Effects of sex hormones on development of physiological and pathological cardiac hypertrophy in male and female rats ASHWANI MALHOTRA, PETER BUTTRICK, AND JAMES SCHEUER Division of Cardiology, Department of Medicine, Montefiore Medical Center and Albert Einstein College of Medicine, Bronx, New York 10467
MALHOTRA, ASHWANI, PETER BUTTRICK, AND JAMES SCHEUER. Effects of sex hormones on development of physiological and pathological cardiac hypertrophy in male and female rats. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H866-H871, 1990.-Previous studies have demonstrated a role for sex hormones in maintaining normal heart weight and myosin isoenzyme balance in the rat. To determine if sex hormones were necessary to elicit cardiac adaptations to the chronic loads of swimming or hypertension, female rats were gonadectomized (X) and then exposed either to a chronic swimming program (SW) or to renal hypertension for 8-10 wk. Because gonadectomy in females increased heart and body weight, separate groups of food-restricted sedentary and SW gonadectomized females (XFR) were included. Swimming resulted in significant increases in both heart weight and in the percent ventricular VI isomyosin in female controls (C), X, and XFR. Hypertension was studied in C, X, and X with estrogen replacement. Cardiac hypertrophy developed in all groups, but estrogen therapy attenuated the decline in percent VI isomyosin in both normotensive and hypertensive X animals. Swimming, which is generally not associated with cardiac hypertrophy in males, was also studied in that sex. Gonadectomy did not alter either the heart weight or the myosin isoenzyme response to SW, although testosterone replacement in gonadectomized males restored ventricular VI myosin levels to or above normal. Measures of serum thyroid levels and of myocardial catecholamines failed to demonstrate a causal relationship between these hormones and the various results. Therefore, although sex hormones are important for maintaining normal heart weights and myosin isoenzyme balance in rats, they do not appear to be important in the adaptations hearts exhibit when exposed to physiological or pathological loads. cardiac contractile proteins; phosphatase activity
myosin
isoenzymes;
adenosinetri-
WE HAVEPREVIOUSLY SHOWN thatgonadectomyinboth pre- and postpubertal male and female rats is associated with depressed contractile performance (23). This is paralleled by decreases in ventricular myosin adenosinetriphosphatase (ATPase) activity and a shift in the myosin heavy chain isoenzyme balance from predominantly V, to Vs. The changes in cardiac mechanics and myosin enzymology can be largely reversed by sex-specific hormone replacement with estrogen or testosterone (26). These studies, coupled with the identification of sex hormone receptors in the myocardium, justify the inclusion of the sex hormones, along with thyroid hormone, insulin, and catecholamines, among others, in the list of H866
0363-6135/90
$1.50 Copyright
cardioregulatory hormones (10-12, 18, 19, 28, 29). The hormones could exert their effects either directly via specific receptors or indirectly (6, 8). In the present study, we asked whether male and female gonads were important in the cardiac biochemical adaptations to imposed physiological and pathological loads. Adult female rats, in which characteristic cardiac adaptations to imposed loads have previously been defined (16), were ovariectomized and then exposed to either conditioning by swimming or to renovascular hypertension. Because we have observed differences in the response to swim conditioning between male and female rats, we also studied the response of castrated males to the physical conditioning program. Myosin isoenzyme distribution was measured to characterize the cardiac response to the various protocols. The data indicate that normal sex hormone levels, while important in regulating myocardial myosin heavy chain content, are not necessary for the myocardium to adapt to imposed pathological and physiological loads. METHODS
Animal models. In all studies, male or female Wistar rats (Charles River Breeding Laboratory) were employed and were gonadectomized or sham operated by the supplier at 10 wk of age. Studies as described below were initiated l-2 wk after gonadectomy, and animals were killed 8-10 wk later. The adequacy of gonadectomy was determined at the time of death by documenting the absence of testicular tissue in males or ovarian tissues in females. We have previously shown that this is sufficient to assure the success of the surgical procedure (30), although in the present study serum hormone levels were also measured in a number of animals. Animals were fed and given water ad libitum, with the exception of a foodrestricted gonadectomized group of female animals. In this group food intake was monitored and adjusted daily so that the animals did not exhibit the weight gain normally associated with ovariectomy. The logistics of conditioning large numbers of animals dictated that these studies be conducted in a sequential fashion, with each intervention evaluated independently. Age-matched control animals were maintained and studied in parallel with each interventional group. In study 1, female gonadectomized or sham-operated control animals were randomized to either cage confinement or to swim conditioning. The physical conditioning
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 8, 2019.
HORMONES
IN
PHYSIOLOGICAL
AND
protocol consisted of swimming 90 min twice a day for 8 wk. The specific details for this regimen and the physiological and biochemical consequences in the rat heart have been described (24). Because we have previously demonstrated that gonadectomy in female rats increases both heart and body weights, a group of food-restricted ovariectomized animals were maintained separately, onehalf of which were cage confined at rest and the other one-half conditioned by swimming, so that weight gain could be avoided. In study 2, the interaction between gonadectomy and renovascular hypertension was evaluated in female rats. In this study the groups were control, ovariectomized animals, and ovariectomized animals given estrogen replacement (2 pg/day for 10 wk). The group of ovariectomized animals with estrogen replacement was added because hypertension and ovariectomy both lower myosin V1 content. The purpose of this group was to define the relative contribution of both hormonal and the hypertensive intervention and to see if estrogen replacement could limit the effect of hypertension on myosin heavy chain distribution. A set of animals from each condition was made hypertensive so that six groups were investigated. Renovascular hypertension was created using the two-kidney, one-clip model that is standard in our laboratory (4). Blood pressure was measured by the tail-cuff method under light ether anesthesia 3 wk after the surgical procedure and then every 2-3 wk for the duration of the study. Only those hypertensive animals whose blood pressures were consistently X50 mmHg were included in the final analysis. Because of the time lag necessary to develop sustained hypertension, the animals in this protocol were 3-4 wk older than those in the swimming study at the time of death. In study 3, male animals were evaluated to assess the interaction between the male sex hormones and the physiological adaptations that develop in response to conditioning by swimming. Three groups of controls and three groups of swim-conditioned male animals were studied: sham-operated controls, gonadectomized, and gonadectomized animals treated with testosterone administered in an oil-based vehicle at a dose of 3 mg/day, 5 days/wk. Animals were followed for 8-10 wk before they were killed. Because we have previously demonstrated that gonadectomy in male animals is not associated with a marked weight gain (23), a pair-weighted cohort of male animals was not maintained. Biochemical analysis. When the animals were killed, hearts were removed, the great vessels and atria were trimmed, and the ventricles were weighed. A piece of the left ventricle was dried to constant weight, and the remainder was stored for enzymology at -80°C in 50% glycerol containing 50 mM KC1 and 10 mM K3P04 (pH 7.0) before preparation of the extracts. Myosin was extracted and fractionated as previously described (15) and was shown by sodium dodecyl sulfate gel electrophoresis to be free of actin, troponin, and tropomyosin and without evidence of proteolytic degradation. In some of the preparations, crude myosin was extracted and isolated as previously described (9, 14). For myosin isoenzymes, purified or crude myosin was subjected to polyacrylamide
PATHOLOGICAL
HYPERTROPHY
H867
gel electrophoresis using Na-pyrophosphate buffer under nondissociating conditions at 2°C as reported by d’Albis et al. (7). Densitometric scans of the gels were recorded at 575 nm on an EC apparatus attached to a HewlettPackard integrator (model 3390A) and semiquantitative estimates of the isoenzymes were calculated from the area under the peak height by integration. The amount of VZ isoenzyme was equally distributed between V1 (aa) and V3 (pp) to reflect the myosin heavy chain distribution. Serum hormone LeueLs. Blood was obtained from some of the animals at the time of death, and serum estrogen levels were measured after extraction according to the method of Pang et al. (21). Serum testosterone levels were determined with a commercially available radioimmunoassay (RIA) kit. Thyroid levels (T3 and T4) were measured in plasma using a commercially available RIA (Pantex, La Jolla, CA), which we have previously validated for use with rat plasma. Myocardial catecholamine levels. Right and left ventricular myocardium was assayed for levels of epinephrine and norepinephrine. The tissue was homogenized in 5 ml of 0.1 M HC104 plus 5 mg of NazSz05 and centrifuged (5,000 g) for 30 min. The supernatant was frozen for subsequent high-performance liquid chromatography-EC analysis as described previously (22). StatisticaL analysis. Results were submitted to a twofactor analysis of variance using conditioning status (i.e., sedentary-swimmer or normotensive-hypertensive) and gonadectomy as the two factors. The mean-square error within groups was then used in a Newman-Keuls multiple comparison test to evaluate differences between groups (32). To evaluate the effect of food restriction and gonadectomy on myocardial catecholamines, the three groups, control, gonadectomized, and food-restricted ovariectomized, for each state, sedentary or swimming, were analyzed as individual three-group experiments by the above statistical procedure. Unless otherwise stated significance is reported at the 0.05 level. RESULTS
Inspection of all gonadectomized, ovariectomized and given estrogen, and gonadectomized and given testosterone animals when killed revealed no detectable gonadal tissue and in females absence of a developed uterus. In addition, serum estrogen or testosterone levels were measured in randomly selected gonadectomized animals. Gonadectomy reduced serum estrogen in females from 12.4 t 4.2 (SE) pg/ml to
MALHOTRA, ASHWANI, PETER BUTTRICK, AND JAMES SCHEUER. Effects of sex hormones on development of physiological and pathological cardiac hypertrophy in male and female rats. Am. J. Physiol. 259 (Heart Circ. Physiol. 28): H866-H871, 1990.-Previous studies have demonstrated a role for sex hormones in maintaining normal heart weight and myosin isoenzyme balance in the rat. To determine if sex hormones were necessary to elicit cardiac adaptations to the chronic loads of swimming or hypertension, female rats were gonadectomized (X) and then exposed either to a chronic swimming program (SW) or to renal hypertension for 8-10 wk. Because gonadectomy in females increased heart and body weight, separate groups of food-restricted sedentary and SW gonadectomized females (XFR) were included. Swimming resulted in significant increases in both heart weight and in the percent ventricular VI isomyosin in female controls (C), X, and XFR. Hypertension was studied in C, X, and X with estrogen replacement. Cardiac hypertrophy developed in all groups, but estrogen therapy attenuated the decline in percent VI isomyosin in both normotensive and hypertensive X animals. Swimming, which is generally not associated with cardiac hypertrophy in males, was also studied in that sex. Gonadectomy did not alter either the heart weight or the myosin isoenzyme response to SW, although testosterone replacement in gonadectomized males restored ventricular VI myosin levels to or above normal. Measures of serum thyroid levels and of myocardial catecholamines failed to demonstrate a causal relationship between these hormones and the various results. Therefore, although sex hormones are important for maintaining normal heart weights and myosin isoenzyme balance in rats, they do not appear to be important in the adaptations hearts exhibit when exposed to physiological or pathological loads. cardiac contractile proteins; phosphatase activity
myosin
isoenzymes;
adenosinetri-
WE HAVEPREVIOUSLY SHOWN thatgonadectomyinboth pre- and postpubertal male and female rats is associated with depressed contractile performance (23). This is paralleled by decreases in ventricular myosin adenosinetriphosphatase (ATPase) activity and a shift in the myosin heavy chain isoenzyme balance from predominantly V, to Vs. The changes in cardiac mechanics and myosin enzymology can be largely reversed by sex-specific hormone replacement with estrogen or testosterone (26). These studies, coupled with the identification of sex hormone receptors in the myocardium, justify the inclusion of the sex hormones, along with thyroid hormone, insulin, and catecholamines, among others, in the list of H866
0363-6135/90
$1.50 Copyright
cardioregulatory hormones (10-12, 18, 19, 28, 29). The hormones could exert their effects either directly via specific receptors or indirectly (6, 8). In the present study, we asked whether male and female gonads were important in the cardiac biochemical adaptations to imposed physiological and pathological loads. Adult female rats, in which characteristic cardiac adaptations to imposed loads have previously been defined (16), were ovariectomized and then exposed to either conditioning by swimming or to renovascular hypertension. Because we have observed differences in the response to swim conditioning between male and female rats, we also studied the response of castrated males to the physical conditioning program. Myosin isoenzyme distribution was measured to characterize the cardiac response to the various protocols. The data indicate that normal sex hormone levels, while important in regulating myocardial myosin heavy chain content, are not necessary for the myocardium to adapt to imposed pathological and physiological loads. METHODS
Animal models. In all studies, male or female Wistar rats (Charles River Breeding Laboratory) were employed and were gonadectomized or sham operated by the supplier at 10 wk of age. Studies as described below were initiated l-2 wk after gonadectomy, and animals were killed 8-10 wk later. The adequacy of gonadectomy was determined at the time of death by documenting the absence of testicular tissue in males or ovarian tissues in females. We have previously shown that this is sufficient to assure the success of the surgical procedure (30), although in the present study serum hormone levels were also measured in a number of animals. Animals were fed and given water ad libitum, with the exception of a foodrestricted gonadectomized group of female animals. In this group food intake was monitored and adjusted daily so that the animals did not exhibit the weight gain normally associated with ovariectomy. The logistics of conditioning large numbers of animals dictated that these studies be conducted in a sequential fashion, with each intervention evaluated independently. Age-matched control animals were maintained and studied in parallel with each interventional group. In study 1, female gonadectomized or sham-operated control animals were randomized to either cage confinement or to swim conditioning. The physical conditioning
0 1990 the American
Physiological
Society
Downloaded from www.physiology.org/journal/ajpheart by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 8, 2019.
HORMONES
IN
PHYSIOLOGICAL
AND
protocol consisted of swimming 90 min twice a day for 8 wk. The specific details for this regimen and the physiological and biochemical consequences in the rat heart have been described (24). Because we have previously demonstrated that gonadectomy in female rats increases both heart and body weights, a group of food-restricted ovariectomized animals were maintained separately, onehalf of which were cage confined at rest and the other one-half conditioned by swimming, so that weight gain could be avoided. In study 2, the interaction between gonadectomy and renovascular hypertension was evaluated in female rats. In this study the groups were control, ovariectomized animals, and ovariectomized animals given estrogen replacement (2 pg/day for 10 wk). The group of ovariectomized animals with estrogen replacement was added because hypertension and ovariectomy both lower myosin V1 content. The purpose of this group was to define the relative contribution of both hormonal and the hypertensive intervention and to see if estrogen replacement could limit the effect of hypertension on myosin heavy chain distribution. A set of animals from each condition was made hypertensive so that six groups were investigated. Renovascular hypertension was created using the two-kidney, one-clip model that is standard in our laboratory (4). Blood pressure was measured by the tail-cuff method under light ether anesthesia 3 wk after the surgical procedure and then every 2-3 wk for the duration of the study. Only those hypertensive animals whose blood pressures were consistently X50 mmHg were included in the final analysis. Because of the time lag necessary to develop sustained hypertension, the animals in this protocol were 3-4 wk older than those in the swimming study at the time of death. In study 3, male animals were evaluated to assess the interaction between the male sex hormones and the physiological adaptations that develop in response to conditioning by swimming. Three groups of controls and three groups of swim-conditioned male animals were studied: sham-operated controls, gonadectomized, and gonadectomized animals treated with testosterone administered in an oil-based vehicle at a dose of 3 mg/day, 5 days/wk. Animals were followed for 8-10 wk before they were killed. Because we have previously demonstrated that gonadectomy in male animals is not associated with a marked weight gain (23), a pair-weighted cohort of male animals was not maintained. Biochemical analysis. When the animals were killed, hearts were removed, the great vessels and atria were trimmed, and the ventricles were weighed. A piece of the left ventricle was dried to constant weight, and the remainder was stored for enzymology at -80°C in 50% glycerol containing 50 mM KC1 and 10 mM K3P04 (pH 7.0) before preparation of the extracts. Myosin was extracted and fractionated as previously described (15) and was shown by sodium dodecyl sulfate gel electrophoresis to be free of actin, troponin, and tropomyosin and without evidence of proteolytic degradation. In some of the preparations, crude myosin was extracted and isolated as previously described (9, 14). For myosin isoenzymes, purified or crude myosin was subjected to polyacrylamide
PATHOLOGICAL
HYPERTROPHY
H867
gel electrophoresis using Na-pyrophosphate buffer under nondissociating conditions at 2°C as reported by d’Albis et al. (7). Densitometric scans of the gels were recorded at 575 nm on an EC apparatus attached to a HewlettPackard integrator (model 3390A) and semiquantitative estimates of the isoenzymes were calculated from the area under the peak height by integration. The amount of VZ isoenzyme was equally distributed between V1 (aa) and V3 (pp) to reflect the myosin heavy chain distribution. Serum hormone LeueLs. Blood was obtained from some of the animals at the time of death, and serum estrogen levels were measured after extraction according to the method of Pang et al. (21). Serum testosterone levels were determined with a commercially available radioimmunoassay (RIA) kit. Thyroid levels (T3 and T4) were measured in plasma using a commercially available RIA (Pantex, La Jolla, CA), which we have previously validated for use with rat plasma. Myocardial catecholamine levels. Right and left ventricular myocardium was assayed for levels of epinephrine and norepinephrine. The tissue was homogenized in 5 ml of 0.1 M HC104 plus 5 mg of NazSz05 and centrifuged (5,000 g) for 30 min. The supernatant was frozen for subsequent high-performance liquid chromatography-EC analysis as described previously (22). StatisticaL analysis. Results were submitted to a twofactor analysis of variance using conditioning status (i.e., sedentary-swimmer or normotensive-hypertensive) and gonadectomy as the two factors. The mean-square error within groups was then used in a Newman-Keuls multiple comparison test to evaluate differences between groups (32). To evaluate the effect of food restriction and gonadectomy on myocardial catecholamines, the three groups, control, gonadectomized, and food-restricted ovariectomized, for each state, sedentary or swimming, were analyzed as individual three-group experiments by the above statistical procedure. Unless otherwise stated significance is reported at the 0.05 level. RESULTS
Inspection of all gonadectomized, ovariectomized and given estrogen, and gonadectomized and given testosterone animals when killed revealed no detectable gonadal tissue and in females absence of a developed uterus. In addition, serum estrogen or testosterone levels were measured in randomly selected gonadectomized animals. Gonadectomy reduced serum estrogen in females from 12.4 t 4.2 (SE) pg/ml to
