Accelerated Cardiomyopathy in Mice With Overexpression of Cardiac Gsα and a Missense Mutation in the α-Myosin Heavy Chain
Background— To understand further the pathogenesis of familial hypertrophic cardiomyopathy, we determined how the cardiomyopathy induced by an Arg403→Gln missense mutation in the α-myosin heavy chain (403) is affected by chronically enhancing sympathetic drive by mating the mice with those overexpressing Gsα (Gsαx403).
Methods and Results— Heart rate in 3-month-old conscious mice was elevated similarly (P<0.05) in mice overexpressing Gsα (Gsα mice; 746±14 bpm) and Gsαx403 mice (718±19 bpm) compared with littermate wild-type mice (WT; 623±18 bpm) and 403 mice (594±16 bpm). Left ventricular ejection fraction (LVEF), as determined by echocardiography, was enhanced in Gsαx403 mice (88±1%, P<0.001) compared with WT (69±1%), 403 (75±1%), and Gsα (69±2%) mice. Isolated cardiomyocytes from Gsαx403 mice also exhibited higher (P<0.001) baseline percent contraction (11.9±0.5%) than WT (7.0±0.5%), 403 (5.5±0.5%), and Gsα (7.8±0.3%) cardiomyocytes. Relaxation of myocytes was impaired in 403 mice compared with WT but enhanced in Gsα and normalized in Gsαx403 mice. This was also observed in vivo. In vivo isoproterenol (0.1 μg · kg−1 · min−1) increased LVEF to maximal levels in Gsαx403 and Gsα, whereas in 403, the response was attenuated compared with WT. At 10 months of age, Gsαx403 had significantly depressed LVEF (57±4%). Histopathological examination demonstrated that myocyte hypertrophy and fibrosis were already present in young Gsαx403 mice and that old animals had severe cardiomyopathy. By 15 months of age, the survival of Gsαx403 was 0% compared with 100% for WT, 71% for Gsα, and 100% for 403 mice (P<0.05).
Conclusions— These results show that the cardiomyopathy developed by Gsαx403 mice is synergistic rather than additive, most likely owing to the elevated baseline function combined with enhanced responsiveness to sympathetic stimulation.
Received September 10, 2001; revision received October 30, 2001; accepted November 19, 2001.
Familial hypertrophic cardiomyopathy (FHC) is considered to be a disease of the sarcomere,1,2⇓ and several mutations in different sarcomeric proteins have been identified in affected individuals.3–5⇓⇓ In some cases, FHC is associated with premature death and frequent arrhythmias,6 but why this occurs in some patients and not in others is not known. Indeed, even though the genetic mutations have been widely identified, the pathophysiological mechanisms leading to the clinical phenotype are still poorly understood. Genetically altered mice bearing myosin heavy chain mutations that mimic FHC in humans are a useful model to clarify the processes that lead to the common phenotype of FHC, eg, mice with an Arg403→Gln missense mutation in the α-myosin heavy chain (403 mice),7 which appear morphologically normal when they are young but develop hypertrophy and cardiomyopathy when they age.8 On the other hand, transgenic mice with overexpression of Gsα (Gsα mice) exhibit a hyperadrenergic phenotype (specifically, enhanced contractile and relaxation responses to β-adrenergic stimulation) but remarkably also develop cardiomyopathy when they age.9,10⇓
The goal of this investigation was to determine whether a new mouse model created by mating 403 mice with Gsα mice (Gsαx403) affected the development of cardiomyopathy in these mice and specifically (1) whether the cardiomyopathy was accelerated, (2) whether the diastolic dysfunction in 403 mice was corrected, and (3) whether β-adrenergic responsiveness was altered in dual transgenic animals, in view of the fact that β-adrenergic responsiveness is enhanced in Gsα10 and depressed in 403 mice (unpublished observations). To address these goals, we examined Gsαx403 mice in vivo using echocardiography and left ventricular (LV) pressure measurements to assess LV function and in vitro using isolated cardiomyocytes, and we performed postmortem histological examinations of the hearts. These results could potentially explain why some individuals with FHC fare more poorly than others.
Gsα-transgenic mice, which overexpress cardiac Gsα 3- to 5-fold, were mated with heterozygous 403 mice to produce a dual transgenic mouse line. Four groups of the first-generation offspring were generated, including wild-type (WT) mice, Gsα-overexpressing mice, 403 mice, and Gsαx403 mice, with each accounting for 25% of the total offspring, as confirmed by genotyping by methods described previously.7,9⇓ Thus, the WT littermates were true controls for the Gsαx403 mice. Initial studies were performed in 3- to 4-month-old mice. To determine changes in LV function over time, a total of 32 mice from the 4 groups (6 to 9 per group) were examined by echocardiography at 3 and 6 months of age; thereafter, the same mice were studied serially each month by echocardiography up to 10 months of age. Survival was followed up in these mice up to 15 months of age. Histological analysis was performed in 3- to 4-month-old and in 10- to 11-month-old mice.
Heart Rate in Conscious Mice
Heart rate was measured by telemetry techniques under conscious and unrestrained conditions in 3-month-old animals. Mice were anesthetized with an intraperitoneal injection of ketamine (0.065 mg/g), acepromazine (0.02 mg/g), and xylazine (0.013 mg/g), and a telemetry transducer (TA10EA-F20; Data Science Co) was implanted subcutaneously into the back, with paired electrodes placed around the thorax. Experiments were initiated 3 to 5 days after recovery from surgical instrumentation. Mice with implanted telemetry devices were housed in individual cages with free access to food and water and were exposed to 12-hour-light/12-hour-dark cycles.
Mice were anesthetized as described above. The procedure for echocardiography has been reported previously.10–12⇓⇓ To assess responses to β-adrenergic stimulation, LV function was measured during intravenous infusion of isoproterenol (0.1 μg · kg−1 · min−1 for 5 minutes) via a chronically implanted jugular catheter.
LV Hemodynamics and Arterial Pressure
The mice (3 months old) were anesthetized as described above, and a bipolar transesophageal pacing electrode was introduced. A 1.4F micromanometer catheter (Millar Instruments Inc) was inserted into the right carotid artery, and LV hemodynamics and arterial blood pressures were measured at spontaneous heart rates and during atrial pacing with 550 bpm.
Measurement of Contractile and Relaxation Function
Cardiac myocytes were prepared as described previously.13,14⇓ Myocytes were transferred to a warmed and continuously perfused cell chamber located on an inverted microscope stage (Nikon Inc). Myocyte length was measured with a video motion edge detector (VED103; Crescent Electronics), and the data were acquired at 240 images per second. Myocyte contraction was also examined in response to isoproterenol (10−8 mol/L) to determine the extent to which β-adrenergic function is altered in Gsαx403 mice. To examine whether the inotropic response of myocytes to non–β-adrenergic-receptor–mediated stimulation was altered, myocyte function was assessed with Ca2+ (4 mmol/L). Myocyte contractile and relaxation function was also assessed with Rp-cAMP (200 μmol/L), which blocks protein kinase A. Myocytes were loaded with 3.8 mmol/L of fura-2 (Sigma Chemical Co), and intracellular free Ca2+ was measured as the fluorescence ratio (340/380 nm).13,14⇓
Histological studies of the hearts were conducted in perfusion-fixed hearts from 28 younger mice (3 to 4 months old) and 24 older mice (10 to 11 months old). Myocardial connective tissue was analyzed quantitatively in a cross section of the LV obtained midway between base and apex and stained with picric acid sirius red F36 as described previously.12 DNA fragmentation of myocyte nuclei was detected in situ by use of terminal dUTP nick end-labeling (TUNEL) on paraffin sections of the mouse hearts, as described previously.15
Western blot analysis of Gsα and β-receptor density was performed from mouse hearts as described previously.16
All data were reported as mean±SEM. Group comparisons were analyzed by multiple analysis of variance (MANOVA), followed by post hoc comparisons between groups with the least significant difference test or by ANOVA for repeated measurements. A P value <0.05 was considered statistically significant.
In Vivo Physiology
Heart rate (bpm) measured in the conscious state was elevated similarly (P<0.05) in 3-month-old Gsα (746±14) and Gsαx403 (718±19) mice compared with either WT (623±18) or 403 (594±16, n=6 to 10 per group) mice. Heart rate was also elevated in these 2 groups compared with WT and 403 mice under anesthesia (Table 1). LV dP/dtmax (mm Hg/s), measured in anesthetized animals, was enhanced only in Gsαx403 mice (10 836±513, P<0.05; Table 2) compared with WT (6065±347), Gsα (7192±449), and 403 (5927±373) mice. To exclude potential confounding effects of different heart rates, measurements were repeated during transesophageal pacing at 550 bpm. With heart rate held constant, LV dP/dtmax (mm Hg/s) was still only higher (P<0.05) in Gsαx403 mice (12 400±551) compared with WT (9259±607), Gsα (8157±629), and 403 (10 843±652) mice. Impaired relaxation (LV dP/dtmin, mm Hg/s) in anesthetized 403 mice in spontaneous rhythms (3767±186) was restored in Gsαx403 (6255±382, P<0.05; Table 1).
Ejection fraction was one LV functional parameter used, recognizing that there are limitations to its interpretation because only 1 LV axis was measured. In Gsαx403 mice, LV ejection fraction (LVEF) was enhanced in young animals and depressed by 10 months of age (Figure 1). LVEF was enhanced in Gsαx403 mice (88±1%, P<0.001) compared with WT (69±1%), 403 (75±1%), and Gsα (69±2%) mice (Table 1). The Vcfc (velocity of circumferential fiber shortening corrected for heart rate) versus afterload relation also revealed enhanced contractility in Gsαx403 mice compared with all other groups (Figure 2, left). Relaxation half-time, the time required for LV pressure at end ejection to be reduced by 50%, which is another index of diastolic function, was prolonged in 403 mice and shortened in Gsα and Gsαx403 mice (Figure 2, right). At 6 months of age, echocardiographic measurements were unchanged, whereas at 10 months of age, LVEF was depressed significantly only in Gsαx403 mice (57±4%) compared with LVEF in these same mice at 6 months of age (Figure 3). LV end-diastolic diameter was similar among groups at 3 months of age (Table 1). By 10 months of age, only the Gsαx403 mice exhibited LV dilatation as reflected by LV end-diastolic diameter, which was significantly higher (4.9±0.1 mm) than for 403 (4.5±0.1 mm), Gsα (4.4±0.1 mm), and WT (4.6±0.1 mm) mice (P<0.05), but end-diastolic LV anterior wall thickness was significantly increased only in 403 mice. Isoproterenol (0.1 μg · kg−1 · min−1) increased LVEF to maximal levels in Gsαx403 (91±1%) and Gsα (90±2%) mice but not in 403 (80±1%) or WT (80±2%) mice (Figure 4, left).
By 15 months of age, mortality was 100% in Gsαx403 mice compared with 0% in WT, 29% in Gsα, and 0% in 403 mice (P<0.05). Postmortem analysis revealed overt signs of heart failure (ascites, pleural effusion, and enlargement of LV and right ventricle) in all Gsαx403 mice that died spontaneously.
In parallel to the in vivo findings, Gsαx403 cardiomyocytes had a significantly higher baseline contraction (11.9±0.5%) than WT (6.7±0.6%), 403 (5.5±0.5%), and Gsα (7.8±0.3%) cardiomyocytes (Table 2). Consistent with the in vivo findings, the time of relaxation required for myocytes to lengthen to 70% of end-diastolic baseline, an indicator of lusitropic properties, was prolonged (P<0.05) in 403 (82±6 ms) compared with WT (68±4 ms) mice but normalized in Gsαx403 mice (62±3 ms) and enhanced (P<0.05) in Gsα mice (56±3 ms; Table 2). The maximal rate of relengthening (+dL/dtmax), another index of diastolic properties, was enhanced in Gsα and Gsαx403 mice but diminished in 403 mice (Table 2). Similar to in vivo results, isoproterenol (10−8 mol/L) also elicited enhanced responsiveness in isolated myocytes from Gsα and Gsαx403 mice and a blunted response to β-adrenergic stimulation in 403 myocytes (Figure 4, middle). In contrast, nonspecific inotropic stimulation with increased extracellular Ca2+ (4 mmol/L) increased percent contraction similarly in all groups (Figure 4, right). Baseline contractility after incubation with Rp-cAMP did not significantly change in Gsαx403 myocytes (12.6±1.2%) and was still higher than percent contraction of WT (8.6±0.6%), Gsα (9.5±1.3%), and 403 (6.6±0.3%) myocytes. Ca2+ transients were significantly higher in Gsαx403 mice than in WT and 403 mice (Table 2). There was no difference among groups for diastolic Ca2+ content of the myocytes. Ca2+ reuptake, as assessed by the time for recovery to 70% of baseline, was prolonged in 403 myocytes (154±12 ms) compared with WT myocytes (138±7 ms) and was shorter in Gsα (103±11 ms) and Gsαx403 (111±4 ms) myocytes.
Gsα-protein was upregulated in Gsα (3.1±0.4) and Gsαx403 (2.7±0.3) mice compared with WT mice (1.0±0.04) and was consistently downregulated in 403 mice (0.7±0.01; P<0.05, n=10 animals per group). In contrast, Gi protein remained unchanged among groups (data not shown). β-Receptor density was unchanged in 403 mice (WT, 34.4±3.2 fmol/mg; 403, 30.0±5.1 fmol/mg).
Heart weight/body weight ratio (mg/g) of fresh tissue was significantly higher (P<0.05) in 3-month-old Gsαx403 mice (4.8±0.2) than in WT (4.3±0.1), Gsα (4.1±0.1), and 403 (4.2±0.1) mice, whereas body weight was similar among the groups. Histopathological examination demonstrated that mild to moderate myocyte hypertrophy and fibrosis were present in 3-month-old Gsαx403 mice, whereas lesions were essentially absent in the other groups (data not shown). Heart weight/body weight ratio was significantly increased in 10- to 11-month-old Gsαx403 mice (4.7±0.3) compared with WT (3.6±0.1), Gsα (3.6±0.1), and 403 (3.9±0.1) mice. Among the 10- to 11-month-old mice (Figure 5), the WT mice appeared normal and the 403 group had occasional focal lesions of increased interstitial fibrosis, enlarged myocytes, and evidence of myofiber disarray in ventricular myocytes, similar to that reported previously.7 The Gsα group also had lesions similar to those reported previously in mice aged 10 months or older,12 including multifocal regions of interstitial fibrosis; myocyte enlargement; enlarged bizarre nuclei, sometimes with vacuoles; focal cytoplasmic vacuoles in myocytes; and evidence of myocyte degeneration, atrophy, and cell loss. These lesions were markedly accentuated in the Gsαx403 mice (Figure 5d). Myocyte size was increased in Gsα and 403 mice compared with WT and was greater in Gsαx403 mice than in all other groups (Figure 6A). Quantitative evaluation of volume percent collagen (Figure 6B) revealed a large increase in the Gsαx403 mice compared with the other groups. The percent of TUNEL-positive myocytes was also markedly increased in Gsαx403 mice (Figure 6C, indicating enhanced apoptosis).
We investigated the effects of targeted overexpression of Gsα in the hearts of 403 mice, a transgenic animal model of FHC. Our results indicate that the mechanisms that cause cardiomyopathy in Gsα and 403 mice act in concert in the development of cardiomyopathy, which is accelerated and more severe in Gsαx403 mice than in either of the parents and ultimately leads to premature death in these animals, potentially offering an explanation for the adverse consequences in some patients with FHC but not others. The Gsαx403 mice exhibit evidence of beginning cardiomyopathy, eg, fibrosis and hypertrophy, at 3 months of age, which becomes more severe and includes depressed LV function, LV dilatation, and increased apoptosis as the animals age. There are several possible explanations for the mechanism of this accelerated cardiomyopathy. The most intriguing possibility relates to the enhanced cardiac function in vivo and myocyte function in vitro at baseline and in response to β-adrenergic stimulation. Impaired relaxation and blunted responses to β-adrenergic stimulation characteristic of 403 mice were not only corrected but actually enhanced in the Gsαx403 mice in vivo. Similar findings were observed in isolated cardiomyocytes, ie, enhanced contraction and relaxation in Gsαx403 myocytes, which was also reflected in the Ca2+ transients. These findings in isolated myocytes are important because this occurs in the absence of neurohormonal stimulation. The increased baseline levels of heart rate, systolic function, and diastolic function, in combination with enhanced β-adrenergic responsiveness, provide a framework for understanding the mechanism of the earlier and more severe development of cardiomyopathy.
The question arises, why is baseline function enhanced in Gsαx403 compared with Gsα alone? The answer to this question will be essential for understanding the mechanism of the accelerated cardiomyopathy. A clue may be found in the LVEF, which is also enhanced in the 403 mice. These mice have impaired diastolic function, which is corrected in the Gsαx403 cross. It is our contention that by correcting the impaired diastolic relaxation, systolic function is permitted to increase further, which becomes even more noticeable in isolated myocytes. This concept is consistent with the point of view that the 403 mutation exhibits a gain of function, as has been suggested from single-molecule mechanics of whole cardiac myosin,17 as well as pressure-volume relationships.18 Thus, in the Gsαx403 model, the gain of function is augmented by correcting the diastolic dysfunction of 403 mice when mated with Gsα mice, which results in enhanced systolic function at baseline even in the absence of neurohormonal stimulation. This concept is supported by data with Rp-cAMP, a protein kinase A inhibitor, which did not affect baseline myocyte function. Thus, the elevated LV function and heart rate at baseline combined with augmented contractility on β-adrenergic stimulation conceivably provide the framework for the accelerated cardiomyopathy observed in the present study. At least in this mouse model of FHC, there is no evidence that loss of function is the cause of cardiomyopathy.19
As expected, protein levels of Gsα were increased similarly in Gsα and Gsαx403 compared with WT mice, but interestingly, Gsα-protein was downregulated in 403 mice. Because β-adrenergic receptor density was unchanged in 403 mice, our findings suggest that in 403 mice, a 30% reduction in the protein level of Gsα may be responsible for the decreased responsiveness to β-adrenergic stimulation in vivo and in isolated cardiomyocytes. However, it is just as conceivable that the correction of diastolic function and enhanced systolic function at baseline provide the framework for the augmented responses to β-adrenergic stimulation, both at the level of isolated myocyte function and in vivo. Enhanced inotropic responses were not observed with the nonspecific inotropic agent Ca2+, which also confirms that enhanced β-adrenergic signaling is critical to the development of the cardiomyopathy. Although we cannot entirely exclude the possibility that other G-protein–coupled receptors or distal mechanisms may play a role in the development of cardiomyopathy in Gsαx403 mice, our previous study demonstrated that the cardiomyopathy in Gsα mice was blocked completely by propranolol,12 which further supports the concept that the cardiomyopathy in Gsαx403 mice is mediated through the β-adrenergic signaling pathway.
The data from the present study show that the combination of overexpression of Gsα and the α-myosin heavy chain mutation results in both accelerated and more severe cardiomyopathy than observed in either parent alone. It is likely that enhanced baseline LV function combined with elevated heart rate in young Gsαx403 mice accounts for the accelerated cardiomyopathy. These results provide a molecular underpinning to clinical studies, which have suggested that sympathetic stimulation may be deleterious both in the short and long term in patients with FHC and that β-blocker therapy is beneficial in some patients with FHC.20,21⇓
This study was supported in part by NIH HL-33107, HL-33065, HL-69020, HL-59139, HL-65182, HL-65183, and AG-14121.
Guest Editor for this article was Robert Roberts, MD, FACC, FRSM, FRC, Baylor College of Medicine, Houston, Tex.
- ↵Bonne G, Carrier L, Richard P, et al. Familial hypertrophic cardiomyopathy: from mutations to functional defects. Circ Res. 1998; 83: 580–593.
- ↵Doevendans PA. Hypertrophic cardiomyopathy: do we have the algorithm for life and death? Circulation. 2000; 101: 1224–1226.
- ↵Marian AJ, Roberts R. Recent advances in the molecular genetics of hypertrophic cardiomyopathy. Circulation. 1995; 92: 1336–1347.
- ↵Geisterfer-Lowrance AA, Christe M, Conner DA, et al. A mouse model of familial hypertrophic cardiomyopathy. Science. 1996; 272: 731–734.
- ↵McConnell BK, Fatkin D, Semsarian C, et al. Comparison of two murine models of familial hypertrophic cardiomyopathy. Circ Res. 2001; 88: 383–389.
- ↵Gaudin C, Ishikawa Y, Wight DC, et al. Overexpression of Gs alpha protein in the hearts of transgenic mice. J Clin Invest. 1995; 95: 1676–1683.
- ↵Iwase M, Bishop SP, Uechi M, et al. Adverse effects of chronic endogenous sympathetic drive induced by cardiac Gs alpha overexpression. Circ Res. 1996; 78: 517–524.
- ↵Iwase M, Uechi M, Vatner DE, et al. Cardiomyopathy induced by cardiac Gs alpha overexpression. Am J Physiol. 1997; 272: H585–H589.
- ↵Kim SJ, Iizuka K, Kelly RA, et al. An alpha-cardiac myosin heavy chain gene mutation impairs contraction and relaxation function of cardiac myocytes. Am J Physiol. 1999; 276: H1780–H1787.
- ↵Geng YJ, Ishikawa Y, Vatner DE, et al. Apoptosis of cardiac myocytes in Gs alpha transgenic mice. Circ Res. 1999; 84: 34–42.
- ↵Tyska MJ, Hayes E, Giewat M, et al. Single-molecule mechanics of R403Q cardiac myosin isolated from the mouse model of familial hypertrophic cardiomyopathy. Circ Res. 2000; 86: 737–744.
- ↵Loogen F, Kuhn H, Gietzen F, et al. Clinical course and prognosis of patients with typical and atypical hypertrophic obstructive and with hypertrophic non-obstructive cardiomyopathy. Eur Heart J. 1983; 4 (suppl F): 145–153.