Cardiac hypertrophy associated with impaired regulation of cardiac ryanodine receptor by calmodulin and S100A1

Abstract

The cardiac ryanodine receptor (RyR2) is inhibited by calmodulin (CaM) and S100A1. Simultaneous substitution of three amino acid residues (W3587A, L3591D, F3603A; RyR2^ADA) in the CaM binding domain of RyR2 results in loss of CaM inhibition at submicromolar (diastolic) and micromolar (systolic) Ca²⁺, cardiac hypertrophy, and heart failure in Ryr2^ADA/ADA mice. To address whether cardiac hypertrophy results from the elimination of CaM and S100A1 inhibition at diastolic or systolic Ca²⁺, a mutant mouse was generated with a single RyR2 amino acid substitution (L3591D; RyR2^D). Here we report that in single-channel measurements RyR2-L3591D isolated from Ryr2^D/D hearts lost CaM inhibition at diastolic Ca²⁺ only, whereas S100A1 regulation was eliminated at both diastolic and systolic Ca²⁺. In contrast to the ∼2-wk life span of Ryr2^ADA/ADA mice, Ryr2^D/D mice lived longer than 1 yr. Six-month-old Ryr2^D/D mice showed a 9% increase in heart weight-to-body weight ratio, modest changes in cardiac morphology, and a twofold increase in atrial natriuretic peptide mRNA levels compared with wild type. After 4-wk pressure overload with transverse aortic constriction, heart weight-to-body weight ratio and atrial natriuretic peptide mRNA levels increased and echocardiography showed changes in heart morphology of Ryr2^D/D mice compared with sham-operated mice. Collectively, the findings indicate that the single RyR2-L3591D mutation, which distinguishes the effects of diastolic and systolic Ca²⁺, alters heart size and cardiac function to a lesser extent in Ryr2^D/D mice than the triple mutation in Ryr2^ADA/ADA mice. They further suggest that CaM inhibition of RyR2 at systolic Ca²⁺ is important for maintaining normal cardiac function.

cardiac muscle ryanodine receptors (RyR2s) are large, ligand-gated ion channels composed of four 560-kDa RyR2 subunits, four 12.6-kDa FK506 binding protein subunits, and various associated proteins that include protein kinases, protein phosphatases, S100A1, and calmodulin (CaM) (11, 17, 19, 30). Binding of CaM to RyR2 inhibits the release of Ca²⁺ from sarcoplasmic reticulum (SR) at diastolic and systolic Ca²⁺ (2, 8, 19, 31). Deletion of RyR2 amino acid residues 3583–3603 and mutations in this region eliminate CaM binding and inhibition of the recombinant RyR2 by CaM in vitro (36). An in vivo role of RyR2 regulation by CaM was established by preparing mutant mice with three amino acid substitutions in the CaM binding domain of RyR2 (W3587A, L3591D, F3603A; RyR2^ADA). Ryr2^ADA/ADA mice die within ∼2 wk after birth because of the rapid development of cardiac hypertrophy and heart failure (35). Their early death is associated with abnormal SR Ca²⁺ release and altered gene regulation (33, 35).

S100A1 is a small Ca²⁺ binding protein expressed at high levels in cardiomyocytes (13). S100A1 targets multiple sites that include sarcomeric proteins, mitochondrial proteins, and Ca²⁺ handling proteins Ca_v1.2, RyR2, and sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA)2a and SERCA2a regulatory protein phospholamban (PLN) (10, 21, 24, 29). S100A1 also regulates skeletal muscle RyR1 and binds to the conserved CaM binding domain of RyR1 (25, 27). S100A1^−/− mice have normal life span and cardiac function under normal conditions (7). However, transgenic overexpression of S100A1 increases SR Ca²⁺ content and Ca²⁺ release and improves myocardial contractile performance (21). More recent studies have shown that molecular targeting of S100A1 to myocardially infarcted animals (23) and human failing cardiomyocytes (4) restores cardiac function by improving mitochondrial function, SR Ca²⁺ handling, and contractile performance. This suggested that S100A1 gene targeting may be a promising strategy to treat heart failure.

To further define the role of RyR2 and Ca²⁺ in cardiac development and function, a second mouse model was prepared with a single amino acid substitution in the CaM and S100A1 binding domain of RyR2 (L3591D; RyR2^D). This mutation was selected on the basis of earlier studies showing that the RyR2-L3591D mutant is inhibited by CaM at micromolar Ca²⁺ but not submicromolar Ca²⁺ concentrations when expressed in human embryonic kidney 293 (HEK293) cells (36). Here we report that, in contrast to Ryr2^ADA/ADA mice, Ryr2^D/D mice live longer than 1 yr and develop a less severe impairment of cardiac function than Ryr2^ADA/ADA mice.

MATERIALS AND METHODS

Ethics statement.

This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocols were approved by the University of North Carolina at Chapel Hill and the Medical University of South Carolina Institutional Animal Care and Use Committees. Animals were anesthetized and killed with 1–4% isoflurane (drop method) or intraperitoneal injection of Avertin (160 mg/kg) followed by physical euthanasia.

Materials.

[³H]ryanodine was obtained from PerkinElmer and unlabeled ryanodine from EMD Biosciences. Complete protease inhibitors were from Roche Applied Science and phospholipids from Avanti Polar Lipids. Unlabeled CaM was prepared as described previously (2). Rabbit polyclonal antibody RyR2-F9221 was prepared by New England Peptide (Gardner, MA). Unless otherwise specified, all other chemicals were obtained from Sigma-Aldrich.

Preparation of mutant mice.

Mutant mice carrying the RyR2-L3591D mutation were prepared with established methods (35). Briefly, linearized targeting vector was transfected into E14 embryonic stem (ES) cells by electroporation. Cells were selected in the presence of G418 and gancyclovir and screened by both PCR and Southern blot analysis. Mutations were confirmed by DNA sequencing. Targeted ES cells were injected into the blastocysts of C57BL/6J mice in the Animal Models Core Facility of the University of North Carolina at Chapel Hill. A male chimera mouse carrying the mutation in the Ryr2 gene was mated with a 129/SvEv female mouse to obtain heterozygous offspring. The neomycin-resistant gene flanked by loxP sequences was removed by Cre recombinase driven by the testis-specific angiotensin-converting enzyme promoter (Fig. 1). Germ line transmission of the mutation to heterozygous offspring caused removal of a gene cassette flanked by loxP sequences (5).

Fig. 1.

Generation of mice with RyR2-L3591D mutation. A: schematic representation of the mouse Ryr2 gene and targeting construct. S represents SphI restriction site. Neomycin-resistant gene (neo), Cre recombinase gene driven by testis-specific angiotensin-converting enzyme promoter (tACE-cre), and thymidine kinase gene (TK) are indicated. Arrows and x indicate the position of primers for screening and a mutation site, respectively. B: Southern blot of genomic DNA. After restriction enzyme digestion of genomic DNA with SphI, 5′ and 3′ probes identify 13.5-kb and 11-kb fragments in the targeted allele, respectively. Both probes hybridize with 20.5-kb fragments in the wild-type (WT) allele. C: DNA sequence analysis of RyR2-L3591D. cDNA encoding the CaM binding site of RyR2 was amplified from total RNA from homozygous (hybrid genetic background) mouse cardiac muscle and sequenced. The RyR2-L3591D mutation was confirmed. A HinfI site created by the L3591D mutation (GACTC) was used for screening the mutant allele.

Homozygous gene-targeted animals (Ryr2^D/D) were obtained by mating heterozygous mice (Ryr2^+/D). Previous experiments with Ryr2^ADA/ADA mice (35) and experiments with Ryr2^D/D mice were performed with animals that were backcrossed at least five times into the 129/SvEv genetic background. Offspring were genotyped by PCR followed by restriction digestion with HinfI.

Morphological analyses.

Hearts from 6- to 7-mo-old mice were fixed with 4% (wt/vol) paraformaldehyde in PBS (pH 7.4) and dehydrated with increasing concentrations of ethanol in water. Paraffin-embedded hearts were sectioned into 6- to 9-μm thickness and stained with hematoxylin and eosin. Tetramethylrhodamine isothiocyanate (TRITC)-conjugated wheat germ agglutinin was used to visualize the cell membranes in the sections (35). TRITC fluorescence was captured by confocal microscopy (Leica TCS SP5), and cross-sectional areas of individual cells in left and right ventricles and interventricular septum were calculated with ImageJ software. Thirty cells were randomly selected from wild-type and mutant mice.

Echocardiography.

Left ventricular cardiac parameters were determined by transthoracic M-mode echocardiography on restrained mice with a Vevo 2100 high-resolution imaging system (VisualSonics) with a 40-MHz probe. Unanesthetized 4- and 6-mo-old mice were restrained on a warmed mouse board (Indus Instruments for VisualSonics).

Transverse aortic constriction.

Pressure overload in the left ventricle was induced by transverse aortic constriction (TAC) of 3-mo-old wild-type and Ryr2^D/D mice as described previously (3). The aorta was ligated between the innominate and left common carotid arteries with a 7-10 silk suture and a tapered 27-gauge needle placed on top of the aorta. The tapered needle was removed, leaving the suture to produce a defined stenosis of the vessel. Control mice were subjected to a sham operation in which a band was twined around the aorta, but not ligated, and subsequently removed. The skin was closed with separate sutures, and buprenorphine was administered for analgesia. Echocardiography was performed after 4 wk of constriction.

Quantitative RT-PCR.

Gene expression was measured by quantitative RT-PCR with the ABI Prism 7700 Sequence Detector (Applied Biosystems) (15). RNA was isolated from the left ventricle of 4- and 6-mo-old mice with the ABI Prism 6700 Automated Nucleic Acid Workstation according to the manufacturer's protocol. Forward and reverse primers and corresponding fluorogenic probes were β-actin: AAGAGCTATGAGCTGCCTGA, ACGGATGTCAACGTCACACT, FTCACTATTGGCAACGAGCGGTTCGCQ; α-myosin heavy chain (α-MHC): AGCTCAGCCAGGCCAATAGA, TGGGTGTCCTTCAAGTGAGC, FAGAATTCTTCAGGTGTTTCTGTGCCTCAGQ; β-MHC: AGCTCAGCCATGCCAACCGT, TGAGTGTCCTTCAGCAGACT, FAGGCTCTTCACTTGTTTCTGGGCCTCAQ; atrial natriuretic peptide (ANP): GAGAAGATGCCGGTAGAAGA, AAGCACTGCCGTCTCTCAGA, FATGCCCCCGCAGGCCCGGQ; brain natriuretic peptide (BNP): CTGCTGGAGCTGATAAGAGA, TGCCCAAAGCAGCTTGAGCT, FCTCAAGGCAGCACCCTCCGGGQ; Ca_v1.2: AGACATCTCTCAGAAGACAGC, TCAAGCCTGCCACTGCCAAT, FTGCCCTTGCATCTGGTTCATCACCAGQ; RyR2: CTGTCCCTGAGGTACAAGAA, AGCTTTCTCTGGTTCCGACT, FTTCAGGAGCAGAAGGCGAAAGAGGAQ; inositol trisphosphate receptor (IP₃R): TCCTCAAGACAACCTGCTTC, CTGACTTGATGTGCTCCTCA, FCTGTGGCCTGGAGAGAGACAAGTTTGQ; calsequestrin (CSQ2): AAGAGTCACCCAGATGGCTA, ATGCTCAAGTCAGGATTGTCA, FCCTAGAGATCCTGAAACAGGTTGCCCQ; SERCA2a: TCCTCTACGTGGAACCTTTG, AAGGAGATTTTCAGCACCATC, FCCAGATCACACCGCTGAATCTGACCCQ; and sodium-calcium exchanger (NCX)1: AAGCAGCCACTGACCAGCA, TGGTGTGCTCGCCTAGGAT, FCCCATTTCCGCAATGCGCCTCTCCQ [F is 5′-fluorescein (FAM) and Q is quencher (TAMRA)]. Levels of gene expression were quantitated relative to β-actin.

Preparation of heart homogenates and crude membrane fractions.

Hearts were homogenized in 20 mM imidazole, pH 7.0, 0.3 M sucrose, 0.15 M NaCl, 0.1 mM EGTA, protease inhibitors (Complete; Roche Applied Science), and 1 mM glutathione (oxidized form) with a Tekmar Tissumizer twice for 5 s at 13,500 rpm (35). Homogenates were centrifuged for 45 min at 100,000 g in a type 75Ti rotor (Beckman). Pellets were suspended in the aforementioned buffer without EGTA and glutathione to obtain a crude membrane fraction. Homogenates and membranes were stored in small aliquots at −80°C.

Immunoblot analyses.

Homogenates (10 μg protein/lane) were separated by 3–12% gradient SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blotted with 2% ECL Advance blocking reagent (Amersham Biosciences) in 0.5% Tween 20-Tris-buffered saline (TBS), pH 7.4 at 24°C for 1 h and probed with primary rabbit polyclonal antibody RyR2-F9221 and secondary peroxidase-conjugated anti-rabbit IgG antibody. Immunoblots were developed with enhanced chemiluminescence and quantified with ImageQuantTL Analysis Software. Glyceraldehyde-3-phosphate dehydrogenase was the loading control.

Single-channel recordings.

Single-channel measurements of wild-type and mutant channels were performed in planar lipid bilayers (32). Measurements were made with symmetrical 0.25 M CsCl, 20 mM Cs HEPES, pH 7.4 with the indicated concentrations of Ca²⁺. A strong dependence of single-channel activities on the cis Ca²⁺ concentration indicated that the large cytosolic “foot” region faced the cis chamber of the bilayers. The trans SR luminal side of the bilayer was defined as ground. Exogenous CaM and S100A1 were added to the cis, cytosolic solution. Electrical signals were filtered at 2 kHz, digitized at 10 kHz, and analyzed as described previously (32). Channel open probability (P_o) was obtained from 2-min recordings by setting the threshold level at 50% of the current amplitude between the closed and open channel states.

[³H]ryanodine binding.

Ryanodine binds specifically to RyRs. Hence, [³H]ryanodine binding can be used as a probe of channel activity and concentration (26). Ca²⁺ dependence of [³H]ryanodine binding was determined by incubating homogenates with a relatively low [³H]ryanodine concentration (3 nM) at 24°C for 20 h in 0.25 M KCl, 20 mM imidazole, pH 7.0, protease inhibitors, and indicated free Ca²⁺ concentrations. Maximum binding capacity (B_max) of [³H]ryanodine binding was determined by incubating homogenates for 4–5 h at 24°C with a near-saturating concentration of 20 nM [³H]ryanodine in 20 mM imidazole, pH 7.0, 0.6 M KCl, protease inhibitors, and 0.1 mM Ca²⁺. Nonspecific binding was determined with a 1,000-fold excess of unlabeled ryanodine. Aliquots of samples were diluted ninefold with ice-cold water and placed on Whatman GF/B filters saturated with 2% polyethyleneimine. Filters were washed with ice-cold 0.1 M KCl, 1 mM K-PIPES, pH 7.0, and radioactivity remaining on the filters was determined by liquid scintillation counting to obtain bound [³H]ryanodine.

⁴⁵Ca²⁺ uptake rate.

ATP-dependent ⁴⁵Ca²⁺ uptake rates by homogenates were determined by a filtration assay (33). ⁴⁵Ca²⁺ uptake rates were determined in 0.15 M KCl, 20 mM imidazole, pH 7.0 solution containing 5 mM ATP, 8 mM Mg²⁺, and 5 mM potassium oxalate, a Ca²⁺ precipitating agent to increase Ca²⁺ uptake capacity, 10 μM ruthenium red to inhibit RyR2, 5 mM NaN₃ to inhibit mitochondrial Ca²⁺ uptake, 1 mM EGTA, and Ca²⁺ and ⁴⁵Ca²⁺ to yield a free Ca²⁺ concentration of 0.5 μM.

Biochemical assays and data analysis.

Free Ca²⁺ concentrations were measured with a Ca²⁺-selective electrode. Results are expressed as means ± SE. Differences between wild-type and mutant samples were determined by one-way ANOVA followed by Tukey's test. P < 0.05 was considered significant.

RESULTS

Mice carrying three mutations in the CaM binding domain of RyR2 (W3587A, L3591D, F3603A; RyR2^ADA) rapidly develop cardiac hypertrophy and die at ∼2 wk of age (35). The triple mutation eliminated CaM-dependent inhibition of RyR2 activity at both diastolic and systolic Ca²⁺. To determine whether the phenotype was caused by loss of CaM inhibition at diastolic or systolic Ca²⁺, another genetically modified mouse carrying a single amino acid mutation in Ryr2 gene (L3591D) was prepared by gene targeting (Fig. 1). This mutation was hypothesized to eliminate CaM inhibition of RyR2 at diastolic Ca²⁺ only (36).

Single-channel measurements indicate loss of CaM and S100A1 regulation of RyR2-L3591D in mice at diastolic Ca²⁺.

We previously showed that the RyR2-L3591D mutation caused loss of CaM-dependent regulation of channel activity at submicromolar (diastolic) but not micromolar (systolic) Ca²⁺ when expressed in HEK293 cells (36). To determine whether CaM regulation of the RyR2-L3591D channel was maintained in mice, membranes isolated from hearts of wild-type and Ryr2^D/D mice were fused with a lipid bilayer (Fig. 2). RyR2 regulation by CaM, determined by changes in channel P_o, was recorded in the presence of 0.2 μM or 2 μM free cis cytosolic Ca²⁺ in the absence and on the subsequent addition of cis cytosolic CaM. At 0.2 μM cytosolic Ca²⁺, addition of 50 nM and 1 μM cytosolic CaM resulted in significant inhibition of channel P_o of wild-type RyR2, with no changes in P_o for RyR2-L3591D (Fig. 2A). In contrast, in the presence of 2 μM Ca²⁺, P_o of both wild type and RyR2-L3591D decreased at 50 nM and 1 μM CaM (Fig. 2B). Thus CaM regulation of RyR2-L3591D was lost at low Ca²⁺ but maintained at higher Ca²⁺ levels.

Fig. 2.

Effects of calmodulin (CaM) on wild-type and RyR2-L3591D single-channel activities. Membranes isolated from the hearts of wild-type and Ryr2^D/D mice were fused with a lipid bilayer. Top: single-channel currents were recorded at −35 mV in the presence of 0.2 μM (A) and 2 μM (B) cis cytosolic Ca²⁺ as described in materials and methods in the absence of CaM (top) and on the subsequent addition of 50 nM (middle) and 1 μM (bottom) cis CaM. P_o, open channel probability. Bottom: mean ± SE P_o of number of recordings indicated in parentheses, normalized to control. *P < 0.05, **P < 0.001 compared with control (no CaM added). Mean P_o in absence of CaM were 0.08 ± 0.02 (n = 10) and 0.10 ± 0.07 (n = 8) at 0.2 μM Ca²⁺ and 0.55 ± 0.05 (n = 15) and 0.47 ± 0.12 (n = 8) at 2 μM Ca²⁺ for wild-type and RyR2-L3591D channels, respectively.

S100A1 is a Ca²⁺ binding protein reported to bind to the conserved CaM binding region of RyR1 (25) and regulate RyR2 activity (10, 21, 24, 29). Single-channel measurements in Fig. 3A show that at 0.2 μM Ca²⁺ the addition of 0.3 μM and 1 μM cytosolic S100A1 reduced P_o of wild type. At 2 μM Ca²⁺, addition of 1 μM S100A1 was required for significant wild-type inhibition (Fig. 3B). No inhibition of RyR2-L3591D channels by S100A1 was observed under any of these conditions. The single-channel data suggest that the L3591D mutation impairs CaM and S100A1 regulation of RyR2 at 0.2 μM Ca²⁺. At 2 μM Ca²⁺, RyR2-L3591D inhibition by CaM was retained whereas that by S100A1 was lost. One possible explanation is that, in contrast to RyR1 (34), S100A1 and CaM do not compete for the same CaM binding site in RyR2. Nitu et al. (22), using fluorescence resonance energy transfer, recently reported that S100A1 likely interacts allosterically with the CaM binding domain on RyR2 rather than by a direct interaction.

Fig. 3.

Effects of S100A1 on wild-type and RyR2-L3591D single-channel activities. Top: single RyR2 channels isolated from hearts were recorded in the presence of 0.2 μM (A) and 2 μM (B) cis cytosolic Ca²⁺ as described in Fig. 2 in the absence of S100A1 (top) and on the subsequent addition of 0.3 μM (middle) and 1 μM (bottom) cis S100A1. Bottom: mean ± SE P_o of number of recordings indicated in parentheses, normalized to control (−S100A1). *P < 0.05, **P < 0.001 compared with control (no S100A1 added).

Ryr2^D/D mice show modest signs of cardiac hypertrophy.

Unlike Ryr2^ADA/ADA mice (35), Ryr2^D/D mice live longer than 1 yr without a significant difference in cardiac performance compared with Ryr2^+/+ mice, as determined by echocardiography (n = 3, not shown). No major changes in cardiac gross morphology were observed for 6- to 7-mo-old homozygous mutant mice (Fig. 4A). In homozygous mice, heart weight-to-body weight ratio increased by 11% (P < 0.003) compared with Ryr2^+/D mice and by 9% (P < 0.02) compared with wild-type mice (Fig. 4B). Histological examination showed modest but significant changes in right ventricular wall thickness and cross-sectional area in homozygous mutant mice compared with heterozygous mice (Fig. 4, C and D). However, the three groups of mice showed no significant changes in septal and left ventricular wall thickness and cross-sectional area.

Fig. 4.

Phenotypic characterization of wild-type and mutant hearts. A: hematoxylin- and eosin-stained heart sections from 6- to 7-mo-old wild-type and Ryr2^D/D mice. B: body weights and heart weight-to-body weight ratios of wild-type and mutant mice. Data are means ± SE for number of mice shown in parentheses. *P < 0.05 compared with Ryr2^+/+ and Ryr2^+/D mice. C: wall thickness of ventricles of 6- to 7-mo-old wild-type and mutant mice. RV, right ventricle; LV, left ventricle. D: cross-sectional areas of individual cardiomyocytes in ventricles were measured as described in materials and methods. C and D: data are means ± SE for number of mice shown in parentheses. *P < 0.05 compared with Ryr2^+/D mice.

To determine the effect of the RyR2-L3591D mutation on cardiac function, echocardiography was performed on conscious 6-mo-old animals (Table 1). Significant increases in left ventricular end-diastolic and end-systolic dimensions were observed in heterozygous mice compared with wild-type mice. Left ventricular posterior diastolic and systolic wall thicknesses were increased in homozygous mice compared with heterozygous mice. Heart rate, fractional shortening, interventricular septum, and left ventricular diastolic and systolic posterior wall thicknesses were all similar among the three genotypes, suggesting limited alterations in cardiac morphology and contractility of the mutant mice.

Table 1.

Echocardiography in 6-mo-old wild-type and mutant mice

Parameters	Ryr2+/+ (n = 13)	Ryr2+/D (n = 15)	Ryr2D/D (n = 11)
HR, beats/min	542 ± 30	449 ± 30	491 ± 23
LVEDD, mm	2.82 ± 0.12*	3.34 ± 0.08	3.00 ± 0.12
LVESD, mm	1.40 ± 0.10*	1.90 ± 0.09	1.54 ± 0.12
FS, %	50.6 ± 2.6	43.2 ± 1.8	49.3 ± 2.4
IVSD, mm	1.23 ± 0.07	1.06 ± 0.06	1.22 ± 0.07
IVSS, mm	1.77 ± 0.10	1.54 ± 0.06	1.83 ± 0.10
LVPWD, mm	1.10 ± 0.07	0.97 ± 0.04	1.34 ± 0.11*
LVPWS, mm	1.69 ± 0.09	1.47 ± 0.04	1.79 ± 0.13*

Data are means ± SE for n mice.

HR, heart rate; LVEDD, left ventricular end-diastolic dimension; LVESD, left ventricular end-systolic dimension; FS, fractional shortening [LVEDD − LVESD)/LVEDD]; IVSD, interventricular septum diastolic thickness; IVSS, interventricular septum systolic thickness; LVPWD, left ventricular posterior wall diastolic thickness; LVPWS, left ventricular posterior wall systolic thickness.

^*P < 0.05 compared with heterozygous mice.

Relevant gene expression in 6-mo-old wild-type and Ryr2^D/D mice was determined by quantitative RT-PCR. ANP mRNA increased 1.8-fold in the hearts of homozygous mice compared with wild type (Fig. 5). This suggests that the RyR2-L3591D mutation elicited a modest hypertrophic response in Ryr2^D/D mice (16). However, mRNA levels of other genes associated with cardiac remodeling and Ca²⁺ handling were not altered compared with wild-type (Fig. 5). These included cardiac hypertrophy-associated genes β-MHC and BNP and genes for Ca²⁺ handling proteins Ca_v1.2, RyR2, IP₃R, CSQ2, SERCA2a, and NCX1.

Fig. 5.

Quantitative RT-PCR of 6-mo-old wild-type and Ryr2^D/D mice. RT-PCR was performed with total RNA from left ventricles of 6-mo-old mice. RNA levels were normalized to those of wild type for each gene. α-MHC, α-myosin heavy chain; β-MHC, β-myosin heavy chain; ANP, atrial natriuretic peptide; BNP, brain natriuretic peptide; Ca_v1.2, L-type Ca²⁺ channel; RyR2, cardiac ryanodine receptor; IP₃R, inositol trisphosphate receptor; CSQ2, calsequestrin; SERCA2a, sarco(endo)plasmic reticulum Ca²⁺-ATPase 2a; NCX1, sodium-calcium exchanger 1. Data are means ± SE of 9–16 samples. *P < 0.05 compared with Ryr2^+/+.

Cardiac muscle SR Ca²⁺ handling.

The SR Ca²⁺ handling properties of wild-type and Ryr2^D/D mice were compared in heart homogenates of 6-mo-old mice. In Fig. 6A, a similar Ca²⁺ dependence of [³H]ryanodine binding suggested that the leucine-to-aspartic acid substitution in the CaM binding domain did not alter the Ca²⁺ sensitivity of the receptor in Ryr2^+/D and Ryr2^D/D mice. B_max of [³H]ryanodine binding, a measure of RyR protein concentration, indicated a modest decrease (not significant) in RyR2 concentration in hearts of homozygous mice compared with heterozygous and wild-type mice (Table 2). Likewise, immunoblot analysis did not reveal significant differences in RyR2 expression among the three genotypes (Fig. 6B). ⁴⁵Ca²⁺ uptake rates, a measure of SERCA activity by cardiac muscle isolates, were not altered (Table 2).

Fig. 6.

Ca²⁺ dependence of [³H]ryanodine binding and RyR2 expression in hearts of wild-type and mutant mice. A: specific [³H]ryanodine binding to homogenates containing wild-type and mutant RyR2s was carried out at the indicated free Ca²⁺ concentrations as described in materials and methods. Data are means ± SE of 4 samples each. B, left: immunoblots of RyR2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) of heart homogenates from 6-mo-old Ryr2^+/+, Ryr2^+/D, and Ryr2^D/D mice. Right: levels of RyR2 expression, quantitated relative to GAPDH, were normalized to RyR2^+/+. Data are means ± SE of 3 preparations done in triplicate.

Table 2.

Calcium handling by heart homogenates of 6-mo-old wild-type and Ryr2^D/D mice

Genotype	Bmax of [3H]Ryanodine Binding, pmol/mg protein	45Ca2+ Uptake Rate. pmol•mg protein−1•min−1
Ryr2+/+	0.16 ± 0.02 (10)	3.67 ± 0.26 (6)
Ryr2+/D	0.17 ± 0.02 (6)	3.44 ± 0.49 (4)
Ryr2D/D	0.13 ± 0.02 (13)	3.76 ± 0.34 (6)

Values are means ± SE for number of heart preparations indicated in parentheses.

B_max, maximum binding capacity.

Pressure overload induces a modest hypertrophic response in Ryr2^D/D mice.

To determine whether Ryr2^D/D mice are capable of a cardiac hypertrophic response under conditions of pressure overload, 3-mo-old Ryr2^D/D and wild-type mice were subjected to TAC for 4 wk. TAC induced a significant 1.4-fold increase in heart weight-to-body weight ratio in Ryr2^D/D mice compared with 1.1-fold increase in wild-type mice (Table 3). Echocardiography indicated that TAC increased interventricular septum dimensions and left ventricular posterior wall diastolic thickness in homozygous mice compared with sham-operated wild-type mice, without significantly affecting overall cardiac performance.

Table 3.

Echocardiography of Ryr2^+/+ and Ryr2^D/D mice after TAC

	Sham		TAC
Parameters	Ryr2+/+ (n = 8)	Ryr2D/D (n = 8)	Ryr2+/+ (n = 6)	Ryr2D/D (n = 8)
HW/BW, %	0.55 ± 0.01	0.56 ± 0.01	0.63 ± 0.03	0.79 ± 0.07*†
HR, beats/min	583 ± 33	573 ± 27	650 ± 10	587 ± 19
LVEDD, mm	2.67 ± 0.15	2.18 ± 0.09*	2.59 ± 0.10	2.42 ± 0.08
LVESD, mm	1.25 ± 0.17	0.67 ± 0.07*	1.12 ± 0.09	0.96 ± 0.15
FS, %	55.0 ± 5.1	69.4 ± 2.4	57.1 ± 2.1	60.4 ± 5.6
IVSD, mm	1.12 ± 0.13	1.37 ± 0.06	1.36 ± 0.08	1.59 ± 0.14*
IVSS, mm	1.75 ± 0.18	2.10 ± 0.07	2.04 ± 0.09	2.33 ± 0.11*
LVPWD, mm	1.07 ± 0.14	1.32 ± 0.06	1.39 ± 0.11	1.67 ± 0.13*
LVPWS, mm	1.62 ± 0.21	1.89 ± 0.09	1.88 ± 0.13	2.22 ± 0.13

Data are means ± SE for n mice.

HW/BW, heart weight-to-body weight ratio; Sham, sham operated; TAC, transverse aortic constriction.

^*P < 0.05 compared with Sham Ryr2^+/+;

^†P < 0.05 compared with Sham Ryr2^D/D.

To further investigate the response of Ryr2^D/D mice to pressure overload, mRNA levels of cardiac hypertrophy-related genes were measured by quantitative RT-PCR. No significant changes in α-MHC, β-MHC, and BNP mRNA levels were observed (Fig. 7). However, there was a significant two- to threefold increase in ANP levels of sham-operated Ryr2^D/D animals compared with sham-operated wild-type mice. In pressure-overloaded mice, 1.3- and 2.1-fold increases in ANP mRNA levels were observed in wild-type and homozygous mutant mice, respectively, compared with sham-operated animals. Together, the data suggest that small differences between mutant and wild-type mice under baseline conditions become more prominent during pressure overload of Ryr2^D/D mice.

Fig. 7.

Quantitative RT-PCR analysis of marker genes for cardiac hypertrophy in mice subjected to transverse aortic constriction (TAC). RT-PCR was performed with total RNA from left ventricles of 4-mo-old mice. RNA levels were normalized to β-actin. Abbreviations are as in Fig. 5. Data are means ± SE of 6–8 samples. *P < 0.05 compared with sham-operated (Sham) wild type; ^#P < 0.05 compared with Sham Ryr2^D/D.

DISCUSSION

Accumulating evidence suggests that a dysfunctional RyR2 calcium channel has a major role in the genesis of cardiomyopathy. Aberrant phosphorylation, oxidation, and S-nitrosylation of RyR2, which cause leaky Ca²⁺ channels with an increased activity at diastolic Ca²⁺, have been implicated in arrhythmias and heart failure (1, 6, 18, 30, 37). Three simultaneous mutations in the CaM binding site of RyR2 (W3587A, L3591D, F3603A; RyR2^ADA) attenuated CaM regulation of RyR2 at diastolic and systolic Ca²⁺ concentrations (35). Ryr2^ADA/ADA knockin mice carrying the three mutations have severe cardiac hypertrophy and early death. Cardiomyocytes derived from homozygous Ryr2^ADA/ADA mice showed prolonged Ca²⁺ release. This suggests that CaM inhibition of RyR2 likely contributes to the termination of intracellular Ca²⁺ transients. To determine at which Ca²⁺ concentration, at diastolic or systolic Ca²⁺, CaM inhibition of RyR2 is of physiological significance, an additional knockin mouse carrying a single mutation (L3591D; RyR2^D) was generated. In contrast to Ryr2^ADA/ADA mice, Ryr2^D/D mice showed only modest changes in heart size and function. Single RyR2^ADA and RyR2^D channels lost CaM inhibition at diastolic Ca²⁺; however, unlike RyR2^ADA channels, RyR2^D channels maintained CaM inhibition at systolic Ca²⁺. The results suggest that loss of CaM regulation does not have as much an impact in Ryr2^D/D mice as in Ryr2^ADA/ADA mice, hence implying that CaM inhibition of RyR2 at systolic Ca²⁺ is necessary for normal cardiac growth and function.

Ryr2^D/D mice showed modest changes in heart contractility, morphology, and gene expression compared with Ryr2^ADA/ADA mice. Heart weight-to-body weight ratio of 6-mo-old Ryr2^D/D mice increased 9% compared with wild-type mice. In comparison, a more than twofold increase in heart weight-to-body weight ratio was observed for 10-day-old Ryr2^ADA/ADA mice compared with wild-type mice (35). ANP gene expression increased 1.8-fold in 6-mo-old Ryr2^D/D mice compared with ∼40-fold increase in 10-day-old Ryr2^ADA/ADA mice. In [³H]ryanodine binding experiments, functional RyR2 protein content decreased moderately by ∼20% (not significant) in hearts of 6-mo-old Ryr2^D/D mice compared with an ∼60% (significant) decrease in 10-day-old Ryr2^ADA/ADA mice.

Some differences were noted between the two mutant genotypes as well. Heterozygous mice had a lower heart weight-to-body weight ratio but increased right ventricular wall thickness and cross-sectional area and reduced posterior wall thickness as determined by echocardiography compared with homozygous mice. RyR2 is composed of four identical subunits, and, assuming a random distribution, 15 of 16 RyR2 have at least one wild-type subunit in heterozygous mice. Since the binding of about one CaM is sufficient to inhibit the tetrameric RyR (2), a majority of RyR2s in heterozygous mice were likely inhibited by CaM at systolic Ca²⁺. Additional studies are required to learn how the changes in CaM occupancy of heterozygous and homozygous mutant RyR2s affect cardiac morphology.

Stimulation of neonatal cardiomyocytes with endothelin-1 (9) and pacing-induced heart failure in canines (12) resulted in partial dissociation of CaM from RyR2. Expression of CaM isoform with higher binding affinity restored normal SR Ca²⁺ handling in cardiomyocytes from failing hearts (12). In the present study, TAC was used to learn how Ryr2^D/D mice respond to conditions of stress. We found that pressure overload increased heart weight-to-body weight ratio in Ryr2^D/D mice, with only small changes in cardiac parameters of wild-type and homozygous mice subjected to TAC compared with sham-operated mice. RT-PCR indicated a modest increase in ANP mRNA levels in mice subjected to TAC. The results suggest that inhibition of RyR2 by CaM and S100A1 at diastolic Ca²⁺ may preserve heart weight-to-body weight ratio and ANP expression during stress such as exercise. However, it should be noted that even under pressure overload of Ryr2^D/D mice, changes in heart growth and function were smaller than those in 10-day old Ryr2^ADA/ADA mice not subjected to TAC.

S100A1 is an EF hand Ca²⁺ binding protein expressed at high levels in the heart (13). Evidence indicates that S100A1 regulates intracellular Ca²⁺ homeostasis and cardiac contraction. S100A1, like CaM, regulates cardiac Ca²⁺ handling proteins such as RyR2, voltage-dependent L-type Ca²⁺ channel, and the SERCA2a-PLN complex (10, 21, 24, 29). [³H]ryanodine binding and Ca²⁺ spark measurements indicated that S100A1 has multiple effects on RyR2 channel activity. In [³H]ryanodine binding studies, low concentrations of S100A1 increased RyR2 activity at diastolic 0.1 μM Ca²⁺ but high concentrations decreased activity. At systolic Ca²⁺, S100A1 activated RyR2 (14, 20). In permeabilized cardiomyocytes, Ca²⁺ spark frequency decreased in response to exogenously added S100A1 (28). Similar to previous reports, we observed S100A1 effects on RyR2 function. However, this is the first study that describes a genetically modified mouse with a loss of S100A1 binding to RyR2, thus avoiding the multiple effects of gene delivery such as altering the regulation of SERCA and L-type Ca²⁺ channel. Single-channel recordings showed that 1 μM S100A1 inhibited wild-type RyR2 at 0.2 and 2 μM Ca²⁺. The RyR2-L3591D mutation eliminated the inhibitory effect of S100A1 on RyR2 at both Ca²⁺ concentrations. Nevertheless, Ryr2^D/D mice did not demonstrate major changes in cardiac function and growth. Together, the results suggest that S100A1 has a limited role in regulating RyR2 in vivo under normal conditions.

In conclusion, our studies with Ryr2^D/D mice show that severe cardiac hypertrophy and early death in Ryr2^ADA/ADA mice (33, 35) can be ascribed to impaired RyR2 inhibition by CaM at systolic Ca²⁺. Single-channel measurements showed that S100A1 and CaM share a single regulatory domain in RyR2. Loss of inhibition of RyR2 by CaM at diastolic Ca²⁺ and by S100A1 at diastolic and systolic Ca²⁺ in mice had only modest effects on cardiac morphology and function.

GRANTS

The work was supported by American Heart Association Grant 10SDG3500001 and National Science Foundation Grant EPS-0903795 to N. Yamaguchi and National Heart, Lung, and Blood Institute Grant HL-073051 to G. Meissner.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

Author contributions: N.Y., A.C., T.H., L.X., and G.M. conception and design of research; N.Y., A.C., T.H., L.X., A.C.G., and D.A.P. performed experiments; N.Y., A.C., T.H., L.X., A.C.G., D.A.P., and G.M. analyzed data; N.Y., A.C., T.H., L.X., D.A.P., and G.M. interpreted results of experiments; N.Y., L.X., and G.M. prepared figures; N.Y. and G.M. drafted manuscript; N.Y., A.C., T.H., L.X., A.C.G., D.A.P., and G.M. edited and revised manuscript; N.Y., A.C., T.H., L.X., A.C.G., D.A.P., and G.M. approved final version of manuscript.