|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 2003; 41:666-673, doi:10.1016/S0735-1097(02)02860-7 © 2003 by the American College of Cardiology Foundation |
,*
* Department of Clinical Methodology and Clinical-Surgical Technologies, University of Bari, Bari, Italy
Department of Human Anatomy and Physiology (Section of Anatomy), University of Padova, Padova, Italy
Institute of Clinical Pharmacology and Toxicology, Freie University of Berlin, Berlin, Germany
Department of Clinical and Experimental Medicine, Clinica Medica 4, University of Padova, Padova, Italy
Manuscript received June 19, 2002; accepted October 17, 2002.
* Reprint requests and correspondence: Dr. Gian Paolo Rossi, Dept. of Clinical and Experimental Medicine, Clinica Medica 4, University Hospital, via Giustiniani, 2, 35126 Padova, Italy.
gianpaolo.rossi{at}unipd.it
 |
Abstract |
|---|
 Top Abstract Materials and methodsResultsDiscussionReferences |
|---|
BACKGROUND: Angiotensin II (Ang II) can induce cardiac fibrosis, but the underlying mechanisms are incompletely understood.
METHODS: Four-week-old transgenic (mRen2)27 rat (TGRen2) received for four weeks a placebo, the mixed ETA/ETB endothelin receptor antagonist bosentan, the angiotensin II type I receptor (AT-1) antagonist irbesartan, the ETA endothelin receptor antagonist BMS-182874, and a combined treatment with irbesartan plus BMS-182874. We measured collagen density on Sirius red–stained serial sections of the left ventricle (LV) with a photomicroscope equipped with specific software and assessed the gene expression of procollagen 
1(I), atrial natriuretic peptide (ANP), transforming growth factor-beta 1 (TGFß1), endothelin converting enzyme, and ETB receptor.
RESULTS: In the placebo group, hypertension was associated with LV hypertrophy and cardiac fibrosis (LV weight: 4.0 ± 0.3 mg/g body weight; collagen density: 2.21 ± 0.16%), which were all prevented with irbesartan (2.3 ± 0.1, 1.30 ± 0.13, p < 0.001), but not with BMS-182874 (4.0 ± 0.2, 2.41 ± 0.22). Bosentan also prevented fibrosis (1.39 ± 0.18) but not hypertension and LV hypertrophy (3.38 ± 0.27). Combined irbesartan and BMS-182874 treatment prevented LV hypertrophy (2.9 ± 0.1) but not fibrosis (2.52 ± 0.16). Collagen density correlated (r = 0.414, p < 0.05) with plasma aldosterone levels. In TGRen2 with LV hypertrophy, the gene expression of ANP and ETB but not that of TGFß1 and procollagen 1(I) was increased.
CONCLUSIONS: In Ang II–dependent hypertension, cardiac fibrosis was associated with LV hypertrophy and was hindered by both mixed ETA/ETB blockade and AT-1 blockade. Only the latter treatment prevented both hypertension and LV hypertrophy. Thus, there is a dissociation between the mechanisms of cardiac fibrosis and hypertension, which do and do not entail ET-1, respectively.
| ||||||||||||||||||||||||||
The renin-angiotensin-aldosterone system (RAAS) is deemed to play a pivotal role in regulating the deposition of extracellular matrix in the heart and causing cardiac fibrosis. In vitro experiments on fibroblasts initially showed that both angiotensin II (Ang II) and aldosterone concentration-dependently promoted collagen synthesis; in vivo studies in experimental models of hypertension with either an excess mineralocorticoid activity or an activated RAAS subsequently confirmed these findings (1,2).
The transgenic (mRen2)27 rat (TGRen2) features an overexpression of the Ren2 transgene in several tissues, with ensuing increased Ang II concentrations (3); accordingly, it is regarded as a paradigm of severe Ang II–dependent hypertension and cardiovascular disease. In the myocardium, the amount of Ren2 messenger ribonucleic acid (mRNA) correlates not only with the transgene dose (4) but also with the content of Ang II, which might act in a paracrine fashion to cause both cardiac hypertrophy and fibrosis. Thus, the high tissue Ang II levels in the myocardium might interact synergistically with the stress imposed by the high afterload in stimulating protein synthesis and cardiomyocyte and fibroblast growth (2). Because cardiac fibrosis was found to correlate with impaired diastolic function more closely than cardiac hypertrophy (5,6), it is likely to play a major role in impairing cardiac function.
In the adrenal cortex, expression of the Ren2 transgene and high tissue Ang II (7) lead to enhanced aldosterone secretion (8) that, along with Ang II (3,9), can enhance collagen synthesis, thus inducing a progressive collagen accumulation with aging (10). Nonetheless, it remains uncertain if and by which mechanisms cardiac fibrosis occurs in TGRen2 (10,11).
The existence of interactions between the RAAS and endothelin (ET)-1 (12) suggests an involvement of the latter in cardiac fibrosis. Angiotensin II activates the transcription of the preproET-1 gene in different cell types of the heart, including myocytes, vascular smooth muscle, and endothelial cells (13). Thus, the high cardiac tissue Ang II content might increase locally the ET-1 synthesis. Endothelin-1 activates the procollagen I promoter (14) and collagen synthesis in fibroblasts; it can modulate collagenase activity, although it is unclear which ET receptor subtype mediates these effects (15–18). Moreover, ET-1 potently stimulates, via ETB receptors, aldosterone release (19) and therefore might increase collagen synthesis via enhanced aldosterone secretion. Thus, we investigated the deposition of myocardial collagen and the gene expression of procollagen 1(I), ETB, transforming growth factor-beta 1 (TGFß1), endothelin converting enzyme (ECE), and atrial natriuretic peptide (ANP) in TGRen2, an obvious experimental model in which to test this hypothesis. We also assessed the effects of selective blockade of the angiotensin II type I receptor (AT-1), ETA, and ETB receptor subtypes on myocardial collagen deposition.
|
Materials and methods |
|---|
|
TopAbstractMaterials and methodsResultsDiscussionReferences |
|---|
Each treatment was given individually to each rat as a chocolate-flavored agarose gel tablet. The latter was cut into pieces of proper size in order to obtain the amount required to provide each individual rat with its BW-tailored daily dosage of drug (3). The study protocol and animal handling followed our institutional guidelines for animal studies and were previously reported (20). At weekly intervals, we measured BW and systolic BP levels (by a tail-cuff method, as the mean value of at least three readings). When the rats were eight weeks old, they were weighed and euthanized by cervical dislocation. The hearts were quickly removed and weighed, then the left ventricle (LV) was rapidly dissected from atria and right ventricle, weighted, snap-frozen in isopentane precooled on dry ice, and stored in liquid nitrogen until analyzed. After removal of the heart and lungs, blood was collected into heparinized tubes for the measurement of plasma aldosterone, as reported (3).
Collagen density and types. 5-µm equatorial and serial sections obtained from the LV tissue were stained with Sirius red (0.5% Sirius red F3BA in saturated picric acid) to visualize fibrillar collagen. Quantitative analysis was performed blindly by a single examiner (T.M.S.), using a photomicroscope Leica DM equipped with QWin Standard Leica Image Software. To minimize operator-dependent variability, a specific routine was used to automatically detect chromatic tonalities corresponding to Sirius red–stained collagen fibrils. Views were randomly selected from each section. Ten random views were captured and analyzed for each section at magnification x10, because a pilot study showed that they represented accurately the entire section. To estimate perivascular fibrosis, at least one vessel with external diameter between 50 and 150 µm was also included in 4 of the 10 views considered for each section. All these measurements were performed on at least four sections for each rat. We estimated collagen density by measuring in each field the percent of total surface area pertaining to fibrillar collagen, identified as specified earlier (21). This morphometric approach for measurement of fibrillar collagen within the cardiac interstitium was previously validated (22); collagen density assessed by this technique was also shown to correlate with hydroxyproline concentration in the LV (23,24).
For measurements of the media cross-sectional areas of the cardiac arterioles, sections of four to six arterioles (ranging between 50 and 250 µm in diameter) for each rat were cut across their transverse axis and examined at x20 magnification. The aforementioned software was used to automatically compute all direct measurements.
Polarization microscopy was used to measure strongly (red-orange) and weakly (greenish-yellow) birefringent fibers in picrosirius-stained sections (25). These fibers roughly correspond to collagen type I, 1.6 to 2.4 µm diameter-thick, and collagen type III, less than 0.8 µm diameter-thick fibers, respectively (26). Specific routines were created with the QWin Standard Leica Image Software and were used to measure total and red-orange or greenish-yellow birefringent areas. The latter were expressed as percent of the total area examined in each section, as described previously.
Immunohistochemistry to laminin
Serial sections obtained from the LV tissue were also immunostained with a mouse anti-laminin monoclonal antibody (MAB 1920, Chemicon International, Temecula, California) in a Ventana Full Nexes IHC Unit (Ventana, Strasbourg, France) using biotin-streptavidin-conjugated AffiniPure donkey antimouse immunoglobulin G (Jackson ImmunoResearch, ListarFish, Carugate, Italy). The extent of immunostaining was scored blindly from 1 to 4 by two independent observers.
Gene expression studies
Total RNA was extracted according to the Trizol protocol (Gibco-Life Technology, Grand Island, New York). Gene expression for ETB, TGFß1, ECE, and ANP was assessed by Northern blot analysis, as described (27). These experiments were carried out in the Department of Pharmacology and Toxicology of the Freie University of Berlin.
Gene expression for procollagen 1(I) was measured by real-time fluorescence reverse transcriptase/polymerase chain reaction. Primers for this gene (Access code: Z78279) were selected with Primer3 software: Forward: GGCAACAAAGGAGACACTG, Reverse: CAACACCATCAGCACCAG. The housekeeping gene GAPDH (Access code: AF106860) was amplified in parallel with the following primers: Forward: CCCTTCATTGACCTCAACTA, Reverse: GCCAGTGAGCTTCCCGTTCA and used for normalization purposes. Quantification of mRNA levels was carried out using a thermal cycler (iCycler, Biorad, Milan, Italy) equipped with an online fluorescence-based detection system of amplification products. Melting curve analysis was used to confirm the specificity of the amplification products.
Statistical analysis
Results are expressed as mean ± SEM. Comparisons of active treatments with placebo were performed with non-parametric Mann-Whitney U test. Plasma aldosterone was analyzed after log transformation. To investigate the relationship between collagen density and plasma aldosterone or total birefringence, and between plasma aldosterone and selective red-orange or greenish-yellow birefringence, we used Pearson correlation coefficient and a stepwise regression analysis (backward method with an inclusion cutoff value of 0.05). A p value <0.05 was considered statistically significant. Analyses were carried out with the SPSS for Windows statistical package (version 10.0, SPSS Inc., Chicago, Illinois).
|
Results |
|---|
|
TopAbstractMaterials and methodsResultsDiscussionReferences |
|---|
|
|
|
|
|
Immunostaining to laminin
The immunostaining to laminin was located in the plasma membranes of cardiomyocytes and in the subendothelial lining. Results of the semiquantitative assessment showed no discernible differences between experimental groups (Table 1).
Gene expression studies
Results of Northern blots are summarized in Table 1. The ETB exhibited a significant decrease with irbesartan compared with BMS-182874. No significant difference between groups was observed for mRNA levels of ANP, TGFß1, ECE, and procollagen 1(I).
|
Discussion |
|---|
|
TopAbstractMaterials and methodsResultsDiscussionReferences |
|---|
Our findings, with a painstaking operator-independent quantitative evaluation of collagen density on serial sections of the LV, fully support the conclusion that fibrillar collagen deposition occurs at an early stage of cardiac hypertrophy in the LV of TGRen2 (27).
Polarization microscopy evaluation of the sections demonstrated that Sirius red–stained fibrillar collagen mostly represents collagen types I and III, as indicated by the close correlation between collagen density and total birefringence in each section. Furthermore, accumulation of fibrillar collagen was largely accounted for by thick, strongly birefringent red-orange fibers that correspond to a large extent to collagen type I. In contrast, thin, weakly birefringent greenish fibers, mostly entailing collagen type III, were predominantly found in those hearts that did not show overt cardiac fibrosis, such as those of the irbesartan and bosentan groups. The relative abundance of these types of fibers and collagens might be of utmost importance in determining the diastolic properties of the heart. Of note, we did not find any differences in the amount of immunostaining to laminin between our experimental groups, suggesting that changes of this basement membrane protein do not play a major role in cardiac fibrosis.
Mechanisms of cardiac fibrosis.
We used a combined strategy based on gene expression experiments and on pharmacologic intervention with drugs capable of providing specific blockade of the AT-1 or of the ETA and ETB receptors to gain information on their potential involvement and on the mechanisms underlying cardiac fibrosis. Our gene expression experiments showed an increased expression of ANP mRNA and a significant correlation between ANP mRNA and LV mass index, which accord well with previous findings (11,21), as well as with the contention that this gene is a marker of cardiac hypertrophy. We also found an increase of ETB mRNA in the groups with overt LV hypertrophy (Table 1). The ETB mRNA correlated significantly with the media cross-sectional area of the intracardiac arterioles (r = 0.636, p = 0.001) and with LV mass index. Accordingly, an increased ETB gene expression, which might therefore be due to an increase in vascular endothelial cells, could also be a marker of cardiac hypertrophy. The expression of the TGFß1 gene was previously found to occur with the development of cardiac hypertrophy, but only early and transiently (21,30). Hence, it is not surprising that its expression was not found to be enhanced under the chronic conditions of this study. The expression of procollagen 1(I) did not reflect the marked differences in collagen density that were seen between groups. Thus, it might be that the early activation of Ang II production due to the transgene expression from birth and the ensuing recruitment of the ET-1 system stimulate collagen synthesis only early on and that a blunted collagen-degrading activity may account for extracellular matrix deposition later on. We could not measure the activity of enzymes involved in collagen degradation, therefore this hypothesis remains to be conclusively proven.
We found that the AT-1 receptor antagonist irbesartan prevented both hypertension and cardiac hypertrophy. It has recently been contended that pressure overload is the only factor responsible for development of LV hypertrophy and cardiac fibrosis in TGRen2 (10). This proposal has been contradicted by a reevaluation of the data obtained with amlodipine in the same study (31) and is not supported by the present results with the different receptor antagonists. Our regression analysis showed that plasma aldosterone, besides BP levels, significantly predicted collagen density, thus supporting the view that there are multiple determinants of cardiac fibrosis. This conclusion agrees also with findings in another model of hypertension caused by inhibition of nitric oxide synthesis, where bosentan prevented the severity of glomerular and interstitial fibrosis in the kidney (14), even despite that there was no effect on the development of hypertension (32).
ET-1 and cardiac fibrosis
The role of ET-1 in Ang II–dependent hypertension remains unclear because there are studies implicating ET-1 in models of Ang II infusion (33,34) and others contradicting this contention in models with enhanced synthesis of endogenous Ang II, such as the 2 kidney 1 clip (2K1C) and the TGRen2 models (3,12). Nonetheless, in these latter conditions, as well as in post-myocardial infarction models of congestive heart failure (35), the ET system would appear to play a role in the progression of cardiovascular damage, including cardiac fibrosis and LV dysfunction (27). This detrimental role of ET-1 is deemed to be even more important in models with excess mineralocorticoid activity, where the vascular synthesis of ET-1 is enhanced (1). In deoxycorticosterone acetate (DOCA)–salt hypertensive rats, an ETA-selective antagonist prevented cardiac fibrosis and activation of procollagen 1(I) synthesis (21), thus implicating the ETA receptor subtype. Experiments in 2K1C rat with the ETA antagonist BQ-123 and with the ETB antagonist IRL-1038 led to the hypothesis that the ETB subtype might mediate cardiac fibrosis when the RAAS is activated (36). However, the implication of the ETB receptor in cardiac fibrosis remained to be conclusively demonstrated (37), because IRL-1038 was publicly retreated because of its inconsistent selectivity for the ETB receptor by those who synthesized it (38).
The present findings of prevention of cardiac fibrosis with the mixed ETA/ETB antagonist bosentan, but not with the ETA-selective antagonist BMS-182874, are consistent with previous findings in chronic heart failure (35) and point to the ETB as mediator of cardiac fibrosis in TGRen2. A role of ETB receptors in collagen deposition was previously shown in the liver (15,18). In addition, in chronic heart failure ETB blockade reduced cardiac fibrosis despite the lack of hemodynamic effects (18,39). The present findings in TGRen2 are consistent with this observation and shed new light on the role of this receptor subtype in cardiac fibrosis.
It must be acknowledged that our pharmacologic experiments provided evidence for a role of ETB receptors in cardiac fibrosis by "subtraction," i.e., by comparison of a mixed ETA/ETB with an ETA-selective antagonist. However, this is due to the current unavailability of an ETB-selective antagonist for chronic in vivo studies. Furthermore, our findings of an increased expression of the ETB gene in the heart of TGRen2 treated with the ETA-selective antagonist that exhibited marked cardiac fibrosis, compared with all other groups (Table 1), are consistent with a role of ETB receptors in cardiac fibrosis.
ETB receptor subtype, cardiac fibrosis, and aldosterone synthesis
Based on the fact that the correlation between cardiac ETB receptor gene expression and collagen density was weak and not significant, we would like to suggest, however, an alternative explanation, which does not postulate changes of ETB receptors in the heart. Endothelin-1 acts as a potent secretagogue of aldosterone via ETB receptors (19,40), and compelling evidence implicates both aldosterone and Ang II in cardiac fibrosis (2). Thus, it might be that the major mechanism leading to cardiac fibrosis entails an Ang II- and ET-1-driven stimulation of aldosterone secretion. The beneficial effect of bosentan and the lack of effect of BMS-182874 could be explained on this ground. The predominant perivascular distribution of cardiac fibrosis and the significant correlation between collagen density and plasma aldosterone levels (Fig. 2) also support a major role of blood-borne aldosterone. This contention might also reconcile the present results with those obtained with A-127722, an ETA-selective antagonist that lowered BP and prevented cardiac fibrosis in DOCA–salt hypertensive rats (21). In this latter model, the secretagogue effect of ET-1 on aldosterone is likely to be negligible because of the concomitant suppression of the RAAS induced by the exogenously administered salt and DOCA. In addition, the lowering of BP seen with A-127722 made it difficult to attribute the prevention of cardiac fibrosis only to ETA. We hypothesize that the relative importance of ETA and ETB receptors in causing cardiac fibrosis could markedly differ between experimental models, depending on the prevailing degree of activation of the RAAS and on the concomitant changes in BP. Thus, development of cardiac fibrosis in TGRen2 is likely to derive from a synergistic interaction of the RAAS and ET system. Results in another double transgenic model of severe hypertension and cardiac hypertrophy characterized by activation of RAAS support this view (41). The clinical implications of these findings for the prevention or progression of cardiac fibrosis with ET antagonists in patients with congestive heart failure, who might have an activated RAAS, are obvious.
In conclusion, the AT-1-selective antagonist irbesartan prevented hypertension, cardiac hypertrophy, and cardiac fibrosis in TGRen2. The mixed ETA/ETB antagonist bosentan, but not an ETA-selective antagonist, also hindered cardiac fibrosis despite having no effect on BP and cardiac hypertrophy. Both irbesartan and bosentan corrected the secondary hyperaldosteronism; furthermore, plasma aldosterone levels correlated significantly with collagen density in the LV myocardium. Thus, collectively these results indicate that although cardiac hypertrophy and cardiac fibrosis go hand in hand, the mechanisms regulating the two processes differ and could be differentially modulated pharmacologically.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMaterials and methodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. F. Opitz, R. Ewert, W. Kirch, and D. Pittrow Inhibition of endothelin receptors in the treatment of pulmonary arterial hypertension: does selectivity matter? Eur. Heart J., August 2, 2008; 29(16): 1936 - 1948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Whaley-Connell, J. Habibi, S. A. Cooper, V. G. DeMarco, M. R. Hayden, C. S. Stump, D. Link, C. M. Ferrario, and J. R. Sowers Effect of renin inhibition and AT1R blockade on myocardial remodeling in the transgenic Ren2 rat Am J Physiol Endocrinol Metab, July 1, 2008; 295(1): E103 - E109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. G. DeMarco, J. Habibi, A. T. Whaley-Connell, R. I. Schneider, R. L. Heller, J. P. Bosanquet, M. R. Hayden, K. Delcour, S. A. Cooper, B. T. Andresen, et al. Oxidative stress contributes to pulmonary hypertension in the transgenic (mRen2)27 rat Am J Physiol Heart Circ Physiol, June 1, 2008; 294(6): H2659 - H2668. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Pavlopoulos, J. Grapsa, E. Stefanadi, V. Kamperidis, E. Philippou, D. Dawson, and P. Nihoyannopoulos The evolution of diastolic dysfunction in the hypertensive disease Eur J Echocardiogr, April 29, 2008; (2008) jen145v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wong, F. E. Brennan, M. J. Young, P. J. Fuller, and T. J. Cole A Direct Effect of Aldosterone on Endothelin-1 Gene Expression in Vivo Endocrinology, April 1, 2007; 148(4): 1511 - 1517. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. A. Gould, M. Yii, M. D. Esler, J. M. Power, and D. M. Kaye Atrial fibrillation impairs cardiac sympathetic response to baroreceptor unloading in congestive heart failure Eur. Heart J., December 1, 2005; 26(23): 2562 - 2567. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. Thackaberry, Z. Jiang, C. D. Johnson, K. S. Ramos, and M. K. Walker Toxicogenomic Profile of 2,3,7,8-Tetrachlorodibenzo-p-Dioxin in the Murine Fetal Heart: Modulation of Cell Cycle and Extracellular Matrix Genes Toxicol. Sci., November 1, 2005; 88(1): 231 - 241. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Vaneckova, H. J. Kramer, A. Backer, Z. Vernerova, M. Opocensky, and L. Cervenka Early Endothelin-A Receptor Blockade Decreases Blood Pressure and Ameliorates End-Organ Damage in Homozygous Ren-2 Rats Hypertension, October 1, 2005; 46(4): 969 - 974. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Milliez, X. Girerd, P.-F. Plouin, J. Blacher, M. E. Safar, and J.-J. Mourad Evidence for an increased rate of cardiovascular events in patients with primary aldosteronism J. Am. Coll. Cardiol., April 19, 2005; 45(8): 1243 - 1248. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Pollock Endothelin, Angiotensin, and Oxidative Stress in Hypertension Hypertension, April 1, 2005; 45(4): 477 - 480. [Full Text] [PDF]
|
|
|
|
 |
|
 C. Tikellis, P. J. Wookey, R. Candido, S. Andrikopoulos, M. C. Thomas, and M. E. Cooper Improved Islet Morphology after Blockade of the Renin- Angiotensin System in the ZDF Rat Diabetes, April 1, 2004; 53(4): 989 - 997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G Perez, M. C Villa-Abrille, E. A Aiello, R. A Dulce, H. E Cingolani, and M. C Camilion de Hurtado A low dose of angiotensin II increases inotropism through activation of reverse Na+/Ca2+ exchange by endothelin release Cardiovasc Res, December 1, 2003; 60(3): 589 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.-H. Cheng, P.-Y. Cheng, N.-L. Shih, I.-B. Chen, D. L. Wang, and J.-J. Chen Involvement of reactive oxygen species in angiotensin II-induced endothelin-1 gene expression in rat cardiac fibroblasts J. Am. Coll. Cardiol., November 19, 2003; 42(10): 1845 - 1854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Q. Pu, M. F. Neves, A. Virdis, R. M. Touyz, and E. L. Schiffrin Endothelin Antagonism on Aldosterone-Induced Oxidative Stress and Vascular Remodeling Hypertension, July 1, 2003; 42(1): 49 - 55. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
