A Mutation in Calsequestrin, CASQ2^D307H, Impairs Sarcoplasmic Reticulum Ca²⁺ Handling and Causes Complex Ventricular Arrhythmias in Mice

Abstract

Objective

A naturally-occurring mutation in cardiac calsequestrin (CASQ2) at amino acid 307 was discovered in a highly inbred family and hypothesized to cause Catecholaminergic Polymorphic Ventricular Tachycardia (CPVT). The goal of this study was to establish a causal link between CASQ2^D307H and the CPVT phenotype using an in vivo model.

Methods and Results

Cardiac-specific expression of the CASQ2^D307H transgene was achieved using the α-MHC promoter. Multiple transgenic (TG) mouse lines expressing CASQ2^D307H from 2- to 6-fold possess structurally normal hearts without any sign of hypertrophy. The hearts displayed normal ventricular function. Myocytes isolated from TG mice had diminished I_Ca-induced Ca²⁺ transient amplitude and duration as well as increased Ca²⁺ spark frequency. These myocytes, when exposed to isoproterenol and caffeine, displayed disturbances in their rhythmic Ca²⁺ oscillations and membrane potential and exhibited delayed afterdepolarizations. ECG monitoring revealed that TG mice challenged with isoproterenol and caffeine developed complex ventricular arrhythmias, including non-sustained polymorphic ventricular tachycardia.

Conclusions

The findings of the present study demonstrate that expression of mutant CASQ2^D307H in the mouse heart results in abnormal myocyte Ca²⁺ handling and predisposes to complex ventricular arrhythmias similar to the CPVT phenotype observed in human patients.

1. Introduction

CASQ2 is the most abundant Ca²⁺ buffering protein present in the lumen of the cardiac sarcoplasmic reticulum (SR). It allows calcium to be stored at total concentrations of up to 20 mM, while the free concentration remains ~1 mM ^1–4. It binds calcium with high capacity (40–50 Ca²⁺/CASQ2) and moderate affinity (k_d ~1 mM) ^1,3–5. Increasingly, it has become apparent that CASQ2 might also play an important role in regulating the SR Ca²⁺ release ^6,7. Studies have suggested that CASQ2 can interact functionally with the cardiac ryanodine receptor (RyR2) in vitro, either by binding with the RyR2 directly or through its interactions with junctin and triadin ^8–11. It is hypothesized that this interaction allows CASQ2 to be localized to the terminal cisternae of the junctional SR (jSR) ^8,10,12,13. Also, it was proposed that through interaction with triadin, CASQ2 might be able to communicate with the RyR2 ^6,14. However, the nature of this interaction and its exact role in Ca²⁺ release remain to be understood.

The importance of CASQ2 is highlighted by the fact that mutations in the CASQ2 gene have been recently linked to ventricular arrhythmias and sudden cardiac death ^15–17. In particular, a naturally-occurring mutation in cardiac CASQ2 at amino acid 307 (CASQ2^D307H) was discovered in seven Bedouin families ^18,19, as a cause for Catecholaminergic Polymorphic Ventricular Tachycardia (CPVT) ¹⁹. CPVT is a disease that is characterized by adrenergically-mediated ventricular tachycardia that can rapidly progress into ventricular fibrillation and sudden cardiac death ¹⁸. The characteristic symptoms are recurrent syncope, seizures, or sudden death triggered by exercise or emotional stress, both of which raise the levels of catecholamine ¹⁸.

Sequencing of DNA obtained from CPVT patients showed that CASQ2 exon 9 had a G→C substitution at nucleotide 1038 ¹⁹. This substitution resulted in a change from an aspartate residue to a histidine at position 307 in CASQ2 ¹⁹. This aspartate residue is invariant both in vertebrate and invertebrate CASQ ¹⁹. Acute expression of this mutant protein in isolated adult rat cardiomyocytes by adenoviral gene transfer resulted in decreased SR Ca²⁺ storage, Ca²⁺ channel-gated Ca²⁺ release (Ca²⁺ transients) and local Ca²⁺ release events (Ca²⁺ sparks) ²⁰. In addition, the rhythmicity of the Ca²⁺ transients was disrupted in these myocytes upon catecholamine stimulation.

In the present study, we wished to determine the effect of CASQ2^D307H on heart function by chronically expressing this mutant protein in vivo. This was accomplished by generating TG mouse lines that express CASQ2^D307H in the heart. Transgenic expression of mutant protein does not lead to cardiac hypertrophy. Cardiomyocytes isolated from the TG mice displayed in vitro arrhythmias that are consistent with the CPVT phenotype. Most importantly, complex ventricular arrhythmias, including non-sustained ventricular tachycardia, are triggered in the hearts of TG mice when challenged with isoproterenol and caffeine. Our findings in the TG mouse suggest that CASQ2^D307H mutation can contribute to the CPVT phenotype in humans.

2. Materials and Methods

2.1. Generation of TG mice expressing the CASQ2^D307H mutant protein

The D307H mutation was introduced into the CASQ2 open reading frame, using PCR method. Once the D307→H mutation was confirmed by DNA sequencing, the coding region of the mutant protein was sub-cloned into a vector containing the α-MHC promoter as described previously ^21–23. TG mice (FVB/N) expressing mutant CASQ2^D307H were generated by the TG core facility. This investigation conforms to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication # 85-23, revised 1996).

2.2. Western blotting analyses

Western blotting was performed following standard protocols ²⁴.

2.3. Isolation of mouse myocytes and electrophysiological recordings

Myocytes were isolated using established protocols ²⁵. Transmembrane ionic currents were measured using whole-cell patch-clamp recordings, as described previously ²⁶, with Axopatch 200B amplifier (Axon Instruments, USA) and pClamp-9 software. The external solution contained 140 mM NaCl, 5.4 mM KCl, 1.0 mM CaCl₂, 0.5 mM MgCl₂, 10 mM Hepes, and 5.6 mM glucose (pH 7.3). Patch pipettes (tip resistance of 1–3 MΩ) were filled with a solution that contained 90 mM K-aspartate, 50 mM KCl, 3 mM Na₂ATP, 3.5 mM MgCl₂, 10 mM Hepes, and 0.05 mM Fluo-3 K-salt (pH 7.3). The myocytes were stimulated by application of 400-ms-long voltage pulses to specified membrane potentials from a holding potential of −50 mV at 1-min intervals. For the measurements of calcium currents potassium ions was replaced by cesium.

2.4. Confocal Ca²⁺ measurements

Intracellular Ca²⁺ imaging was performed by using an Olympus Fluoview 1000 laser scanning confocal microscope equipped with an Olympus 60 Ч 1.4NA oil objective. Fluo-3 was excited by the 488-nm line of an argon-ion laser, and the fluorescence was acquired at wavelengths >510 nm in the line-scan mode of the confocal system at the rate of 2 ms per scan. Ca²⁺ spark parameters were quantified with a detection/analysis computer algorithm ²⁶.

2.5. Calcium sparks analysis

To measure Ca²⁺ sparks in intact myocytes, the myocytes were permeabilized by exposure to saponin (0.01% for 45–60 s). The permeabilization solution contained 100 mM K aspartate, 20 mM KCl, 3 mM MgATP 0.5 mM EGTA, 0.114 mM CaCl₂ (free [Ca²⁺], ~100 nM), 0.81 mM MgCl₂ (free [Mg²⁺], ~1 mM), 10 mM phosphocreatine, 10 mM HEPES, 5 U/ml creatine phosphokinase, and 8% dextran (40,000), pH 7.2. The control experimental solution contained 100 mM K aspartate, 20 mM KCl, 3 mM MgATP, 0.5 mM EGTA, 0.114 mM CaCl₂ (free [Ca²⁺], ~100 nM), 0.81 mM MgCl₂ (free [Mg²⁺], ~1 mM), 10 mM phosphocreatine, 10 mM Hepes, 0.03 mM Fluo-3 potassium salt (TefLabs, Austin, TX), and 5 U/ml creatine phosphokinase, pH 7.2.

The free [Ca²⁺] and [Mg²⁺] at given total Ca²⁺, Mg²⁺, ATP, and EGTA concentrations were calculated using WinMAXC 3.2 (Stanford University, Stanford, CA).

2.6. Analysis of cardiac function by echocardiography

Echocardiograms to assess systolic function were performed by using M-mode and two-dimensional measurements. The measurements represented the average of six selected cardiac cycles from at least two separate scans performed in random-blind fashion with papillary muscles used as a point of reference for consistency in the level of scan. End diastole was defined as the maximal left ventricular diastolic dimension and end systole was defined as the peak of posterior wall motion. Fractional shortening (FS), a surrogate of systolic function, was calculated from left ventricle dimensions as follows: FS = ((EDD − ESD)/EDD) ×100%.

2.7. ECG monitoring and induction of arrhythmia

ECG recordings were obtained from mice as previously described ^27–29. Mice over-expressing CSQ^D307H (n=8) and their wild-type littermates (n=8) were anesthetized with isoflurane (1–1.5%) at minimum effective concentrations, and placed on a heating pad (Braintree Scientific, Inc.) to maintain normothermia. ECGs were recorded using a physiologic data acquisition system (MP 100, Biopac Systems) with a sampling rate of 2 kHz for 84±17 min. After baseline recording (10 min), 5 pairs of mice (5 WT and 5 TG: 2 pairs each from 2- and 4-fold lines, and 1 pair from 6-fold line) received 4 doses of isoproterenol (2 mg/kg IP) in 10–20 min intervals, the first 2 of which were combined with caffeine (120 mg/kg IP; see refs. ^30,31). To ensure that the observed arrhythmias were adrenergically mediated, another group of mice (3 pairs from 2-fold line) received incremental doses of isoproterenol alone every 10 min (cumulative dose of 8.75±0.02 mg/kg IP). Both TG and age-matched WT littermates were tested the same day using the same drug protocol.

2.8. Statistical analysis

Cross tabulations with chi-square and yale correction factor or z-test were used as appropriate for categorical variables using a statistical software package (Sigmastat 2.03, Jandel Scientific).

3. Results

3.1. Generation and characterization of TG mice expressing CASQ2^D307H

Cardiac-specific expression of the CASQ2^D307H transgene was achieved using the α-MHC promoter ^21–23. Out of five positive TG lines containing the transgene, three TG lines (F8, F17, and F21) were propagated for further characterization. Three- to six-month-old mice were used for the studies, except as indicated.

The level of CASQ2^D307H protein expression in the three TG lines was determined by western blotting using an antibody specific for cardiac CASQ2 (Fig. 1A). Quantification of CASQ2 bands showed that the CASQ2^D307H protein is expressed 2-, 4- and 6-fold in TG lines F21, F8 and F17, respectively (Fig. 1A and data not shown). These data confirmed that we obtained multiple TG lines with graded expression of the CASQ2^D307H protein, thus allowing us to compare the effect of expressing different amounts of mutant protein. To determine whether expression of CASQ2^D307H protein induced alterations in other SR proteins, we determined the levels of RyR2, L–type calcium channel (DHPR 2α), Na⁺/Ca²⁺ exchanger (NCX1), triadin, SERCA 2a and phospholamban (PLB) in the 4- and 6-fold lines (Fig. 1B and data not shown). Quantitation and correction for loading variations indicated that there were no significant changes in the expression levels of these proteins. Therefore 2- to 6-fold expression of CASQ2^D307H does not alter the expression levels of related SR Ca²⁺ transport proteins. For most of the studies described below we compared the 2-fold and 6-fold overexpressors.

Fig. 1. (A–B) Expression of SR proteins in TG (2- and 6-fold CASQ2^D307H) and WT hearts.

(A) 5–10 μg of whole heart homogenates (2- and 6-fold CASQ2^D307H) were separated on 10% SDS-PAGE gels and probed with antibodies against CASQ2, SERCA 2a, triadin, α-actin and α-actinin. (B) 4–8 μg (10–30 μg for RyR2) of whole heart homogenates (6-fold CASQ2^D307H) were separated on SDS-PAGE gels and probed with antibodies against RyR2 (4–20% gradient gel), SERCA 2a (10% gel), phospholamban (PLB; 15% gel), L-type calcium channel (DHPR 2α; 10% gel) and Na⁺/Ca²⁺ exchanger (NCX1, 10% gel). The 120-kDa full length NCX1 protein is shown ⁴⁴. (C) CASQ2 immunofluorescence in TG and WT myocytes. Mice ventricular cardiomyocytes from WT littermates and TG (6-fold CASQ2^D307H) were immunostained with mouse anti α-actinin (1:500) or rabbit anti-calsequestrin (1:200), followed by Alexa 488-conjugated goat anti-mouse (1:300) and Cy^™3-conjugated Donkey anti-rabbit (1:300 dilution) antibodies.

3.2. TG cardiac myocytes show subtle structural changes

To determine whether the mutant CASQ2^D307H was being properly localized to the junctional SR, confocal and electron microscopy (EM) were performed. Imaging of the stained cells revealed the expected patterns of fluorescence for CASQ2 and α-actinin (Fig. 1C). The CASQ2 staining co-localized with the α-actinin staining, consistent with the fact that the Z-line (α-actinin) and junctional SR (CASQ2) are in close proximity to one another. The only difference between the TG and WT staining pattern was that the intensity of CASQ2 staining appeared brighter in the TG myocytes, which is consistent with the over-expression of CASQ2^D307H in the 6-fold line.

The ultrastructure of myocytes expressing CASQ2^D307H was very similar to that of WT littermate cells but subtle changes were observed. We observed variations in the size and internal configuration of the jSR cisternae associated with the transverse tubules. In WT myocardium, the jSR is typically in the form of flat cisternae with a dense content, due to calsequestrin, that is condensed into periodic densities (Fig. 2A). Most of the cisternae in the mutant hearts have the same general structure as WT ones (Figs. 2B and C). Some however, show a decrease in the dense content, a general loosening of the content disposition and some widening of the cisternae (Fig. 2E and F). Some TG myocytes show evidence for a disturbance of T tubules. Normally, T tubules run transversely at the Z lines and have occasional longitudinal branches. In thin sections, each intermyofibrillar space has either none or a single T tubule profile. Some myocytes from CASQ2^D307H TG hearts show multiple T tubule profiles, indicating a convoluted tubule (data not shown).

Fig. 2. jSR alterations in CASQ2^D307H (6-fold) myocytes.

(A) In WT myocardium, the jSR cisternae are narrow, they contain a condensed form of CASQ2 and are closely apposed to wide T tubule profiles (T). (B,C) Most of the jSR cisternae in the mutant myocytes are quite similar to those in WT. (D–F) Alterations of jSR in mutant myocytes include a looser appearance of the CASQ2 in the lumen and an increase in width of the cisternae. The double arrows in A, E and F indicate the widening of terminal cisternae in TG compared to WT myocytes.

3.3. TG hearts expressing CASQ2^D307H show normal overall structure and contractility

We also wanted to make sure that expression of mutant CASQ2^D307H does not cause cardiac hypertrophy/pathology that may complicate our analysis. Several lines of evidence suggest that the TG hearts are essentially normal. First, there was no change in the heart weight-to-body weight ratios of TG (4.4±0.7 mg/g) compared to WT (4.3±0.3 mg/g) mice. Second, hearts from TG and WT littermates were subjected to standard histological analysis by staining with hematoxylin/eosin. These hearts appear normal and do not exhibit hypertrophy or fibrosis. Myocyte structure is similar in appearance in all hearts (Fig. 3A). Third, we examined the expression level of genes that are known markers of hypertrophy: atrial natriuretic factor (ANF), β-myosin heavy chain (β-MHC) and skeletal α-actin by RT-PCR. These genes are typically re-activated under conditions of cardiac hypertrophy, but we did not see activation of these genes in TG hearts (data not shown).

Fig. 3. TG hearts show normal morphology and function is unaltered.

(A) Hearts from TG mice (~1 year old) were subjected to standard histological analysis by staining with hematoxylin/eosin. (B) Examples of M-mode echocardiography in 4–6 month old WT and TG mice (6-fold CASQ2^D307H) showing normal cardiac chamber and contractility with no evidence of structural abnormalities. Summary of data are shown in Table 1.

M-mode echocardiography in WT (Fig. 3B, top panel) and TG (lower panel) mice showed normal cardiac chamber and contractility with no evidence of structural abnormalities. There were no differences in the left ventricular end systolic dimension (LV-ESD) and end diastolic dimension (LV-EDD) and fractional shortening (FS) in TG mice compared to their littermate controls (see Table 1).

Table 1.

Echocardiographic analysis of WT and TG hearts.

	WT (n = 8)	TG (n = 9)
LV-ESD (cm)	0.15±0.02	0.14±0.01
LV-EDD (cm)	0.32±0.02	0.33±0.01
ES-PW (cm)	0.14±0.01	0.13±0.01
ED-PW (cm)	0.07±0.01	0.07±0.01
FS%	54±3.3	58.1±3.6

Data shown represent mean ± s.e.m.; left ventricular end systolic dimension (LV-ESD), end diastolic dimension (LV-EDD), posterior wall thickness at end systole and end diastole (ES-PW and ED-PW, respectively) and fractional shortening (FS), a surrogate of systolic function.

3.4. SR Ca²⁺ content and release

Calsequestrin is a Ca²⁺ storage protein and expression of mutant CASQ2^D307H could affect calcium store and release properties. Therefore, we studied the calcium handling properties of myocytes isolated from TG and WT hearts. Caffeine applications (10 mM) were used to assess changes in the total SR Ca²⁺ content in each group of isolated myocytes. The relative amounts of Ca²⁺ released from the SR after caffeine administration were assessed from changes in both Fluo-3 fluorescence and Na⁺-Ca²⁺ exchange current (I_NCX) in myocytes dialyzed with the Ca²⁺ indicator Fluo-3 (Fig. 4A). Neither the caffeine-induced transients [2.39±0.24 (WT) vs. 2.06±0.22 (TG)], nor the I_NCX current [0.55±0.12 (WT) vs. 0.52±0.15 (TG)] changed significantly (Fig.4B–C).

Fig. 4. Intracellular Ca²⁺ load, I_Ca and Ca²⁺ transients in WT and TG myocytes (6-fold CASQ2^D307H).

(A) Patch clamp recordings were used to measure Ca²⁺ transients (top traces) and NCX currents (bottom traces); pooled data for caffeine-induced Ca²⁺ transients (B) and NCX currents (C); (D) representative recordings of I_Ca (bottom traces) and calcium transients (top traces) evoked by depolarization steps from −50 mV MP to 0 mV; voltage dependencies of Ca²⁺ transients (E) and I_Ca (F) in ventricular myocytes, isolated from WT (squares and solid lines) and TG (6-fold CASQ2^D307H) hearts (circles and dashed lines).

3.5. Ca²⁺ sparks, Ca²⁺ transients and I_Ca

The effects of expressing mutant CASQ2^D307H on I_Ca and intracellular [Ca²⁺] transients in patch-clamped myocytes are illustrated in Figure 4D–F. There were no apparent changes in the parameters of I_Ca in myocytes expressing CASQ2^D307H and cells from WT littermates (Fig. 4F, Table 2). The peak amplitude of I_Ca was nearly identical for all experimental groups of cells. In addition, the time course of I_Ca decay was similar (Table 2). Thus, expression of CASQ2^D307H did not change the characteristics of the Ca²⁺ trigger for Ca²⁺ release from the SR. However, Ca²⁺ transients from TG mice were smaller and had reduced τ_decay compared to WT myocytes (Fig. 4D–E and Table 2). Measurement of spontaneous Ca²⁺ sparks in permeabilized myocytes from TG mice and their WT littermates showed a trend toward decreased Ca²⁺ spark amplitude and a significant increase in Ca²⁺ spark frequency (Supplemental Figure and Table 3).

Table 2.

Parameters of I_Ca and Ca transients.

	ICa			Ca transients			N of cells
	Peak Amplitude (pA/pF)	Tfast (ms)	Tslow (ms)	F/F0	Rise Time (ms)	Tdecay (ms)
WT	−4.88±0.55	15.7±4.7	76±25	1.96±0.16	28±4	320±16	12 (7 animals)
TG	−4.19±0.55	16.1±4.2	75±23	1.61±0.14*	24±3	268±15*	9 (6 animals)

^*Significantly different at P<0.05, compared with WT

Table 3.

Characteristics of spontaneous Ca²⁺ sparks recorded in permeabilized myocytes isolated from WT mouse hearts and TG mice expressing CASQ2^D307H.

	Amplitude ΔF/F0	Rise time, ms	HA Width, μm	Frequency, 100 μm−1 s−1	N of Sparks	N of Cells	N of Mice
WT	0.78±0.01	8.0±0.1	2.20±0.02	7.7±0.5	1152	80	6
TG	0.75±0.01	8.7±0.1*	2.24±0.02	10.2±0.8*	1077	62	3

^*Significantly different at P<0.05, One Way ANOVA.

3.6. Ca²⁺ cycling in rhythmically paced myocytes

The effects of isoproterenol treatment (1 μmol/L) on periodic Ca²⁺ transients in WT and TG myocytes are illustrated in Figure 5. To induce arrhythmia in myocytes, we used a modified protocol ³⁰ where the bath solution contains 100 μM of caffeine. The myocytes were stimulated at 1 Hz, and membrane potential (MP) changes were recorded in the current-clamp mode. The exposure of WT myocytes to isoproterenol caused an increase in the amplitude of Ca²⁺ transients without any apparent disturbances in periodic Ca²⁺ cycling (Fig. 5A–B; results are representative of six myocytes). However, in TG myocytes expressing CASQ2^D307H, the amplitude of Ca²⁺ current-induced Ca²⁺ transients was reduced (Fig. 5A). The AP duration between control and TG myocytes was not significantly different (APD₉₀ is 57±24 for control animals and 45±19 for TG). Isoproterenol administration in the presence of caffeine caused profound disturbances in Ca²⁺ cycling manifested by extra systolic, spontaneous Ca²⁺ transients. As seen in the line-scan images (Fig. 5B), spontaneous release usually originated locally and then propagated through the cell as a regenerative Ca²⁺ wave. Importantly, the membrane potential traces showed delayed afterdepolarizations (DADs) and the occasional generation of action potentials at the time of spontaneous Ca²⁺ transients. Similar results were obtained in four other myocytes. These results indicate that expression of CASQ2^D307H destabilizes the Ca²⁺ release mechanism, leading to spontaneous, premature discharges of SR Ca²⁺ stores in myocytes undergoing periodic pacing.

Fig. 5. TG myocytes (6-fold CASQ2^D307H) show spontaneous Ca²⁺ transients.

Recording of membrane potential (top traces), along with line-scan confocal images (middle traces) and temporal profiles (bottom traces) in myocytes before (A) and after superfusion (B) with 1 μM of ISO and 100 μM of caffeine.

3.7. Complex ventricular arrhythmias are triggered in TG hearts by isoproterenol and caffeine

Continuous ECG monitoring prior to drug challenge did not reveal any significant spontaneous ventricular arrhythmias in WT mice. In contrast, CASQ2^D307H mice had more spontaneous supraventricular ectopic beats at baseline than WT mice (cumulative numbers of observations were 378 vs. 8 in TG and WT mice, respectively, P<0.001).

After challenge with isoproterenol alone or in combination with caffeine, mutant CASQ2^D307H mice developed significantly more supraventricular and junctional ectopic beats compared to WT (cumulative numbers of observations for junctional ectopic beats was 2862 vs. 39 in TG and WT mice, respectively, P<0.001). CASQ2^D307H mice did not have an increased number of simple ventricular arrhythmias compared to WT (Fig. 6A). However, after drug challenge, the number of total complex ventricular arrhythmias was increased 4-fold (P<0.001) in CASQ2^D307H mice compared to WT mice (Fig. 6A). Furthermore, there was a significant increase (P<0.05) in the number of CASQ2^D307H mice that developed complex ventricular arrhythmias (Fig. 6B), including non-sustained ventricular tachycardia, compared to WT mice (Fig. 6C+D).

Fig. 6. Mice expressing mutant CASQ2^D307 displayed more complex forms of ventricular arrhythmia, including non-sustained ventricular tachycardia, compared to aged-matched WT mice.

(A) Cumulative number of simple and complex ventricular arrhythmias in TG (n=8) from 2-, 4- and 6-fold over-expression and age-matched WT (n=8) mice. Under light anesthesia, mice were challenged with IP isoproterenol injection alone or in combination with caffeine injection. Simple ventricular arrhythmias included single isolated ventricular ectopic beats. Complex ventricular arrhythmias included: couplet, bigeminy, trigeminy and non-sustained ventricular tachycardia. ** P< 0.001 TG vs. WT mice. (B) Higher proportion of mice expressing CASQ^D307H (n=8), which displayed complex forms of ventricular arrhythmias compared to WT (n=8), following treatment with isoproterenol alone or in combination with caffeine injection. * P< 0.05 TG vs. WT mice. (C) Representative ECG tracings (lead 1 top and lead 3 bottom) in a TG mouse (6-fold CASQ2^D307H) displaying a run of non-sustained polymorphic ventricular tachycardias (NSVT). Please note the presence of 2 couplets before and after the run of NSVT, which can be considered as an attempted, but aborted VT. (D) Magnification of section A of the ECG tracings (lead 1 top and lead 3 bottom) displaying 2 non-sustained polymorphic ventricular tachycardias. Please note a junctional escape beat (e.g., compensatory pause) following each NSVT. S: sinus beat; P: compensatory pause; J: Junctional premature beat. VT: ventricular tachycardia.

4. Discussion

CPVT is a potentially devastating condition that can lead to sudden cardiac death even in very young children after physical or emotional stress. CPVT can be caused by mutations in either the ryanodine receptor (RyR2) or the CASQ2 gene ^{15,17,19,32–35}. There are several different RyR2 mutations that are linked to CPVT and they are all dominant, while there are fewer known CASQ2 mutations and these are all recessive ^15,17.

In the present study, we chose to introduce increasing amounts of mutant CASQ2^D307H to compete with the endogenous WT CASQ2 protein to determine its effect on Ca²⁺ handling and heart function. Since the CASQ2^D307H mutation is recessive, we predicted that low amounts of CASQ2^D307H would not significantly affect heart function, while higher amounts of CASQ2^D307H would out-compete the endogenous WT CASQ2 and lead to ventricular arrhythmias. Our findings show that TG mice did not develop a significant amount of spontaneous ventricular arrhythmias under resting conditions. However, when challenged with catecholamines, the TG hearts developed a much higher level of complex ventricular arrhythmias compared to WT hearts (Fig. 6A). These complex VEBs included couplets, bigeminy, trigeminy, as well as non-sustained ventricular tachycardia, both monomorphic and polymorphic, consistent with the CPVT phenotype observed in patients in that they are triggered by catecholamines. We observed arrhythmias in all 3 TG lines and did not observe a graded response between lines.

The predisposition to complex ventricular arrhythmias in our TG mice could be due to the D307H mutation itself and/or increased expression of CASQ2 in the heart. Recent studies showed that over-expression of WT mouse CASQ2 (20-fold) and canine CASQ2 (10-fold) in the mouse hearts can cause mild to severe cardiac hypertrophy and result in an increase in SR Ca²⁺ storage capacity ^36,37. Despite an increase in Ca²⁺ storage capacity, Ca²⁺ release was severely compromised in these mice. These data suggested that excessive amounts of CASQ2 can also lead to abnormally high levels of Ca²⁺ storage but reduced Ca²⁺ release. In contrast, we did not observe cardiac hypertrophy or pathology by increasing CASQ2 levels from 2- to 6-fold in TG hearts. This was supported by several lines of evidence, including histological data showing normal morphology and structure. In addition, expression of CASQ2^D307H did not cause an increase in SR Ca²⁺ storage. These facts represent a striking difference between the CASQ2^WT and CASQ2^D307H expression models suggesting that the mutant protein behaves very differently from the WT protein.

Biochemical studies on purified CASQ2^D307H mutant protein showed that the D307H mutation affects the Ca²⁺-dependent conformation of the protein and the ability of CASQ2 to interact with junctin and triadin ³⁸. An additional possibility is that the mutation may also affect the ability of CASQ2 to polymerize into long polymers that appear to be crucial to its buffering capacity ^39,40. Our findings suggest that the expression of CASQ2^D307H destabilizes the Ca²⁺ release mechanism, possibly by interfering with the CASQ2-triadin and CASQ2-junctin interactions that regulate RyR2 function ⁶, leading to spontaneous, premature discharges of SR Ca²⁺ stores in myocytes undergoing periodic pacing. Indeed, the membrane potential traces of TG myocytes showed delayed afterdepolarizations (DADs) and the occasional generation of action potentials at the time of spontaneous Ca²⁺ transients (Fig. 5). Increased spontaneous Ca²⁺ release leads to the development of DADs through activation of depolarizing membrane currents via the Na-Ca exchange. These oscillations in membrane potential are thought to be the underlying cause of ventricular arrhythmias ^20,41–43 and would also appear to be the mechanism underlying the complex ventricular arrhythmias observed in our TG mice.

In conclusion, the findings of the present study demonstrate that over-expression of CASQ2^D307H in the mouse heart results in abnormal myocyte Ca²⁺ handling and complex ventricular arrhythmias that could contribute to the CPVT phenotype observed in human patients. Our study provides important evidence to establish a link between a mutation in CASQ2 and complex ventricular arrhythmias that are consistent with the CPVT phenotype.

Limitations

A potential limitation associated with our methodology was the lack of evaluation of arrhythmias in conscious mice. However, one of the main advantages of recording ECG in anesthetized mice is the ability to record from multiple leads to improve the diagnostic specificity, which can be technically very difficult with telemetric devices in conscious and exercising mice.

Abbreviations

SR

Sarcoplasmic reticulum

CASQ2

Calsequestrin

CPVT

Catecholaminergic polymorphic ventricular tachycardia

WT

Wild-type

TG

Transgenic

RyR2

Cardiac ryanodine receptor