The cardiac ryanodine receptor (RyR2) is a Ca2+ channel located on the sarcoplasmic reticulum of cardiomyocytes. In regulating Ca2+ release, it maintains both electrical and contractile function, and in these tasks its normal function is modulated by the Ca2+-binding accessory protein calmodulin (CaM)1,2. Several CaM mutations have been linked to cardiac arrhythmias, such as Catecholaminergic Polymorphic Ventricular Tachycardia (CPVT), but their arrhythmogenic mechanisms are not fully resolved. This research investigates the effects of two arrhythmia-linked CaM mutations (D132E and Q136P) on RyR2 function (either directly, or by virtue of CaM’s regulation of its associated kinase: CaMKII) and how this might lead to aberrant Ca2+ signalling3.
RyR2 and CaM were recombinantly co-expressed in HEK293 cells and live-cell confocal Ca2+ imaging was used to assess recursive Ca2+ release dynamics. The effect of wild type (WT) CaM co-expression on RyR2 is inhibitory – decreasing the duration of Ca2+ release events (thereby affecting the frequency of release). Co-expression of D132E and Q136P CaM was shown to reverse this phenomenon by increasing the duration of Ca2+ release events (WT 9.79±0.54 vs D132E 12.62±0.52 vs Q136P 13.20±1.02 seconds (mean±SEM), p<0.05, one-way ANOVA test, n=11, 19 and 10 field of view containing approximately 15 cells each, for WT, D132E and Q136P CaM respectively). This finding suggests a decrease in direct regulation of RyR2 by these mutant CaMs. Additionally, in line with HEK293 data, D132E CaM significantly reduced the endoplasmic reticulum Ca2+ store load in HEK293 cells compared to WT CaM (ΔF/F0= WT 0.47, 95% CI [0.42, 0.55] vs D132E 0.24, [0.22, 0.28], p<0.05, Kruskal-Wallis test, n=150 cells for each construct, values are shown as median, 95% CI [lower confidence limit, upper confidence limit]), as determined by the addition of 10mM caffeine. It is, therefore, likely that D132E variant CaM is causing Ca2+ leakage from the channel which results in smaller Ca2+ stores.
Line-scans were used to detect Ca2+ sparks in permebilised C57BL/6 mouse ventricular myocytes in the presence of WT or mutant CaMs. D132E variant CaM was shown to significantly increase spark duration (WT 51.25, 95% CI [45, 57.5] vs D132E 58.75, 95% CI [52.5, 67.5] ms) and time to peak (WT 10, 95% CI [8, 11] vs D132E 12, 95% CI [10, 13.5] ms) in cardiac myocytes compared to WT CaM (p<0.05, Kruskal-Wallis test, n=39 and 37 cells isolated from 3 mice for WT and D132E CaM respectively, values are shown as median, 95% CI [lower confidence limit, upper confidence limit]).
The time course of CaMKII autophosphorylation showed significantly reduced autophosphorylation in the presence of Q136P CaM compared to WT (p<0.05 at each timepoint, one sample Wilcoxon signed rank test, n=5). Western analysis also revealed that Q136P CaM significantly reduced RyR2 phosphorylation at the S2814 CaMKII phosphorylation site compared to WT CaM (p<0.05, one sample Wilcoxon signed rank test, n=6). These results show that D132E and Q136P exert their dysfunction in different ways, suggesting that calmodulinopathy in CPVT is likely mechanistically complex.