Catecholaminergic-induced arrhythmias in failing cardiomyocytes associated with human HRC^S96A variant overexpression

Abstract

The histidine-rich calcium binding protein (HRC) Ser96Ala polymorphism was shown to correlate with ventricular arrhythmias and sudden death only in dilated cardiomyopathy patients but not in healthy human carriers. In the present study, we assessed the molecular and cellular mechanisms underlying human arrhythmias by adenoviral expression of the human wild-type (HRC^WT) or mutant HRC (HRC^S96A) in adult rat ventricular cardiomyocytes. Total HRC protein was increased by ∼50% in both HRC^WT- and HRC^S96A-infected cells. The HRC^S96A mutant exacerbated the inhibitory effects of HRC^WT on the amplitude of Ca²⁺ transients, prolongation of Ca²⁺ decay time, and caffeine-induced sarcoplasmic reticulum Ca²⁺ release. Consistent with these findings, HRC^S96A reduced maximal sarcoplasmic reticulum calcium uptake rate to a higher extent than HRC^WT. Furthermore, the frequency of spontaneous Ca²⁺ sparks, which was reduced by HRC^WT, was increased by mutant HRC^S96A under resting conditions although there were no spontaneous Ca²⁺ waves under stress conditions. However, expression of the HRC^S96A genetic variant in cardiomyocytes from a rat model of postmyocardial infarction heart failure induced dramatic disturbances of rhythmic Ca²⁺ transients. These findings indicate that the HRC Ser96Ala variant increases the propensity of arrhythmogenic Ca²⁺ waves in the stressed failing heart, suggesting a link between this genetic variant and life-threatening ventricular arrhythmias in human carriers.

arrhythmic events such as ventricular tachycardia or ventricular fibrillation are the most common causes of sudden cardiac death in dilated cardiomyopathy patients (12). During progression of cardiac remodeling, the expression levels of many Ca²⁺ cycling proteins are altered (14), which contribute to defective Ca²⁺ handling in failing hearts. Among these alternations, the current density of Na⁺/Ca²⁺ exchanger (NCX) is enhanced (4, 22) and Ca²⁺ leak from the sarcoplasmic reticulum (SR) is elevated, resulting in disturbances of rhythmic cardiac contraction, either through early afterdepolarization or delayed afterdepolarization (DAD).

SR Ca²⁺ load has been suggested to play an important role in regulating SR Ca²⁺ leak, especially in failing hearts (13). Along these lines, the expression level of histidine-rich calcium binding protein, which has structural analogy to calsequestrin (CSQ), has been found reduced in experimental and human heart failure. Interestingly, HRC appears to directly interact with both triadin (16, 25) and sarco(endo)plasmic reticulum Ca²⁺-ATPase 2a (SERCA2a; Ref. 3) and has been implicated as a regulator of both SR Ca²⁺ release and uptake. Transgenic models with HRC overexpression revealed decreased Ca²⁺ transient amplitude and prolonged transient decay time (8), implicating HRC as a regulator of Ca²⁺ cycling in vivo.

Studies on identification of human mutations in key SR Ca²⁺ cycling genes have provided further insights on the functional role of these proteins in the heart. Actually, several point mutations in the ryanodine receptor (RyR; Refs. 9, 18) and CSQ (21, 23) have been linked to sudden cardiac death and the underlying mechanism involved defective RyR activity. Generally, stress conditions or exercise related catecholaminergic polymorphic ventricular tachycardia have been associated with increased functional activity of RyR channels, which contributed to increased SR Ca²⁺ leak and disturbed intracellular Ca²⁺ homeostasis leading to ventricular arrhythmias. However, parallel studies on human mutations in junctin, triadin, and HRC, known as effective regulators of RyR activity and SR Ca²⁺ release, are not currently available.

Recently, a genetic variant of HRC, Ser96Ala, was shown to have a correlation with ventricular arrhythmogenesis in dilated cardiomyopathy (DCM; Ref. 2). Although HRC^S96A occurred with comparable frequencies in DCM patients and normals, this variant was associated with a history of ventricular tachycardia or ventricular fibrillation before screening entry. Furthermore, the percentage of homozygous HRC^S96A DCM patients with an implantable cardioverter-defibrillator was almost double of the number of HRC^S96A heterozygotes, which suggested a strong dosage effect of this polymorphism on ventricular arrhythmias. To assess the underlying mechanisms, we expressed the human wild-type and S96A variant of HRC in normal or heart failure rat myocytes. Our results demonstrate that the single substitution of Ser96Ala modulates HRC function in normal cardiomyocytes, while it elevates the occurrence of Ca²⁺ waves in heart failure cells under stress conditions.

MATERIALS AND METHODS

Recombinant adenoviral constructs.

Human HRC cDNA was purchased from Open Biosystems. cDNA expressing HRC S96A was generated using the Quik-change-site-directed mutagenesis II kit (Stratagene). Human HRC cDNA was firstly cloned into the pShuttle vector and then transferred into the Adeno-X viral DNA using the Ad-Easy XL system (Stratagene). Adenoviruses containing the cDNA sequence of wild-type HRC (Ad.HRC^WT), mutant HRC (Ad.HRC^S96A), or greenfluorescent protein (Ad.GFP) were assembled and amplified in HEK-293 cells. In the final step, the viruses were purified, using the adenovirus mini purification kit (Virapur) and tittered, using the Adeno-X rapid titer kit (Clonetech).

Myocardial infarction-induced heart failure and echocardiographic evaluation.

After echocardiographic assessment of baseline function, male Wistar rats were subjected to left descending coronary artery ligation as previously described (15). Left ventricular echocardiography was performed to monitor the heart function with a General Electric Vivid 7 dimension instrument equipped with a linear i13L transducer. There was a progressive increase in left ventricular end-diastolic volume. Eight weeks following the surgery, cardiac ejection fraction was reduced to 32%, and this model was defined as a postmyocardial infarction heart failure model. All surgical procedures described in this study were approved by the Institutional Animal Care and Use Committee of Peking University.

Adenoviral infection of cardiomyocytes.

Cardiomyocytes were isolated from either normal or failing rat hearts as previously described (8). After 2 h of being cultured on coverslips or in dishes precoated with 10 μg/ml laminin, myocytes were infected with correspondent adenoviruses. Experiments were performed 48 h after infection and ∼100% of cells appeared infected.

HEK 293 cell culture and coimmunoprecipitation of HRC with SERCA2.

HEK 293 cells (ECACC, Salisbury, UK) were maintained in DMEM supplemented with 10% FBS (Invitrogen, Carlsbad, CA). Full-length human SERCA2 (NIH Mammalian Gene Collection Clone ID 5503508; Invitrogen) and GFP-HRC S96 or GFP-HRC A96 constructs were transiently transfected in HEK293 cells with Lipofectamine 2000 (Invitrogen), according to the manufacturer's instructions. Forty-eight hours after transfection, cells were harvested and lysed in 50 mM Tris·HCl, 150 mM NaCl, and 1% NP40 with protease inhibitors. Cell lysates were precleared with 30 μl protein G-Sepharose (Amersham Biosciences Europe, Uppsala, Sweden) for 1 h at 4°C. The precleared cell lysates were then incubated with 4 μg of SERCA2 antibody (Affinity Bioreagents) and 30 μl protein G-Sepharose at 4°C overnight. The beads were washed twice with lysis buffer and were analyzed by Western blots with either anti-SERCA (Affinity Bioreagents) or anti-HRC (Sigma-Aldrich) antibodies.

SR Ca-uptake rates.

Initial rates of SR Ca uptake were determined in cardiomyocyte homogenates after 48 h of infection, with the use of oxalate and ⁴⁵CaCl₂, as previously described (17). Briefly, 250 μg of homogenate were incubated at 37°C in a reaction buffer containing: 40 mmol/l imidazole pH 7.0, 95 mmol/l KCl, 5 mmol/l NaN₃, 5 mmol/l MgCl₂, 0.5 mmol/l EGTA, and 5 mmol/l potassium oxalate. The initial uptake rates were determined over a range of Ca concentrations (pCa 8 to 5). Ca uptake into SR vesicles was initiated by addition of 5 mmol/l ATP, and aliquots were filtered through a 0.45-μm Millipore filter after 0, 30, 60, and 90 s. The specific Ca-uptake values, Ca affinity (EC₅₀), and maximal Ca-uptake rates (V_max) were analyzed by nonlinear regression, using the OriginLab 5.1 program.

Quantitative immunoblotting.

Cultured cardiomyocytes were harvested and lysed as previously described (8). An appropriate amount of protein sample was separated by SDS-PAGE and transferred to a nitrocellulose membrane. After being blocked with 5% nonfat milk, the membranes were incubated with the corresponding primary antibodies: phospholamban (1:5,000), CSQ (1:1,000), actin (1:5,000; Affinity Bioreagents), SERCA2a (1:1,000; homemade polyclonal rabbit anti-SERCA antibody), junctin (1:1,000; Santa Cruz Biotechnology), triadin (1:1,000; Thermo Scientific), and HRC (1:5,000; a generous gift from Dr. Woo Jin Park). The corresponding horseradish peroxidase linked secondary antibody was purchased from GE Healthcare. Membranes were then developed using SuperSignal West Pico chemiluminescent substrate from Thermo Scientific. Protein loading was normalized to endogenous actin level and quantitated by using AlphaEaseFC software (Alpha Innotech, San Leandro, CA).

Ca²⁺ cycling measurements.

Cardiomyocytes were loaded with Rhod-2 AM (10 μmol/l, 30 min; Invitrogen), and Ca²⁺ transients as well as Ca²⁺ sparks were acquired with Zeiss LSM 510 confocal microscope under line=scan mode. To measure Ca²⁺ transients, myocytes were paced at 0.5 Hz in solution containing (in mmol/l): 137 NaCl, 20 HEPES, 10 d-glucose, 5.4 KCl, 1.2 MgCl₂, 1.2 NaH₂PO₄, and 1 CaCl₂ 1 (pH 7.4). To assess SR Ca²⁺ load, caffeine-induced Ca²⁺ release was initiated by rapid application of 20 mM caffeine after steady-state field stimulation. Stress condition was induced with 1 μM isoproterenol treatment and 1-Hz field stimulation. Rhod-2 was excited at 543 nm, and emitted fluorescence was collected at >560 nm.

For resting Ca²⁺ measurement, myocytes were loaded with indo-1 AM (10 μmol/l, 30 min) and excited with ultraviolet light. The emission fluorescence was recorded by a CCD camera (Luca; Andor Technology) at 405 nm (F405) and 485 nm (F485) separated by an emission splitter. Resting Ca²⁺ was estimated by the ratio of F405/F485. All experiments were performed at room temperature. Image processing and data analysis were performed using IDL program (Research Systems).

Statistical analysis.

All values were expressed as means ± SE. Statistical comparisons among three groups were evaluated with one-way ANOVA. In all analyses, P values <0.05 were considered statistically significant.

RESULTS

Overexpression of human HRC^WT or HRC^S96A in adult rat cardiomyocytes.

Rat cardiac myocytes were infected with Ad.GFP, Ad.HRC^WT, or Ad.HRC^S96A. At 48 h postinfection, quantitative immunoblotting revealed 50% increases in total HRC levels of HRC^WT and HRC^S96A groups (Fig. 1A). Interestingly, the protein level of triadin was increased by twofold in HRC^WT or HRC^S96A, compared with GFP (Fig. 1B), while the expression levels of other key SR Ca²⁺ cycling proteins, such as SERCA2a, CSQ, phospholamban, junctin, and RyR, did not show any alternations (Fig. 1A).

Fig. 1.

Effects of human histidine-rich calcium binding protein wild type (HRC^WT) and HRC^S96A overexpression on intracellular Ca²⁺ transients. A: quantitative immunoblotting showed total HRC protein was increased by ∼50% in both the human Ad.HRC^WT- and Ad.HRC^S96A-infected cells, compared with Ad.GFP. There were no alternations in the expression levels of sarco(endo)plasmic reticulum Ca²⁺-ATPase 2a (SERCA2a), calsequestrin (CSQ), phospholamban (PLN), ryanodine receptor (RyR), and junctin between these groups. B: triadin protein level was increased in HRC^WT and HRC^S96A, compared with GFP. C: confocal line-scan images of Ca²⁺ transients under 0.5-Hz field stimulations. D: corresponding time courses of Ca²⁺ transients. E and F: statistical results of Ca²⁺ transient amplitude and decay time (T50, time from peak amplitude to 50% decrease of Ca²⁺ transients). n = 30 to 36 cells from 6–7 hearts for each group. *P < 0.05, compared with GFP. ^#P < 0.05, compared with HRC^WT.

Regulation of intracellular Ca²⁺ transients by HRC^WT and HRC^S96A.

To assess the functional significance of human HRC protein expression in rat cardiomyocytes, intracellular Ca²⁺ transients were recorded during field stimulations. Figure 1, C and D, is a typical example of line-scan images and their corresponding time courses from Ad.GFP-, Ad.HRC^WT-, or Ad.HRC^S96A-infected cells, respectively. Overexpressing HRC^WT resulted in a marked decrease (29%) in Ca²⁺ transient amplitude, compared with the GFP group, in agreement with earlier studies (8) on mouse HRC overexpression. Interestingly, expression of the human variant, HRC^S96A at similar levels as HRC^WT, caused a further reduction in the Ca²⁺ transient amplitude (50% decrease compared with GFP; Fig. 1E). The transient half-decay time (T50) was also delayed 25% with human HRC^WT overexpression, while the HRC^S96A induced additional decelerated Ca²⁺ decline (62% compared with GFP; Fig. 1F).

These results on human HRC further confirm previous findings (8) on the functional role of this protein in the heart and suggest that Ala substitution at position Ser96 may increase the inhibitory activity of HRC on SR Ca²⁺ uptake.

SR Ca²⁺ load and fractional release.

Previous studies (16) have shown that HRC shares structural similarity with CSQ and may play a role as a Ca²⁺ storage protein. We (8) have also shown that acute overexpression of mouse HRC in rat cardiomyocytes increased SR Ca²⁺ load. In the present study, HRC^WT- or HRC^S96A-infected myocytes were subjected to 20 mM caffeine after steady-state 0.5-Hz field stimulations (Fig. 2A). Expression of HRC^WT resulted in a dramatic increase in SR Ca²⁺ load (40%), while expression of HRC^S96A did not alter SR Ca²⁺ load, compared with the GFP control (Fig. 2B). We also determined the ratio of electrically evoked Ca²⁺ transients to caffeine associated SR Ca²⁺ load, referred to as fractional release, to assess the fraction of Ca²⁺ within the SR that responds to electrical stimulation. This fractional Ca²⁺ release was decreased by 38 and 52% upon HRC^WT and HRC^S96A expression, respectively (Fig. 2C). Moreover, the half-decay time of caffeine-induced Ca²⁺ transient did not change among the three groups, which was an index of similar Na⁺/Ca²⁺ exchanger activity (Fig. 2D).

Fig. 2.

Effects of human HRC^WT and HRC^S96A on sarcoplasmic reticulum (SR) Ca²⁺ storage. A: typical line-scan image and its time course of steady-state Ca²⁺ transients under 0.5-Hz field stimulation, followed by caffeine-induced SR Ca²⁺ release. B: statistics of caffeine-induced SR Ca²⁺ release amplitude. C: field stimulation evoked Ca²⁺ transient expressed as percentage of caffeine-induced SR Ca²⁺ store (fractional release) in each group. D: T50 of caffeine-induced SR Ca²⁺ release. n = 16–18 cells for each group *P < 0.05, compared with GFP. P < 0.05, compared with HRC^WT.

Variant of S96A HRC on its interaction with SERCA2 and modulation of SERCA2 function.

To determine whether HRC^S96A and HRC^WT have different binding affinities for SERCA2, we generated full-length human HRC^WT and HRC^S96A and cotransfected HEK293 cells with either of these constructs together with full-length human SERCA2. We then coimmunoprecipitated the SERCA2/HRC complex and observed that the HRC levels were almost twofold higher from the HRC^S96A cell lysates, compared with the HRC^WT cells, while the levels of SERCA2 were similar between these two groups (Fig. 3A). Furthermore, to assess the effects of HRC^WT and HRC^S96A on SERCA2 function, oxalate-supported SR Ca-transport rates were determined. HRC^WT overexpression resulted in significantly decreased maximal calcium uptake rates compared with GFP, while the HRC^S96A caused a further reduction (Fig. 3, B and C). However, the EC₅₀ of SERCA2 for calcium was similar among the three groups (Fig. 3D).

Fig. 3.

Interaction of human HRC^WT and HRC^S96A with SERCA2 and alterations of SR Ca transport. A: anti-SERCA2 antibody, coupled to protein G-agarose beads, was used for coimmunoprecipitation of HRC from HEK293 cells, coexpressing SERCA2 and HRC^WT or HRC^S96A. Top: immunoblotting with SERCA2 antibody; bottom: immunoblotting with HRC antibody. B: oxalate-supported SR calcium uptake rates in homogenates from HRC^WT- or HRC^S96A-infected cells. C and D: maximal calcium uptake rates and the EC₅₀ of SERCA2 for Ca, respectively (n = 4–6 preparations, each in duplicate).

Modulation of spontaneous Ca²⁺ sparks and resting Ca²⁺ by HRC^WT and HRC^S96A.

To further understand the effects of HRC^S96A on the Ca²⁺ release properties, we measured spontaneous Ca²⁺ sparks in resting intact cardiomyocytes. HRC^S96A and HRC^WT had opposite effects on the occurrence of Ca²⁺ sparks. HRC^WT expression diminished the frequency by 31%, compared with GFP control. In contrast, HRC^S96A induced hyperactive Ca²⁺ sparks, increasing the spark frequency by 40% (Fig. 4, A and B). Pooled data demonstrated unchanged spatial and temporal properties of sparks among the three groups (Fig. 4, C–E).

Fig. 4.

Effects of human HRC^WT and HRC^S96A on spontaneous Ca²⁺ sparks properties. A: representative line-scan images of Ca²⁺ sparks acquired in cells infected with either Ad.GFP, Ad.HRC^WT, or Ad.HRC^S96A. B–E: cumulative data on Ca²⁺ spark frequency (B), amplitude (C), full duration at half-maximal (FDHM; D) as well as full width at half maximal (FWHM; E), respectively. n = 177 to 186 sparks for each group. *P < 0.05, compared with GFP. ^#P < 0.05, compared with HRC^WT.

Resting Ca²⁺ levels were estimated as the ratio of F405/F485 measured by indo-1. HRC^S96A elicited elevated resting Ca²⁺ compared with GFP, while no significant difference was found between HRC^WT and GFP (Supplemental Fig. S1; Supplemental Material for this article is available online at the Am J Physiol Heart Circ Physiol website).

Spontaneous Ca²⁺ waves during rhythmic pacing in failing cardiomyocytes.

To assess the effects of HRC^S96A expression under stress conditions, cardiomyocytes were subjected to 1-Hz electrical field stimulation in the presence of 1 μM isoproterenol. It is interesting to note that almost no spontaneous Ca²⁺ waves were observed in GFP, HRC^WT, or HRC^S96A expressing cardiomyocytes under these conditions (Fig. 5A, left).

Fig. 5.

Spontaneous Ca²⁺ waves during rhythmic pacing in cultured heart failure (HF) myocytes expressing HRC^S96A. A: typical line-scan images and corresponding time courses of Ca²⁺ transients from Ad.GFP, Ad.HRC^WT, or Ad.HRC^S96A infected normal or HF myocytes, which were subjected to 1-Hz field stimulation in the presence of 1 μM isoproterenol. B: analysis of spontaneous Ca²⁺ wave frequency in HF cells. n = 18–30 cells for each group. *P < 0.05, compared with GFP.

To further investigate the function of the HRC genetic variant under DCM conditions, we generated a rat model with myocardial infarction-induced heart failure as previously described (1). Heart failure was verified by the increased end-diastolic volume as well as decreased ejection fraction at 8 wk after infarction surgery. Other cardiac morphological and functional parameters, such as left ventricular inner dimension at end diastole and systole, end-systolic volume, and weight-to-body weight ratio, were also determined (Table 1). Cardiomyocytes isolated from the rat failing hearts exhibited an average cell length of 160.9 ± 2.5 μm (n = 223), compared with 121.7 ± 1.3 μm (n = 150) for normal hearts. The normal and heart failure myocytes were infected with Ad.GFP, Ad.HRC^WT, or Ad.HRC^S96A for 48 h, and the basal Ca²⁺ cycling properties were determined. Generally, these three viruses gave similar alterations in heart failure cells, as in normal myocytes (Supplemental Fig. S2, A and B). Then the cardiomyocytes were subjected to 1-Hz field stimulation in the presence of 1 μM isoproterenol. Notably, only the HRC^S96A cells developed dramatic Ca²⁺ disturbances, manifested as spontaneous extrasystolic Ca²⁺ waves, while the GFP and HRC^WT myocytes did not (Fig. 5). These oscillations of intracellular Ca²⁺ transients were considered as the underlying cause of delayed afterdepolarizations, which could induce malignant arrhythmias. Isoproterenol stimulation of the heart failure myocytes resulted in elevations of SR Ca content in the Ad.GFP (28%)-, Ad.HRC^WT (16%)-, or Ad.HRC^S96A (18%)-infected groups (Supplemental Fig. S2C). However, it is worth noting that the SR Ca content of HRC^WT was still significantly higher than GFP as well as HRC^S96A. Thus these results indicate that expression of HRC^S96A may destabilize the RyR in failing cardiomyocytes, leading to increased SR Ca²⁺ leak, which may be the underlying mechanism of sudden death in human carriers.

Table 1.

Echocardiographic parameters before and 8 wk after left descending coronary artery ligation

	Before Surgery	8 wk After Surgery
LVIDd, mm	7.8 ± 0.2	10.1 ± 0.1*
LVIDs, mm	4.6 ± 0.2	8.8 ± 0.2*
EDV, ml	1.04 ± 0.07	2.12 ± 0.06*
ESV, ml	0.24 ± 0.03	1.46 ± 0.10*
HW/BW, mg/g	2.9 ± 0.1	3.5 ± 0.2*
EF, %	77.3 ± 1.2	31.5 ± 3.4*

Data are means ± SE. LVIDd and LVIDs, left ventricular inner dimension at end diastole and systole; EDV, end-diastolic volume; ESV, end-systolic volume; HW/BW, heart weight-to-body weight ratio; EF, ejection fraction.

^*P < 0.05, compared with before surgery.

DISCUSSION

Molecular mechanisms of arrhythmia in heart failure.

In end-stage DCM patients, beyond contractile dysfunction, malignant arrhythmias constitute a major risk factor for sudden cardiac death (12). These arrhythmias stem mainly from nonreentry mechanisms, such as early afterdepolarization and delayed afterdepolarization, which are all linked to impaired cardiomyocyte Ca²⁺ cycling. During heart failure, elevated SR Ca²⁺ leakage enhances the occurrence of spontaneous Ca²⁺ sparks and Ca²⁺ waves (6), which may activate the inward current through NCX and trigger cell membrane depolarization. Actually, the expression levels and activity of several key calcium transport proteins are altered during heart failure, which may be responsible for the SR leak. For example, the protein levels of junctin, triadin, and HRC, which are components of the macromolecular complex regulating RyR activity, are diminished in human failing hearts (8, 10). Accordingly, downregulation of junctin and triadin in genetic knockout animal models is associated with arrhythmia under stress conditions (7, 24). In addition, the elevated NCX expression and function in heart failure may lead to a greater inward current and further depolarization of cell membrane at a given SR Ca²⁺ release. Indeed, studies in human failing hearts or heart failure animal models have demonstrated the role of NCX in inducing arrhythmias (4, 22). However, the overall remodeling of molecular and cellular networks in heart failure is complicated and the exact mechanisms for sudden cardiac death are not yet entirely clarified.

One of the downregulated SR proteins during heart failure is HRC, a SR luminal protein. Previous studies have provided insights into its regulatory role on SR calcium homeostasis during physiological and pathological conditions. Briefly, adenoviral-mediated acute HRC overexpression in cultured rat cardiomyocytes (8) showed diminished Ca²⁺ transient amplitude although the SR load was increased. Chronic HRC overexpression in vivo led to cardiac hypertrophy and dilated ventricular remodeling, when the mice were subjected to transverse aortic constriction surgery (11). Thus it was hypothesized that decreased HRC expression during heart failure may provide a compensatory mechanism in the deteriorated Ca²⁺cycling.

Effects of S96A on HRC function in normal hearts.

The present study explored the mechanisms underlying a recently discovered genetic variant of HRC^S96A, which was shown to be associated with arrhythmia in DCM patients (2). Expression of human wild-type HRC reduced Ca²⁺ transient amplitude and diminished spontaneous Ca²⁺ spark frequency, while the SR load was increased. This suggests that HRC^WT may immobilize SR luminal Ca²⁺ through its buffering property and inhibit Ca²⁺ release or it may function as a luminal regulator and stabilize RyR. Compared with HRC^WT, human HRC^S96A increased the frequency of Ca²⁺ sparks (nearly twofold) in the setting of reduced Ca²⁺ transient amplitude and lower SR load. This apparent paradox pointed to the possibility that HRC^S96A may enhance RyR activity from the SR luminal side, increasing uncontrolled Ca²⁺ release and induced Ca²⁺ instability, supporting the existence of a luminal gating mechanism relatively independent of the cytosolic gating mechanism (26). The modulation of RyR function could be mediated by direct interaction of HRC with triadin, a component of the RyR Ca²⁺ release complex. Interestingly, the triadin protein levels were significantly increased by either HRC^S96A or HRC^WT expression. A previous study (20) has shown that acute overexpression of triadin in rat cardiomyocytes is associated with lower SR Ca²⁺ load and increased open probability of RyRs in lipid bilayers as well as increased frequency of spontaneous Ca²⁺ sparks, leading to arrhythmias. However, the triadin levels were increased to a similar extent in HRC^S96A and HRC^WT, suggesting that the aberrant SR Ca release was not mediated by triadin overexpression under our experimental conditions.

In addition, HRC^S96A diminished the Ca²⁺ transient decline time even more than HRC^WT, indicating an increased inhibitory effect by the S96A variant on SERCA2a activity. Actually, the reduced Ca reuptake by the SR may be due to increased inhibition of the maximal SR calcium transport rate by HRC^S96A, consistent with its stronger binding to SERCA2, compared with HRC^WT. Collectively, these phenotypes point to HRC as a nodal point, bridging Ca²⁺ release and Ca²⁺ uptake in the heart (3). Furthermore, the S96A substitution may modulate this regulatory role of HRC.

To test if this genetic variant is linked to arrhythmia, we challenged the cardiomyocytes with 1 μM isoproterenol under 1-Hz field stimulation. However, these conditions did not trigger any after contractions in the HRC^S96A cells, suggesting that this single amino acid substitution is not sufficient to introduce arrhythmogenesis in the normal heart.

S96A-HRC is linked to arrhythmias in heart failure.

Further studies were carried out in cardiomyocytes from a heart failure rat model. When challenged with 1 μM isoproterenol and 1-Hz field stimulation, the variant HRC^S96A was associated with a significantly elevated Ca²⁺-wave frequency. This may be a plausible mechanism underlying arrhythmias in HRC S96A carriers under heart failure conditions, when their catecholamine levels are elevated. So why does HRC^S96A elicit arrhythmias in failing cardiomyocytes and not in normal cells? The hypothesis is that heart failure alters the properties of SR Ca²⁺ release as evidenced by enhanced Ca²⁺ spark frequency (Fig. 4B vs. Supplemental Fig. S3A). As a result, the inconspicuous effects of HRC^S96A on extra-systolic Ca²⁺ release during rhythmic pacing in normal cardiomyocytes become dominant under heart failure. Since Ca²⁺ sparks appear to occur at the majority of the sites for Ca²⁺ wave initiation (5), the dramatically elevated Ca²⁺ spark frequency may exceed the stable threshold of RyR local control and elicit arrhythmic Ca²⁺ waves, which could be the triggering mechanism of arrhythmias in human carriers. Thus the HRC^S96A turns out to be a context-sensitive human mutation, which becomes arrhythmogenic in heart failure condition.

However, a limitation of the present study is that expression of human HRC^S96A occurred in the presence of endogenous rat HRC, while the heterozygous or homozygous HRC^S96A patients have only human HRC present and at a lower overall expression level than our study. Thus it will be valuable to generate knockin HRC^S96A mice and obtain more detailed information on how this genetic variant regulates SR Ca²⁺ cycling in vivo. These mice can be subjected to stress conditions including heart failure, which may closer mimic the conditions of human carriers and elucidate the underlying mechanisms of the observed arrhythmias.

Conclusions.

In summary, we have suggested a link between the human HRC S96A variant and clinical arrhythmias in DCM patients carrying this polymorphism. Expression of human HRC^S96A elevated the activity of RyRs in normal as well as failing cardiomyocytes, while it enhanced the disturbances of rhythmic Ca²⁺ transients under stress conditions in heart failure. These arrhythmogenic events may serve as the underlying mechanism of life-threatening ventricular arrhythmias in human carriers.

GRANTS

This work was supported by the National Heart, Lung, and Blood Institute Grants HL-26057, HL-64018, and HL-77101 (to E.G. Kranias), the Leducq Foundation (to E. G. Kranias), the European Community's Seventh Framework Programme FP7/2007-2013 under Grant Agreement No. HEALTH-F2-2009-241526, EUTrigTreat (to E. G. Kranias), and the Chinese National Natural Science Foundation, Major Basic Research Development Programs of China (2007CB512100 and 2011CB809100 to H. Cheng).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).