Author + information
- Received October 8, 1999
- Revision received March 15, 2000
- Accepted April 26, 2000
- Published online September 1, 2000.
- Sofija Jovanović, DVM, PhD∗,†,
- Aleksandar Jovanović, MD, PhD∗,†,*,
- Win K Shen, MD, FACC∗ and
- Andre Terzic, MD, PhD∗,†
- ↵*Reprint requests and correspondence: Dr. Aleksandar Jovanović, Tayside Institute of Child Health, Ninewells Hospital and Medical School, University of Dundee, Dundee, DD1 9SY, Scotland, United Kingdom
The main objective of the present study was to determine whether low physiological levels of estrogen directly protect cardiac cells against metabolic stress.
The beneficial effect of estrogens on the cardiovascular system has been traditionally ascribed to decrease in peripheral vascular resistance and to an antiatherogenic action. Whether physiological concentrations of 17β-estradiol (E2) are also able to protect cardiomyocytes against metabolic insult directly is unknown.
Isolated ventricular cardiomyocytes were loaded with the Ca2+-sensitive fluorescent dye Fluo-3 and imaged by a digital epifluorescence imaging system. In cardiac cells preincubated with hormones and/or drugs for 8 h, metabolic stress was induced by addition and removal of 2,4-dinitrophenol (DNP).
In cardiomyocytes, a 3-min-long exposure to chemical hypoxia, followed by reoxygenation, produced intracellular Ca2+ loading independently of gender (female: 729 ± 88 nmol/liter; male: 778 ± 97 nmol/liter). Pretreatment with E2 (10 nmol/liter) significantly reduced the magnitude of hypoxia/reoxygenation-induced Ca2+ loading in female (E2-treated: 298 ± 39 nmol/liter; untreated: 729 ± 88 nmol/liter), but not in male (E2-treated: 1029 ± 177 nmol/liter; untreated: 778 ± 97 nmol/liter) cardiac cells. The protective action of E2 was not mimicked by the inactive estrogen stereoisomer, 10 nmol/liter 17α estradiol (17α estradiol-treated: 886 ± 122 nmol/liter; untreated: 729 ± 88 nmol/liter), and was abolished by tamoxifen (1 μmol/liter), which acts as an antagonist of E2 on estrogen receptors (E2 plus tamoxifen-treated: 702 ± 98 nmol/liter; untreated: 729 ± 88 nmol/liter).
In a gender-dependent manner, E2 directly protects cardiac cells against hypoxia-reoxygenation injury through an estrogen receptor–mediated mechanism. Such property of E2 may contribute to cardioprotection in the female gender.
Gender-specific differences in the incidence of cardiovascular disease were recognized half a century ago. While the risk of heart disease in men increases constantly with age, premenopausal women have a significantly lower risk, which, however, increases rapidly after menopause to levels comparable to male counterparts (1). Estrogen substitution in postmenopausal women reduces cardiovascular mortality by 30% to 50%, suggesting that estrogens are protective (2). In fact, the antiatherogenic action of estrogens on the lipid profile and arterial wall has been well documented (3). More recently, however, experiments performed in whole heart preparations have revealed that estrogens may also decrease infarct size and protect against ischemia-reperfusion injury without changing hemodynamic parameters (4–7). These findings have implied that estrogens can directly interact with cardiomyocytes. This is supported by the finding that 17β-estradiol (E2), applied in high pharmacological concentrations, inhibits hyperkalemia-induced Ca2+ loading in cardiomyocytes through a “Ca2+ antagonistic action” (8). However, whether low physiological concentrations of E2 protect cardiomyocytes against metabolic insult is unknown. We report that E2 protects cardiomyocytes against hypoxia-reoxygenation, and we present evidence that gender plays a crucial role in the cytoprotective action of estrogens.
Single ventricular cardiomyocytes
Ventricular cardiomyocytes were enzymatically dissociated from pentobarbital-anesthetized, sexually mature female and male guinea pigs (8,9). Hearts were perfused (at 37°C) with medium 199, Ca2+-EGTA-buffered low-Ca2+ medium, and low-Ca2+ medium containing pronase E (8 mg per 100 ml), proteinase K (1.7 mg per 100 ml), bovine albumin (0.1 g per 100 ml), and 200 μmol/liter CaCl2. Ventricles were cut into fragments, and single cells were isolated by stirring tissue in pronase E, proteinase K and collagenase (5 mg per 10 ml). The investigation conformed with the “Guide for the Care and Use of Laboratory Animals” (NIH publication 85-23, revised 1985).
Digital epifluorescent imaging
Rod-shaped cardiomyocytes with clear striations and smooth surface were imaged by digital epifluoresecent microscopy (8,9). Cells were loaded (for 30 min) with the esterified form of the Ca2+-sensitive fluorescent probe Fluo-3 acethoxymethyl ester (5 μmol/liter Fluo-3, dissolved in dimethyl sulfoxide plus pluronic acid; Molecular Probes) and superfused with Tyrode solution (in mmol/liter: 136.5 NaCl; 5.4 KCl; 1.8 CaCl2; 0.53 MgCl2; 5.5 glucose; 5.5 HEPES-NaOH; pH 7.4). Cardiomyocytes were imaged using a digital epifluorescence imaging system coupled to an inverted microscope (Zeiss Axiovert-135 TV) with a ×40 oil-immersion objective lens. A 100-W mercury lamp served as a source of light to excite Fluo-3 at 488 nm. An excitation dichroic mirror with a cutoff of 510 nm, and a long pass emission filter with a cutoff of 520 nm, were used to detect Fluo-3 fluorescence using an intensified charge coupled device camera. Fluorescence was digitized using an imaging software (Attoflor RatioVision, Atto Instruments). An estimate of cytosolic Ca2+ concentration was calculated according to the equation: [Ca2+] = Kd(F − Fmin/Fmax − F), where Fmin and Fmax are minimal and maximal fluorescence intensity, Kd dissociation constant of the Fluo-3-Ca2+ complex (422 nmol/liter) and F intensity of fluorescence. To obtain Fmin and Fmax values, cells were exposed to 100 μmol/liter ionomycin either in the absence of Ca2+ (extracellular Ca2+ was removed and 3 mmol/liter EGTA added to the extracellular solution) or in the presence of saturating concentrations of Ca2+ (10 mmol/liter CaCl2), respectively (8,9).
Chemical hypoxia-reoxygenation injury
Cardiomyocytes, superfused with Tyrode solution, were exposed to 2 mmol/liter 2,4-dinitrophenol (DNP), a metabolic poison that inhibits mitochondrial oxidative phosphorylation. Following 3-min treatment, DNP was removed (∼10 s was required for removal of DNP), and cells reexposed to Tyrode solution (9–11). We have previously established that this protocol induces Ca2+ overload in cardiomyocytes followed by an irreversible cellular hypercontracture and cell death, with the intracellular concentration of Ca2+ reflecting accurately the condition of a cardiac cell (9). Such chemical hypoxia-reoxygenation protocol was applied to cells previously incubated for 8 h in the absence (control) or presence of E2 with or without the antiestrogen tamoxifen, or in the presence of 17α-estradiol, an inactive E2 stereoisomer. Estrogens and tamoxifen were dissolved in alcohol, and Fluo-3AM was dissolved in dimethyl sulfoxide plus pluronic acid. The final concentration of solvents was kept to less than 0.1%. At this concentration, solvents did not affect intracellular Ca2+ levels (8).
Data are presented as mean ± SEM, with n representing number of imaged fields. Mean values were compared by the two-way repeated measures analysis of variance (ANOVA) using SigmaStat program (Jandel Scientific). A p value <0.05 was considered statistically significant.
Hypoxia-reoxygenation induces Ca2+ loading in both female and male cardiomyocytes
Resting cardiomyocytes from both male and female hearts had a low cytosolic Ca2+ concentration (98 ± 11 nmol/liter; n = 13–14; Fig. 1). A 3-min-long exposure to chemical hypoxia induced by the mitochondrial uncoupler DNP, followed by reoxygenation evoked by rapid removal of DNP, produced significant Ca2+ loading independently of gender (female: 729 ± 88 nmol/liter, n = 14; male: 778 ± 97 nmol/liter, n = 13, Fig. 1). The gender difference in the intracellular Ca2+ concentration was not statistically significant (p = 0.110). However, the difference in the Ca2+ concentration under control conditions and following hypoxia-reoxygenation was statistically significant (p < 0.001). No statistically significant interaction existed between gender and cellular response to the metabolic insult (p = 0.846).
17β-estradiol prevents hypoxia-reoxygenation–induced Ca2+ loading in female, but not male, cardiomyocytes
In female cardiac cells, pretreatment with 10 nmol/liter E2 significantly reduced the magnitude of hypoxia-reoxygenation–induced Ca2+ loading (untreated: 729 ± 88 nmol/liter, n = 14; E2-treated: 298 ± 39 nmol/liter, n = 7; Figs. 1 and 2), ⇓ without affecting the basal levels of Ca2+ (see Figs. 1 and 2). The difference between E2-treated and untreated cells was statistically significant (p < 0.001). The interaction between metabolic status of the cell and treatment was also statistically significant (p < 0.001). This effect was concentration-dependent, and at 1 nmol/liter E2 decreased Ca2+ loading to 469 ± 37 nmol/liter (n = 14). In contrast, in male cardiac cells pretreatment with E2 (10 nmol/liter) did not prevent hypoxia-reoxygenation–induced Ca2+ loading (untreated: 778 ± 97 nmol/liter, n = 13; E2-treated: 1029 ± 177 nmol/liter, n = 6; Figs. 1 and 2). The difference between E2-treated and untreated cells was not statistically significant (p = 0.308) and the interaction between the metabolic status of a cell and treatment was also not statistically significant (p = 0.671). The gender-dependent difference in response to the treatment with E2 was statistically significant (p < 0.001).
Tamoxifen blocks the protective action of E2
The protective effect of E2 (10 nmol/liter) was abolished by tamoxifen (1 μmol/liter), a partial agonist of estrogen receptors. Following hypoxia-reoxygenation, in the presence of both E2 (10 nmol/liter) and tamoxifen (1 μmol/liter), cytosolic Ca2+ levels increased to 702 ± 98 nmol/liter in female cardiomyocytes (n = 5; E2-treated: 298 ± 39 nmol/liter, n = 7, Fig. 3). The difference between E2-treated and E2+tamoxifen-treated cells was statistically significant (p = 0.002), and the interaction between the metabolic status of a cell and treatment was also statistically significant (p < 0.001).
17α-estradiol does not prevent hypoxia-reoxygenation–induced Ca2+ loading in female cardiomyocytes
In female cardiac cells, pretreatment with 17α estradiol (10 nmol/liter), an inactive estrogen stereoisomer, did not significantly decrease the magnitude of hypoxia-reoxygenation–induced Ca2+ loading (17α estradiol-treated: 886 ± 122 nmol/liter, n = 6; untreated: 729 ± 88 nmol/liter, n = 14, Fig. 4). The difference between 17α estradiol-treated and untreated cells was not statistically significant (p = 0.107) and the interaction between the metabolic status of a cell and treatment was also not statistically significant (p = 0.852).
In the present study we demonstrate that physiological levels of E2 confer resistance to female, but not male, cardiomyocytes against intracellular Ca2+ loading induced by hypoxia-reoxygenation. These findings indicate that E2 is cytoprotective, and that gender is critical for E2 to exhibit such a property.
In cardiomyocytes, reoxygenation that follows a hypoxic insult initiates a series of cellular reactions (9,12). In particular, hypoxia-reoxygenation induces intracellular Ca2+ loading, which represents a common denominator leading to cell injury (9–11). In cardiomyocytes, used here, basal levels of cytosolic Ca2+ were low and similar to those previously reported for healthy cells (8,9). The significant increase in intracellular Ca2+ following hypoxia-reoxygenation confirms that cardiomyocytes are highly vulnerable to such an insult.
Female gender as a protector against ischemic heart disease has been associated with high levels of circulating estrogens (2,3). The beneficial effect of estrogens has been traditionally ascribed to decrease in peripheral vascular resistance and to an antiatherogenic action (3,13), although more recently additional effects on the myocardium have also been proposed (4–7). In the present study, exposure of cardiomyocytes to nanomolar levels of E2 protected female, but not male, cardiomyocytes against hypoxia-reoxygenation, providing evidence at the cellular level for gender-dependent cardioprotective properties of estrogens.
Our findings that exposure of cardiomyocytes to 17α-estradiol, an inactive E2 stereoisomer, or to E2 in the presence of the antiestrogen tamoxifen, did not protect against hypoxia-reoxygenation further suggests that activation of estrogen receptors is necessary for the prevention of Ca2+ overload. In this regard, the present findings strongly support previous studies, done at the whole heart level, that have indicated a protective action of estrogens on the myocardium itself (4–7). Moreover, the present study also supports the notion that low blood levels of estrogens may have direct cardioprotective properties in females (6). Using single cardiomyocytes, a pure myocardial preparation free of neuronal and vascular elements, we provide first direct evidence that physiological levels of E2 indeed confer resistance toward hypoxia-reoxygenation injury in a gender-dependent manner.
In recent years, it has been reported that estrogen supplementation may be beneficial in males under some conditions of metabolic stress (14). Conversely, disruptive mutation of the estrogen receptor gene in man has been associated with premature coronary artery disease (15). The present study indicates that gender difference in myocardial response to metabolic insult is not solely the consequence of different circulating E2 levels, but rather to a more profound difference in the interaction between cardiomyocytes and E2. In principle, estrogen receptors have been found in cardiac cells in both males and females (16). However, it should be mentioned that estrogen receptors are apparently more functional and present at higher density in female than in male cardiomyocytes, which may explain the gender-dependence of E2-mediated cytoprotection observed in the present study (17,18). In some studies, it has been demonstrated that estrogens may be protective in a gender-independent manner (4,5). Such findings have been obtained by achieving higher blood levels of E2, which further supports our conclusion that gender-dependent difference in the response to low E2 concentration may be due to differences in the density and function of estrogen receptors. Taken together, it is apparent that activation of estrogen receptors in a cardiac cell may protect this cell type against metabolic stress, thus effectively preventing deleterious Ca2+ overload under hypoxia-reoxygenation injury.
Both the role and function of estrogen receptors in the heart are still largely unknown. It is possible that E2 may regulate the expression of genes that encode proteins implicated in endogenous cardioprotection (19–21). In agreement with such a possibility are recent findings that expression of the cardiac calcium channel, an ion conductance central in cardiac excitation-contraction coupling, is regulated by activation of estrogen receptors (16). Moreover, it has been recently demonstrated that estrogens up-regulate the expression of protein kinase C, a signaling cascade protein known to protect cardiomyocytes against Ca2+ overload and associated cellular injury (22,23). Therefore, it is possible that E2-induced expression of genes encoding cytoprotective proteins protects cardiac cells against metabolic stress.
Conclusion and significance
This study demonstrates, for the first time, that E2 directly protects, in a gender-dependent manner, cardiac cells against hypoxia-reoxygenation injury through an estrogen receptor-mediated mechanism. These findings may provide further support for the therapeutic use of estrogens and contribute to an understanding of the resistant phenotype associated with female gender.
To determine the direct effect of E2 on intracellular Ca2+ concentration during hypoxia-reoxygenation, it was necessary to image isolated cardiomyocytes. Such approach provided a direct visualization of the protective effect of E2 at the single-cell level. However, it should be considered that, in intact myocardium, additional cardiac, as well as extracardiac, mechanisms could further modulate the action of E2.
☆ This research was supported by grants from the American Heart Association (9806356X and 9950052N); the British Heart Foundation (PG/99105); the Bruce and Ruth Rappaport Program in Vascular Biology and Gene Delivery; Miami Heart Institute; TENOVUS-Scotland; University of Dundee, and the Wellcome Trust (059528/Z/99/Z/JMW/CP/JF).
- Fluo-3 acethoxymethyl ester
- Received October 8, 1999.
- Revision received March 15, 2000.
- Accepted April 26, 2000.
- American College of Cardiology
- Node K.,
- Kitakaze M.,
- Kosaka H.,
- Minamino T.,
- Funaya H.,
- Hori M.
- Jovanović A.,
- Jovanović S.,
- Lorenz E.,
- Terzic A.
- Jovanović N.,
- Jovanović S.,
- Jovanovic A.,
- Terzic A.
- Kloner R.A.,
- Bolli R.,
- Marban E.,
- et al.
- Chen S.J.,
- Li H.,
- Durand J.,
- Oparil S.,
- Chen Y.F.
- Toung T.J.K.,
- Traystman R.J.,
- Hurn P.D.
- Sudhir K.,
- Chou T.M.,
- Chatterjee K.,
- et al.
- Johnson B.D.,
- Zheng W.,
- Korach K.S.,
- Scheuer T.,
- Cattarall W.A.,
- Rubanyi G.M.
- Grohe C.,
- Kahlert S.,
- Lobbert K.,
- Vetter H.
- Hearse D.J.
- Simkhovich B.Z.,
- Przyklenk K.,
- Kloner R.A.
- Gross G.J.,
- Fryer R.M.