| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 2006;48:579.)
© 2006 American Heart Association, Inc.
Original Articles |
From the Department of Medicine (V.J., K.H., S.K.G.M., G.E., T.P.), University of Würzburg, Würzburg, Germany; Pharmaceutical and Medicinal Chemistry (P.K., E.B., R.W.H.), Saarland University, Saarland, Germany; Schering AG (K.-H.F., K.P., C.H.-H.), Berlin, Germany; and the Division of Cardiology (L.N.), University of Manchester, Manchester, United Kingdom.
Correspondence to Theo Pelzer, Department of Medicine, University of Würzburg, Josef-Schneider Str 2, D-97080 Würzburg, Germany. E-mail pelzer_t{at}klinik.uni-wuerzburg.de
| Abstract |
|---|
|
|
|---|
expression was lower in intact and in ovariectomized senescent compared with young SHRs and increased with estradiol substitution in aged rats. Plasma estradiol and estrone levels were lower not only in sham-operated but surprisingly also in E2-substituted senescent SHRs and associated with a reduction of hepatic 17ß-hydroxysteroid dehydrogenase type 1 enzyme activity, which converts weak (ie, estrone) into potent estrogens, such as E2. Aging attenuates the antihypertrophic effect of estradiol in female SHRs and is associated with profound alterations in cardiac estrogen receptor-
expression and estradiol metabolism. These observations contribute to explain the lower efficiency of estrogen substitution in senescent SHRs.
Key Words: estrogen receptors cardiac hypertrophy aging hypertension
| Introduction |
|---|
|
|
|---|
35% to 50%.2 But controlled clinical end point studies, such as Heart and Estrogen/progestin Replacement Study (HERS) and the recently terminated Womens Health Initiative (WHI) trial, reported no protective function of HRT consisting of conjugated equine estrogens plus medroxy-progesterone-acetate for the primary or secondary prevention of coronary artery disease.3 In contrast, estrogen supplementation in animal models of human heart disease was mostly associated with beneficial hormone effects. But the majority of these studies were conducted in young animals and, thus, do not take the possibility of aging effects into account, which might, in part, explain controversial results between animal and human studies.46 Until recently, these shortcomings could be attributed to the lack of animal models that mimic postmenopausal hypertension. Meanwhile, senescent spontaneously hypertensive rats (SHRs) and aged Dahl salt-sensitive rats have been identified and characterized as suitable models to study postmenopausal hypertension. As reported recently, female SHRs stop cycling at the age of 10 to 12 months and develop further increases in blood pressure thereafter.7 Similar studies have been conducted in Dahl salt-sensitive rats.8 Increased blood pressure in postmenopausal SHRs is associated with the natural cessation of ovarian function and decreasing estrogen plasma levels, which closely resemble observations in postmenopausal women.9 These studies provided substantial and new insight into gender and aging aspects of hypertension. In contrast, aging aspects in the development of cardiac hypertrophy, gene expression, and function are less well documented, although it seems conceivable that estrogen substitution of young and postmenopausal female SHRs might result in different cardiac phenotypes. Therefore, and because most previous studies including our own were conducted in young animals, we determined cardiovascular function, cardiac gene expression, as well as plasma sex hormone profiles in sham-operated and ovariectomized young and senescent female SHRs on long-term treatment with placebo or identical, body weightadjusted doses of 17ß-estradiol (E2). In particular, we speculated whether the efficacy of estrogen substitution to modulate cardiac hypertrophy and gene expression might be blunted in aged SHRs. | Methods |
|---|
|
|
|---|
Hemodynamic Analysis
Hemodynamic measurements were performed according to published protocols after 6 weeks of continuous treatment under light isoflurane anesthesia and spontaneous respiration (isoflurane 1.5 vol percent supplemented by 0.5 L of oxygen per minute).10 Pressure curves were measured via polyethylene tubing and a microtip manometer (Millar Instruments) calibrated to midchest level. Systolic and diastolic blood pressure measurements were obtained on catheter placement in the thoracic aorta. A calibrated flowmeter (2.5 mm; Statham) was placed around the ascending aorta for continuous measurement of aortic blood flow (cardiac output). Measurements were performed by a trained observer blinded for treatment groups. Animals with nonphysiological heart rate <250 bpm were excluded from hemodynamic analysis (senescent rats: sham: n=3; ovx: n=2; ovx+E2: n=2). Blood pressure in senescent SHRs was also monitored by telemetry (online supplement II, please see http://hyper.ahajournals.org).
Morphometry
Body weight, heart weight, uterus weight, and tibia length were determined after hemodynamic analysis. Relative heart weight was determined by normalizing absolute heart weight for tibia length.
Plasma Hormone Level
Plasma samples obtained after hemodynamic analysis and within identical time frames after hormone injection were used for radioimmunoassay (RIA) measurements of E2, estrone (E1), total testosterone, free-testosterone, and 4-androstendione according to the manufacturers instructions (Diagnostic Systems Laboratories, Inc). Cross-reactivity of the third-generation estradiol RIA was 6.9% with E1 and with other hormones was not detectable. Cross-reactivity of the E1 RIA was 1.25% with estradiol and with other hormones was not detectable. Cross-reactivity for androstenedione RIAs with dihydrotestosterone was 0.08%, with E1 0.03%, and with E2 and testosterone not detectable. Cross-reactivity for testosterone RIAs with dihydrotestosterone was 6.6%, with androstenedione 0.9%, and with E2 not detectable. Information on RIA cross-reactivities were provided by the manufacturer.
Cardiac Protein Expression
Cardiac
and ß myosin heavy chain protein (
- and ß-MHC) expression was analyzed by SDS-PAGE electrophoresis with subsequent visualization of protein bands by silver staining according to published protocols.10 Left ventricular samples were homogenized in ice-cold sample buffer containing a protease inhibitor mixture (Roche), centrifuged, and 500 ng of the extract were subjected to SDS-PAGE analysis using a 6% polyacrylamide gel containing 5% glycerol for 18 hours at 4°C before band visualization by silver staining (Silver Stain Plus, Bio Rad). Band intensities for
- and ß-MHC were determined by densitometry using the ScanPack 3.0 software (Biometra).
Cardiac protein expression was also analyzed by Western blots using unfractionated ventricular protein extracts, which were separated on 7% to 10% polyacrylamide sodium dodecyl sulfate gels followed by electrophoretic transfer on nitrocellulose membranes. Nonspecific background was blocked using 5% nonfat milk powder in PBS/Tween 20 (1 hour, room temperature). The following primary antibodies were used: anti-ER
(ER21, rabbit polyclonal; generous gift of Dr G. Greene, University of Chicago, Chicago, IL), anti-ERß (ER3919, rabbit polyclonal; raised against rat ERß ligand binding domain; see online supplement I for details), anti-phospholamban (mouse monoclonal, Alexis), anti-phospho-phospholamban (rabbit polyclonal, Upstate Biotechnology), anti-SERCA 2 (rabbit polyclonal, Abcam), and anti-
B crystallin (rabbit polyclonal, StressGen). Anti-rabbit and anti-mouse secondary antibodies conjugated with horseradish peroxidase (anti-mouse, anti-rabbit; Amersham, 1:5000) and the ECL detection system (Amersham) were used to visualize immunoreactive proteins. Gel loading was normalized to
-B-crystallin expression.
17ß-HSD1 and 17ß-HSD2 Activity Assays
Frozen liver samples were homogenized and fractionated into cytosolic extracts for measurements of 17ß-HSD1 activity and microsomal extracts to assess 17ß-HSD2 activity.11,12 Enzymatic 17ß-HSD activities were measured according to published protocols at 37°C in a phosphate buffer supplemented with 20% glycerol (vol/vol) and 1 mmol/L of EDTA the presence of oxidized nicotinamide-adenine dinucleotide and [3H]-estradiol for 20' in 17ß-HSD2 assays or in the presence of reduced nicotinamide-adenine dinucleotide and [3H]-E1 for 10' in 17ß-HSD1 assays followed by steroid extraction with ether.1315 [3H]-Estradiol and [3H]-E1 were obtained from Perkin Elmer LAS. Substrates and products were separated using acetonitrile/water (47.5:52.5 vol/vol) as a mobile phase on a C18 rp chromatography column (Nucleodur C18 Gravity, 3 µm, Macherey-Nagel) connected to a high-performance liquid chromatography system (Agilent 1100 Series, Agilent Technologies). Steroid detection and quantification was performed using a radioflow detector (Berthold Technologies). Substrate conversion rates were calculated, and the formation of 1 µm of product per minute was defined as 1 U of enzyme activity. Specific activities are given in units per milligram of protein.
Statistics
Statistical significance was calculated by 1-way ANOVA followed by StudentNewmanKeuls post hoc testing in all of the experiments except 17ß-HSD activities, in which statistical significance was determined by 2-sided t tests. Values are mean±SEM, and P<0.05 was considered significant.
| Results |
|---|
|
|
|---|
|
Hemodynamics and Telemetric Blood Pressure Analysis
Ovariectomy and E2 substitution did not affect blood pressure among the groups of aged and among the groups of young SHRs (Table). However, mean blood pressure levels were higher in sham-operated and estrogen-substituted senescent compared with young SHRs receiving estrogen substitution. Left ventricular stroke volume and cardiac output were comparable among all of the groups of senescent and young rats. Telemetric blood pressure analysis was in good agreement with invasive measurements and did not reveal a blood pressurelowering effect of E2 substitution in senescent SHRs (online supplement II).
|
Plasma Sex Hormone Levels
Plasma E2, E1, total, and free testosterone, as well as androstendione plasma levels, were significantly higher in sham-operated young compared with senescent SHRs (Figure 2). Ovariectomy decreased plasma levels of E2 and E1 in young rats but had no such effect in senescent SHRs. Plasma androgen levels were lower in ovariectomized SHRs of both age groups. Estradiol, as well as E1, levels were higher in young SHRs treated with E2 compared with placebo-treated ovariectomized animals. In contrast, plasma E2 and E1 levels in E2-substituted senescent SHRs were not different from placebo-treated aged rats. E1 and estradiol levels were closely correlated among all of the samples (P<0.001; R2=0.71).
|
Cardiac MHC Expression
The
-MHC protein was the predominant cardiac isomyosin in sham-operated young SHRs, whereas senescent hearts contained both myosin subtypes at approximately equal amounts (Figure 3). Ovariectomy decreased cardiac
-MHC expression in young SHRs and shifted the MHC ratio toward ß-MHC accumulation (sham: 1.74±0.12 versus ovx 1.07±0.07; P<0.01; n=10 per group). E2 substitution of ovariectomized young SHRs resulted in increased amounts of
-MHC protein (ovx+E2: 1.52±0.13; P<0.01; versus ovx; n=10 per group). In contrast, ovariectomy and E2 treatment conferred only insignificant effects on cardiac MHC expression in senescent SHRs.
|
Cardiac Protein Expression Analysis
Cardiac estrogen receptor
(ER
) expression was lower in sham-operated and ovariectomized senescent compared with young SHRs (Figure 4A). E2 substitution increased cardiac ER
expression in senescent rats to levels that were no longer different from those observed among young animals. In contrast, cardiac ERß expression was uniform and not different among all of the treatment groups (Figure 4B). The expression levels of cardiac calcium handling proteins such as SERCA2, phospholamban, and phosphorylated phospholamban were not different among all of the animals (data not shown).
|
17ß-HSD1 and 17ß-HSD2 Activity
Hepatic 17ß-HSD1 activity, which catalyzes the reduction of weak estrogens, such as E1, to potent estrogens, such as E2, was significantly lower in senescent compared with young SHRs receiving estrogen substitution (Figure 5A). Activity levels of 17ß-HSD2, which metabolizes E2 to less potent estrogens, such as E1, was comparable among estrogen-treated young and senescent SHRs (Figure 5B).
|
| Discussion |
|---|
|
|
|---|
expression, were responsive to E2 supplementation of senescent SHRs, heart weight and cardiac isomyosin expression were not. Lower E2 plasma levels in sham-operated and in estradiol-substituted senescent rats together with lower 17ß-HSD1 activity indicate that aging in SHRs is associated not only with reduced E2 synthesis but also with altered E2 metabolism. These observations could at least in part explain the blunted effect of E2 on cardiac hypertrophy in aged SHRs. Aging is associated with increased blood pressure and cardiac hypertrophy in postmenopausal women.1,9 Blood pressure levels were comparable and not different among all 3 of the groups of young rats and among the different groups of senescent SHRs, which is in line with previous findings, including our own.10,16 Telemetric blood pressure measurements in senescent SHRs were in good agreement with invasive blood pressure levels and again did not reveal a blood pressurelowering effect of estradiol. In line with previous observations, blood pressure differed, however, between some groups of aged and young SHRs, because systolic and mean blood pressure was higher in intact and in estrogentreated senescent SHRs compared with young rats receiving hormone substitution.17
Cardiac hypertrophy, which is frequently associated with hypertension, represents an independent predictor of cardiovascular mortality.18 Increased cardiac mass in sham-operated senescent compared with young SHRs could result from higher blood pressure levels in aged rats, although statistical significance was not reached in comparisons between treatment groups receiving identical treatment (ie, intact young versus intact senescent, etc). In line with previous observations, E2 efficiently attenuated cardiac hypertrophy in ovariectomized young SHRs, but blood pressure levels were not significantly different among all of the groups of young SHRs.10,19 These findings indicate that estrogens are able to attenuate cardiac growth also by direct and blood pressureindependent mechanisms, such as cardiac ANP expression and mitogen-activated protein kinase activation, as we and others have reported before.20 But in contrast to young animals, estrogen substitution failed to attenuate cardiac hypertrophy in aged SHRs. These observations are supported by measurements of cardiac isomyosin expression. Cardiac hypertrophy in rodents is associated with a prototypical switch in cardiac isomyosin expression from predominant
- to ß-MHC expression. As one would predict from measurements of cardiac mass, estrogen treatment reversed the downregulation of
-MHC in ovariectomized young but not in aged SHRs.
Estrogen effects on cardiac hypertrophy in senescent rats are at present not fully understood. Xu et al21 reported on increased left ventricular remodeling in ovariectomized adult SpragueDawley rats that was prevented by E2 substitution. However, the aim of these studies, which were conducted in normotensive rats, was clearly different from the present study and not designed to compare the efficiency of E2 substitution between young and adult rats.17 The role of estrogens in hypertension and cardiac hypertrophy in aged rats has also been evaluated by Hinojosa-Laborde et al,8 who provided solid evidence for increased blood pressure in ovariectomized compared with sham-operated Dahl salt-sensitive rats that was attenuated by estrogen substitution. But cardiac hypertrophy has so far not been assessed in E2-treated Dahl rats.22 Sharkey et al16 reported that absolute heart weight was not affected by E2 treatment of aged, heart failureprone hypertensive rats (SHHF/Mcc-facp). However the focus of these studies, which were conducted in intact and heart failureprone rats, was on young animals. Moreover, the lack of complete hemodynamic data, estrogen plasma levels, and estrogen receptor expression did not allow for a more comprehensive analysis and a direct comparison between E2-supplemented young and senescent rats.
General inefficiency of estrogen treatment does not exist in aged rats, because established biomarkers of estrogen activity, such as uterine weight and body weight, responded to hormone treatment in young and aged SHRs as observed here and reported before.21 Although uterus weight decreased with ovariectomy in young and in senescent SHRs, the loss of uterine mass was less pronounced in senescent SHRs, which supports the hypothesis that uterus mass in sham-operated aged rats is less representative for estrogen action.
Aging is associated with profound alterations of male and female sex hormone plasma levels in humans and in rodents. Moreover, heart weight and body weight, as well as blood pressure, are subjected to regulation via estrogens and androgens, including their pharmacologically active metabolites.23,24 As expected, plasma estradiol levels were lower in ovariectomized SHRs and restored to physiological levels in E2-treated young SHRs. Lower plasma E2 levels in intact senescent compared with young SHRs match with previous reports and with decreased plasma levels of E1. E1 is generated by different 17ß-steroid dehydrogenases that catalyze the conversion of potent into less potent estrogens in an initial step of E2 metabolism.25,26 Surprisingly, plasma samples from estrogen-substituted senescent SHRs, which were taken within identical time frames after hormone injection in young and aged animals, revealed low E2 levels that were indistinguishable from ovariectomized and placebo-treated senescent SHRs. These results, which are supported also by low E1 plasma levels in E2-substituted senescent SHRs, suggest that estrogen metabolism differs between young and senescent SHRs. Interestingly, E2 plasma levels were also lower in aged compared with young Dahl salt-sensitive rats receiving E2 supplementation via silastic implants (30±2 versus 20±3 ng/mL; young versus old; P<0.05). But these findings in 12-month compared with 4-month-old rats were not explicitly interpreted as an indicator of increased E2 metabolism.8 The present study, thus, substantiates previous findings and provides completely independent evidence for aging-associated alteration in estrogen metabolism.
Estrogen metabolism is initiated by 17ß-HSDs, which are expressed in multiple isoforms in multiple tissues and catalyze not only the conversion of potent into less potent estrogens but also opposite reactions yielding high affinity from low-affinity ER ligands. Oxidative 17ß-HSDs, which include isoforms 2, 4, 8, 10, and 11, convert high-affinity to low-affinity estrogen receptor ligands.25,27,28 Comparable hepatic 17ß-HSD2 activity levels among E2-substituted young and senescent SHRs suggest that 17ß-HSD2 is most likely not responsible for lower estradiol plasma levels in estradiol-substituted senescent SHRs. However, we cannot rule out that oxidative 17ß-HSD isoforms other than 17ß-HSD2 might show increased activities in senescent SHRs. In contrast to oxidative enzymes, reductive 17ß-HSDs, such as the type 1 enzyme, catalyze the synthesis of potent, high-affinity estrogen receptor ligands, such as E2, from low-affinity precursors, such as E1. E1 plasma levels in senescent SHRs provide a suitable substrate pool for 17ß-HSD1 to catalyze the formation of E2, and it is therefore conceivable that lower hepatic 17ß-HSD1 activities in senescent compared with young SHRs might have contributed to lower E2 plasma levels. Although the extent to which 17ß-HSD1 regulates estradiol plasma levels has up to now not been determined, different hepatic 17ß-HSD1 activity profiles in young and senescent SHRs further support the idea that aging is associated with functionally important alterations in estrogen metabolism.
The biological function of estrogens is mediated by 2 different estrogen receptor subtypes, ER
and estrogen receptor ß (ERß), which are functionally expressed in numerous tissues and cell types, including cardiac myocytes and vascular cells.29,30 In contrast to previous studies, which failed to detect measurable amounts of ER
and ERß protein in the mouse heart, several independent studies reported on robust cardiac expression levels of both ER subtypes in rats and humans.10,31,32 However, cardiac expression levels of ER
and ERß may not only vary between different species but also among young and senescent SHRs. Therefore, it is interesting to note that cardiac ER
expression was significantly lower in sham-operated and ovariectomized aged rats, whereas ERß was detected at comparable amounts in the heart of young and senescent rats. The observation that estradiol upregulates ER
expression in vascular cells suggests that lower plasma E2 levels might result in a downregulation of cardiac ER
expression.33 This hypothesis is supported by increased cardiac ER
content in E2 supplemented senescent SHRs; although plasma E2 levels remained low in these rats, ER
-mediated signal transduction seems, at least in part, to be functional in aged rats.
Perspectives
The current study provides first evidence for lower efficacy of estrogen substitution and altered E2 metabolism in senescent SHRs. Estrogen substitution in postmenopausal women is frequently initiated during the transition phase to menopause. Because SHRs stop cycling at
12 months of age, estrogen substitution in ovariectomized 24-month-old senescent SHRs does not resemble estrogen replacement at an early postmenopausal stage. Further studies will, thus, be required to determine whether initiation of estrogen replacement in SHRs during the decline of endogenous estrogen plasma levels is more efficacious than in senescent SHRs. Further studies will also be required to determine the functional relevance of individual 17ß-HSD isoforms that have not yet been studied in young and in senescent SHRs.
| Acknowledgments |
|---|
T.P. has received support from the Interdisciplinary Center for Clinical Research Würzburg ([IZKF] TP F-11); V.J. received support from the German Academic Exchange Service (DAAD); S.K.G.M. received support from the Deutsche Forschungsgemeinschaft "DFG" (Ma2252); L.N. has received support from the Medical Research Council (United Kingdom), the British Heart Foundation, and the EUMORPHIA consortium (European Union framework program 5); and G.E. has received support from the Deutsche Forschungsgemeinschaft "DFG" (SFB 688).
Disclosures
T.P. and L.N. have received financial support from the Schering AG Berlin that relates to the evaluation of novel steroid hormone receptor ligands.
Received February 23, 2006; first decision March 8, 2006; accepted July 27, 2006.
| References |
|---|
|
|
|---|
Related Article:
This article has been cited by other articles:
![]() |
V. Jazbutyte, P. A. Arias-Loza, K. Hu, J. Widder, V. Govindaraj, C. von Poser-Klein, J. Bauersachs, K.-H. Fritzemeier, C. Hegele-Hartung, L. Neyses, et al. Ligand-dependent activation of ER{beta} lowers blood pressure and attenuates cardiac hypertrophy in ovariectomized spontaneously hypertensive rats Cardiovasc Res, March 1, 2008; 77(4): 774 - 781. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Murphy and C. Steenbergen Gender-based differences in mechanisms of protection in myocardial ischemia-reperfusion injury Cardiovasc Res, August 1, 2007; 75(3): 478 - 486. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Hinojosa-Laborde and M. L. Lindsey Aging Modifies the Cardiac Response to Estrogen: A New Dimension to Hormone Replacement Therapy Hypertension, October 1, 2006; 48(4): 558 - 559. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |