The importance of biological sex and estrogen in rodent models of cardiovascular health and disease

Abstract

Nearly one-third of deaths in the United States are caused by cardiovascular disease (CVD) each year. In the past, CVD was thought to mainly affect men leading to the exclusion of women and female animals from clinical studies and preclinical research. In light of sexual dimorphisms in CVD, a need exists to examine baseline cardiac differences in humans and the animals used to model CVD. In humans, sex differences are apparent at every level of cardiovascular physiology from action potential duration and mitochondrial energetics to cardiac myocyte and whole heart contractile function. Biological sex is an important modifier of the development of CVD with younger women generally being protected, but this cardioprotection is lost later in life, suggesting a role for estrogen. While endogenous estrogen is most likely a mediator of the observed functional differences in both health and disease, the signaling mechanisms involved are complex and are not yet fully understood. To investigate how sex modulates CVD development, animal models are essential tools and should be useful in the development of therapeutics. This review will focus on describing the cardiovascular sexual dimorphisms that exist both physiologically as well as in common animal models of CVD.

Introduction

Premenopausal women experience lower rates of cardiovascular disease (CVD) compared to age-matched men; a benefit thought to be mediated, at least in part, by the female sex hormone estrogen.^{1, 2} Consistent with that notion, despite a lower CVD in younger women, the rate of CVD development and mortality in women after menopause exceeds that of men.^{2, 3} These findings underscore the critical need to understand both baseline differences in cardiovascular function and responses to pathological cardiac insults between the sexes.⁴ Indeed, major funding agencies in the US, Canada, and Europe have emphasized the inclusion of both sexes; however, women and female animals remain vastly underrepresented in all stages of CVD research.^5–8 Until only recently, consideration of both sexes was not required in clinical and preclinical studies that focus on CVD.^{9, 10} Characterization of baseline differences between the sexes is required to appropriately assess the utility of animal models of CVD in understanding mechanisms responsible for CVD and its treatment in women and men. In light of the need for a comprehensive understanding of sex differences in both cardiac health and pathology, we will review molecular and functional differences between healthy men and women and will relate these findings to healthy rodents used in CVD research. Although both male and female sex hormones modulate cardiac function, we focus on the genomic effects of estrogen as the major mediator of sex differences. Finally, we present molecular and functional characteristics of animal models of CVD, emphasizing sex differences in their phenotypes.

Sexual dimorphisms in human and rodent cardiovascular physiology at baseline

Although baseline characteristics of cardiac function in healthy men and women differ in terms of heart rate, left ventricular ejection fraction (LVEF) and stroke volume (SV), cardiac functional advantages in healthy men compared to women have been debated in the literature for many years.^{11, 12} Higher LVEF and SV in men develop after adolescence, suggesting a role for sex hormones in the regulation of cardiac function.^{13, 14} The mechanisms mediating these differences in the healthy heart have not been fully elucidated. The need for experimental manipulation of potential modulators requires the use of animal models to study on molecular, cellular, and systemic levels, the complicated interactions among sex, cardiac myocytes, and the cardiovascular system. As in humans, sex differences in baseline cardiovascular function are observed in many experimental rodents. In light of the dramatic effects sex has on cardiovascular function in humans, the National Institutes of Health has called for the inclusion of both male and female animals in preclinical CVD research.⁹

The question of whether cardiovascular function/dysfunction in animals translates to human cardiovascular physiology has been a challenge in research for nearly a century. Basic cellular and functional differences in some experimental models are similar to those of humans, making the use of animals to model human disease possible. For example, the development of the mouse heart remarkably recapitulates that of the human heart.¹⁵ Notably, the role of sex hormones, particularly estrogen, in cardiovascular function is also similar in animals compared to humans.^{16, 17} Use of animal models has therefore allowed elucidation of specific effects of sex hormones on cardiac function that include direct and indirect modulation of contractility, ion channel expression and function, reactive oxygen species (ROS) production, and substrate utilization, among others.¹⁸

Cardiac contractility and ion channels

Consistent with cardiac functional differences observed in humans, cardiac myocytes isolated from male rodents contract more strongly and rapidly than female cardiac myocytes, a difference that is reduced with age.^19–21 Additionally, relaxation rates differ between the sexes. Although conflicting results have been reported, when differences are observed, it is the female cardiac myocyte that relaxes more slowly.²¹ Sex differences in cardiac ion channel expression and function have been a focus of many studies and are implicated in the cellular basis for these differences in cardiac contractility. For example, the amplitude and frequency of calcium sparks produced by release of calcium from the SR is higher in male rat cardiac myocytes.²² Detailed electrophysiological measurements of action potentials in individual cardiac myocytes have revealed specific roles of many different types of ion channels that may mediate differences in conductivity and contraction in the hearts of men and women, as well as animal models of CVD.

Expression and activity of sodium, calcium, and potassium channels dramatically affect the contractility of the heart through modulation of components of the action potential (Figure 1); the sequence and duration of ionic movement in the cardiac myocyte dictates the strength and frequency of contraction. When a sufficiently depolarizing current reaches the cardiac myocyte, sodium channels open to rapidly depolarize the cell, which promotes opening of voltage-gated L-type calcium channels.²³ Increased calcium concentrations in the cardiac myocyte prolong repolarization and refractory durations and promote contraction through the interaction between calcium and troponin. The plateau phase, during which contraction takes place, is achieved through continued slow inward movement of calcium through L-type calcium channels and flow of potassium out of the cell through slow delayed rectifier channels that begin repolarizing the membrane potential.²⁴ The cardiac myocyte returns to the resting membrane potential by removal of calcium from the sarcoplasm and continued flow of potassium out of the cell through potassium channels. Importantly, this plateau and slow repolarization produces a refractory period in which the cell cannot be depolarized, thus preventing tetanus.²⁵

Figure 1.

Cardiac action potential waveform in men (blue) and women (red). I_Na, sodium; I_CaL, L-type slow inward calcium current; I_Ks, slow delayed potassium rectifier; I_kr, rapid delayed potassium rectifier. Adapted from Zipes DP, et al., 2013.²⁰²

Each of these currents, particularly those mediated by calcium and potassium channels, exhibits significant sexual dimorphisms. Female hearts, for example, exhibit a longer repolarization phase mediated by potassium channels than males (Figure 1), a characteristic that develops shortly after puberty and leaves women more prone to cardiac arrhythmias than men.^{26, 27} Interestingly, tissue samples from healthy hearts of female donors reveal lower levels of potassium channel proteins than male donor hearts including the Kv1.4, Kv channel-interacting protein 2 (KChIP2), sulfonylurea receptor 2 (SUR2) and human ether-a-go-go (hERG) subunits that are responsible for cardiac myocyte repolarization;²⁸ lower levels of proteins that contribute to I_Kr and I_Ks prolongs the QT interval.²⁹ In women, longer repolarization during normal sinus rhythm is likely attributable to a higher contribution of slow rectifying potassium channels, based on the morphology of the action potential (reviewed in ²⁴). Similar to humans, action potential duration in female rodents is longer than in males caused by a longer repolarization segment. However, mice and rats do not exhibit a plateau phase and rapidly repolarize without the delayed rectifiers I_Kr and I_Ks observed in human hearts (Figure 2A,B).^{30, 31}

Figure 2.

Action potential waveform in (A) adult human and (B) mouse ventricular cardiac myocytes. Currents (I) contributing to each component of the action potential are shown below. I_Na, sodium current; I_CaL, L-type slow inward calcium current; I_tof, fast transient outward potassium current; Itos, slow transient outward potassium current; IKs, slow delayed potassium rectifier; IKr, rapid delayed potassium rectifier. Adapted from Nerbonne, J 2004.³⁰

Movement of calcium into the cell during depolarization of the cardiac myocyte activates release of calcium from the sarcoplasmic reticulum (SR) in both humans and rodents, thus producing higher intracellular calcium and initiating contraction. Although this process of electrochemical signaling through the heart is recognized as the process by which mammalian hearts contract, excitation-contraction coupling and the mechanisms that mediate it vary among species and between the sexes. Expression of several types of pumps and channels mediate movement of calcium including the sodium-calcium exchanger (NCX), SR calcium-ATPase, sarcolemmal calcium-ATPase, and a calcium uniporter expressed on the membranes of mitochondria. SR calcium-ATPase pumps are responsible for approximately 70% of calcium removal to produce relaxation in humans.³² By contrast, in rats and mice, more than 90% of calcium removal is mediated by the SR calcium-ATPase due to higher expression of the transporter.³³ Interestingly, in female rat hearts, diastolic concentrations of calcium inside the SR and the rate of reuptake after contraction is lower than in males, whereas in mice, sex differences are not observed.²¹ Other studies in rats have revealed weaker and slower contractions in isolated female cardiac myocytes that could be explained by sex differences in calcium stored in the SR rather than differences in influx mediated by NCX or altered myofilament calcium sensitivity.²¹ Despite evidence in rodents supporting a role for SR-mediating calcium flux as a mechanism responsible for differences in contractility, these studies have not been fully supported by human data.

Significant differences also arise between human and animal depolarizing currents mediated by calcium and sodium channels. In female dogs, for example, the density of depolarizing calcium currents mediated by L-type calcium channels is higher than in males, consistent with a more rapid depolarizing current that causes more efficient excitation-contraction coupling in female hearts.³⁴ Indeed, estrogen-receptor deficient male mice also exhibit higher expression of L-type calcium channels.³⁵ Although Cav1.2, which mediates calcium entry into the cardiac myocyte and plays an important role in depolarization, is reduced by estrogen, the NCX, which is involved in repolarization is increased by estrogen; ovariectomy (OVX) reversed this expression pattern in females rats,³⁶ which is significant in light of prolonged repolarization in women. NCX mediates calcium influx and maintains the action potential plateau.³⁷ In rats, calcium concentrations in the cytosol are predominantly mediated by the sarcoplasmic reticulum rather than by NCX, as in humans.³¹ Currents mediated by L-type calcium channels are rapidly inhibited by exposure to estrogen or testosterone in rat and guinea pig cardiac myocytes.^38–40 Opposing effects on ion currents by steroid hormones is emphasized by the decrease in potassium currents by estrogen and increase in potassium currents by testosterone could account for longer repolarization durations in the hearts of women.^38–40 Despite striking differences between the action potential segment durations and heart rates, mechanisms mediating the ionic currents are conserved and sex differences observed in humans are also present in rodents. Thus, if one were interested in mechanisms mediating sex differences, rodent models are clinically relevant.

Molecular biology of the heart and the cardiac myocyte

Sex differences in cardiac gene expression are present in both humans and rodents. In humans (< 40 years old) and mice (2 months old), expression of a similar number of cardiac genes is different between males and females, many of which are expressed on sex chromosomes, a mechanism requiring more study.^{41, 42} Cardiac genes expressed on autosomal chromosomes also differ between the sexes, and in mice, these differences appear to be independent of the estrous cycle.⁴¹ Notably, several GeneOntology categories that are different between the sexes are similar when humans and mice are compared including genes mediating chemotaxis and inflammation. However, in humans, these categories were overrepresented in males, while in mice, enriched genes were primarily observed in the hearts of females.⁴¹

Sexual dimorphisms that are species-specific are also observed in expression of genes encoding contractile proteins. In the human heart, cardiac myosin expression is dominated by β-myosin heavy chain (βMyHC), with greater than 90% of total MyHC composed of βMyHC, and is higher in the atria and ventricles of healthy women compared to men.^{43, 44} By contrast, in mice and rats, expression of αMyHC dominates the ventricles, while in the hearts of larger animal models with slower heart rates like dogs and rabbits, MyHC isoform distribution more closely resembles that of humans.^{31, 45} On average, the healthy murine heart is composed of more than 95% αMyHC.⁴⁶ However, as in humans, rodent hearts exhibit sex differences in MyHC isoforms; female rat hearts, for example, express higher αMyHC and lower βMyHC compared to males.⁴⁷ In fact, the absence of estrogen in females reduces the levels of αMyHC,⁴⁸ again supporting a molecular advantage in premenopausal females. Conversely, removal of testosterone through gonadectomy in rats reduces expression of βMyHC.⁴⁹ αMyHC uses ATP more efficiently than βMyHC, and despite proportional differences in the isoforms between humans and rodents, increases in βMyHC are associated with reduced cardiac function in both.⁴⁹ Expression of several other genes is higher in the hearts of female rats compared to males including skeletal actin, connexin 43, phospholamban, collagens, and transforming growth factor (TGF)-β.^{28, 29, 50} In fact, several of these genes have been shown to be regulated by the presence of estrogen.^{51, 52} Despite dramatic differences in heart rate and ATP use, MyHC isoform ratios change in response to pathological stimuli in a manner similar to humans, therefore making smaller animals appropriate for use as CVD models.

Mitochondrial bioenergetics

Alterations of cardiac bioenergetics are indicators of physiological or pathological responses to work load. Under normal conditions, the heart utilizes fatty acids as an energy substrate. However, when pathology is introduced, a switch in glucose utilization occurs, to an anaerobic process that allows ATP production even under conditions of low oxygen like ischemia⁵³. Interestingly, substrate utilization in the absence of increased workload differs between men and women. In a small study of men and women, cardiac utilization of glucose was significantly higher in males, suggesting that substrate preference may play a role in women being affected more negatively by ischemic disease like myocardial infarction.⁵⁴

Baseline mitochondrial function in many female animal models is consistent with cardioprotection mediated by healthy cardiac metabolism. Older female rat hearts have higher oxidative phosphorylation (OXPHOS) capacity compared to age-matched males, whereas this capacity is nearly equal in younger rats of either sex.^{55, 56} Additionally, lower production of ROS is observed in young female rats, which is promoted by higher level of aldehyde dehydrogenase expression and activity and reduced production of ROS by α-ketoglutarate dehydrogenase, supporting observations in women, suggesting that the hearts of women suffer less oxidative damage over a lifetime.^{57, 58} Young adult female Fischer 344 rat hearts also exhibit higher expression of nuclear genes with a role in β-oxidation of fatty acids compared to the hearts of young males,⁵⁶ similar to humans where the hearts of premenopausal women favor fatty acid substrates over glucose compared to age-matched men. In fact, estrogen decreases the expression of glucose transporter type 4 via increased expression of nitric oxide synthase in both humans and rats,^{54, 59} thereby forcing the utilization of fatty acids under normoxic conditions. Much remains to be learned about how sex influences mitochondrial number and function in the heart; interactions among approximately 1,500 genes contribute to normal function.⁶⁰ At a basic level, however, rodents exhibit similar cardiac metabolic characteristics to humans.

Hormones implicated in sexually dimorphic cardiac function

In humans, it has long been thought that the female sex hormone estrogen and signaling through estrogen receptors (ERs) expressed in the vasculature and the heart are primarily responsible for the cardiac protection experienced by pre-menopausal women compared to age-matched men.⁶¹ Estradiol is synthesized through the aromatization of testosterone and interacts with two main receptors that are localized in the nucleus, cytoplasm, at the plasma membrane, and on mitochondria: ERα and ERβ. ERs are primarily localized in the nucleus where they bind to estrogen response elements (EREs) or to other transcription factors and regulate transcription of genes mediating cell growth, contractility, apoptosis, and energy substrate utilization.^{18, 62} A third protein, G protein-coupled receptor (GPR30 or GPER), has been implicated in mediating rapid, non-genomic estrogen signaling independently of the canonical ERs.^63–65 However, studies suggest that GPER plays a role in estrogen signaling by regulating the expression of an extranuclear ERα isoform (ERα36), not by GPER itself binding to estrogen.⁶³ Due to multiple conflicting findings regarding GPER’s ER status, this review will focus on more classical estrogen signaling pathways mediated through either ERα or ERβ.

An abundance of studies in pre- and post-menopausal women including those receiving hormone replacement therapy (HRT) implicate estrogen in altering cardiovascular function. The Women’s Health Initiative clinical trial suggested that HRT initiation in postmenopausal women is associated with increased CVD; the trial was halted early (at 5.2 years) due to concerns about adverse events in the heart, pulmonary emboli, and in cancer. Additionally, results from the Heart and Estrogen/Progestin Replacement Study revealed that HRT in post-menopausal women with existing CVD did not improve, and again, a significant trend toward worse cardiac outcomes was apparent. However, most of these women had been post-menopausal for significant periods of time and with additional health conditions, complicating interpretation of the results. ⁶⁶ Recently, a randomized study involving 727 women within three years of their last menses revealed that earlier hormone replacement improves some aspects of CVD risk, and participants receiving HRT did not differ in their rates of atherosclerosis progression.⁶⁷ Other steroid hormones like progesterone and testosterone, that are expressed in both men and women and are also implicated in modulating cardiac function and disease phenotype, exhibit cyclic concentrations in serum, and decrease with age.⁶⁸

The attention given to HRT initiated interest in the potential therapeutic abilities of products that contain phytoestrogens, such as soy, that have the ability to initiate estrogenic signaling by binding ERα or ERβ.⁶⁹ However, after several clinical trials with postmenopausal women, the cardiovascular benefits of soy supplementation remained controversial and the American Heart Association reversed its endorsement of soy products for this purpose.^{69, 70} Additionally, phytoestrogens present in standard laboratory rodent chow, such as genistein, directly alter contractility and inhibit tyrosine kinase activation in cardiac myocytes, which should be considered when performing animal experiments. ^71–73

Clinical studies that identified sex differences in CVD where the findings are abrogated in postmenopausal women have motivated a large number of studies that manipulate levels of sex hormones, particularly estrogen, in experimental animals. Comparisons between women and female experimental animals are challenging due to significant differences in estrous cycle duration and serum estrogen levels. In mice, for example, the estrous cycle varies from 2–7 days, and decreased cycling is observed in older females (>13 months old).⁷⁴ Female rodents do not go through a significant decline in estrogen that resembles human menopause, and there are no changes in estrogen levels between young C57Bl6 mice and older (14 month old), acyclic mice.⁷⁵ However, a major caveat to these studies is the notorious difficulty with which low serum estrogen levels can be measured in rodents.⁷⁶ Removal of ovaries from mature female animals may reproduce systemic conditions in postmenopausal women. However, this strategy in mice leads to substantial weight gain beyond that which is observed in postmenopausal women. Additionally, the majority of women do not undergo menopause surgically, making ovariectomy animal models of limited use to effectively study perimenopause, which is experienced by most women and is characterized by a gradual change in ovarian hormones.⁷⁷ Therefore, these differences should be taken into consideration in studies that utilize systemic manipulations of sex hormone levels in animals to identify mechanisms of sex differences in CVD models.

Intracellular signaling in the cardiac myocyte of males and females

Data in our lab suggest a sexually dimorphic role for estrogen in the modulation of signaling activity in cardiac myocytes. Activation of receptor tyrosine kinases, for example, has been implicated in several important cardiac functions including cardiac myocyte hypertrophy, contractility, and vascular growth.^78–80 Adult rat ventricular myocytes (ARVMs) isolated from male and female rats were treated with physiological doses of estradiol. Alterations in the phosphorylation of 39 receptor tyrosine kinases (RTKs) and 46 intracellular signaling molecules were measured by comparing lysates of estrogen- to vehicle-treated ARVMs. A consistent observation among experiments is basal levels of RTK phosphorylation are significantly higher (approximately 4.5-fold) in untreated ARVMs isolated from females compared to males (n = 2–3 separate isolations per sex)⁸¹ (Figure 3A). In response to estradiol treatment, phosphorylation of many RTKs was reduced in female compared to male cells (Figure 3B). When considered collectively, RTK phosphorylation with estradiol treatment in female ARVMs was reduced by 61.8% but increased 55.9% in male cells, indicating that estrogen exhibits opposing effects on RTK phosphorylation in male and female cardiac myocytes. Interestingly, addition of estradiol to female ARVMs normalized the difference between activation of intracellular kinases between male and female vehicle-treated cells; 3.9-fold higher activation in female cells was reduced to less than 5% with the addition of estradiol. The addition of physiological doses of estradiol did not affect mRNA levels of seven of the most highly inhibited RTKs (data not shown), suggesting that transcriptional or post-transcriptional regulation by estradiol does not significantly impact the data. These data are supported by inhibition of growth factor signaling by estrogenic compounds that signal through ERs and require further investigation, but suggest a possible novel mechanism by which estrogen modulates signaling in cardiac myocytes.⁸²

Figure 3.

Phosphorylation of RTKs in ARVMs isolated from males and females. A. Baseline phosphorylation levels of RTKs in female and male ARVMs. B. Phosphorylation levels of RTKs in female and male ARVMs treated with 300nmol/L estradiol, relative to vehicle treated. Red bars, female; blue bars, male. A: n = 2 pooled preparations, 2–3 rats per preparation. B: n = 3 pooled preparations, 1 rat per preparation.

Animal models of cardiovascular disease

I. Hypertension

More than 95% of hypertension cases are classified as essential hypertension without a single cause.⁸³ Sex differences in blood pressure, like many cardiovascular features, originate during adolescence with persistently higher systolic and diastolic pressures observed in males.^{84, 85} Additionally, sex differences are also observed in pulmonary hypertension which progressively leads to right heart failure with women being at greater risk than men.⁸⁶ The review by Mair et al (2014)⁸⁶ provides an in depth account for these differences as it is beyond the scope of this article to discuss pulmonary hypertension. The key organ for long-term control of blood pressure and body fluid volume is the kidney.^{87, 88} According to the renal-body fluid feedback concept, chronic increases in arterial pressure occur as the result of abnormalities in the relationship between renal perfusion pressure and sodium excretion. That is, in order for long-term increases in arterial pressure to occur, a reduction in the kidneys’ capacity to excrete sodium and water must be present. A common defect that has been found in all forms of hypertension examined to date, including genetic and experimental animal models and human essential hypertension, is a rightward shift (toward increased blood pressure) in the chronic renal pressure-natriuresis relationship. The renin-angiotensin system (RAS) and other hormones including sex steroid hormones modulate baseline sexual dimorphisms in blood pressure.^{89, 90} In men, plasma renin activity is approximately 27% higher than women, regardless of age, but renin activity increases in postmenopausal women, an effect that may be mediated by lower estrogen or influenced by increases in testosterone.⁹¹ Additionally, dietary intake of salt can dramatically alter blood pressure, especially in African Americans. It is now recognized that blood pressure is salt-sensitive in up to 30–50% of hypertensive individuals.⁹⁰ Interestingly, blood pressure in premenopausal women is relatively unaffected by sodium intake, but blood pressure becomes more salt-sensitive with the onset of menopause in many women.⁹² The precise mechanisms of sexual dimorphisms in hypertension have been extensively examined in several different rodent models that have revealed novel pathways for the development of hypertension (Table 1).

Table 1.

Overview of sex differences in animal models of hypertension

Model	Age	Treatment	Reported in Males	Reported in Females	Molecules/Pathways
SHR 96, 102	12–20 weeks	NA	Higher systolic and mean arterial blood pressure		Testosterone; RAS signaling
SHR 98	9–12 weeks	NA	Isolated cardiac myocytes displayed reduced diastolic and systolic sarcomere dynamics		Not Addressed
SHR 97	3–30 months	NA	Progressed to failure by 15 months	Less hypertensive until 18 months; better mortality rates	Not Addressed
SHR 103	16 months	Losartan (30mg/kg/day): 3 weeks	Displayed an enhanced depressor response after angiotensin receptor antagonism		RAS
DSS 106	4–5 weeks	Low salt (0.3%) or high salt (8%) diet: 3 weeks	Higher blood pressure and increased mortality in response to high salt diet	Lower basal systolic blood pressure	Vasodilatory prostaglandins which were higher in females
DSS 108	17 weeks	Low salt (0.28%) or high salt (8%) diet: 4 weeks	Expressed higher levels of renal angiotensinogen mRNA and protein in response to high salt diet; this response was attenuated by castration		Testosterone
DSS 100	14 weeks	Low salt (0.4%) or high salt (4%) diet: 2 weeks		OVX rats displayed increased blood pressure at baseline and in response to high salt diet	Estrogen potentially regulating NO
DSS 107	8–9 weeks	Varying salt diets: 0.15% (7days), 1% (14 days), 4% (14 days) 8% (14 days), 0.15% (14 days)	Salt dependent hypertension was greater	OVX rats developed hypertension similar to males, but their blood pressure did not normalize upon return to normal diet	Estrogen
L-NAME treatment - rats 110	13–14 weeks	L-NAME (75mg/100mL drinking water): 5 weeks	Blood pressure in males was higher after treatment; this sex difference was abolished upon castration of males	OVX had no effect	Testosterone
L-NAME treatment - rats 111	12 weeks	L-NAME (50mg/100mL drinking water): 4 weeks	No difference	No difference	NA
L-NAME treatment - rats 112	7 weeks	10 weeks L-NAME (20mg/100mL drinking water); withdrawal observed for 7 weeks		Developed more severe and rapid hypertension; took longer to respond to withdrawal of treatment	Not Addressed

Spontaneously hypertensive rat model (SHR)

In 1963, a male rat with essential hypertension that was 25 mmHg higher than normal was identified and bred to produce a line of spontaneously hypertensive rats.⁹³ While this is a genetic model of hypertension, the underlying genetic variations responsible for disease development are extremely complex as quantitative trait locus mapping experiments have identified multiple genes that may be associated essential hypertension.^{94, 95} As male and female SHR animals mature, both sexes exhibit increased systolic blood pressure; however, by 12 weeks of age this increase is significantly higher in the males compared to females, similar to humans.⁹⁶ This sexual dimorphism persists through adulthood; male SHR are more hypertensive than females until after females stop estrus cycling (10–12 months of age), when, by 16 months of age, blood pressure is higher in females than males.⁹⁷ Male SHRs also develop signs of heart failure (HF) by 24 months, but female animals do not develop the ventricular stiffness or dilation that is observed in the males.⁹⁷ Sexual dimorphisms are also observed at the level of the cardiac myocyte as SHR left ventricular myocyte diastolic and systolic sarcomere dynamics were reduced compared to normotensive controls; this observation was more pronounced in male SHR myocytes.⁹⁸ The progress of disease and sexual dimorphisms are similar to those observed in men and women, making this model particularly useful.⁹⁹

Sex hormones appear to play an important role in mediating these observed sex differences. When young male and female SHRs were castrated or OVX to deplete endogenous sex hormones, castrated male SHRs exhibited reduced blood pressure similar to that of females.⁹⁶ In this study, OVX alone did not affect blood pressure. However, supplementing OVX females with testosterone increased blood pressure by 10 percent, suggesting that androgens may mediate the observed sex difference between SHRs,⁹⁶ and that female sex hormones are not protective for hypertension in the female SHR.^{100, 101} Interestingly, sex differences are also influenced by the RAS. When male and female SHR, both intact and gonadectomized, were treated with an angiotensin converting enzyme (ACE) inhibitor, sex differences were abrogated as reduced blood pressure was most significant in males and OVX females supplemented with testosterone after treatment. ¹⁰² The RAS also mediates hypertension in aged SHRs; treatment of 16 month male and female SHRs with the angiotensin receptor antagonist, losartan, decreased blood pressure in both sexes.¹⁰³ However, the decreases were greater in male SHRs, suggesting that RAS may be more important in aging male SHRs than females for maintaining blood pressure.

Dahl Salt-Sensitive (DSS) Rat Model

Increases in dietary sodium can lead to dramatic increases in blood pressure. Men that have developed salt-sensitive hypertension are at greater risk of early death than women, even though women are more salt-sensitive than men in terms of blood pressure.¹⁰⁴ The DSS rat, a genetic model of hypertension induced by feeding the animals a high-sodium diet, demonstrates sex differences similar to those observed in men and women.¹⁰⁵ Prior to the addition of a high salt diet, female salt-sensitive rats have significantly lower basal systolic blood pressure compared to males.¹⁰⁶ Female DSS rats have also been shown to display significantly lower systolic blood pressure than males after being fed a high salt diet for three weeks.¹⁰⁶ In this same study, hypertension induced by four weeks of a high salt diet in both male and female DSS rats resulted in 50% mortality in males only.^{104, 106}

Unlike the SHR model, female sex hormones contribute to protection against hypertension, as OVX of female DSS rats increased basal blood pressure compared to intact females.¹⁰⁰ In addition, OVX female DSS rats fed a high salt diet developed hypertension in a manner that was not significantly different from males.¹⁰⁷ However, when dietary sodium was decreased to normal levels, blood pressure in the male and intact female DSS decreased, whereas in the OVX female DSS animals, it remained unchanged, suggesting that removing the female sex hormones predisposes the DSS female rats to develop hypertension, independent of sodium intake.¹⁰⁷ Testosterone also appears to have a role this process as castration of male DSS rats fed a high salt diet attenuated the development of both hypertension and the increased expression of renal angiotensinogen.¹⁰⁸ In addition, castrated DSS male rats fed a high salt diet supplemented with testosterone, had elevated blood pressure and increased renal injury.¹⁰⁸ Future studies are necessary to completely understand the role of male and female hormones during development of salt sensitive hypertension.

N^ω-nitro-L-arginine methyl ester (L-NAME) model of hypertension

Nitric oxide (NO) exerts complex actions on the cardiovascular system through the regulation of vascular tone and renal function. In humans, inflammatory molecules cause NO synthase (NOS) II (inducible or iNOS) to produce NO, which has significant hypotensive effects. However, it is NOS III (endothelial or eNOS) that is most important in control of blood pressure. Indeed, treatment of rats with a non-selective NOS inhibitor, L-NAME, induces a NO-deficient animal model of hypertension.¹⁰⁹ Within five weeks of treatment, intact male rats developed a greater amount of hypertension compared to females.¹¹⁰ In addition, castration of males attenuated the development of hypertension, whereas OVX of L-NAME-treated females did not affect blood pressure.¹¹⁰ These results suggest that estrogen is not mediating the protective effect observed in females with L-NAME. While these results appear similar to other models of hypertension discussed earlier, there are contradictory reports regarding the sexually dimorphic response to L-NAME treatment. Other groups have observed that female animals actually develop more hypertension than males or report that there is no difference in blood pressure in response to NOS inhibition between the sexes.^{111, 112} These conflicting data could be due to the type of L-NAME treatment as each of these studies used a different dose and length of treatment to induce hypertension. Even with these discrepancies, sex differences with respect to hypertension development after L-NAME treatment are apparent in other rat models, such as SHR animals and normotensive Sprague Dawley rats. While blood pressure in SHR males is greater than females at baseline, females exhibited a greater increase in blood pressure after L-NAME treatment, ¹¹³ suggesting that female SHRs are more sensitive to NOS inhibition than males. However, these data are consistent with studies showing that estradiol increases NOS III and NOS I synthesis. ¹¹⁴

The mechanisms responsible for hypertension are a multifactorial with a combination of genetic and environmental influences; therefore, no one animal model will completely mimic human disease development. While the SHR model is a commonly used animal model of essential hypertension, the DSS rat is a model of salt sensitive hypertension observed in humans. However, SHRs are not prone to strokes or vascular thrombosis ¹¹⁵, and high salt diets cause significant renal injury and mortality within 4–6 weeks in DSS rats. Rats also do not develop signs of atherosclerosis. However, all of the discussed rodent models demonstrate increases in blood pressure in a relatively rapid and reproducible manner, providing reliable experimental systems for hypertension induced by independent mechanisms. Overall, these animal models mirror sex differences observed in humans with females developing less hypertension than males, aging confounding the hypertension in females, and estrogen signaling playing a critical role in mediating certain aspects of this protection.

II. Myocardial infarction (MI) and ischemic injury

Each year, over 600,000 Americans will experience a new MI event and nearly 40% of these cases will progress to HF.¹¹⁶ Sex differences with respect to MI and HF are observed, with women seemingly protected in that they develop the disease later in life compared to men.^{117, 118} However, men and women often present different disease symptoms, and young women hospitalized for acute MI actually have worse outcomes than their male counterparts.^{118, 119} Additionally, after menopause, the prognosis for women with MI is significantly worse than for age-matched men.¹²⁰ Cardiac remodeling in response to ischemic injury exhibits hallmarks of cardiac myocyte death, inflammatory cell infiltrations, and the development of fibrosis within the injured area in both men and women.¹²¹ Remodeling also occurs in the surrounding healthy tissue including cardiac myocyte hypertrophy and altered ion channel expression that causes arrhythmias such that, in humans, the degree of remodeling negatively correlates with mortality.¹²² Additionally, trends of decreased sarcoplasmic reticulum calcium ATPase, phospholamban, and ryanodine receptors are observed, indicating reduced calcium transients and decreased contractility in human HF.¹²³ Similar to humans, animal models also display cardiac remodeling and sexually dimorphic characteristics with respect to ischemic injury development, severity, and response to reperfusion (Table 2).

Table 2.

Sex differences in animal models of myocardial infarction

Model	Age	MI Stage/Ischemia Protocol	Reported in Males	Reported in Females	Molecules/Pathways
Coronary Artery Ligation -C57BL/6J mice 125	12 weeks	Acute: 1,2,4,7 or 14 days after MI Chronic: 12 weeks after MI	Acute: Mortality was higher; Chronic: displayed worse cardiac function and more dilation		Not Addressed
Coronary Artery Ligation-C57BL/6J mice 127	12–15 weeks	2, 3 or 60 days post MI	No difference in infarct size	Exhibited better survival and were less likely to progress to dilation	Greater induction of genes involved in angiogenesis, ECM remodeling and immune response observed in infarcted female hearts
Coronary Artery Ligation -Sprague Dawley rats 131	12 weeks	1 & 6 weeks post MI	Greater increases in thickness of non-infarcted regions and restrictive diastolic filling patterns	No difference in infarct size	Not Addressed
Coronary Artery Ligation -Sprague Dawley rats 132	10–12 weeks	4 weeks post MI	No difference in mortality rates	Females exhibited more pronounced LV dilation than males	Not Addressed
I/R injury -129J & C57BL/6 mice 146, 148	12–24 weeks	Ex vivo: 30 min perfusion, 20 minutes ischemia followed by 40 minutes reperfusion	Injury was greater in Iso or calcium treated hearts	No difference under basal conditions	NOS Signaling
I/R injury - NCX Tg mice & non Tg littermates 147	32–36 weeks	Ex vivo: 30 min perfusion, 20 minutes ischemia followed by 40 minutes reperfusion	Post ischemic function was decreased in Tg males	No difference under basal conditions; Tg sex difference was abolished by OVX	Estrogen
I/R injury - Sprague Dawley rats 150	11–15 weeks	Ex vivo: 30 minutes occlusion followed by 150 reperfusion	Infarct sizes were larger; gonadectomy of both sexes produced opposite results.		Sex hormones, particularly androgens down-regulating apoptosis in response to MI
I/R injury-Sprague Dawley rats 151	10 weeks	In vivo: 30 minutes of ischemia by clamping a coronary artery, 24 hrs reperfusion	Infarct area and percentage of apoptosis was greater		Differential regulation of apoptosis and autophagy pathways by an unknown mechanism
I/R injury - Sprague Dawley rats 154	not specified	Ex vivo: 15 minutes equilibration, 27 minutes ischemia followed by 40 minutes of reperfusion	Increased inflammatory response in response to injury	Post-ischemic cardiac function was significantly improved	Inflammatory signaling mediated by p-38 MAPK activation
Isolated cardiac myocytes - C57BL/6J mice 156	8–10 weeks	Cells treated with 100uM H2O2 for 30 minutes		Treated cells exhibited greater survival, decreased LDH release, apoptosis and necrosis	Akt and caspase signaling

Models of Myocardial injury induced by coronary artery ligation

The complex nature of MI and HF has made developing a single animal model difficult. However, ligation of the left coronary artery in a variety of different rodent species recapitulates much of what is observed in human patients. Previous studies have utilized rodents to investigate sex differences observed both in the initial response to MI as well as the development of HF. For example, the rate of cardiac rupture and mortality within the first week after MI was greater in male mice compared to females, regardless of infarct size.^124–127 In addition, immediately following MI, the hearts of male mice have increased neutrophil infiltration, damage to the interstitial collagen network, and matrix metalloproteinase (MMP) activity compared to their female counterparts.^{124, 126} Furthermore, despite infarct size being equal at intermediate time points, during the chronic phase twelve weeks after MI, males displayed worse cardiac function, more cardiac myocyte hypertrophy, and increased ventricular dilation compared to females.^{125, 128} However, female mice were able to maintain contractile function over time, whereas males displayed progressive declines in contractile function associated with maladaptive cardiac remodeling.¹²⁸ The cardiac phenotypes of mice with experimentally-induced MI are consistent with humans with MI in that the hearts of women with MI exhibit lower rates of myocardial cell death and progress to HF more slowly than men.¹²⁹ Additionally, as in humans, alterations in potassium and calcium channel expression and currents that prolong the repolarization segment led to arrhythmogenicity in post-MI rats, and estrogen reduces post-MI arrhythmias associated with these ionic changes in mice.^{122, 130}

Rats with experimental MI show variable and contrasting results compared to what has been reported in mice. In response to MI, male and female rats developed similar size infarcts and did not exhibit differences in contractile function six weeks after injury.¹³¹ However, six weeks after MI, males displayed restriction of left ventricular filling, as well as greater increase in LV posterior wall thickness and myocyte diameter compared to female animals.¹³¹ In contrast, Chen et al., demonstrated that four weeks after MI, female rats displayed more dilation than males.¹³² Analysis of isolated cardiac myocytes from infarcted rat hearts also exhibited no differences in morphometrics in response to MI between sexes.¹³² In a similar study that analyzed scar composition four weeks post-MI, males developed larger MI scars than females, but the overall structural composition of the scars was not different between sexes.¹³³ Aged male and female rats displayed similar patterns of overall LV remodeling, but differences were apparent in regional cardiomyocyte hypertrophy and arteriole expansion four weeks post-MI.¹³⁴ Together, these studies demonstrate that sex differences do exist in rodents with experimental MI, but inconsistencies are observed between mice and rats.

The underlying causes of sex differences in MI and HF animal models have been examined at the transcriptome level. In a microarray study analyzing cardiac gene expression changes three days after MI, female mice displayed increased induction of genes involved in angiogenesis, immune response, and extracellular matrix remodeling compared to males.¹²⁷ In addition, female animals had a decreased amount of pathologic cardiac remodeling.¹²⁷ While this result may provide a general explanation for sex differences in cardiac remodeling observed after MI in men and women, the mechanism is undoubtedly multifactorial.

ER signaling appears to play a role in mediating post-MI responses. Female ERβ knockout (βERKO) mice experiencing chronic HF after MI exhibited increased mortality and altered expression of calcium handling proteins, suggesting that ERβ plays a critical role during HF.¹³⁵ Additionally, estrogen treatment of OVX ERα knockout (αERKO) mice resulted in smaller infarcts.¹³⁶ This was not observed in wild-type controls, and the opposite was observed in βERKO animals, further implicating that ERβ is important for mediating estrogen’s effects in the heart post-MI.¹³⁶ These results are complicated by the observation that estrogen treatment increased post-MI mortality in wild-type controls and αERKO animals, suggesting that estrogen signaling is not universally protective.¹³⁶ While reports using ERKO mice suggest these receptors have important and distinct cardiac roles, these studies are confounded by the systemic effects of global ER deletion, as αERKO mice have increased levels of circulating estrogen, are obese, and insulin resistant, and βERKO mice exhibit hypoxia and hypertension.^137–141 Other models, however, provide evidence of the cardioprotective effects of estrogen. For example, estrogen treatment of OVX rats 24 hours after MI resulted in increased expression of connexin 43, which allowed for critical cell-gap junctions to be maintained and reduced fatal ventricular arrhythmias.¹⁴² OVX alone in wild type mice worsened LV function and dilation, suggesting a protective role for estrogen.¹⁴³ In contrast, testosterone worsened cardiac function in both intact and OVX wild type females, and castration of wild type males decreased the amount of cardiac ruptures and improved cardiac function.¹⁴³

Histone deacetylases (HDACs) are key modulators of MI by regulating the activity of cardiac transcription factors, such as myocyte enhancer factor 2 (MEF2).¹⁴⁴ While HDACs have become therapeutic targets for cardiac hypertrophy, they appear to have sex specific effects, which should be considered. Female, but not male, HDAC 5 or 9 KO mice were protected against pathological cardiac remodeling after MI in an ERα dependent manner.¹⁴⁴ Thus, active class II HDACs repress ERα expression and appear to promote pathological cardiac remodeling specifically in females. Increased ERα expression due to HDAC inhibition could protect females by regulating expression of genes such as vascular endothelial growth factor (VEGF), thus promoting angiogenesis.¹⁴⁴ These data provide a novel explanation for how cardiac ERα expression is regulated post-MI, and give insight into the potential protective effects of HDAC inhibitors in humans, particularly women.

Ischemia/reperfusion (I/R) studies

Animal models utilizing I/R provide valuable mechanistic information in terms of surgical reconstitution of blood flow to damaged cardiac tissue after MI in humans.¹⁴⁵ Under basal conditions, there was no difference in susceptibility to I/R injury or infarct size between male and female mice.^146–148 Despite this lack of difference at baseline, over-expression of NCX increased I/R injury in male but not female mice, suggesting that females are less prone to injury when calcium homeostasis is perturbed.¹⁴⁶ Exposure to isoproterenol (Iso) or calcium pre-treatment supports this conclusion; I/R injury is increased to a greater extent in male mice compared to females.¹⁴⁶ The hearts of male mice treated with Iso prior to I/R accumulate more intracellular sodium than do female hearts.¹⁴⁹ In contrast, several other studies report that under basal conditions female hearts displayed preserved contractile function, smaller infarct size, and fewer apoptotic cells compared to males after I/R.^150–154 These differences are also apparent at the cellular level as intracellular sodium tended to be higher in isolated cardiac myocytes from male mice, and female cardiac myocytes survived at a higher rate than males when exposed to an oxidative stressor.^{155, 156} The variation in results of I/R injury between sexes could be due to differences in I/R experimental protocol as many of these studies used different lengths of time to induce an ischemic event.

Examination of the mechanisms mediating sexual dimorphisms in response to I/R also support a role for calcium homeostasis and involves differences in NO signaling, specifically S-nitrosylation between sexes.¹⁵⁷ NO production was higher in the hearts of female mice at baseline, and female cardioprotection in I/R injury models after Iso exposure was lost upon treatment with the L-NAME.¹⁴⁶ Furthermore, female hearts lacking NOS I or NOS III exhibited increased contractility and were also not protected from I/R injury.¹⁴⁸ NO may be mediating this effect by altering intracellular sodium levels as L-NAME treatment of hypercontractile hearts blocked differences in intracellular sodium levels, which were higher in males.¹⁴⁹ Non-specific NOS inhibition also increased calcium in female hypercontractile hearts to levels similar to males, a process that is regulated by the S-nitrosylation state of the L-type calcium channel.¹⁴⁸ Thus, sex differences in I/R injury animal models may be attributable to increased intracellular calcium in males.

Survival and apoptotic signaling pathways also play important roles in mediating sex differences during ischemic events. For example, cardiac myocytes isolated from female mice displayed higher levels of phosphorylated Akt both before and after oxidative stress.¹⁵⁶ Caspase 3 activity was also lower in female rat cardiac myocytes treated with an oxidative stressor compared to males, possibly contributing to the greater survival observed in the female cells.¹⁵⁶ Apoptotic signaling differences are also observed in the whole heart. After I/R injury, levels of the anti-apoptotic protein, Bcl-2, were significantly lower, while levels of the pro-apoptotic protein, Bax, were unchanged in males, but not females.¹⁵¹ Furthermore, phospho-p38 levels that promote apoptosis were significantly increased in males after I/R whereas in the hearts of females, increased autophagy was observed.¹⁵¹

In I/R injury studies, sex hormones promote opposing effects in terms of infarct size. OVX of female rats led to larger infarct sizes, but estrogen supplementation attenuated this response.¹⁵⁰ Testosterone, however, aggravated the response to I/R by down-regulating the anti-apoptotic protein Bcl-xL leading to the enhanced cardiac injury observed in males.¹⁵⁰ βERKO mice (but not αERKO mice) displayed less functional recovery if exposed briefly to Iso before I/R injury making them similar to what was observed in wild type males, suggesting that cardioprotection in females is mediated by ERβ signaling.¹⁵⁸ Furthermore, hearts of βERKO mice had altered expression of multiple metabolic genes compared to wild-type and αERKO females, which could explain functional differences in response to I/R injury.¹⁵⁸ The protective effects of estrogen are also observed in a cellular model of ischemia as estrogen treatment led to decreases in intracellular calcium and sodium during metabolic inhibition in male cardiac myocytes, abolishing observed sex differences.¹⁵⁵

The complexity of MI or I/R injury in mouse models discussed here parallels that of humans by producing more severe phenotypes in male mice, but this is less apparent in rats. Additionally, female mouse hearts display less cell death after injury and are less likely to progress to HF, which is in agreement with human studies.^{129, 151} The role of estrogen in mediating the observed protection in females continues to be unclear since estrogen supplementation has conflicting effects on MI outcome, which is also representative of reports in humans.^{66, 136}

III. Cardiac Hypertrophy

Pathological stimuli, such as hypertension, aortic stenosis, or cardiac injury, results in cardiac hypertrophy that may initially compensate for disrupted function; however, prolonged exposure to these pathological stressors leads to decreased cardiac function, increased fibrosis and an increased risk of HF.¹⁵⁹ The phenotypic appearance and development of cardiac hypertrophy are distinct between males and females (Table 3) and are modulated by hormones as well as rodent diets that contain high levels of phytoestrogenic compounds.^{160, 161}

Table 3.

Sexual dimorphisms in animal models of pathological hypertrophy

Model	Age	Hypertrophic Stimulus	Reported in Males	Reported in Females	Molecules/Pathways
Pressure overload -Wistar rats 165	3–4 weeks	Aortic banding: 6 & 20 weeks	By 20 weeks progressed to heart failure	Hypertrophic response is similar after 6 weeks	Not addressed
Pressure overload -Wistar rats 166	Not specified	Aortic banding: 6 weeks	Exhibited increased expression of fetal genes	Extent of hypertrophy was similar between sexes	Not addressed
Pressure overload -C57BL/6J WT & ERβ KO mice 167, 168	8 weeks	Aortic banding: 9 weeks	WT males developed more hypertrophy, fibrosis and heart failure; this difference was abolished upon ERβ KO		Estrogen signaling through ERβ regulating apoptosis and fibrosis
Pressure overload -B6D2/F1 mice 169	5 weeks	Aortic banding: 4 weeks	Exhibited more fibrosis	Extent of hypertrophy was similar between sexes	CaMKII activation regulating MEF2 transcription
Volume overload - Sprague Dawley rats 179	8 weeks	AV fistula: 8 weeks	Increased mortality and development of heart failure		Not addressed
Volume overload - Sprague Dawley rats 180, 181, 183	6 weeks	AV shunt: 4 or 16 weeks	By 16 weeks, exhibited decreased cardiac function which progressed to heart failure	Maintained cardiac function by 16 weeks	Estrogen up regulating phospho-Bcl2 to attenuate apoptosis & β-adrenergic signaling in female hearts
Chemical - C57BL/6J mice 184	4 months	Isoproterenol treatment: 7 days		Developed less hypertrophy	Estrogen broadly regulating the activation of kinase signaling
Chemical - Sprague Dawley rats 186	Weight matched (270–290g)	Acute Isoproterenol treatment (1uM) of isolated myocytes	In response to Iso, cells displayed greater cell shortening, calcium current density and cAMP production.		β-adrenergic signaling differences between sexes
Genetic HCM - mice 194	4 & 10 months	R403Q mutation in alpha MYHC	By 10 months, developed cardiac dilation and dysfunction,	Developed hypertrophy by 4 weeks, maintained function by 10 weeks	Not addressed
Genetic HCM - mice 197	12 weeks	Truncated cTnT or missense R92Q mutation in cTnT	In both mouse models, treatment with Iso or PE resulted in sudden cardiac death in all males	At baseline, R92Q hearts were larger, exhibited decreased hypertrophic gene expression and fibrosis.	Not addressed
Genetic HCM - mice 199	13 weeks	Knock-in model with a heterozygous point mutation in MYBPC3	Isolated cardiac myocytes and myofilaments displayed reduced maximal force generating capacity		Not addressed

Pathological hypertrophy induced by pressure overload

In humans with aortic valve stenosis or hypertension, left ventricular hypertrophy develops to maintain functional cardiac output and up to 50% of those patients will progress to HF.¹⁶² Pressure overload is commonly studied in animal models by banding either the ascending or transverse aorta and is particularly useful since development of cardiac hypertrophy is gradual and progresses to HF.^{163, 164} As with other CVD animal models, sexual dimorphisms are apparent in pressure overload studies with males consistently developing more severe disease symptoms in a variety of experimental settings, consistent with studies in men and women (Figure 4).^165–170 In studies of both mice and rats, male hearts developed eccentric cardiac hypertrophy, while female hearts exhibited concentric hypertrophy, as observed in men and women with aortic stenosis.^{167, 171} The hearts of males exposed to pathological stimuli were more likely to exhibit decreased contractility and increased fibrosis compared to their female counterparts. ^165–170 While female hearts hypertrophied after aortic constriction, sometimes even to a similar degree as the males, cardiac function was preserved over time, and cardiac expression of fetal genes, such as atrial naturietic peptide, β-MyHC, and MMP, was lower compared to males.^{165, 166} In addition, the proteomic response of the heart to pressure overload was very different between the sexes, with increased expression of cytoskeletal proteins in females whereas mitochondrial proteins were more prominently regulated in males, suggesting that male and female animals utilize different mechanisms to respond to the same pathological stimulus.¹⁶⁸

Figure 4.

Male and female rodent hearts adapt differently to pathological stimuli. While left ventricles of both sexes increase in size to response to increases in metabolic demand, male hearts are more likely to develop fibrosis, apoptosis and progress to heart failure. Heart image copyright ©2016 Abcam.

Consistent with clinical studies that demonstrate postmenopausal loss of cardioprotection in women, estradiol supplementation of OVX rats ameliorated cardiac dysfunction as well as hypertrophic development induced by transaortic constriction, suggesting that estrogen provides a cardioprotective role in female animals.¹⁷² Furthermore, in female mice lacking either ERα or ERβ, the hearts of βERKO mice displayed a more pronounced response to aortic banding, whereas αERKO mice were indistinguishable from wild type mice, suggesting that signaling through ERβ may be mediating the protection observed in normal females.^{170, 173} Female βERKO mice also exhibited increased fibrosis relative to wild-type controls in response to pressure overload, and ERβ appears to mediate this effect by regulating the expression of repressors of the MAPK-ERK1/2 pathway.^{167, 174} The cardiac proteomic response to pressure overload was also dramatically different between βERKO and wild-type animals of both sexes, suggesting that ERβ also plays a role in the male heart.¹⁶⁸ However, ERβ gene expression was not detectable in adult cardiac myocytes from male mice or rats of either sex, suggesting this ERβ signaling is mediated in cardiac fibroblasts, endothelial, or vascular cells. ¹⁷⁵ (Pugach & Blenck, unpublished data, 2016)

Evidence exists for other signaling pathways involved in mediating sex differences observed in response to pressure overload. The calcium-calmodulin-dependent kinase-MEF2 (CaMKII-MEF2) pathway may be important as CaMK-phosphatase compartmentalization differed between sexes after pressure overload, leading to differences in MEF2 activation, which can promote cardiac hypertrophy.¹⁶⁹ In addition, cardiac expression of a dominant-negative form of p38α MAPK resulted in severe hypertrophic development and mortality in female, but not male mice after pressure overload.¹⁷⁶ However, OVX abolished the increased hypertrophic responses observed in transgenic females.¹⁷⁶ Additionally, male rats exhibited increases in NOS1 expression, a factor that has consistently been up-regulated in HF, after aortic banding much earlier than females; once again demonstrating the importance of NO in mediating cardiac sexual dimorphisms.¹⁷⁷ Complex signaling mechanisms mediate sexual dimorphisms associated with pressure overload hypertrophy, and further studies are required to elucidate interactions among those implicated.

Pathological hypertrophy induced by volume overload

Anatomical defects that cause conditions such as mitral or aortic valve regurgitation result in increased ventricular blood volume and cause the thickness of ventricular walls to increase to maintain cardiac function.¹⁷⁸ Similar to pressure overload, the consequences of volume overload were less severe in females; males exhibited decreased contractile function, increased mortality rates, and were more likely to progress to HF.^{179, 180} Males also displayed higher levels of plasma catecholamines, and increased cardiac expression of angiotensin II (Ang II) type 1 receptor as well as pro-apoptotic proteins such as BAX, caspase 3 and 9.^{180, 181} This sex-specific alteration in apoptotic signaling in animals agrees with increased fibrotic gene expression observed in cardiac biopsies from men, but not women experiencing LV hypertrophy due to aortic stenosis.¹⁸²

As in pressure overload, estrogen protects from development of pathological hypertrophy induced by volume overload. OVX of female rats resulted in more pronounced hypertrophy that progressed to ventricular dilation.¹⁷⁹ Interestingly, supplementation of OVX rats with 17β-estradiol did not fully attenuate the development of hypertrophy, suggesting that other ovarian hormones may be mediating protection.¹⁸⁰ However, this contradicts studies that demonstrated rescue of cardiac function with estrogen administration in similar volume overload models.^{181, 183} Additionally, estrogen regulates apoptosis during this hypertrophic response as OVX rats exhibited increased cardiac pro-apoptotic signaling similar to levels observed in males, but this response was abrogated with estrogen supplementation.¹⁸¹ Estrogen also regulates fibrosis in a sexually dimorphic manner in that collagen gene expression increased upon estrogen treatment in male, but decreased in female adult rat cardiac fibroblasts, which is consistent with sex differences observed clinically.¹⁸²

Estrogen may also be regulating other signaling pathways in response to volume overload. Male but not female rats displayed decreased cardiac expression of β-adrenergic receptors (β-ARs) and adenylyl cyclase in response to arteriovenous shunt.¹⁸³ However, OVX females also exhibited decreases in β-AR and adenylyl cyclase expression in response to volume overload, and estrogen supplementation brought these values back to levels observed in intact females, suggesting that estrogen may maintain cardiac function in response to volume overload stress by up regulating the β-AR signaling pathway.¹⁸³

Chemical induction of cardiac hypertrophy

Chemical agonism of cardiac pathways, such as the β-adrenergic and Ang II pathways, also promotes pathological cardiac hypertrophy. After one week of Iso treatment, male mice developed greater cardiac hypertrophy than females and also displayed altered contractile function.¹⁸⁴ Sex differences are also apparent at the cellular level as Iso treatment increased SR calcium levels in male but not female cardiac myocytes.¹⁸⁵ Additionally, Iso treatment elicited a greater increase in cAMP production, cell shortening, and intracellular calcium transients in male cardiac myocytes.^{186, 187} However, at higher Iso concentrations, male but not female cardiac myocytes exhibited signs of calcium overload.¹⁸⁷ Ang II promoted a different sex-specific response; cardiac myocytes isolated from aged female mice overexpressing Ang II were more prone to contractile dysfunction than their male counterparts.¹⁸⁸ However, these female cells also exhibited stable SR calcium stores, whereas the male cells displayed increased spontaneous contractility, once again suggesting female myocytes are protected from calcium overload by unknown mechanisms.¹⁸⁸

Acute estrogen treatment of isolated male cardiac myocytes also treated with Iso inhibited cAMP production and increased peak calcium.¹⁸⁹ Similarly, myocytes isolated from the hearts of OVX rats treated with Iso exhibited increased calcium transients, force of contraction, and PKA activity compared to sham controls, but estrogen supplementation restored these parameters to sham levels.¹⁹⁰ Myocytes from OVX rats displayed altered expression of β-ARs compared to sham controls, which could explain functional differences in response to Iso.¹⁹¹ Estrogen appears to protect from Ang II-induced hypertrophy as OVX mice supplemented with estrogen during Ang II treatment developed less hypertrophy and fibrosis than the vehicle treated animals.¹⁹² This response appears to be mediated by ERβ signaling because OVX βERKO animals supplemented with estrogen were no longer protected from developing hypertrophy or fibrosis, as was observed in αERKO or wild type OVX animals.¹⁹²

Genetic hypertrophic cardiomyopathy animal models

Hypertrophic cardiomyopathy (HCM) is a well-characterized autosomal dominant genetic disease, which affects about 1 in 500 individuals. Disease-causing mutations have been found in at least 11 different genes that are important for maintaining contractile function, such as components of the sarcomere and calcium handling genes. A variety of different mouse models have been created that harbor disease-causing mutations reported in humans.¹⁹³ One of the most commonly used mouse models has a missense mutation (R403Q) in α-myosin heavy chain (α-MyHC,) which when present in the β-MyHC human gene produces a severe form of HCM.¹⁹⁴ These mice develop HCM similar to human patients in a manner that is modulated by sex over time.¹⁹⁴ While males and females both developed cardiac hypertrophy to a similar extent at four months of age, only the males displayed LV dilation and systolic dysfunction at ten months.¹⁹⁴ In young HCM mice, the R403Q α-MyHC mutation enhanced LV myofilament performance, but myofilament function was not different between sexes.¹⁹⁵ However, in a more recent study analyzing older mice with established HCM, 10-month-old female HCM mice had larger hearts, and their trabeculae were more sensitive to calcium than their male counterparts.⁵¹ This difference in calcium sensitivity could be due to higher expression of sarco/endoplastic reticulum calcium transport ATPase (SERCA2A) in female HCM hearts, or the reduced phosphorylation of cardiac troponin T observed in the male HCM hearts.⁵¹ Furthermore, the signaling pathways activated in this HCM model appear to be different between sexes as HCM males also expressing a constitutively activated glycogen synthase kinase 3β exhibited contractile dysfunction, decreased SERCA2A expression, and premature death, but this was not observed in their female counterparts.¹⁹⁶ Interestingly, removal of phytoestrogens from rodent chow abrogated the severe dilated cardiomyopathy observed in HCM males.¹⁶¹ However, female HCM mice did not progress to HF on either phytoestrogen-free or soy-based diets.¹⁶¹ The predominant phytoestrogen in soy, genistein, activates apoptotic pathways in the male HCM heart contributing to the development of cardiac dysfunction.⁷³ Additionally, estrogen treatment was not protective in either male or female HCM animals and actually increased mortality in phytoestrogen-fed male HCM animals.⁷³

Mice harboring a truncated cardiac troponin T (cTnT) protein displayed smaller ventricles and contractile dysfunction, but not fibrosis, where transgenic mice with a missense mutation (R92Q) in the cTnT gene exhibited severe fibrosis and induction of hypertrophic markers.¹⁹⁷ In addition, while no significant differences at baseline were observed in the truncated cTnT animals from either sex, the male R92Q animals displayed smaller heart weights, increased expression of hypertrophic markers, as well as increased fibrosis compared to their female counterparts.¹⁹⁷ Both of these cTnT models displayed sex differences with respect to exposure to adrenergic stimuli. Treatment with either Iso or phenylephrine (PE) resulted in sudden cardiac death of all male, but not female HCM animals.¹⁹⁷ Unlike the R403Q α-MHC model, estrogen appears to be cardioprotective in the R92Q cTnT female animals. OVX further decreased contractile function as well as myocardial energy metabolism, but estrogen supplementation restored these parameters and reduced cardiac oxidative stress. ¹⁹⁸ In another model of HCM, in which mice carry a point mutation in myosin-binding protein C, isolated cardiomyocytes and myofilaments from male animals exhibited reduced maximal force generating capacity compared to females.¹⁹⁹

While the mechanisms for the observed sex differences in genetic HCM models are not well understood, they should be taken into account when choosing a genetic model of the disease. Directly extrapolating results from genetic HCM animal studies to humans should also be done carefully due to the phenotypic diversity of HCM patients, which is not as apparent in the animal models. ²⁰⁰ Whether this limitation of HCM animal models is due to unknown modifiers of the human disease or a result of the difference in MyHC isoform predominance between mice and humans, caution should used when relating results to the human population. Additionally, it is not quite clear how the observed sex differences in animal HCM models relate to humans since understanding the effect of sex on the development and mortality associated with HCM is complicated by women being diagnosed later in life due to multiple environmental factors, such as clinical screening biases.²⁰¹

Conclusions

Over the past 20 years, awareness of heart disease in women has dramatically increased, but gaps remain in knowledge among women about their cardiovascular risk. Increasing our knowledge of sex differences in basic cardiac physiology is critical for effectively treating patients of both sexes suffering with CVD. Human and animal studies have demonstrated that cardiovascular sexual dimorphisms exist in normal and diseased states from the level of cardiac myocyte to the whole heart. Despite several differences in cardiac physiology compared to humans, small rodent models have been extremely useful for better understanding the mechanisms responsible for mediating these observed sex differences. At baseline, differences in excitation-contraction coupling and mitochondrial function are apparent between the sexes, providing evidence that mechanisms involved in maintaining cardiac function are in place prior to the onset of any disease. Estrogen is believed to mediate this protection both prior to and after disease onset, but the exact mechanism is still not well understood and appears to be context dependent since estrogen supplementation can also be detrimental in some cases. As observed in humans, female animals are generally protected from developing multiple CVDs in genetic, surgical, and chemically-induced models. In many disease states, such as cardiac hypertrophy or hypertension, this protection in females or increased risk in males is lost upon the depletion of endogenous sex hormones. To appropriately investigate the mechanisms underlying CVD development, biological sex is an important experimental variable that needs to be addressed both in basic and clinical research studies. Although this review cited many different studies that focused on cardiac sex differences, most studies have not taken sex into account. More research needs to include animals of both sexes; not only to better understand the responsible signaling mechanisms, but to also ensure that therapeutics will work effectively in both men and women. With the new guidelines recently released by the National Institutes of Health that require biological sex to be included as a potential experimental variable in vertebrate animal and human studies, more insight into mechanistic studies of the sexual dimorphisms observed in the cardiovascular system may be gained.

List of Non-standard Abbreviations and Acronyms

ACE

angiotensin converting enzyme

Ang II

angiotensin II

AR

adrenergic receptor

ARVMs

Adult rat ventricular myocytes

CaMKII

calcium-calmodulin-dependent kinase

cAMP

cyclic AMP

cTnT

cardiac troponin T

CVD

cardiovascular disease

DSS

dahl salt-sensitive

ER

estrogen receptor

ERE

estrogen reponse element

HCM

hypertrophic cardiomyopathy

HDACs

histone deacetylases

hERG

human ether-a-go-go

HF

heart failure

HRT

hormone replacement therapy

I/R

ischemia-reperfusion

Iso

isoproterenol

KChIP2

Kv channel-interacting protein 2

KO

knock-out

L-NAME

N^ω-nitro-L-arginine methyl ester

LVEF

left ventricular ejection fraction

MAPK

mitogen-activated protein kinase

MEF2

myocyte enhancer factor 2

MI

myocardial infarction

MyHC

myosin heavy chain

NCX

sodium-calcium exchanger

NO

nitric oxide

NOS

NO synthase

OVX

ovariectomy/ovariectomized

OXPHOS

oxidative phosphorylation

PE

phenylephrine

RAS

renin-angiotensin system

ROS

reactive oxygen species

RTKs

receptor tyrosine kinases sarco/endoplasmic reticulum calcium transport

SERCA2A

ATPase

SHR

spontaneously hypertensive rat

SR

sarcoplasmic reticulum

SUR2

sulfonylurea receptor 2

SV

stroke volume

TGF-β

transforming growth factor-β

WT

wild-type