of WT-Cav1.2 channels expressed with CaM_WT and CaM₁₂₃₄ at 10 mV. Note that the CaM₁₂₃₄ trace has been normalized to that of CaM_WT and that the arrows denote peak of activation for WT-Cav1.2 with either CaM_WT (black) or CaM₁₂₃₄ (grey). D) Sample traces of A39V-Cav1.2 channels expressed with CaM_WT and CaM₁₂₃₄ at 10 mV. Note that the CaM₁₂₃₄ trace has been normalized to that of CaM_WT and that the arrows denote peak of activation for A39V-Cav1.2 with either CaM_WT (black) or CaM₁₂₃₄ (grey).