In Situ Confocal Imaging in Intact Heart Reveals Stress-Induced Ca²⁺ Release Variability in a Murine CPVT Model of RyR2^R4496C+/− Mutation

Abstract

Background

Catecholaminergic polymorphic ventricular tachycardia (CPVT) is directly linked to mutations in proteins (e.g., RyR2^R4496C) responsible for intracellular Ca²⁺ homeostasis in the heart. However, the mechanism of Ca²⁺ release dysfunction underlying CPVT has only been investigated in isolated cells but not in the in situ undisrupted myocardium.

Methods and Results

We investigated in situ myocyte Ca²⁺ dynamics in intact Langendorff perfused hearts (ex vivo) from wildtype (WT) and RyR2^R4496C+/− mice using laser scanning confocal microscopy. We found that myocytes from both WT and RyR2^R4496C+/− hearts displayed uniform, synchronized Ca²⁺ transients. Ca²⁺ transients from beat to beat were comparable in amplitude with identical activation and decay kinetics in WT and RyR2^R4496C+/− hearts, suggesting that excitation-contraction (EC) coupling between the sarcolemmal Ca²⁺ channels and mutated RyR2^R4496C+/− channels remains intact under baseline resting conditions. Upon adrenergic stimulation, RyR2^R4496C+/− hearts exhibited a high degree of Ca²⁺ release variability (CRV). The varied pattern of Ca²⁺ release was absent in single isolated myocytes, independent of cell cycle length, synchronized among neighboring myocytes, and correlated with CPVT. A similar pattern of action potential variability, which was synchronized among neighboring myocytes, was also revealed under adrenergic stress in intact hearts but not in isolated myocytes.

Conclusions

Our studies using in situ confocal imaging approach suggest that mutated RyR2s are functionally normal at rest but display a high degree of CRV upon intense adrenergic stimulation. CRV is a Ca²⁺ release abnormality resulting from electrical defects rather than the failure of the Ca²⁺ release response to action potentials in mutated ventricular myocytes. Our data provide important insights into Ca²⁺ release and electrical dysfunction in an established model of CPVT.

Introduction

Catecholaminergic polymorphic ventricular tachycardia (CPVT) is a lethal genetic disease characterized by exercise/stress-induced malignant ventricular arrhythmias and sudden cardiac death in young individuals with structurally normal hearts.¹ The genetic foundation of CPVT is linked to mutations within two important Ca²⁺ handling proteins: autosomal dominant mutations of intracellular Ca²⁺ release channels or type 2 ryanodine receptors (RyR2),^{2, 3} and autosomal recessive mutations of calsequestrin (CASQ), a Ca²⁺ binding protein in the sarcoplasmic reticulum (SR).⁴ Patients with CPVT mutations are typically healthy at rest but develop arrhythmias under emotional or physical stress. Electrocardiograms of CPVT patients reveal bidirectional ventricular tachycardia and polymorphic ventricular tachycardia.²

Both RyR2 and CASQ play important roles in controlling intracellular Ca²⁺ handling and homeostasis.^{5, 6} Ca²⁺ release dysfunction is believed to be the underlying mechanism for the above mentioned ventricular arrhythmias in human patients due to the critical importance of the two key proteins in Ca²⁺ regulation of cardiomyocytes.^{3, 7, 8} Indeed, many studies have identified Ca²⁺ handling defects using mutant channels in isolated lipid bilayers,^9–11 cultured myocytes, HEK 293 cells stably transfected with these mutants,^12–15 or in ventricular myocytes isolated from mouse models that carry mutant RyR2.^16–19 Notably, a recent study by Fernandez-Velasco, et al., showed that, compared to wild type (WT) myocytes, untreated RyR2^R4496C+/− myocytes have an enhancement of Ca²⁺ sensitivity and an increase in spontaneous Ca²⁺ release in diastole during electrical pacing, which is augmented by isoproterenol and increasing the pacing frequency.²⁰ Their results were further supported by Kang, et al., who demonstrated spontaneous Ca²⁺ release events in both RyR2^R4496C+/− ventricular and Purkinje cells.²¹ However, these in vitro studies are controversial because it is unclear if these mutant proteins behave abnormally under resting conditions (i.e., in the absence of catecholamine stimulation).^{22, 23} One unresolved issue, however, is that the resting defect in Ca²⁺ dynamics does not correspond to the clinical manifestations of CPVT patients with the same mutation, whose hearts are structurally normal and free of arrhythmias unless under stress.

A second question is related to the pathophysiology of CPVT RyR2 mutations, in particular how RyR2 mutations cause CPVT. Several previous studies demonstrated that ventricular myocytes harboring the RyR2 R4496C mutation are prone to spontaneous Ca²⁺ release and delayed after depolarizations (DADs).^{13, 16, 20, 21} One of the leading hypotheses is that the mutation induces SR calcium leak in ventricular myocytes during diastole, thereby generating DADs that, in turn, trigger fatal cardiac arrhythmias.²⁴ Those previous results were obtained from isolated myocytes, and it remains unclear whether mutated RyR2s in physiologically coupled myocytes in intact heart behave differently from isolated myocytes during steady state beating. It is therefore important to study the nature of Ca²⁺ handling under physiological conditions in RyR2-mutated hearts.

In the present study, we aimed to identify the in situ feature of Ca²⁺ mis-handling in CPVT hearts and to enhance our understanding of Ca²⁺-dependent arrhythmogenesis using a RyR2^R4496C+/− mouse model. We performed in situ, ex vivo Ca²⁺ / action potential (AP) imaging in Langendorff-perfused intact hearts under near physiological conditions, using laser scanning confocal microscopy at baseline and following catecholamine stimulation. We also mapped Ca²⁺ dynamics to simultaneously recorded ECGs. Our data demonstrate that myocyte EC coupling between the sarcolemmal Ca²⁺ channels and mutated RyR2^R4496C+/− channels remains intact under baseline resting conditions. However, under intense adrenergic stress, we identified a previously unappreciated pattern of Ca²⁺ handling dysfunction in physiologically-coupled ventricular myocytes with the RyR2 R4496C mutation. Interestingly, highly variable SR Ca²⁺ release in RyR2^R4496C+/− hearts was synchronized among neighboring myocytes and correlated with CPVT occurrence as measured by ECG. Similarly, with in situ AP imaging, we detected stress-induced beat-to-beat variability in AP, which was also synchronized among neighboring mutated myocytes.

Methods

Animal experiments were performed in accordance with the protocol approved by the Institutional Animal Care and Use Committee at the University of Iowa. In situ confocal Ca²⁺ imaging in intact hearts with / without ex vivo electrocardiogram, in situ confocal Aaction Potential (AP) imaging in intact hearts were adapted from published reports.^25–27 Ca²⁺ imaging in adult single isolated ventricular myocytes and in primary cultured neonatal myocytes were performed as previously described.^{28, 29} Confocal Ca²⁺ / AP images were analyzed offline with custom routines composed with IDL image analysis software (ITT VIS Inc., Boulder, CO).³⁰ Pseudo ECG data were processed offline with Clampfit 10. Data were expressed as mean ± SE, and median with interquartile range in boxplots. Multiple regression analysis was performed to determine the correlation coefficient and significance. Student’s t-tests were applied for pair-wise comparison. Bonferroni procedure following a global test based on linear mixed effects model was performed for multiple group comparisons (NCSS, Kaysville, Utah). A compound symmetry correlation structure was assumed for linear mixed effects model tests. A p value of <0.05 was considered statistically significant. Expanded methods are available in Supplementary Material.

Results

Normal EC coupling in in situ RyR2^R4496C+/− myocytes at rest

RyR2^R4496C+/− mutant mice are susceptible to CPVT under catecholamine stimulation.³¹ We performed in situ Ca²⁺ imaging of ventricular myocytes from intact hearts attached to an oxygenated Langendorff-perfusion system.²⁵ We examined autonomous Ca²⁺ signals, initiated by sinus rhythm, in ventricular myocytes. In linescan mode, each myocyte from both WT and RyR2^R4496C+/− hearts displayed uniform, synchronized Ca²⁺ transients. Ca²⁺ transients from beat-to-beat were comparable in amplitude with identical activation and decay kinetics between these groups (Figure 1), suggesting that EC coupling between the sarcolemmal Ca²⁺ channels and mutated RyR2^R4496C+/− channels (e.g. calcium-induced calcium release) remains intact under baseline resting conditions. In addition, spontaneous Ca²⁺ sparks or waves were rarely observed at diastolic phase during steady state beating in both WT and RyR2^R4496C+/− myocytes, indicating mutated RyR2s are not leaky under resting condition.

Figure 1.

In situ confocal Ca²⁺ imaging in WT and RyR2^R4496C+/− hearts: normal Ca²⁺ transients under resting conditions. A–B, Autonomous, synchronized Ca²⁺ transients driven by sinus rhythm in (A) WT and (B) RyR2^R4496C+/− hearts. Spontaneous Ca²⁺ waves or Ca²⁺ sparks were rarely detected during steady state beating under resting conditions. Bottom panels are spatial averages of Ca²⁺ signals from corresponding images. C, Boxplots of Ca²⁺ transients (amplitude, time to peak, and T₅₀ relaxation), ◆ and error bars inside the box denotes mean±SE. N=5–7 hearts, n=54 or 82 frames of transients (most of frames include multiple myocytes) for WT and RyR2^R4496C+/−, respectively. Green bars indicate edges of myocytes in this and subsequent figures.

High beat-to-beat variability of Ca²⁺ transients in RyR2^R4496C+/− myocytes under adrenergic stress

CPVT occurs under physical or emotional stress in patients with the RyR2^R4496C+/− mutation and are elicited exclusively by adrenergic stress.² RyR2^R4496C+/− mice also display similar ECG abnormalities under high catecholamine and caffeine stimulation.³² To further understand Ca²⁺ performance in RyR2^R4496C+/− myocytes during high adrenergic stress, we next compared in situ Ca²⁺ dynamics of WT and RyR2^R4496C+/− myocytes in intact hearts under sinus rhythm with perfusion of epinephrine (1 μM) and caffeine (0.6 mM), a condition that induces CPVT in RyR2^R4496C+/− mice.^{16, 31, 32} In WT myocytes, epinephrine plus caffeine increased the amplitude of Ca²⁺ transients and accelerated both the activation and decay kinetics (Figure 2A, D–F). Epinephrine with caffeine also significantly accelerated the kinetics of Ca²⁺ transients (both activation and relaxation) in RyR2R^4496C+/− myocytes (Figure 2D–F). Strikingly, RyR2^R4496C+/− myocytes under high sympathetic stress exhibited high beat to beat Ca²⁺ release variability (CRV) in the amplitude of Ca²⁺ transients (Figure 2BC). While some Ca²⁺ transients with a higher amplitude were detected in mutant myocytes after adrenergic stimulation, many of the Ca²⁺ transients were markedly reduced, which resulted in a lower average amplitude as compared to WT myocytes (Figure 2D). It should be noted that these transients with reduced amplitude differed from asynchronous Ca²⁺ waves. Specifically, they were synchronized with fast rising kinetics (T_peak), similar to that of WT myocytes (Figure 2E). CRV activity often followed an irregular pattern of amplitude fluctuation of Ca²⁺ transients. In RyR2^R4496C+/− myocytes, CRV was reversible and disappeared upon washout of epinephrine and caffeine, suggesting that this CRV defect is induced by adrenergic stress. Variance analysis from thousands of Ca²⁺ transient samples (5–7 hearts in each group) further supports our observation that Ca²⁺ transients are highly varied in amplitude from beat-to-beat under high sympathetic stress conditions, but not at baseline (Figure 3AB). Importantly, this CRV pattern was not observed in single isolated RyR2^R4496C+/− myocytes under field stimulation with the same adrenergic stimulation (Supporting Information, Figure S1). Our data suggest that adrenergic stress-induced CRV requires physiological cellular coupling in RyR2^R4496C+/− hearts.

Figure 2.

High degree of Ca²⁺ release variability in RyR2^R4496C+/− myocytes from intact heart as demonstrated by in situ confocal imaging after adrenergic stimulation. A, Autonomous Ca²⁺ transients in myocytes from WT hearts under epinephrine (1 μM) and caffeine (0.6 mM) perfusion. Beat-to-beat Ca²⁺ transients were stable in WT myocytes even under adrenergic stress challenge. B–C, Ca²⁺ transients from RyR2^R4496C+/− myocytes of intact hearts, exhibiting high degree of variability in Ca²⁺ transient amplitude. These Ca²⁺ transients with varied amplitude were action potential-triggered events with a quick rising and decay phase, which is distinct from asynchronous propagated Ca²⁺ wave events. D–F, Boxplots of Ca²⁺ transients (amplitude, time to peak - T_peak, and decay rate T₅₀). ◆ and error bars inside the box denotes mean±SE. The averaged amplitude of Ca²⁺ transients of RyR2^R4496C+/− myocytes under epinephrine plus caffeine perfusion was less than that of control. N=5–7 hearts for each group, n=107 or 74 frames of transients for WT and RyR2^R4496C+/− myocytes, respectively. * p<0.05 and ** p<0.01 vs WT or RyR2^R4496C+/− without epinephrine and caffeine (control), ^## p<0.01 vs WT under the same condition (epinephrine and caffeine). Global test: p=0.00001, 0.005, 0.02 among the 4 groups, for D, E, F, respectively.

Figure 3.

Ca²⁺ release variability in RyR2^R4496C+/− myocytes from intact hearts is independent of the beating interval preceding each transient. A–B, Variance analysis of CRV. A, Variance was calculated as the square of variation (sample value minus mean value) of each transient amplitude. Scatter plots of F/F₀ variance from WT and RyR2^R4496C+/− myocytes under resting (control) and stress (epinephrine plus caffeine) conditions, respectively, are shown. B, Boxplot of F/F₀ variance from each group. RyR2^R4496C+/− myocytes exhibited significant higher variance comparing to other groups, ◆ and error bars inside the box denotes mean±SE. n=5–7 hearts, n=1947, 1502, 2405, or 2449, respectively, for each group. **p<0.01 vs RyR2^R4496C+/− under control condition, ^## p<0.01 vs WT under the same condition (epinephrine and caffeine). p=0.008 among the 4 groups (global test). C–F, No Correlation between Ca²⁺ transient amplitude (F/F₀) and the beating interval preceding each transient in WT (C) and mutant (D) hearts. C–D, Examples displaying Ca²⁺ transient amplitude – time interval correlation from a representative Ca²⁺ image of WT (C) and RyR2^R4496C+/− (D) heart, respectively, under epinephrine and caffeine stimulation. E–F, Multiple regression analysis on data from 3–4 hearts of WT and RyR2^R4496C+/−(n=163, 227 events, respectively), indicating no correlation between Ca²⁺ transient amplitude and time interval preceding each transient in both WTs and mutant hearts.

High CRV is not due to beating interval variability

We postulated that the high CRV observed in mutated myocytes from intact hearts is due to variable beating intervals preceding each Ca²⁺ transient, which would cause varied SR Ca²⁺ loading under ex vivo conditions. To test this hypothesis, we measured the time intervals preceding each Ca²⁺ transient and correlated it with the amplitude of the corresponding transient from WT and RyR2 mutant hearts. As shown in Figure 3C–F, the range of amplitudes of Ca²⁺ transients from WT myocytes (after epinephrine + caffeine) was narrow, while the amplitudes of Ca²⁺ transients from mutated myocytes were dispersed over a much wider range. More importantly, there was no correlation between the beating intervals and Ca²⁺ transient amplitudes in either WT or mutated myocytes under stress. This analysis clearly indicates that the high level of CRV observed in RyR2^R4496C+/− myocytes under ex vivo conditions and with high sympathetic stimulation is not caused by beating interval variability and is independent of stimulation frequency.

CRV is synchronized among neighboring myocytes with mutant RyR2

An advantage of in situ confocal imaging is that it allows examination of the Ca²⁺ dynamics of multiple, physiologically interconnected myocytes simultaneously. The data in Figure 2B&C revealed a coordinated pattern of CRV between two neighboring cells from RyR2^R4496C+/− hearts. To further confirm this phenomenon, we evaluated the pattern of CRV among multiple myocytes from RyR2^R4496C+/− hearts. Surprisingly, we found that CRV is indeed present in a coordinated pattern among many neighboring myocytes (Figure 4). This novel pattern of Ca²⁺ release dysfunction was consistently observed in RyR2^R4496C+/− hearts during adrenergic stress but not in WT hearts. Interestingly, CRV was not observed in RyR2^R4496C+/− cultured neonatal myocytes or among neighboring cells under both control and adrenergic stimulation conditions (Supporting Information, Figure S2). Collectively, these data strongly suggest that CRV is an integrated, tissue level response to adrenergic stress, and that the source of CRV is likely not ventricular myocytes themselves, but from an abnormality in the upstream electrical signal.

Figure 4.

Ca²⁺ release variability in RyR2^R4496C+/− myocytes under sympathetic stress: synchrony among neighboring myocytes. A, A typical Ca²⁺ image of 5 myocytes from a RyR2^R4496C+/− heart. B, Spatial profile of Ca²⁺ signals from each myocyte showing that stress-induced Ca²⁺ release variability was coordinated among many neighboring mutated myocytes.

CRV is associated with CPVT

In order to test if increased CRV was related to CPVT, we performed simultaneous recordings of in situ confocal linescan Ca²⁺ images and real time ex vivo ECG measurements in Langendorff perfused intact hearts. WT hearts exhibited normal ECGs and regular, stable Ca²⁺ transients from beat-to-beat among different neighboring myocytes (Figure 5A) in both the absence and presence of epinephrine and caffeine. RyR2^R4496C+/− hearts also displayed normal ECG and Ca²⁺ dynamics at baseline. However, under high sympathetic stimulation, CRV occurred in RyR2^R4496C+/− hearts simultaneously with CPVT (Figure 5B). These data suggest that CRV, a tissue level measure of SR Ca²⁺ release abnormality in RyR2^R4496C+/− hearts, is associated with electrical abnormalities, including CPVT.

Figure 5.

Ca²⁺ release variability and CPVT in RyR2^R4496C+/− hearts. Confocal Ca²⁺ imaging was performed with simultaneous recordings of ex vivo ECG in hearts with continuous Langendorff perfusion. A, A WT heart exhibited normal ex vivo ECG and stable Ca²⁺ transients from beat-to-beat under stress (Note: p wave is visible but has very small magnitude). B, RyR2^R4496C+/− hearts concurrently displayed CRV and CPVT.

Physiologically coupled RyR2^R4496C+/− myocytes have abnormal APs

To investigate the underlying mechanism of CRV, we then measured the in situ APs under the same recording conditions as for in situ Ca²⁺ imaging,²⁷ except that a fast voltage-sensitive dye, ANNINE-6plus, was loaded into the intact hearts through Langendorff perfusion to track the dynamic changes of transmembrane potential. We applied this imaging technique to examine the in situ electrical properties of RyR2^R4496C+/− myocytes in intact hearts under adrenergic stress. Surprisingly, we found that RyR2^R4496C+/− myocytes also exhibited high heterogeneity in AP morphology under adrenergic stress in comparison to WT myocytes (Figure 6AB). On average, RyR2^R4496C+/− myocytes had similar ratio of fluorescence change (indicative of AP amplitude) and shorter AP duration than those of WT myocytes (Figure 6CD). Variance analysis showed that mutant myocytes had a greater variability in both AP amplitude and duration (Figure 6EF). More importantly, AP variability was also synchronized among neighboring myocytes from the RyR2^R4496C+/− hearts (Figure 6B & Supporting Information, Figure S3). By contrast, AP studies in single isolated myocytes using conventional current clamp recordings showed no difference in variance in these parameters between WT and RyR2^R4496C+/− hearts under adrenergic stress (Supporting Information, Figure S4), suggesting that in situ electrical abnormalities in RyR2^R4496C+/− hearts are non-ventricular in origin. Taken together, our data support the notion that electrical abnormalities of non-ventricular origin underlie the mechanism of CRV in physiologically coupled ventricular myocytes.

Figure 6.

In situ imaging of myocyte APs in intact hearts. A–B, Representative examples of ANNINE-6plus fluorescence from WT and RyR2^R4496C+/− hearts perfused with epinephrine plus caffeine. Top to bottom: raw images, images after normalization, and spatial average of ANNINE-6plus fluorescence. RyR2^R4496C+/− myocytes displayed beat-to-beat variation in AP morphology. C–D, Summary of AP amplitude (ΔF/F₀) and APD₅₀ in boxplots. E–F, Boxplots of variance of AP amplitude and APD₅₀. * p<0.05, n=4 and 5 for WT and RyR2^R4496C+/− hearts, and 326 or 213 AP events for WT or RyR2^R4496C+/− hearts, respectively.

Discussion

Ca²⁺ handling abnormalities play an important role in the pathophysiology of heart disease, including heart failure, arrhythmias, and sudden cardiac death.³³ Patients with a specific mutation (R4496C) in the cardiac Ca²⁺ release channel, RyR2, suffer from exercise/stress induced CPVT and sudden cardiac death. The genetically modified mouse model bearing the same mutation provides an excellent model for studying the mechanisms of Ca²⁺-dependent arrhythmogenesis and the pathophysiology of CPVT.³² In this study, we applied the in situ confocal Ca²⁺ imaging techniques to study Ca²⁺ handling in intact hearts.²⁵ This approach allows investigation under near physiological conditions. To our surprise, we obtained distinctly different results as compared to results in ventricular myocytes isolated from these same mice.^{20, 21, 34} We found that 1) mutated RyR2s are functionally normal during non-stressed, resting condition; 3) we identified a new pattern of Ca²⁺ release dysfunction, CRV, in RyR2^R4496C+/− but not WT myocytes under adrenergic stress; 4) CRV is an integrated, tissue level response of mutated myocytes to adrenergic stress and is observed in intact hearts but not in single isolated myocytes; 5) CRV is synchronized among neighboring myocytes, independent of beating interval or stimulation frequency; 6) CRV is tightly associated with CPVT; and 7) AP variability in intact hearts (but not in single isolated myocytes) is the likely cause of CRV.

Normal CICR function in ventricular myocytes with RyR2^R4496C+/− mutation at rest

Whether mutant RyR2 channels are functionally normal at rest is a controversial yet fundamentally important question. Particularly for patients with this mutation, it is critical to determine if the dysfunctional Ca²⁺ handling is constantly present and is exacerbated with stress, or if the mutated channels are normal at resting state and become dysfunctional only with adrenergic stimulation. The answer to this issue may provide important insights in developing therapeutic strategies for CPVT patients. Marks and colleagues demonstrated using single channel recording in lipid-bilayers that RyRs harboring human mutations (including R4496C) show similar gating features as WT RyRs at baseline conditions but are much more leaky upon high protein kinase A stimulation (10-fold increase in channel open probability).^{9, 10} George and Lai also showed unaltered baseline Ca²⁺ activity in HL-1 myocytes transfected with mutated RyR2s.³⁵ However, these results were challenged by other groups who reported an increase in abnormal Ca²⁺ release and higher sensitivity of mutant RyR2s compared to WT at resting conditions. Jiang et al. reported that CPVT mutations enhance the sensitivity of the RyR2 channel to luminal Ca²⁺ activation and lower the threshold for spontaneous Ca²⁺ release in the setting of Ca²⁺ overload.^{11, 13} Recently, Fernandez-Velasco et al. showed an increased incidence of Ca²⁺ sparks and Ca²⁺ waves at baseline in RyR2^R4496C+/− myocytes, which is further enhanced by either isoproterenol or high pacing rates.²⁰ Liu et al. reported similar results of increased spark frequency at baseline and after isoproterenol challenge in isolated RyR2^R4496C+/− myocytes.¹⁹ However, when we examined Ca²⁺ signals from intact WT and mutant hearts using the in situ imaging method, we found RyR2 mutated myocytes exhibit no abnormal Ca²⁺ release, e.g., Ca²⁺ waves, at resting conditions (Figure 1). These discrepancies may be explained by differences in experimental conditions that resulted in exposure of RyR2 to different intra-luminal Ca²⁺ levels. Enzymatic and mechanical dissociation of heart tissue exerts a significant stress to the myocytes, causing myocyte Ca²⁺ overload and spontaneous Ca²⁺ release.^{34, 36, 37} Instead, under conditions which did not exist or cause SR Ca²⁺ overloading such as in lipid bilayers, ^{9, 10} cultured HL-1 myocytes ³⁵ and intact hearts (of the present study), findings were surprisingly consistent, that is, RyR2s carrying R4496C mutation are not leaky at baseline. It has been shown that CPVT-linked RyR2 mutations, including R4496C, increase the sensitivity of the RyR2 channel to SR Ca²⁺ overload.¹³ In the absence of SR Ca²⁺ overload (i.e., in situ under resting conditions or studies with lipid bilayers ^{9, 10} or cultured HL-1 myocytes ³⁵), the R4496C mutation-linked Ca²⁺ release defect would not be apparent. Importantly, our in situ findings are also consistent with the clinical characteristics of CPVT patients whose hearts are structurally normal throughout their lives but only develop fatal arrhythmias under emotional or physical stress. If mutated RyR2 channels expressed in myocytes are continuously leaky at resting conditions, this persistent abnormality of intracellular Ca²⁺ signaling is predicted to lead to alterations in myocyte gene transcription that could eventually promote cardiac pathophysiology (e.g., cardiomyopathy).

Ca²⁺ release variability and electrical abnormalities in physiologically coupled myocytes with RyR2^R4496C+/− mutation

In this study, we identified a new pattern of Ca²⁺ release dysfunction - a high degree of variation in Ca²⁺ transient amplitude in RyR2^R4496C+/− myocytes. This pattern of Ca²⁺ release abnormality was present in mutated myocytes under adrenergic stimulation; it was not observed in WT cells, in RyR2 mutated myocytes under baseline conditions, or in mutated neonatal myocytes in culture. This result was only detectable by in situ imaging methods. Ca²⁺ release variability is distinct in nature from rapid pacing-induced Ca²⁺ alternans,²⁵ in that 1) CRV is independent of the cycle length; 2) the change in Ca²⁺ transient amplitude is predominantly very irregular; 3) CRV is not inducible in isolated myocytes (either WT or mutated myocytes); and 4) rapid high-rate pacing-induced Ca²⁺ alternans may not be coordinated or synchronized among different myocytes from beat to beat.²⁵ These features suggest that proarrhythmic defects in RyR2^R4496C+/− hearts are not limited to ventricular myocytes. Recently, two elegant studies using different approaches (optical mapping and confocal imaging) both concluded that the RyR2^R4496C mutation associated CPVT originated from abnormalities in Purkinje cells. Cerrone et al. studied the mechanisms and origin of CPVT arrhythmias in Langendorff-perfused RyR2^R4496C+/− mouse hearts using optical mapping.³⁸ Their data provide compelling evidence that CPVT arrhythmias originate from the specialized conduction system. Very recently, Fishman and colleagues examined the frequency and severity of spontaneous Ca²⁺ activities in Purkinje cells in comparison to ventricular myocytes with the aid of a novel fluorescent reporter in the cardiac conduction system (including the distal Purkinje fiber network). These studies revealed that Ca²⁺ handling defects in Purkinje cells are more pronounced and frequent than those of ventricular myocytes with the same mutation.²¹ Our data are consistent with these studies by demonstrating that CRV occurs in intact hearts but not in single isolated myocytes with RyR2 mutation, and that CRV is synchronized among different neighboring myocytes.

Our data suggest that CRV may originate from an abnormality in electrical activity in other regions, such as Purkinje fibers or ectopic ventricular foci. Providing additional support for this conclusion, our in situ AP optical recordings and single cell AP studies indicate that this defect is in electrical conduction system rather than ventricular in origin. Therefore, it is postulated that under stress conditions, defective RyR2-mediated abnormal Ca²⁺ release in Purkinje cells leads to aberrant electrical activities, thereby triggering variability in Ca²⁺ release in ventricular myocytes. We speculate that the mechanism is as follows: Under stress condition, spontaneous Ca²⁺ releases in Purkinje fibers with RyR2^R4496C+/− mutation cause DADs, which conduct and produce abnormal APs in ventricular myocytes, as we have observed using in situ AP imaging technique. These abnormal APs, specifically, high variability of AP amplitude and duration but synchronized among neighboring ventricular myocytes, may lead to a varied magnitude of Ca²⁺ influx, causing coordinated Ca²⁺ release variability among mutated ventricular myocytes. Also contributing to the mechanism, it is very possible that upon stress the Purkinje fibers initiate the arrhythmia, which is further sustained by ventricular myocytes carrying defective mutant of RyR2s.

Limitations

In human patients with this or other similar CPVT-linked mutations, intense emotional or physical stress may trigger the appearance of symptoms, such as ventricular arrhythmias, syncope, and sudden cardiac death. However, in RyR2^R4496C+/− mice, the combination of caffeine and epinephrine, instead of epinephrine alone, is necessary to induce CPVT. This may suggest that humans with the same mutation are more susceptible to Ca²⁺ disorder-induced CPVT than mice. This difference does not appear to be related to basal catecholamine levels since β-adrenergic blockade failed to completely prevent sudden cardiac death in patients with CPVT mutations.³⁹ Future studies towards understanding the mechanisms underlying this difference are warranted.

Summary

Our study using the in situ confocal imaging approach provides compelling evidence that the RyR2^R4496C+/− mutants are functionally normal in situ under resting conditions, but display high degree of CRV upon intense adrenergic stimulation. CRV is an integrated, tissue level response of mutated myocytes to adrenergic stress and is closely correlated with CPVT. Our data reveal that CRV results from electrical defects rather than the failure of Ca²⁺ release response to action potentials in mutated ventricular myocytes. This study provides important insights into Ca²⁺ release and electrical dysfunction in an established model of CPVT and has important implications in understanding the mechanism of CPVT in patients.

Ca²⁺ handling abnormalities play important roles in the pathophysiology of heart failure, arrhythmias, and sudden cardiac death. Patients with a specific mutation (R4496C) in the cardiac Ca²⁺ release channel, RyR2, suffer from exercise/stress-induced catecholaminergic polymorphic ventricular tachycardia (CPVT) and sudden cardiac death, but are typically healthy at rest. The reason that the resting defect in Ca²⁺ dynamics in vitro in isolated myocytes does not correspond to the clinical manifestations of CPVT patients with the same mutation is unexplained. In a genetically modified mouse model of CPVT (RyR2^R4496C+/−), we applied in situ confocal Ca²⁺ imaging techniques to study Ca²⁺ handling in undisrupted myocardium of intact hearts. This approach allows investigation under near physiological conditions. The RyR2^R4496C+/− mutants were found to be functionally normal in situ under resting conditions, but had a high degree of Ca²⁺ release variability upon intense adrenergic stimulation. This new pattern of Ca²⁺ handling abnormality is an integrated, tissue level response of mutated myocytes to adrenergic stress, closely correlated with CPVT. Our data reveal that Ca²⁺ release variability results from electrical defects, likely originating from Purkinje fibers, rather than the failure of Ca²⁺ release response to action potentials in mutated ventricular myocytes. This study provides insights into Ca²⁺ release and electrical dysfunction in CPVT.