Previously, we demonstrated that intact female rats fed a standard rodent diet containing soybean products exhibit essentially no adverse left ventricular (LV) remodeling in response to aortocaval fistula-induced chronic volume overload. We hypothesized that phytoestrogenic compounds in the diet contributed to the female cardioprotection. To test this hypothesis, four groups of female rats were studied: sham-operated (Sham) and fistula (Fist) rats fed a diet with [P(+)] or without [P(−)] phytoestrogens. Eight weeks postfistula, systolic and diastolic cardiac function was assessed by using a blood-perfused, isolated heart preparation. High-phytoestrogen diet had no effect on body, heart, and lung weights, or cardiac function in Sham rats. Fistula groups developed LV hypertrophy, which was not reduced by dietary phytoestrogens [1,184 ± 229 mg Fist-P(−) and 1,079 ± 199 mg Fist-P(+) vs. 620 ± 47 mg for combined Sham groups, P < 0.05]. Unstressed LV volume increased in Fist-P(−) rats (428 ± 16 vs. 300 ± 14 μl Sham, P < 0.0001), but it was not different from Sham for Fist-P(+) animals (286 ± 17 μl). Fist-P(−) rats developed increased ventricular compliance (5.3 ± 0.8 vs. 2.3 ± 0.3 μl/mmHg Sham, P < 0.01), whereas Fist-P(+) rats had no change in compliance (2.8 ± 0.4 μl/mmHg). Intrinsic ventricular contractility was maintained in the Fist-P(+) rats, but it was reduced (P < 0.001) in the Fist-P(−) rats [systolic pressure-volume slope: 1.04 ± 0.03, 0.60 ± 0.06, and 0.99 ± 0.08 mmHg/μl, for Fist-P(+), Fist-P(−), and Sham, respectively]. These data indicate that dietary phytoestrogens contribute significantly to female cardioprotection against volume overload-induced adverse ventricular remodeling and that studies evaluating gender differences in cardiovascular remodeling must consider the influence of dietary phytoestrogens.
- ventricular function
- heart failure
our laboratory has recently reported gender differences in the development of heart failure using the aortocaval (AV) fistula rat model (9). In this study, male rats experienced 25% mortality following 8 wk of chronic volume overload, and the remaining males had marked left ventricular (LV) dilatation and increased compliance. In contrast, female rats had only 2% mortality, and surviving females had no dilatation or increase in ventricular compliance. These findings indicate that almost all of the female rats were able to compensate for the chronic volume overload. A possible mechanism for this female cardioprotection is the influence of estrogen or other sex hormones. Also, in that study, both male and female rats were fed a standard rodent chow (Prolab RMH 3000) containing soybean and alfalfa meal. These plant-derived products contain phytoestrogens (e.g., genistein), which are known to be estrogen receptor agonists which may affect cardiac remodeling (13, 32). Accordingly, the purpose of this study was to determine the contribution of dietary phytoestrogens to the cardioprotection exhibited by female rats in response to chronic volume overload. The results indicate that dietary phytoestrogens attenuate ventricular remodeling and contribute significantly to female cardioprotection in the AV fistula model of heart failure.
Studies were performed by using 8-wk-old female Sprague-Dawley (Harlan Hsd:SD) rats, weighing ∼200 g at surgery. Rats were housed under standard environmental conditions and were fed ad libitum either Prolab RMH 3000, containing soybean and alfalfa meal [considered “high” phytoestrogen with a total isoflavone content of 417 ± 76 parts/million (ppm) (aglycone units), 196 ± 33 ppm genistein, and 183 ± 7 ppm daidzein], or TestDiet 5K96 casein-based diet (considered phytoestrogen free with <1.0 ppm total isoflavones). The crude protein content of these diets was similar (∼19–22%). A single lot of RHM 3000 diet was used for this study to avoid lot-to-lot variability of phytoestrogen content. Rats were fed their respective diets for 1 wk before the AV fistula surgery and throughout the 8-wk experimental protocol.
The investigation conformed with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996) and was approved by Auburn University's Institutional Animal Care and Use Committee. Anesthesia for surgical procedures was induced with xylazine (7 mg/kg) and ketamine (62 mg/kg), administered by intraperitoneal injection. At the experimental end point, animals were anesthetized with pentobarbital sodium (50 mg/kg intraperitoneal), and hearts were removed for evaluation of ventricular size and function. Plasma samples from each animal were assayed for estrogen content using a commercial radioimmunoassay (Coat-a-Count kit, DPC, Los Angeles, CA).
Infrarenal AV fistula was created in rats, as previously described (7). Briefly, a ventral laparotomy was performed to expose the aorta and caudal vena cava below the renal arteries. An 18-gauge needle was inserted into the exposed ventral abdominal aorta and advanced through the medial wall into the vena cava to create the fistula. The needle was withdrawn, and the ventral aortic puncture site was sealed with cyanoacrylate. Creation of a successful AV fistula was visually evident by the pulsatile flow of oxygenated blood into the vena cava. Abdominal musculature and skin incisions were closed by standard techniques with absorbable suture and autoclips, respectively. The sham surgical procedure was identical to the fistula surgery procedure, with the exception that the fistula was not created.
All rats were studied 8 wk after sham surgery or AV fistula-induced volume overload. This time period was chosen because, in previous studies using male rats, a significant number had developed symptomatic congestive heart failure (CHF) by 8 wk postfistula (7). A total of four groups of age-matched female rats were studied, including sham-operated controls (Sham) and rats with fistula (Fist) fed either a phytoestrogen-free [P(−)] or high phytoestrogen [P(+)] diet. At the experimental end point, each rat was weighed and anesthetized. Following laparotomy, fistula patency was visually confirmed by the pulsatile flow of oxygenated blood from the aorta into the vena cava. The rat was then ventilated, and the heart exposed. Cardiac output was measured was using a Doppler flow probe (model 3SB, Transonic Systems, Ithaca, NY). The heart was then removed and attached to a perfusion apparatus for evaluation of ventricular function (described below). After completion of the functional studies, the atria and great vessels were removed, and the LV (including septum) and right ventricle (RV) were separated and weighed. Lung wet weight was measured after the esophagus and trachea were trimmed away and the pleural surface blotted dry.
Assessment of ventricular size and function.
LV volume and function were evaluated by using a blood-perfused isolated heart preparation, as previously described (7). Arterial blood from the carotid artery of a support rat was pumped to a pressurized reservoir for retrograde perfusion of the isolated heart. The coronary venous effluent was collected and returned to the support rat through a jugular vein catheter to filter and oxygenate the blood supply to the isolated heart. Coronary perfusion pressure was maintained at 100 ± 5 mmHg. Before removal of the heart from the anesthetized rat, the carotid arteries were ligated, and a cannula was inserted into the thoracic aorta at a level just proximal to the first pair of intercostal arteries and secured with a silk ligature. Retrograde perfusion of the coronary arteries with blood from the perfusion reservoir was begun as soon as the cannula was secured. The heart was then quickly removed from the chest and attached to the apparatus. A highly compliant latex balloon was inserted through the mitral valve orifice into the LV to measure intraventricular pressure. Once the heart developed stable isovolumetric contractions, the balloon volume that produced an LV end-diastolic pressure (EDP) of 0 mmHg (V0) was determined. Balloon volume was then increased in 5- to 10-μl increments from this point until an LVEDP of 25 mmHg was attained. The EDP and peak isovolumetric pressures were recorded following each increase in balloon volume.
Diastolic and systolic stress-strain relations were calculated for each isolated heart as previously described (25). For this calculation, the LV is assumed to be thick-walled and spherical in shape, and the midwall stress and strain are calculated as follows: (1) (2) where P is LV pressure (mmHg), V is LV volume (ml), V0 is LV volume (ml) at EDP of 0 mmHg, and M is LV mass (g). By using peak isovolumetric pressures and EDPs in the above equations, peak systolic and end-diastolic stresses were calculated, respectively.
Data and statistical analysis.
Statistical analyses were performed using Graphpad analysis software (Graphpad 4.0, Prism, San Diego, CA). LV volumes were adjusted to account for the volume displaced by the balloon material in the LV lumen. Balloon volume was calculated from the weight of the balloon divided by the density of the latex (0.898 g/ml). Pressure-volume (P-V) relations generated for the balloon alone indicated that the balloon's contribution to LVEDP was negligible over the range of experimental volumes. Diastolic LV P-V and stress-strain data were fitted with a third-order polynomial regression, and V0 for each P-V curve was calculated. Using this regression, LV end-diastolic volume (EDV) was calculated at discrete EDP intervals (each 2.5 mmHg) for graphical comparison. A minimum of three consistent P-V relations were averaged for each heart. Nonlinear relations were compared by using an analysis of excess variance, which allows calculation of an F-value based on the sum-of-squares error of two data sets regressed as separate and combined data. The resulting F-value and degrees of freedom for the combined and individual data sets allow calculation of a P value, which indicates significantly different relations if P < 0.05 (23). Systolic LV P-V and stress-strain data were fitted with linear regression, and the slopes of these relations were used as an index of intrinsic chamber contractility and myocardial contractility, respectively (i.e., neurohormonal influence removed). Linear relations were statistically compared by analysis of covariance, and further comparison of slopes and intercepts were made by using the method of dummy variables (11). Grouped data comparisons were made by one-way ANOVA, and comparison of discrete data points on nonlinear relations were analyzed using two-way ANOVA. When a significant F ratio (P < 0.05) was obtained, intergroup comparisons were made by using a modified t-test and Bonferroni bounds. Statistical significance was taken to be P < 0.05/k, where k is equal to the number of groups compared (23).
Dietary phytoestrogens did not alter uterine weights or plasma estrogen levels (data not presented). Group-averaged cardiac output, heart rate, and LV, RV, lung, and body weights for the Sham and 8-wk Fist groups are listed in Table 1. There were no significant dietary-related differences in these parameters between the Sham groups. Cardiac output was significantly increased roughly fourfold in the Fist animals, confirming the successful creation of the AV shunt and chronic volume overload conditions. Both Fist groups had elevated heart rates vs. the Sham groups (P < 0.05). Fist animals developed marked ventricular hypertrophy, regardless of diet. LV weights were significantly increased in both the Fist-P(−) and Fist-P(+) animals (91 and 75%, respectively, relative to corresponding Sham group, P < 0.05). Likewise, RV weights in both Fist groups were increased relative to Sham groups [128% Fist-P(−) and 118% Fist-P(+)]. Additionally, both Fist groups developed an increase in lung weight relative to corresponding Sham controls [32 and 19% increases, Fist-P(−) and Fist-P(+), respectively, P < 0.05]. The P(+) diet did not significantly reduce lung weight or LV and RV hypertrophy in the Fist rats.
Differences in diastolic function as assessed using the blood-perfused isolated heart preparation are depicted in Fig. 1 as plots of the average LVEDP-LVEDV relationships for each group (parameters summarized in Table 2). Note that isolated heart function was not obtained for every animal listed in Table 1 due to experimental complications (e.g., aortic valve insufficiency or isolated heart ischemia due to support rat failure). An analysis of excess variance determined that there was no statistical difference in the Sham-P(−), Sham-P(+), and Fist-P(+) diastolic P-V curves, whereas the Fist-P(−) group was determined to be statistically different from all other groups (P < 0.001). Dilatation of the LV was assessed by the change in unstressed LV volume (V0). This structural remodeling is depicted in the LVEDP-LVEDV relation as a parallel rightward shift. Animals fed a P(−) diet developed a significant increase in V0 (39% above Sham, P < 0.05), whereas female Fist rats fed the P(+) diet were not different from Sham rats. In addition to structural dilatation, changes in compliance, as measured by the volume required to increase LVEDP from 0 to 25 mmHg, occurred. Increased chamber compliance is reflected by the degree to which the rightward shift of the LVEDP-LVEDV relation was nonparallel. Hearts from Fist-P(−) animals exhibited significantly increased ventricular chamber compliance [103% relative to Sham-P(−), P < 0.05]. However, Fist rats that were fed the phytoestrogen diet maintained their chamber stiffness and did not develop increased ventricular chamber compliance.
Differences in systolic function are depicted in Fig. 2. Analysis of covariance determined a significant F value, i.e., unequal slopes. Slope and intercept values for each group, as determined by the method of dummy variables (11), are listed in Table 2. Only the Fist-P(−) group had significantly altered systolic function with reduced slope of the systolic P-V relationship [43% reduction vs. Sham-P(−), P < 0.05]. Suga et al. (28) have validated the slope of the systolic P-V relationship as an accurate index of LV contractility. The slope of Fist-P(+) rat hearts was not different from those of the Sham hearts, and there were no significant differences in the y-intercepts of the systolic P-V relations between groups.
A plot of the averaged LV mass-to-volume ratio (M/V) for each study group (Fig. 3) indicates that Fist-P(+) animals were able to significantly increase their M/V over the full range of LVEDP, compared with the Fist-P(−) and Sham groups (P < 0.05). Diet had no significant effect on the M/V for Sham rats.
Diastolic and systolic stress-strain relations are depicted in Figs. 4 and 5, respectively. These curves represent the passive and active mechanical properties of the myocardial tissue alone, normalized for differences in LV chamber size and weight. Using an analysis of excess variance, diastolic and systolic stress-strain curves for the Sham groups were not found to be significantly different; thus the Sham groups were combined for comparison purposes. Diastolic and systolic stress-strain curves for both groups of Fist hearts were significantly different from those of Sham hearts, with the myocardium of Fist rats found to be more compliant. As determined by the analysis of excess variance, there was a significant difference (P < 0.05) in diastolic stress at higher strain in the Fist-P(−) group vs. the Fist-P(+) group. However, systolic stress in the Fist groups was not affected by diet. As depicted in Fig. 5, the systolic stress-strain relationships for both Fist groups were nearly identical. Although the slope of the systolic stress-strain relations was not different between Sham and Fist hearts, the y-intercept of the Sham group was significantly different from those of the Fist groups.
Previously, our laboratory has reported gender differences in the patterns of cardiac remodeling induced by volume overload (9). Male rats subjected to chronic volume overload developed myocardial remodeling characterized by eccentric hypertrophy, marked dilatation, increased ventricular compliance, and consistent onset of CHF associated with significant mortality (7). In contrast, intact female rats developed concentric hypertrophy with no impairment of cardiac function, no significant changes in ventricular dilatation or ventricular chamber compliance, and minimal mortality (9). Also, in a longer term study from our laboratory, ovariectomy completely abolished female cardioprotection, in that the myocardial remodeling of ovariectomized females with a fistula was nearly identical to that of males at 21 wk postfistula (6). These findings suggest that ovarian hormones are capable of modulating ventricular remodeling in response to the increased cardiovascular burden imposed by chronic volume overload. In these previous studies, both male and female rats were fed a standard rodent diet that which contained soybean and alfalfa meal. Soybeans, grapes, and many other plant-derived products contain flavonoids, some of which are capable of interacting with estrogen receptors and causing “estrogen-like” effects, hence the term phytoestrogens (4, 16, 18). Of particular interest to this study are genistein and daidzein, which are the primary isoflavones found in soybeans (22). These compounds have been implicated in the low incidence of cancer and cardiovascular diseases in cultures, such as Japanese, that consume diets high in soy content. In the current study, we chose to assess the contribution of soy-derived dietary isoflavones to cardioprotection against adverse remodeling using the AV fistula model of heart failure in female rats. We hypothesized that dietary isoflavones contribute to the apparent cardioprotection in intact female rats subjected to chronic volume overload.
Heart, lung, and body weights, as well as functional cardiac parameters for the P(+) Sham and Fist groups were consistent with our previous results (9). Alteration of dietary phytoestrogen content did not affect the body, LV, RV, and lung weights of Sham rats. In all groups, uterine weight and serum estrogen levels were not affected by the P(+) diet. Following 8 wk of AV fistula-induced chronic volume overload, rats developed significant LV hypertrophy, regardless of diet. We believe that this development of increased LV mass is not pathological, but rather it is a necessary adaptation for the normalization of increased wall stress caused by the volume overload condition. However, rats fed a P(−) diet developed significantly greater LV chamber dilatation and chamber compliance than animals fed a diet containing phytoestrogens. In fact, female rats on P(+) diet experienced no changes in LV chamber compliance, and both diastolic and systolic P-V relationships were nearly indistinguishable from Sham females. An analysis of systolic P-V relations determined that the highly dilated Fist-P(−) hearts were not able to maintain peak function, whereas the Fist-P(+) hearts were able to compensate and maintain their peak function at control levels. It is interesting to note, however, that the ovariectomized rats with fistula from our previous study developed much greater dilatation and increased LV chamber compliance, albeit at 21 wk postfistula, than that developed by the intact females herein on P(−) diet (6). This observation indicates that dietary phytoestrogens alone, in lieu of ovarian hormones, cannot provide the cardioprotection exhibited by the intact female rats fed the P(+) diet.
In our previously published study, male rats with fistula were unable to increase their LV mass to end-diastolic volume ratio (M/V) relative to controls (9). An increase in LV M/V indicates an attempt to normalize wall stress. Failure to sufficiently increase LV M/V is indicative of maladaptive hypertrophy and adverse ventricular remodeling. Figure 3 shows that both sets of female Fist rats increased their LV M/V relative to Sham animals. Rats fed the P(+) diet developed a greater increase in LV M/V, indicating a more successful myocardial remodeling response to the volume overload. The reduced ability of females on a P(−) diet to sufficiently increase their M/V indicates that the increase in mass was insufficient for the degree of ventricular dilatation. This inability to fully compensate eventually leads to adverse ventricular remodeling (i.e., further dilatation and increased compliance) and CHF.
The purpose of converting the P-V data to stress-strain was to assess the mechanical properties and contractility of the myocardial tissue. As can be seen in Fig. 5, the systolic stress-strain relation for both Fist groups was shifted downward relative to the Sham group, indicative of depressed myocardial contractility. Interestingly, this finding is in contrast to the LV systolic P-V analysis, which indicated that chamber contractility was maintained in the Fist-P(+) group. The ability of the Fist-P(+) group to maintain chamber contractility in the face of reduced myocardial contractility is clearly due to hypertrophy without chamber dilatation (i.e., concentric hypertrophy). However, a similar degree of hypertrophy in the Fist-P(−) group together with significant chamber dilatation (i.e., eccentric hypertrophy) is insufficient to compensate for the reduced myocardial contractility, resulting in a depression of chamber contractility.
The in vitro findings of no differences in chamber volume and contractility between the Fist-P(+) and Sham groups raise the question of how the Fist-P(+) group can generate a fourfold increase in cardiac output in vivo. This increase in cardiac output can be partially explained by the elevated heart rate occurring postfistula. Another factor contributing to the observed increase in cardiac output is the in vivo influence of sympathetic drive and circulating neurohormones; however, the assessment of intrinsic cardiac contractility obtained from the isolated heart preparation is void of these influences. Additionally, significant in vivo increases in LVEDP would produce a functional dilatation (i.e., operating on a higher portion of the P-V curve), as opposed to the structural dilatation achieved by remodeling in the Fist-P(−) group. This is supported by data obtained in a preliminary study of Fist-P(+) rats at 2-wk postfistula. Although systolic arterial pressures were not different in these animals, LVEDPs were elevated to 12–17 mmHg in Fist-P(+) rats vs. 5–8 mmHg in Sham rats. Also, the maximum change in pressure with respect to time was found to be increased to ∼8,500 mmHg/s in Fist-P(+) rats vs. ∼6,500 mmHg/s in Sham rats. Thus several factors influencing in vivo function, including contractility, sympathetic tone, preload, and changes in the ventricular M/V, are all increased in the Fist-P(+) group, and these, together with the elevated heart rate, result in the increased cardiac output.
The soy-derived isoflavone genistein has demonstrated anti-inflammatory effects and alters various cell signaling pathways. In addition to its antioxidant effect (20, 22), genistein inhibits several key signal transduction proteins, including mitogen-activated protein kinase and protein tyrosine kinases (2, 21, 30), as well as the NF-κB pathway (24). Of particular interest to our cardiac remodeling studies are the effects of genistein on transforming growth factor-β (TGF-β) (14). TGF-β is involved in the development of various cardiovascular fibrotic disorders and in the regulation of collagen synthesis and myofibroblast formation (1, 29). Additionally, reduced serum levels of TGF-β were associated with advanced atherosclerosis in humans (10). Exposure to genistein stimulated TGF-β production in epithelial cells, and it is worth noting that, unlike many in vitro studies regarding the effects of isoflavones, this TGF-β effect was induced by using physiologically attainable levels of genistein (8, 14, 15, 22). Genistein has been shown to provide protection from endotoxin-induced organ damage in both rats and mice (26, 27). In mice, dietary soy supplementation suppressed the inflammatory response to endotoxin, and further in vitro studies determined that genistein attenuated IL-6 secretion and signaling pathways (26). The anti-inflammatory properties of soy compounds represent a possible cardioprotective mechanism, as we have identified that the progression of adverse remodeling in fistula-induced heart failure is set in motion by the release of inflammatory compounds from mast cells in the myocardium (5, 12). Observations in our laboratory and by others (3) that proinflammatory cytokines are critically involved in modulating the initial myocardial remodeling in response to insult (e.g., increased LVEDP or wall stress) suggests that a reduction or prevention of an inflammatory response by phytoestrogenic compounds, such as genistein, provides cardioprotection and prevent the development of adverse ventricular dilatation. In addition, isoflavones may attenuate ventricular dilatation and increased compliance by reducing the expression and/or activation of matrix metalloprotienases (MMPs), which degrade the extracellular matrix. Genistein has been shown to inhibit the transcription of MMP genes (17, 19), prevent the activation of MMP-2 and -9, and increase the message of tissue inhibitors of MMP-1 (31). However, it should be noted that these findings were in cancer cell lines and not cardiac tissue.
The underlying mechanisms responsible for the lower incidence of cardiac disease in premenopausal women are not understood. Dietary modification and supplementation may represent a possible alternative to hormone replacement therapy, but further studies are needed to determine the beneficial compounds involved and specific mechanisms of action. This is the first study to demonstrate that dietary-derived phytoestrogens can have marked effects on cardiovascular remodeling. These findings call into question the results of many cardiac remodeling studies regarding gender differences in which dietary phytoestrogen content was not taken into account. Also, they indicate that future studies to assess gender differences in cardiovascular remodeling must consider the influence of diet on the resulting response of the vasculature and myocardium. From our previous study, it is clear that a diet high in phytoestrogens does not prevent male rats with fistula from developing CHF (9). However, phytoestrogenic compounds have been shown to be effective in males in reducing cardiac injury (13), and male rats on a P(−) diet might experience significantly more adverse remodeling in response to cardiovascular insult. These findings emphasize the importance of dietary considerations in biomedical research.
In summary, dietary phytoestrogens prevented LV dilatation and increased compliance developed in intact female rats in response to AV fistula induced chronic volume overload. Diastolic and systolic cardiovascular function in rats fed a diet with phytoestrogens was not significantly different than that of Sham rats. Dietary phytoestrogen modulated remodeling and allowed these animals to develop a stable compensated state by normalizing LV wall stress through hypertrophy that was appropriate relative to LV chamber size. Although the isoflavone genistein remains a likely candidate, further studies are warranted to determine whether genistein is indeed responsible for the observed effects on cardiac remodeling and to elucidate the cellular pathways involved.
This study was supported in part by National Heart, Lung, and Blood Institute Grant 1R01 HL-073990 (J. S. Janicki) and American Heart Association National Affiliate Grant 0435298N (J. D. Gardner) and Southeast Affiliate Grant 0160195B (J. D. Gardner).
We are grateful to Kimberly Jagnandan and Kathryn A. Oberle for providing technical support.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 the American Physiological Society