No abstract

Interaction between N-terminal 1– 600 and central domains 2000 –2500 of ryanodine receptor (RyR2), where many mutations have been found in patients with arrhythmogenic right ventricular cardiomyopathy type 2 (ARVC2) or catecholaminergic polymorphic ventricular tachycardia (CPVT), was recently found to play a critical role in channel regulation. Using the domain peptide approach, here, we investigated the role of calmodulin (CaM), one of the accessory proteins of the cardiac ryanodine receptor (RyR2), on Ca 2+ release kinetics. Sarcoplasmic reticulum (SR) vesicles were isolated from dog LV muscles (n=4), then RyR2 was fluorescently labeled with methylcoumarin acetate (MCA) using DP 163–195 , which harbors a human mutation site in ARVC (R176Q), as a site-directing carrier. DP 163–195 mediated a specific MCA fluorescence labeling of central domain (60Kd) of RyR2. Addition of DP 163–195 to the MCA-labeled SR competitively induced the domain unzipping between N-terminal and central domains, as evidenced by an increased accessibility of the bound MCA to a large-size fluorescence quencher. In saponin-permeabilized cardiomyocytes, the addition of DP 163–195 markedly increased the frequency of Ca 2+ sparks (SpF; s −1 ·100μm −1 :13.1±0.9, p<0.01), compared with normal cells (6.9±0.3). Addition of recombinant CaM (100nM), in the presence of KN-93 (CaMKII inhibitor), inhibited the DP 163–195 -induced increase in SpF (7.2±0.5, p<0.01). This effect of CaM was, however, abolished by co-addition of the antibody against the binding site of CaM within RyR2 (3583–3603) (SpF:12.9±0.7, p=ns), strongly suggesting that the binding of CaM to RyR2 corrects the abnormal Ca 2+ release induced by DP 163–195 . The mutation made in the domain peptide, mimicking the same human mutation in ARVC (R176Q) abolished all of the effects that would have been produced by DP 163–195 . In conclusion, the mutation-linked defective inter-domain interaction between N-terminal and central domains within RyR2 (viz. domain unzipping) may increase spontaneous Ca 2+ release, perhaps by weakening CaM-binding to RyR2. Restored binding of CaM to RyR2 may correct defective channel gating of the mutant RyR2.

Mutations in cardiac ryanodine receptor 2 (RyR2) have been shown to be associated with catecholaminergic polymorphic ventricular tachycardia (CPVT). To study the underlying mechanism of this lethal arrhythmia, we developed knock-in (KI:RyR2 S2246L/+ ) mice model with the human CPVT-associated RyR2 mutation (S2246L). The KI mice revealed no structural or histological abnormality in hearts, and also had no contractile or relaxation dysfunction at rest. In all KI mice (n=6), however, bi-directional ventricular tachycardia (VT) was observed after exercise with treadmill (Ex:6/6), but not observed in wild-type (WT) mice (Ex:0/7). In isolated cardiomyocytes, line scan images were obtained to measure local Ca 2+ release events using a confocal microscopy with fluo-4 AM as a Ca 2+ indicator. In the KI cardiomyocytes, the frequency of Ca 2+ sparks (SpF: s −1 ·100μm −1 ) was much more increased in response to 100 nM isoproterenol than in WT cardiomyocytes (KI:4.7±0.5 vs WT:1.9±0.2, p<0.01). Using the canine cardiac SR, we fluorescently labeled RyR2 with methylcoumarin acetamido (MCA), using either DP 2232–2266 or DP 2232–2266 mut; Ser is mutated to Leu (S2246L), as a carrier; DP 2232–2266 harbors the same CPVT mutation site as KI mice (S2246L). The binding affinity of DP 2232–2266 mut to RyR2 (Kd=0.08μM) was higher than that of DP 2232–2266 (Kd=0.32μM), suggesting that abnormally tight interaction of the domain pair (between the domain2232–2266 and another putative domain) may be formed by the S2246L mutation. Interestingly, addition of DP 2232–2266 to the SR dose-dependently inhibited the cAMP (30μM)-induced Ca 2+ leak; IC50=0.1μM, although DP 2232–2266 had no effect on the cAMP-induced increase in 2808Ser phosphorylation. In conclusion, some type of RyR2 mutation in CPVT may causatively induce hyper-activated channel gating by forming abnormally tight domain-domain interaction, triggering diastolic Ca 2+ release and hence lethal arrhythmia. Interruption of such abnormal domain-domain interaction (by competing with native domain) may lead to a new therapeutic strategy against CPVT.

We previously reported that interaction of N-terminal (0–600) and central domains (2000–2500) of the cardiac ryanodine receptor (RyR2), harboring many mutations associated with CPVT, is defective (i.e. domain unzipping) in failing hearts. Here, we examined the pathogenic role of the inter-domain interaction within RyR2 in the mutation-linked channel disorder in human CPVT-associated RyR2 R2474S/+ knock-in (KI) mice model. In all KI mice (6/6), bi-directional ventricular tachycardia was observed during or after exercise with treadmill, but not in wild-type (WT) mice (0/7). Sarcoplasmic reticulum (SR) vesicles were isolated (n=4), then RyR2 was fluorescently labeled with methylcoumarin acetamido (MCA) using DP 2460–2495 (DPc10), harboring the same CPVT mutation site as KI mice; R2474S, as a site-directing carrier. Only in KI (but not in WT) mice, partial domain unzipping was taken place at baseline, and further domain unzipping occurred in response to cAMP (1 μM), as evidenced by an increased accessibility of the bound MCA to a large-size fluorescence quencher. In the saponin-permiabilized KI cardiomyocytes; [Ca 2+ ] was buffered at 30 nM by 0.5 mM EGTA, the relationship curve between the frequency of Ca 2+ sparks (SpF: s −1 ·100μm −1 ; by Rhod- 2) and SR Ca 2+ content {Ca 2+ con ; by caffeine application}, obtained by incremental addition of cAMP (0.1–1 μM), was markedly shifted to the left (towards lower SR Ca 2+ content) compared to WT cardiomyocytes, although RyR2 2808Ser phosphorylation is similarly increased in WT and KI. For instance, at 1 μM of [cAMP], Ca 2+ con decreased by 43 % from WT, although SpF was more increased in KI than WT (KI: 19.6±0.5; WT: 14.7±0.4, p<0.01). Interestingly, addition of DPc10 (50 μM), which was found to induce similar domain unzipping (competing with native domain) in WT as KI (+cAMP), to WT cardiomyocytes reproduced the leftward shift of the SpF-Ca 2+ con curve seen in KI. In conclusion, single point mutation at specific region may causatively induce defective inter-domain interaction between N-terminal and central domains in response to PKA phosphorylation, and in turn enhance the sensitivity of the channel to activation by luminal [Ca 2+ ]: i.e. decreased threshold [Ca 2+ ] L to induce spontaneous Ca 2+ sparks, leading to CPVT.

Calmodulin (CaM), one of the accessory proteins of the cardiac ryanodine receptor (RyR2), has been shown to play a significant role in regulating the channel gating in RyR2. However, the role of calmodulin on the channel gating in diseased hearts is unclear. Here, using knock-in (KI) mice model with CPVT-associated RyR2 mutation (R2474S), we investigated whether CaM plays a role in the pathogenesis of CPVT. Sarcoplasmic reticulum (SR) vesicles were isolated from mouse hearts (n=4). To evaluate the characteristics of CaM binding to RyR2, SR was mixed with various concentrations of CaM-SANPAH conjugate (16nM-256nM) , followed by UV photolysis. Then, the RyR2-bound CaM was detected by Western blotting using anti-CaM antibody. There was no significant difference in the affinity of CaM binding to RyR2 at baseline between KI (Kd=26nM) and WT (Kd=16nM). However, compared to WT, CaM binding was markedly decreased in response to cAMP (1 μM) in KI (18±1.9 % decrease, at 128nM CaM), although 2808Ser phosphorylation in RyR2 was similarly increased by cAMP in WT and KI. This decrease of CaM binding in KI was, however, markedly diminished by co-addition of dantrolene which was previously found to prevent the abnormal Ca 2+ leak by the correction of the defective inter-domain interaction between N-terminal (1–600) and central (2000–2500) domains in both MH-type and CPVT-type RyR2. Moreover, in the presence of FK506 (10μM), which was found to dissociate FKBP12.6 from RyR2 and subsequently to induce domain unzipping, CaM binding was indeed decreased in WT, whereas unchanged by FK506 in KI. In saponin-permeabilized, isolated cardiomyocytes, line scan images in cardiomyocytes were obtained to measure local Ca 2+ release events using a confocal microscopy with Rhod-2 as a Ca 2+ indicator; [Ca 2+ ] was buffered at 30 nM by 0.5 mM EGTA. In the KI cardiomyocytes, the frequency of Ca 2+ sparks (SpF:s −1 ·100μm −1 ) was more increased than in WT, in response to cAMP (at 1μM; +33%, p<0.01 vs WT). In conclusion, CaM binding ability to RyR2 seems to be markedly decreased in response to PKA phosphorylation in KI mice, perhaps via defective inter-domain interaction, causing spontaneous diastolic Ca2+ sparks that may lead to triggered activity and hence lethal arrhythmia.