Abnormal Termination of Ca²⁺ Release Is a Common Defect of RyR2 Mutations Associated With Cardiomyopathies

Abstract

Rationale

Naturally occurring mutations in the cardiac ryanodine receptor (RyR2) have been associated with both cardiac arrhythmias and cardiomyopathies. It is clear that delayed afterdepolarization resulting from abnormal activation of sarcoplasmic reticulum Ca²⁺ release is the primary cause of RyR2-associated cardiac arrhythmias. However, the mechanism underlying RyR2-associated cardiomyopathies is completely unknown.

Objective

In the present study, we investigate the role of the NH₂-terminal region of RyR2 in and the impact of a number of cardiomyopathy-associated RyR2 mutations on the termination of Ca²⁺ release.

Methods and Results

The 35-residue exon-3 region of RyR2 is associated with dilated cardiomyopathy. Single-cell luminal Ca²⁺ imaging revealed that the deletion of the first 305 NH₂-terminal residues encompassing exon-3 or the deletion of exon-3 itself markedly reduced the luminal Ca²⁺ threshold at which Ca²⁺ release terminates and increased the fractional Ca²⁺ release. Single-cell cytosolic Ca²⁺ imaging also showed that both RyR2 deletions enhanced the amplitude of store overload-induced Ca²⁺ transients in HEK293 cells or HL-1 cardiac cells. Furthermore, the RyR2 NH₂-terminal mutations, A77V, R176Q/T2504M, R420W, and L433P, which are associated with arrhythmogenic right ventricular displasia type 2, also reduced the threshold for Ca²⁺ release termination and increased fractional release. The RyR2 A1107M mutation associated with hypertrophic cardiomyopathy had the opposite action (ie, increased the threshold for Ca²⁺ release termination and reduced fractional release).

Conclusions

These results provide the first evidence that the NH₂-terminal region of RyR2 is an important determinant of Ca²⁺ release termination, and that abnormal fractional Ca²⁺ release attributable to aberrant termination of Ca²⁺ release is a common defect in RyR2-associated cardiomyopathies.

The cardiac ryanodine receptor (RyR2) is the major Ca²⁺ release channel of the sarcoplasmic reticulum (SR) and plays an essential role in excitation– contraction coupling and SR Ca²⁺ homeostasis.¹ Abnormal SR Ca²⁺ handling attributable to defective RyR2 function is a well-known cause of ventricular tachyarrhythmias and sudden death.^2,3 Naturally occurring RyR2 mutations have been linked to catecholaminergic polymorphic ventricular tachycardia (CPVT) and catecholaminergic idiopathic ventricular fibrillation.^4–6 More than 150 disease-associated RyR2 mutations have been identified to date.^6,7 Most of these RyR2 mutations are associated with stress-induced ventricular tachyarrhythmias and sudden death in structurally normal hearts. However, some of them are associated with cardiomyopathies as well as cardiac arrhythmias.^6,7 For example, an in-frame deletion of 35 amino acid residues (Asn57-Gly91) in the NH₂-terminal region, corresponding to exon-3 of the RYR2 gene, was identified in several unrelated families. This deletion is associated with an expanding spectrum of phenotypes that include sinoatrial nodal dysfunction, atrial fibrillation, atrioventricular block, decreased left ventricular function, increased trabeculation, CPVT, and dilated cardiomyopathy (DCM).^7–9 Furthermore, a number of RyR2 NH₂-terminal point mutations, including A77V, R176Q/T2504M, R420W, and L433P, are associated with arrhythmogenic right ventricular displasia type 2 (ARVD2).^6,7,10–14 RyR2 mutations also may be associated with hypertrophic cardiomyopathy (HCM).¹⁵ A definitive link between RyR2 mutations and a specific type of cardiomyopathy has not been firmly established because of the small number of RyR2 mutation carriers and their variable clinical phenotypes. However, an increasing body of evidence suggests that defective RyR2 function can lead not only to cardiac arrhythmias but also to cardiomyopathies.

An important question is how the various RyR2 mutations produce all the different phenotypes. One possibility is that different RyR2 mutations may alter different aspects of SR Ca²⁺ release. The initiation/activation of SR Ca²⁺ release is normally well-controlled by membrane depolarization via a mechanism known as Ca²⁺-induced Ca²⁺ release.¹ Under conditions of SR Ca²⁺ overload, however, SR Ca²⁺ release in the form of Ca²⁺ waves can occur spontaneously in the absence of membrane depolarization.^3,16–21 These spontaneous Ca²⁺ waves, also termed store overload-induced Ca²⁺ release (SOICR),^22,23 may result in delayed afterdepolarizations (DADs) and triggered arrhythmia.^2,3,20 We have recently demonstrated that a number of RyR2 mutations associated with CPVT reduce the threshold for SOICR.^22,23 A reduced SOICR threshold will increase the propensity for DADs and, thus, triggered arrhythmias. Therefore, abnormal activation of Ca²⁺ release attributable to inappropriate opening of RyR2 represents a common mechanism for RyR2-associated CPVT.⁶

The mechanism for RyR2-associated cardiomyopathies remains unknown. Recent studies demonstrate that SR Ca²⁺ stores are only partially depleted during local or depolarization-induced global Ca²⁺ transients,^24–26 indicating that SR Ca²⁺ release terminates well before the store is empty. The process that terminates Ca²⁺ release is thought to be critical for maintaining stable excitation– contraction coupling and controlling the cytosolic Ca²⁺ transient.^27–30 Abnormal cytosolic Ca²⁺ transients have been proposed to trigger the cardiac remodeling^31,32associated with cardiomyopathies.^33–35 It is therefore possible that RyR2 mutations associated with cardiomyopathies may alter the process of Ca²⁺ release termination. To test this possibility, we assessed the impact on Ca²⁺ release of a number of NH₂-terminal RyR2 mutations that are associated with cardiomyopathies. We report that the NH₂-terminal region of RyR2 is an important determinant of Ca²⁺ release termination. Further, we show that altered Ca²⁺ release termination is a common defect of RyR2 mutations associated with cardiomyopathies.

Methods

Site-Directed Mutagenesis and DNA Transfection

All RyR2 mutations were generated using the overlap extension method as previously described.^23,36 HL-1 cardiac cells were transfected with the GFP-tagged RyR2 wild-type (WT) or the GFP-tagged RyR2 mutant, Del-Exon-3, using the Nucleofector kit (Amaxa) according to the manufacturer’s instructions.

Generation of Stable Inducible HEK293 Cell Lines Expressing RyR2 WT and Mutants

Stable inducible HEK293 cell lines expressing RyR2 WT or mutants were generated using the Flp-In T-REx Core kit (Invitrogen Life Technologies) according to the manufacturer’s instructions.

Single-Cell Ca²⁺ Imaging

Luminal Ca²⁺ transients in HEK293 cells expressing RyR2 WT or mutant channels were measured using the Ca²⁺-sensitive fluorescence resonance energy transfer (FRET)-based cameleon protein D1ER.^37,38 Stable inducible HEK293 cells were transfected with the D1ER cDNA 24 hours before the induction of RyR2 expression. Fluorescent images were captured using an inverted microscope (Nikon TE2000-S) with S-Fluor 20×/0.75 objective. The amount of FRET was determined from the ratio of the emissions at 535 and 470 nm (excitation at 430 nm). Cytosolic Ca²⁺ transients in these cells were monitored using the fluorescent Ca²⁺ indicator dye Fura-2 acetoxymethyl ester as previously described.^22,23 Time-lapse images (0.5 frame/s) were captured and analyzed with the Compix Simple PCI 6 software (Compix). Fluorescence intensities were measured from regions of interest centered on individual cells.

Detailed Methods are provided in the Online Supplement.

Results

The NH₂-Terminal Region of RyR2 Is an Important Determinant of Ca²⁺ Release Termination

There are more than 20 disease-causing mutations in the NH₂-terminal region of RyR2.^6,7 The functional role of this region and the consequences of these mutations are unclear. To gain insights into these unknowns, we used a deletion approach in which we removed the first 305 NH₂-terminal residues of RyR2 (Del-305) (Online Figure I). Ca²⁺ release assays and immunoblotting analysis revealed that the Del-305 mutant was expressed in HEK293 cells and remained functional (Online Figure IB, IC). Thus, deletion of the first 305 NH₂-terminal residues of RyR2 did not abolish its expression or function.

We used a FRET-based endoplasmic reticulum (ER) luminal Ca²⁺-sensing protein D1ER³⁷ to monitor the ER luminal Ca²⁺ dynamics and to determine whether the NH₂-terminal deletion alters SOICR. As shown in Figure 1, elevating extracellular Ca²⁺ from 0 to 2 mmol/L induced SOICR in HEK293 cells expressing RyR2 WT (observed as the downward deflections of the FRET signal; Figure 1A). SOICR occurred when the ER Ca²⁺ increased to a threshold level (the SOICR activation threshold, F_SOICR) and terminated when the ER Ca²⁺ declined to another threshold level (the SOICR termination threshold, F_termi; Figure 1A). This SOICR in HEK293 cells expressing RyR2 WT terminated at a threshold of 57% store capacity, which is similar to that (60%) observed in cardiomyocytes.²⁶ Note that SOICR is mediated by RyR2, not by any endogenous inositol-1,4,5-trisphosphate receptors (IP3Rs) that may be present, because xestospongin C, an inhibitor of IP3Rs, had no effect on SOICR in HEK293 cells expressing RyR2 WT (Online Figure II).

Figure 1. Effect of deleting the first 305 NH₂-terminal residues of cardiac ryanodine receptor (RyR2) on store overload-induced Ca²⁺ release (SOICR).

Stable inducible HEK293 cell lines expressing RyR2 wild-type (WT) or RyR2-Del-305 were transfected with the fluorescence resonance energy transfer (FRET)-based endoplasmic reticulum (ER) luminal Ca²⁺ sensing protein D1ER 48 hours before single-cell FRET imaging. The expression of RyR2 WT and Del-305 was induced 24 hours before imaging. The cells were perfused with KRH buffer containing increasing levels of extracellular Ca²⁺ (0–2 mmol/L) to induce SOICR. This was followed by the addition of 1.0 mmol/L tetracaine to inhibit SOICR, and then 20 mmol/L caffeine to deplete the ER Ca²⁺ stores. FRET recordings from representative RyR2 WT (A) and Del-305 (B) cells (113–139) are shown. The activation threshold (C) and termination threshold (D) were determined using the equations shown (A). F_SOICR indicates the FRET level at which SOICR occurs, whereas F_termi represents the FRET level at which SOICR terminates. The fractional Ca²⁺ release (E) was calculated by subtracting the termination threshold from the activation threshold. The maximum FRET signal F_max (F) is defined as the FRET level after tetracaine treatment. The minimum FRET signal F_min (G) is defined as the FRET level after caffeine treatment. The store capacity (H) was calculated by subtracting F_min from F_max. Data shown are mean±SEM (n=7–9). *P<0.01 vs WT.

SOICR was also observed in HEK293 cells expressing the RyR2 Del-305 mutant (Figure 1B). The SOICR in the Del-305 mutant cells exhibited a marked reduction in the termination threshold (35% vs 57% in WT; P<0.01) and a slightly lowered activation threshold (90% vs 94% in WT; P<0.01; Figure 1C, 1D). As a result, the fractional Ca²⁺ release during SOICR (activation threshold – termination threshold) is significantly greater in the Del-305 cells (55%) than in the WT cells (36%; P<0.01; Figure 1E). There were no significant differences in the maximum FRET signal (F_max) obtained after tetracaine treatment, the minimum FRET signal (F_min) obtained after 20 mmol/L caffeine application, or the store capacity (F_max − F_min) between the WT and Del-305 mutant cells (Figure 1F–H). Note that the D1ER probe was not saturated in HEK293 cells (data not shown), similar to that reported previously.³⁹ These studies show that the NH₂-terminal region of RyR2 has an important role in the termination of Ca²⁺ release.

Deletion of Exon-3 in RyR2 Reduces the Threshold for Ca²⁺ Release Termination

We next determined whether the cardiomyopathy-associated deletion of exon-3 (Del-exon-3) within the NH₂-terminal region (Online Figure IIIA) alters Ca²⁺ release termination. Online Figure III shows that the Del-exon-3 mutant formed functional RyR2s in HEK293 cells (Online Figure IIIB) and was expressed at a level similar to that of RyR2 WT (Online Figure IIIC). The ER luminal Ca²⁺ dynamics in these Del-exon-3 cells was assessed. SOICR in the Del-exon-3 cells had a slightly lower activation threshold (88% vs 94% in WT; P<0.01) and a markedly reduced termination threshold (39% vs 57% in WT; P<0.01; Figure 2A–D). The fractional Ca²⁺ release in the Del-exon-3 cells (49%) was also significantly greater than in the WT cells (36%; P<0.01; Figure 2E). There were no significant differences in F_max, F_min, or store capacity between the WT and Del-Exon-3 cells (Figure 2F–H). These results demonstrate that the cardiomyopathy-associated exon-3 deletion in RyR2 reduces the threshold for Ca²⁺ release termination and increases fractional release in HEK293 cells.

Figure 2. Disease-causing cardiac ryanodine receptor (RyR2) mutation, exon-3 deletion, reduces the threshold for Ca²⁺ release termination.

D1ER fluorescence resonance energy transfer (FRET) signals were recorded in HEK293 cells expressing RyR2 wild-type (WT) (A) or the RyR2 exon-3 deletion mutant (Del-exon-3) (B) using single-cell FRET imaging. The activation threshold (C), termination threshold (D), fractional Ca²⁺ release (E), F_max (F), F_min (G), and store capacity (H) in RyR2 WT (139) and Del-exon-3 (111) cells were determined as described in Figure 1. Data shown are mean±standard error of the mean (SEM; n=6–9). *P<0.01 vs WT.

Exon-3 Deletion Enhances the Propensity for SOICR and the Magnitude of Ca²⁺ Transients

A reduced termination threshold would increase the magnitude of Ca²⁺ release into the cytosol. To directly test this, we monitored the cytosolic Ca²⁺ transient using the cytosolic Ca²⁺ dye, Fura-2 acetoxymethyl ester. In both the WT (Figure 3A) and Del-Exon-3 (Figure 3B) cells, increasing extracellular Ca²⁺ induced SOICR observed as oscillatory cytosolic Ca²⁺ transients. Importantly, the Del-exon-3 mutation increased the propensity (Figure 3C), amplitude (127%±1.3% of WT; P<0.05; Figure 3D), and frequency (108%±2.2% of WT; P<0.05; Figure 3E) of SOICR. Similarly, we found that the Del-305 deletion also enhanced the propensity and amplitude of SOICR (Online Figure IV). Note that the store Ca²⁺ contents in the RyR2 WT (100%), Del-exon-3 (98%±1.0% of WT), and Del-305 (104%±3.9% of WT) cells were not significantly different.

Figure 3. Exon-3 deletion enhances the occurrence, amplitude, and frequency of store overload-induced Ca²⁺ release (SOICR) in HEK293 cells.

Stable inducible HEK293 cells expressing ryanodine receptor (RyR2) wild-type (WT) and the RyR2 exon-3 deletion mutant (Del-Exon-3) were loaded with 5 µmol/L Fura-2 acetoxymethyl ester (Fura-2 am) in KRH buffer. The cells were then perfused continuously with KRH buffer containing increasing levels of extracellular Ca²⁺ (0–2 mmol/L) to induce SOICR. Fura-2 ratios of representative RyR2 WT (A) and Del-exon-3 (B) cells were recorded using single cell Ca²⁺ imaging. C, The percentages of RyR2 WT (337) and Del-exon-3 (443) cells that display Ca²⁺ oscillations at various extracellular Ca²⁺ concentrations. The amplitude (D) and frequency (E) of SOICR in RyR2 WT and Del-exon-3 cells were determined by measuring the averaged peak amplitude and frequency of Ca²⁺ oscillations at 2 mmol/L extracellular Ca²⁺ and was normalized to that in the RyR2 WT cells (100%). Data shown are mean±standard error of the mean (SEM; n=8–9). ^#P<0.05. *P<0.01 vs WT.

To determine whether the Del-exon-3 deletion alters SOICR in cardiac cells, we transfected the HL-1 cardiac cells (a mouse atrial cell line) with the GFP-tagged RyR2 WT or GFP-tagged Del-exon-3 mutant. Cytosolic Ca²⁺ transients indicative of SOICR are shown in Figure 4. The Del-exon-3 deletion significantly increased the occurrence (Figure 4C) and amplitude (193%±1.0% of WT; P<0.01; Figure 4D), but not the frequency (133%±13% of WT; P=0.06; Figure 4E) of SOICR in HL-1 cardiac cells. There was no significant difference in the store Ca²⁺ contents between the WT (100%) and Del-exon-3 (104%±1.0% of WT; P=0.74) transfected HL-1 cardiac cells. Overall, these results suggest that the cardiomyopthy-associated exon-3 deletion promotes SOICR by reducing its activation threshold and increases the amplitude of Ca²⁺ release (fractional release) by reducing its termination threshold.

Figure 4. Effect of the cardiac ryanodine receptor (RyR2) exon-3 deletion on store overload-induced Ca²⁺ release (SOICR) in mouse HL-1 cardiac cells.

Mouse HL-1 cardiac cells were transfected with a GFP-tagged RyR2 wild-type (WT) or a GFP-tagged Del-exon-3 mutant using the Nucleofection method (Amaxa). Transfected cells were loaded with 5 µmol/L Fura-2 acetoxymethyl ester (Fura-2 am) and were then perfused continuously with KRH buffer containing increasing levels of extracellular Ca²⁺ (0.3–5 mmol/L). Fura-2 ratios of representative HL-1 cardiac cells expressing the GFP-tagged RyR2 WT (A) and the GFP-tagged Del-exon-3 mutant (B), identified based on their GFP fluorescence, were recorded using single-cell Ca²⁺ imaging. C, The percentages of the GFP-tagged RyR2 WT (142) and GFP-tagged Del-exon-3 (153) cells that display Ca²⁺ oscillations at various extracellular Ca²⁺ concentrations. SOICR amplitude (D) and frequency (E) in the GFP-tagged RyR2 WT and GFP-tagged Del-exon-3 cells were determined by measuring the averaged peak amplitude and frequency of Ca²⁺ oscillations at 5 mmol/L extracellular Ca²⁺ and normalized to that in the GFP-tagged RyR2 WT cells (100%). Data shown are mean±standard error of the mean (SEM; n=6–8). ^#P<0.05. *P<0.01 vs WT.

RyR2 NH₂-Terminal Mutations Associated With ARVD2 Reduce the Threshold for Ca²⁺ Release Termination

A number of point mutations in the NH₂-terminal region of RyR2 are associated with ARVD2 cardiomyopathy.^6,7,10–14 To determine whether these NH₂-terminal ARVD2-associated RyR2 mutations also affect Ca²⁺ release termination, we generated HEK293 cell lines that express the RyR2 mutations, A77V, R176Q/T2504M, R420W, and L433P. As shown in Figure 5, each of these RyR2 mutations reduced the SOICR activation (Figure 5A) and termination (Figure 5B) thresholds. Each mutation also increased the fractional Ca²⁺ release because of a greater reduction in the termination threshold than in the activation threshold (Figure 5C). Interestingly, CPVT mutations E189D and R4496C that are not associated with cardiomyopathy^40,41 reduced the activation and termination thresholds to a similar extent (Figure 5A, B). As a result, this mutation did not significantly alter fractional release (Figure 5C). There were no significant differences in F_max, F_min, or store capacity between WT and any of these mutant cells (Online Figure V). Also, all these mutants were expressed in HEK293 cells at a level comparable to that of WT (Figure 5D). Together, these data show that increased fractional Ca²⁺ release attributable to reduced Ca²⁺ release termination is a common defect of ARVD2-associated RyR2 mutations.

Figure 5. Arrhythmogenic right ventricular displasia type 2 (ARVD2)-associated cardiac ryanodine receptor (RyR2) NH₂-terminal mutations decrease the termination threshold and increase the fractional Ca²⁺ release.

D1ER fluorescence resonance energy transfer (FRET) imaging was performed in single HEK293 cells expressing the RyR2 wild-type (WT), ARVD2-associated RyR2 mutations (A77V, R176Q/T2504M, R420W, or L433P), or the RyR2 E189D and R4496C mutations associated with catecholaminergic polymorphic ventricular tachycardia (CPVT) only. The activation threshold (A), termination threshold (B), and fractional Ca²⁺ release (C) in RyR2 WT (250) and mutant (87–110) cells were determined as described in Figure 1. Data shown are mean±standard error of the mean (SEM; n=5–13). *P<0.01 vs WT. D, Immunoblotting of RyR2 WT and RyR2 mutants from the same amount of cell lysates using the anti-RyR antibody (34c; n=3).

The Mouse RyR2 A1107M Mutation Increases the Threshold for Ca²⁺ Release Termination

The human RyR2 T1107M mutation is associated with HCM.¹⁵ We generated a stable inducible HEK293 cell line expressing the mouse RyR2 A1107M mutation, which corresponds to the human HCM-associated RyR2 T1107M mutation. Interestingly, the A1107M mutation significantly increased the termination threshold (61% vs 57% in WT; P<0.05), but it did not significantly affect the activation threshold (Figure 6A–C). This resulted in a significant reduction in the fractional Ca²⁺ release (32% vs 36% in WT; P<0.01; Figure 6D). There were no significant differences in F_max, F_min, and store capacity between the WT and A1107M mutant cells (Figure 6E–G). The expression levels of the RyR2 WT and A1107M mutant were similar (Figure 6H). Therefore, in contrast to the RyR2 mutations associated with DCM and ARVD2, the A1107M mutation associated with HCM increases the threshold for Ca²⁺ release termination and decreases the fractional Ca²⁺ release.

Figure 6. Hypertrophic cardiomyopathies (HCM)-associated cardiac ryanodine receptor (RyR2) A1107M mutation increases the termination threshold and decreases the fractional Ca²⁺ release.

D1ER fluorescence resonance energy transfer (FRET) signals were recorded in HEK293 cells expressing the RyR2 mutation A1107M (A) using single-cell FRET imaging. The activation threshold (B), termination threshold (C), fractional Ca²⁺ release (D), F_max (E), F_min (F), and store capacity (G) in RyR2 WT (250) and the A1107M mutant (111) cells were determined as described in Figure 1. Data shown are mean±standard error of the mean (SEM; n=7–13). ^#P<0.05 and *P<0.01 vs wild-type (WT). H, Immunoblotting of RyR2 WT and the A1107M mutant from the same amount of cell lysates using the anti-RyR antibody (34c; n=3). Note that the images were from the same gel.

Discussion

Mutations in RyR2 are associated with stress-induced ventricular tachyarrhythmias and cardiomyopathies. Extensive investigations over the past decade have demonstrated that spontaneous Ca²⁺ wave (SOICR)-evoked DADs are the major cause of RyR2-associated CPVT.^4–6 We and others have shown that CPVT RyR2 mutations reduce the threshold for SOICR.^{22,23,40,42–45} This reduced SOICR threshold increases the likelihood of spontaneous SR Ca²⁺ release during SR Ca²⁺ overload and, thus, the likelihood of SOICR-evoked DADs and triggered arrhythmias.^3,16–21 In contrast, it is unclear how RyR2 mutations lead to cardiomyopathies. We show here that the DCM-associated RyR2 exon-3 deletion and several ARVD2-associated RyR2 NH₂-terminal mutations (A77V, R176Q/T2504M, R420W, and L433P) all reduce the threshold for Ca²⁺ release termination, whereas an HCM-associated RyR2 mutation (A1107M) increases the termination threshold. These data demonstrate for the first time to our knowledge that RyR2 mutations associated with cardiomyopathies alter the termination of Ca²⁺ release.

The exact mechanism by which SR Ca²⁺ release is terminated remains unclear. Localized SR luminal Ca²⁺-dependent inactivation of RyR2 is likely to be involved.^{24,26,29,46,47} Elegant studies by Zima et al²⁶ showed that spontaneous (diastolic) and stimulated (systolic) SR Ca²⁺ release in cardiomyocytes terminate at the same SR luminal Ca²⁺ threshold. This implies that systolic and diastolic Ca²⁺ release is terminated by a similar process. We demonstrate here that cardiomyopathy-associated RyR2 mutations alter the termination threshold of SOICR or spontaneous Ca²⁺ release. Based on the finding of Zima et al,²⁶ these RyR2 mutations are likely to alter the termination threshold of systolic Ca²⁺ release. Furthermore, Zima et al²⁶ showed that spontaneous SR Ca²⁺ release terminates when luminal Ca²⁺ decreases to approximately 60% of the original SR Ca²⁺ content. Interestingly, we found that SOICR in HEK293 cells expressing RyR2 WT terminates when luminal Ca²⁺ decreases to approximately 57% of the ER store capacity. The similarity of these termination thresholds could be coincidental. It could also indicate that the termination of Ca²⁺ release is an intrinsic property of the RyR2 channel. The latter is consistent with the various RyR2 mutations differentially shifting the termination threshold.

An important question is how abnormal termination of Ca²⁺ release could lead to cardiomyopathies. Cardiomyopathies are generally associated with mutations in sarcomeric proteins. Most disease-associated sarcomeric mutations alter the sensitivity of myofilaments to Ca²⁺.^33,35,48,49 Because the myofilaments represent a major pool of intracellular Ca²⁺ binding sites, changes in the Ca²⁺ sensitivity of myofilaments can alter their response to Ca²⁺ release, which can, in turn, change the amplitude and dynamics of the cytosolic Ca²⁺ transient (but not the total release of Ca²⁺).^35,48,49 HCM-associated sarcomeric mutations tend to increase the myofilament Ca²⁺ sensitivity and thus reduce cytosolic Ca²⁺ transients. However, DCM-associated sarcomeric mutations tend to decrease the myofilament Ca²⁺ sensitivity and thus increase cytosolic Ca²⁺ transients.^33,35,48,49 The abnormal cytosolic Ca²⁺ transient resulting from altered myofilament Ca²⁺ sensitivity is thought to trigger the cardiac remodeling (via Ca²⁺/calmodulin-dependent signaling pathways, the calcineurin/NFAT pathways, or apoptotic signaling) that can lead to HCM or DCM.^31–35,50

Abnormal cytosolic Ca²⁺ transients also could result from altered Ca²⁺ release termination. A reduced termination threshold would delay the termination of SR Ca²⁺ release and thus increase the cytosolic Ca²⁺ transient. An increased termination threshold would cause premature termination of Ca²⁺ release and consequently would limit the size of the cytosolic Ca²⁺ transient. Here, we show that a DCM-associated RyR2 mutation reduces the termination threshold and increases the fractional Ca²⁺ release, whereas an HCM-associated RyR2 mutation increases the termination threshold and decreases fractional release. These changes in fractional release may alter cytosolic Ca²⁺ transients. These effects are similar to those of sarcomeric mutations associated with DCM and HCM. It is important to note that the fractional Ca²⁺ release is determined by both the termination and activation thresholds. We show that CPVT-only RyR2 mutations (E189D and R4496C) reduce the activation and termination thresholds to a similar extent, resulting in no significant changes in the fractional Ca²⁺ release. Therefore, our data suggest a link between abnormal fractional Ca²⁺ release resulting from aberrant Ca²⁺ release termination and RyR2-associated cardiomyopathies.

Aberrant Ca²⁺ release termination may also contribute to CPVT. It has been estimated that a diastolic Ca²⁺ wave (SOICR) that liberates 50% to 70% SR Ca²⁺ content is required to produce DADs of sufficient amplitude to induce triggered arrhythmias.⁵¹ A reduced termination threshold would increase the fractional Ca²⁺ release and, thus, the amplitude of Ca²⁺ waves during SR Ca²⁺ overload. Larger Ca²⁺ waves would, in turn, produce more robust DADs and enhance the propensity for triggered activities. Thus, the reduced termination threshold for SOICR of the DCM-associated and ARVD2-associated RyR2 mutations combined with their lowered SOICR activation threshold may explain why these mutations also enhance the susceptibility to CPVT.

The recently solved three-dimensional structures of the NH₂-terminal region of RyR^52–55 have provided some new insights into how mutations in the NH₂-terminal region of RyR2 alter Ca²⁺ release termination. The NH₂-terminal region of RyR contains three domains that interact with each other to form a cytoplasmic vestibule at the center of the channel. Many disease-causing RyR mutations are located in interfaces between these three domains or between these domains and other parts of the channel.⁵⁴ Interestingly, the cytoplasmic vestibule formed by the NH₂-terminal domains undergoes conformational changes during channel gating.⁵⁶ It has been proposed that the NH₂-terminal domains are allosterically coupled to the transmembrane pore forming domains of the channel.⁵⁵ Therefore, mutations in the NH₂-terminal region may affect the gating of the channel and, thus, Ca²⁺ release by altering the allosteric coupling between the NH₂-terminal domains and the channel pore forming domains. These structural studies also suggest that in addition to the NH₂-teminal domains, other regions of RyR2 are likely involved in Ca²⁺ release termination.

The importance of Ca²⁺ release termination in cardiac physiology and pathophysiology has become increasingly clear.^27,29,30 Besides the link between cardiomyopathies and abnormal Ca²⁺ release termination established here, reduced Ca²⁺ release termination threshold also has been shown in heart failure.^28,57 This suggests that, like activation of Ca²⁺ release, termination of Ca²⁺ release is also an important target of regulation. Aberrant Ca²⁺ release termination may be a common defect associated with cardiomyopathies and other cardiac abnormalities.

Limitations

Our studies in HEK293 cells demonstrate that cardiomyopathy-associated RyR2 mutations alter the termination of Ca²⁺ release. However, HEK293 cells lack many cardiac-specific proteins, and thus the Ca²⁺ release termination defects of cardiomyopathy-associated RyR2 mutations have yet to be confirmed in cardiac cells. At the moment, such studies would require the development of mouse models harboring cardiomyopathy-associated RyR2 mutations. These animal models are presently unavailable. Our studies also do not define the molecular mechanisms underlying the activation/termination thresholds or how these might be regulated by various proteins (eg, calsequestrin) and factors (eg, cytosolic or luminal Ca²⁺). Further comprehensive and detailed investigations will be required to address these important issues.

Novelty and Significance.

What Is Known?

• Naturally occurring mutations in the cardiac ryanodine receptor (RyR2) are associated with stress-induced ventricular tachyarrhythmias (VTs) and cardiomyopathies.

• RyR2 mutations associated with VTs cause abnormal activation of RyR2 and, thus, aberrant sarcoplasmic reticulum (SR) Ca²⁺ release.

• SR Ca²⁺ release self-terminates, and this termination process critically controls normal cytosolic Ca²⁺ signaling.

What New Information Does This Article Contribute?

• The process of SR Ca²⁺ release termination is controlled by intrinsic properties of the RyR2 channel.

• The NH₂-terminal region of RyR2 contains an important determinant of SR Ca²⁺ release termination.

• Abnormal Ca²⁺ release termination is a common defect of RyR2 mutations associated with cardiomyopathies.

RyR2 mutations have been linked to both stress-induced cardiac arrhythmias and cardiomyopathies. It has become clear that spontaneous Ca²⁺ waves resulting from abnormal activation of RyR2 are the major cause of cardiac arrhythmias. However, the causal mechanism of RyR2-associated cardiomyopathies is completely unknown. We examined the impact of a number of cardiomyopathy-associated RyR2 mutations and NH₂-terminal deletions on the termination of Ca²⁺ release. Our results show that the deletion of exon-3 in RyR2, which is associated with dilated cardiomyopathy, and a number of NH₂-terminal point mutations, which are associated with arrhythmogenic right ventricular displasia type 2 (A77V, R176Q/T2504M, R420W, L433P), all reduce the luminal Ca²⁺ threshold at which Ca²⁺ release terminates. However, the RyR2 A1107M mutation associated with hypertrophic cardiomyopathy increases the luminal Ca²⁺ termination threshold. These data demonstrate, for the first time to our knowledge, that impairment of SR Ca²⁺ release termination is common to RyR2-linked cardiomyopathies. This work also shows that the NH₂-terminal region of RyR2 plays an important role in SR Ca²⁺ release termination. Identifying the common and pathologically significant RyR2 structure (NH₂-terminal domain) and function (Ca²⁺ release termination) represent a key step toward understanding how abnormal RyR2 function leads to cardiac arrhythmias or cardiomyopathies.

Non-standard Abbreviations and Acronyms

ARVD2

arrhythmogenic right ventricular displasia type 2

CPVT

catecholaminergic polymorphic ventricular tachycardia

DAD

delayed afterdepolarization

DCM

dilated cardiomyopathy

FRET

fluorescence resonance energy transfer

HCM

hypertrophic cardiomyopathies

HEK293

human embryonic kidney 293 cells

RyR2

cardiac ryanodine receptor

SOICR

store overload-induced Ca²⁺ release

SR

sarcoplasmic reticulum