Biological Determinants of Aldosterone-Induced Cardiac Fibrosis in Rats
Abstract To determine the events leading to cardiac fibrosis in aldosterone-salt hypertensive rats, we studied protein and mRNA accumulation of procollagens I and III for 60 days. After 3 and 7 days of treatment systolic pressure was normal, and no histological or biochemical changes were seen in rat hearts. At day 15 arterial pressure was raised (+40%) and left ventricular hypertrophy was +15%. Cardiac examination after hemalun-eosin staining and immunolabeling with anticollagen I and III antibodies showed no structural alterations, but an 83% increase in right ventricular type III procollagen mRNA levels was found. At 30 and 60 days we found progressive cardiac fibrosis, with inflammatory cells, myocyte necrosis, and elevation of both types I and III procollagen mRNA levels in both ventricles. To determine whether aldosterone had effects on Na,K-ATPase that might lead to ionic disturbances and induce myocyte necrosis, we studied the major cardiac Na,K-ATPase isoform genes. Although Na,K-ATPase α1- and β1-subunit mRNA levels were elevated in kidney at day 1, neither of these cardiac transcripts nor the specific α2 isoform was altered between 1 and 15 days. These results show that accumulation of procollagen mRNAs occurs before collagen deposition. Cardiac alterations are late and not preceded by changes in Na,K-ATPase cardiac gene expression, precluding a direct modulation of cardiac collagen synthesis and Na,K-ATPase by aldosterone.
Long-term treatment of rats with aldosterone-salt results in arterial hypertension, left ventricular hypertrophy, and fibrosis in both the right and left ventricles.1 2 3 The effect of aldosterone on fibrosis development is blocked by low and high doses of spironolactone, even though the low dose does not reverse the hypertension and hypertrophy.4 These results argue for a predominant hormonal stimulus as inducer of the fibrous tissue response. We have previously shown that in addition to the fibrosis, which is histologically evident at 2 months of aldosterone treatment, the levels of cardiac types I and III collagen mRNAs are greatly increased,3 suggesting that the process of fibrosis is still ongoing at this time. In vitro, aldosterone has been reported to increase hydroxyproline incorporation into collagen of isolated rat cardiac fibroblasts within 24 hours,5 although these results could not be confirmed in another laboratory.6 Therefore, the relationship between aldosterone and cardiac fibrosis remains to be clarified. It is of interest to determine in vivo whether aldosterone is able to rapidly induce the accumulation of collagen mRNAs, which would suggest a direct control of collagen genes by the hormone. If collagen synthesis was not induced directly, ie, rapidly, by aldosterone, it would be important to identify the biological determinants involved in the cardiac fibrosis induced by aldosterone-salt treatment. Aldosterone upregulates the expression of major cardiac Na,K-ATPase isoform genes in isolated cardiomyocytes.7 Ionic disturbances arising from alterations of cardiac Na,K-ATPase activity and/or synthesis might explain myocyte necrosis and subsequent reparative fibrosis, as seen in the aldosterone-salt model.
We quantified the levels of types I and III collagen mRNAs over a 60-day period and followed the progression of cardiac fibrosis by histological examination and immunolabeling of types I and III collagens in both ventricles. We also measured the ventricular content of Na,K-ATPase α1-, α2-, and β1-subunit mRNAs between 1 and 15 days of treatment. The findings provide clear evidence for a pretranslational control of collagen accumulation in the aldosterone-salt model, but the late accumulation of collagen mRNAs precludes a direct action of aldosterone on fibroblasts. Unlike kidney, cardiac Na,K-ATPase gene expression was not modulated by aldosterone-salt treatment, suggesting that other mechanisms that remain to be defined are involved in the necrotic process.
Animal Models and Tissues
Male Sprague-Dawley rats weighing 150 to 170 g at the beginning of experiments were used. All sham-operated and treated rats were anesthetized (60 mg/kg pentobarbital) and uninephrectomized. For aldosterone-salt hypertensive rats an osmotic minipump (Alzet, Charles River, Paris, France) was subcutaneously implanted to deliver 0.75 μg/h d-aldosterone (Sigma Chemical Co) and changed every 15 days. Rats received 1% NaCl and 0.3% KCl in the drinking water. Sham-operated rats were also implanted with an osmotic minipump containing only physiological serum and received no salt in the drinking water. Nonoperated rats of the same age were used as controls. Sham-operated rats were not different from controls for any of the parameters studied. All rats were fed ad libitum with M25 pellets (Extralabo). All animal procedures followed were in accordance with institutional guidelines. Systolic pressure was determined by the tail-cuff method. Rats were treated for 3, 7, 15, 30, and 60 days. At the time of euthanasia the heart and kidney were removed, rinsed in ice-cold saline solution, blotted, and weighed. Left ventricles with septum were separated from right ventricles. The cardiac samples were stored at −70°C until use.
Plasma Thyroxine and Aldosterone Levels
Blood samples were collected at the time of euthanasia in tubes containing 0.5% EDTA. After centrifugation, plasma samples were collected and stored at −70°C. Plasma concentrations of aldosterone, free thyroxine (T4), and free triiodothyronine (T3) were measured by radioimmunoassay.
Total RNA Extraction
Total RNA from ventricles and kidney was prepared according to Chomczynski and Sacchi.8 RNA concentrations were measured by 260 nm absorbance, assuming 40 μg/mL for 1 absorbance unit. The RNAs were resuspended in 0.1% sodium dodecyl sulfate (SDS), and aliquots were stored at −70°C until use.
Northern Blots and Slot Blots
For Northern blots 20-μg samples of RNA were denatured in 50% formamide, 2.2 mol/L formaldehyde, and 1× morpholinopropanesulfonic acid buffer (pH 8. 0) and electrophoresed in a 1% agarose gel. Total RNA was then transferred to a Hybond-N membrane (Amersham). For dot blot analysis 1, 2, 4, and 10 μg RNA of each sample were spotted on the membranes.
Slots were hybridized with the following oligomers or cDNA probes: a rat collagen α1(I) cDNA of 1600 bp complementary to the carboxy-terminal propeptide9 and a rat collagen α1(III) cDNA containing 1300 bp of the 3′ noncoding and coding regions (kindly provided by Dr Vuorio).10 Another series of slots were hybridized with α isoform Na,K-ATPase–specific cDNAs (approximately 300 bp) and a β1-subunit Na,K-ATPase probe as described previously by Charlemagne et al.11 All slots were hybridized with a 24-mer oligonucleotide specific for rat 18S rRNA (Institut Pasteur) for relative quantitation as previously described.3 Oligonucleotides were 5′-end labeled with [32P]ATP (NEN) and T4 polynucleotide-kinase (Promega). Hybridizations with collagen and Na,K-ATPase cDNAs were carried out in 50% formamide, 5× Denhardt’s, 200 mg/mL herring sperm DNA, 5× SSC, 0.1% SDS, and 0.1 mol/L phosphate buffer (pH 7. 0) at 42°C. Slots were prehybridized in this solution for 12 hours and hybridized for 24 hours with added cDNA radiolabeled by random primer extension with an Amersham Megaprime DNA labeling system. [32P]dCTP (3000 Ci/mmol) was incorporated to obtain a specific activity of 2×108 to 8×108 counts per milligram. For collagen hybridization slots were washed in 0.1× SSC and 0.1% SDS at room temperature and twice more in the same solution at 50°C. For Na,K-ATPase hybridization slots were washed twice in 2× SSC and 0.1% SDS at room temperature and in 0.1× SSC and 0.1% SDS at 50°C. Membranes were exposed to Amersham Hyperfilm at −70°C with intensifying screens. Different exposures of all autoradiograms were done to ensure that the signal stayed within the linear range of densitometry. The relative amounts of mRNAs were quantified on slot blots by dividing the optical densities measured for the different probes by the optical density measured for the 18S probe. Collagen mRNAs were standardized against a synthetic RNA spotted on every slot. Standard synthetic RNA collagens were prepared after subcloning a collagen cDNA fragment in a transcription vector (pJLP 1, 1600-bp fragment in pGEM3 for type I collagen; pJLP 3, 1300-bp fragment in pGEM3 for type III collagen). After linearization of the plasmids by appropriate enzymatic digestion, RNA transcripts were synthesized with SP6 promoter generating 500-bp and 660-bp RNAs for types I and III collagens, respectively (Promega kit). The concentration of each collagen mRNA was expressed in picograms per microgram of total RNA after dividing values by a factor of 3.2 (1600/500=3.2) and 1.97 (1300/660=1.97) for types I and III collagens, respectively.
Ribonuclease Protection Assay
The level of the mRNA coding for the α3 isoform of Na,K-ATPase was studied by ribonuclease protection assay. Use of this sensitive technique was necessary given the extremely low abundance of this isoform in the adult heart. The antisense RNA complementary to nucleotides −51 to 331 bp was transcribed by T3 RNA polymerase in the presence of [32P]UTP (3000 Ci/mmol) with in vitro transcription kits (Promega). Labeled α3 RNA probe was purified on a polyacrylamide (5%)/urea (8 mol/L) gel. An amount of 0.1 ng of the probe with a specific radioactivity of 3×108 cpm/μg was hybridized with 25 μg total cardiac RNA. The assay was carried out as described (Kit Ambion, Cliniscience). The length of the protected fragment was 382 bp.
Histology and Histochemistry
Cardiac samples were included in mounting solution (Ystosystem) and frozen at −155°C in isopentane. Equatorial sections (5 μm) were cut in a cryostat at −20°C. One slide for each specimen was stained with hemalum-eosin for histological assessment, and the counterpart was used for double immunofluorescence of isocollagens I and III. The primary antisera to human skin type I collagen (Chemicon) and to rat skin type III collagen (Institut Pasteur) were raised in sheep and rabbit, respectively. Heart sections were first overlaid with 30 μL of type I collagen antiserum (1/100) and incubated overnight at 4°C. Sections were washed in 1× phosphate-buffered saline and then incubated at 37°C for 30 minutes with 30 μL of type III collagen antiserum (1/50). All sections were rinsed in phosphate-buffered saline and incubated for 30 minutes with biotinylated anti-sheep IgG (1/100). The biotinylated antibodies were revealed with streptavidin–Texas red complex (1/50). After washing, sections were incubated with anti-rabbit immunoglobulin antibodies conjugated to fluorescein isothiocyanate fluorochrome (Amersham).
Results are expressed as mean±SEM. Comparisons between groups were performed by the unpaired Student’s t test and Wilcoxon Mann-Whitney test after variance analysis. All probability values were two-tailed; a value of P<.05 was considered statistically significant.
Blood Pressure and Anatomic Data
Aldosterone-salt treatment induced a marked increase (+40%, P<.001) of systolic pressure at 15 days (Fig 1⇓). Systolic pressure continued to rise for 60 days compared with sham-operated control rats. As a consequence, the ratio of heart weight to body weight was increased by 15% (P<.05) as early as 15 days, reaching 42% at 30 days and 60% at 60 days of treatment (P<.001). In agreement with previous studies using the same model,1 2 3 hypertensive rats developed left ventricular hypertrophy.
Aldosterone administration at a rate of 0.75 μg/h induced a fourfold to fivefold increase in its plasma concentration above control (2160±525 versus 394±69 pmol/L measured at 3 days, P<.001). We and others have previously shown that aldosterone-salt treatment is also accompanied by a decrease in plasma concentrations of thyroid hormones.3 12 The Table⇓ shows that both T4 and T3 plasma concentrations decreased as early as 15 days and remained low thereafter.
Histological and Immunologic Examination
Equatorial heart sections including left and right ventricles were examined after hemalum-eosin staining. No structural alteration was observed at days 3 to 15 of aldosterone-salt treatment. Among six rat hearts treated for 15 days, a small focus of mononucleated inflammatory cells, colocalized with numerous capillaries, was detected in only one heart.
In contrast, all heart tissue sections of rats treated for 30 days exhibited several lesions (Fig 2⇓). All hearts displayed a marked increase in the number of inflammatory cells, mainly located in thickened fibrous periarterial adventitial layers. These cells were largely lymphocytes, macrophages, and fibroblasts, as identified by their cellular shape (Fig 2F⇓). Some lymphocytes had also infiltrated limited areas of perivascular nonfibrotic myocardium. The endocardium was mildly thickened and contained a few mononucleated inflammatory cells.
In four of six hearts from rats treated for 30 days, several lesions of segmental or circumferential fibrinoid necrosis of the arterial media were observed in both ventricles. The necrosis was associated with panarterial inflammation and endothelial swelling (Fig 2F⇑).
Small foci of recent or healing myocyte necrosis were also observed, containing myocyte cellular debris or iron-loaded macrophages, respectively. In one rat, a subepicardial focus of replacement fibrosis containing iron-loaded macrophages was present.
At 60 days of treatment additional interstitial fibrosis and numerous large areas of hypervascularized replacement fibrosis were seen in both left and right ventricles (Fig 2D⇑ and 2E⇑). At this time fibrinoid necrosis was observed in only one rat among six examined. Moreover, prominent fibromuscular thickening of the wall of intramyocardial arteries was consistently present, with severe narrowing of some arterial lumina.
The immunofluorescence analysis of types I and III collagens showed deposition of both isocollagens, mainly around blood vessels, at day 30; it also extended to the interstitium between days 30 and 60 (Fig 3⇓). The double immunolabeling did not reveal any specific redistribution of an isocollagen relative to the other in fibrotic areas.
Ventricular Collagen mRNA Levels
Fig 4⇓ presents the time course of changes in mRNA levels for ventricular α1(I) and α1(III) procollagens after aldosterone-salt treatment as quantified by dot blot analysis. It should be noted that in sham-operated rats, α1(I) and α1(III) procollagen mRNA levels declined significantly between 3 and 60 days, ie, with increasing age, as previously reported.13 14
In contrast, α1(I) and α1(III) procollagen mRNA levels increased as a consequence of aldosterone-salt treatment after the second week. At 15 days the increases were not yet significant, except in the right ventricle, in which the α1(III) mRNA level increased by 83% (P<.05). The levels of both collagen transcripts were markedly increased in both ventricles at 30 days compared with age-matched sham-operated rats and were slightly diminished at 60 days. One prominent feature of aldosterone-salt treatment was the marked accumulation of collagen transcripts in the right ventricle. For example, procollagen I and III mRNA levels in the right ventricle were increased by 184% and 230%, respectively, at 30 days, whereas in the left ventricle these transcripts were increased by only 78% and 162%. This difference was still evident at 60 days.
Ventricular Na,K-ATPase Subunit mRNA Levels
Na,K-ATPase α1-subunit (+28%, P=.01) and β1-subunit (+24%, P=.04) mRNA levels were increased in the kidney as early as 1 day after onset of aldosterone-salt treatment (Fig 5⇓). This increase was transient because at day 3 both mRNA levels in treated rats were equivalent to those of sham-operated rats. The renal levels of α1 and β1 mRNAs of uninephrectomized rats that drank salt water (n=3) were not altered, indicating a specific effect of aldosterone on renal Na,K-ATPase gene expression. To determine whether cardiac Na,K-ATPase subunit mRNA levels were also changed by the aldosterone-salt treatment before collagen mRNA increases, we measured the Na,K-ATPase α1- and β1-subunits and tissue-specific α2-subunit mRNA levels at 1, 3, and 15 days of treatment. As seen in Fig 6⇓ neither α1 and α2 nor β1 mRNA levels in the left ventricle were affected by treatment, which was also the case for the right ventricle (data not shown). The α3-subunit mRNA was not detected by RNase protection assay in either control or treated groups (data not shown).
This study demonstrates that long-term aldosterone-salt treatment in the rat induced (1) an increase in arterial pressure and left ventricular hypertrophy beginning at 15 days, (2) an increase in cardiac procollagen mRNAs beginning at 15 days, (3) cardiac fibrosis histologically evident at 30 days with the presence of inflammatory cells and myocyte necrosis, and (4) no change in cardiac Na,K-ATPase subunit mRNA levels between 1 and 15 days.
As previously described, cardiac fibrosis induced by a 2-month treatment of rats with aldosterone-salt was associated with an upregulation of both types I and III procollagen mRNA levels. To understand the determinants of aldosterone-induced fibrosis, we investigated the time course of both events. We observed that the increase of collagen mRNAs appeared before collagen deposition. Indeed, the type III procollagen mRNA level was significantly increased in the right ventricle at 15 days. At this time, the levels of type I procollagen in both ventricles and those of type III procollagen in the left ventricle were not significantly elevated. In contrast, histological evidence of cardiac fibrosis was obtained only at 1 month. Therefore, the present study shows that collagen synthesis is regulated at the pretranslational level in aldosterone-salt hypertension. Since there is a dissociation between the hypertrophy restricted to the left ventricle and the fibrosis observed in both the left and right ventricles,1 2 3 we followed the development of fibrosis in both ventricles. Interestingly, the temporal patterns of changes in types I and III procollagen mRNA levels, expressed as a percentage of values in age-matched rats, were very similar in the left ventricle of aldosterone-salt hypertensive rats. In the right ventricle this pattern was slightly different because, as mentioned above, type III procollagen transcripts accumulated before those of type I. However, both types I and III procollagen mRNA levels increased in parallel between 1 and 2 months. These results correlate well with the immunohistochemical study, which did not reveal differences in the spatial deposition of types I and III collagens. Thus, there was no time difference in the stimulation of collagen synthesis (both mRNA and protein) in the ventricles. This eliminates the hypothesis that fibrosis in the right ventricle is a delayed consequence of left ventricular hypertrophy. Since we were able to obtain absolute values for collagen mRNA, we also showed that both the levels of types I and III collagen mRNAs in control rats and the increase of these levels as induced by aldosterone-salt treatment were higher in the right ventricle than the left. This is in agreement with the fact that the concentration of collagen proteins is greater in the right ventricle than in the left15 and may indicate further basal collagen synthesis in the right ventricle.
A major finding of this study is that the upregulation of cardiac collagen mRNAs is delayed in aldosterone-salt–treated rats. Several results from the literature are consistent with a stimulatory effect of aldosterone on cardiac collagen. Cardiac fibrosis is obtained in vivo by long-term treatment with aldosterone-salt but also by aldosterone alone (unpublished results from this laboratory, 1995). In vivo, spironolactone prevents the increase of collagen synthesis by aldosterone-salt.4 However, in vitro studies give conflicting results. Brilla et al5 reported that aldosterone stimulated collagen synthesis in isolated fibroblasts within 24 hours, but Fullerton and Funder6 were not able to confirm this. The reason for this discrepancy is unknown. The hypothesis that aldosterone acts directly on cardiac fibroblasts in vivo needs to be reconsidered in light of the time course of ventricular collagen mRNA changes described in the present work. Indeed, the first signs of increased collagen synthesis occurred around day 15, whereas the plasma concentration of aldosterone was elevated as early as day 1. Thus, our findings clearly suggest an indirect effect of aldosterone on collagen synthesis.
The delayed induction of collagen synthesis we observed is in contrast with other models of hypertension in which an early and transient stimulation of collagen gene expression has been reported. The work of Chapman et al,16 confirmed by that of Villarreal and Dillmann,17 indicates a transient increase of the ventricular content of types I and III procollagen mRNAs on the first day after abdominal stenosis in the rat. In hypertension obtained by continuous infusion of angiotensin II, the levels of cardiac fibronectin and type I collagen mRNA increase rapidly and return to control levels within 1 week.18 These differences in the onset of collagen rise may thus be related to the particular stimulating agent. Because the rise of arterial pressure parallels collagen mRNA accumulation, it might be argued that hypertension is the trigger for collagen synthesis. In all models of hypertension cardiac fibrosis appears concomitant with the elevation of blood pressure and the subsequent left ventricular hypertrophy and originates in the perivascular compartment to spread out thereafter into neighboring interstitial space. Our observation that fibrosis is perivascular at 30 days and perivascular and interstitial at 60 days is in agreement with this mechanogenic hypothesis.
However, several observations indicate that aldosterone induces fibrosis independently of its effects on arterial pressure. First, cardiac fibrosis is elicited in both ventricles of rats chronically treated by aldosterone,13 whereas atrial natriuretic peptide gene expression is stimulated only in the hypertrophied left ventricle and not in the nonhypertrophied right ventricle.3 Second, even at sub-antihypertensive doses, spironolactone prevents the appearance of cardiac fibrosis by aldosterone.4 Third, the intracerebroventricular injection of the mineraloreceptor antagonist RU 28318 in aldosterone-treated rats suppresses the increase in arterial pressure but not the interstitial cardiac fibrosis.19
Thus, there are indications that aldosterone by itself is responsible for at least part of the fibrosis. It also should be mentioned that other biochemical changes resulting in fibrosis arise before the increase in collagen synthesis.17 20 This suggests that the process of aldosterone-induced fibrosis begins before the increase in blood pressure.
Another potentially important factor for the development of cardiac fibrosis in aldosterone-salt–induced hypertension is cellular necrosis. Although the cardiac histological structure was normal at 15 days, it was profoundly altered at 30 days. At the latter time point we observed fibrinoid necrosis; direct and indirect signs of myocyte necrosis, such as the presence of damaged myocytes or cellular debris in the fibroinflammatory areas; and intramacrophagic iron deposition, which is known to originate from myoglobin released by the phagocytosis of damaged myocytes. Injured myocytes and activated macrophages21 are known to produce growth factors such as transforming growth factor–β, which increases collagen synthesis. In addition, in spontaneously hypertensive rats fibroblasts expressing collagen mRNA are colocalized with lymphocytes and macrophages in fibrotic and necrotic areas.22 The presence of lymphocytes and macrophages in the fibrotic areas of hearts of aldosterone-treated rats suggests the possibility that transforming growth factor–β synthesis might contribute to the fibrous tissue response we observed. It may be worthy to note that aldosterone-salt treatment was accompanied by a decreased plasma concentration of thyroid hormones. This decrease was statistically significant at 15 days of treatment and was maintained thereafter. These results need to be considered for interpreting collagen mRNA increase because previous studies have demonstrated that thyroid hormone partly inhibits collagen gene expression.23 The sustained decrease of thyroid hormone level in the plasma may thus suppress the partial inhibition of collagen synthesis by T4 and participate in synergy with other fibrotic factors in the genesis of cardiac fibrosis.
Since our results support an indirect mechanism of induction of cardiac fibrosis by aldosterone, one can conclude that binding of aldosterone to mineralocorticoid receptors in the heart24 25 induces intermediary biochemical steps that remain to be defined. One possibility is that of intracellular ionic disturbances. Mineralocorticoids have a known early stimulatory action on renal Na,K-ATPase activity and gene expression.26 27 Aldosterone has been reported to also increase Na,K-ATPase α1- and β1-subunit mRNA levels in isolated cardiac myocytes at 6 hours.7 Thus, it is conceivable that aldosterone binding in heart could induce changes in myocyte Na,K-ATPase gene expression, with possible subsequent ionic disturbances causing myocyte necrosis and reparative fibrosis. The hypothesis of ionic imbalance is reinforced by the observation that the potassium-sparing agent amiloride prevents microscopic scarring associated with aldosterone-salt treatment but appears ineffective on reactive fibrosis.28
To investigate early cardiac effects of aldosterone, we have studied Na,K-ATPase gene expression between 1 and 15 days, ie, before stimulation of collagen gene expression. We observed an increase of Na,K-ATPase α1- and β1-subunits mRNA levels in the kidney 1 day after onset of aldosterone perfusion, indicating that the aldosterone plasma level reached in our experiments was adequate to stimulate Na,K-ATPase gene expression in vivo. In contrast, neither α1 and α2 nor β1 mRNA levels were affected by the aldosterone treatment in heart between 1 and 15 days, nor was the α3 isoform, which is expressed in the neonatal rat heart,29 reinduced. Similar results have already been reported in deoxycorticosterone acetate–salt hypertensive rats.30 In this model as in other models of hypertrophy,11 only a decrease of the relative level of the α2 mRNA was detected and related to increased blood pressure. Thus, our results indicate that cardiac Na,K-ATPase is not a target for aldosterone and demonstrate a tissue-specific response to this hormone. The present results also emphasize the fact that experiments using isolated cells may give results that are different from those of experiments in intact animals.
However, even though the cardiac Na,K-ATPase remained unaltered by aldosterone-salt treatment, it is possible that other cardiac ionic pumps known to be regulated by aldosterone, such as the Na-H exchanger,26 might be modified. To avoid myocyte damage by secondary effects of aldosterone-salt treatment such as hypokalemia31 32 we supplemented drinking water with KCl at appropriate doses.3
Another possibility is that myocyte necrosis is ischemic in nature and related to fibrinoid necrosis or fibromuscular narrowing of coronary arteries. It has been proposed that angiotensin II plays a major role in these lesions. Indeed, blockade of the angiotensin II action with losartan is effective in preventing fibrinoid necrosis in the heart of stroke-prone spontaneously hypertensive rats despite no decrease in blood pressure.33 Furthermore, in the heart of aldosterone-salt–treated rats, Sun and Weber34 found an increase in cardiac angiotensin II binding between 2 and 8 weeks of treatment. Thus, it is possible that the action of aldosterone is to potentiate the vasoconstrictor and/or fibrogenic effects of angiotensin II.
In conclusion, we have obtained evidence that cardiac collagen gene expression is increased before that of mature proteins, suggesting transcriptional regulation of collagen in aldosterone-salt hypertensive rats. However, the time course of cardiac fibrosis development in our experiments suggests that rather than a direct induction of collagen in cardiac cells, aldosterone has an unknown cardiac action that can induce or facilitate fibrosis. Alternatively, possible secondary effects of the aldosterone treatment, which remain to be defined, may participate in the observed fibrogenic response.
This study was supported by grants from Institut National de la Santé et de la Recherche Medicale, Centre National de la Recherche Scientifique, and Laboratoires Searle France (No 92036-2). The authors thank Drs J. Perennec and J.L. Samuel and N. Van Thiem for helpful discussions, C. Mouas and T. Dakhli for animal handling, and G. Ondola and M. De Villedon for secretarial support.
Reprint requests to Claude Delcayre, INSERM U127, Hôpital Lariboisière, 41 Bd de la Chapelle, 75010 Paris, France.
- Received August 7, 1995.
- Revision received September 7, 1995.
- Accepted September 18, 1995.
Brilla CG, Pick R, Tan LB, Janicki JS, Weber KT. Remodeling of the rat right and left ventricles in experimental hypertension. Circ Res.. 1990;67:1355-1364.
Young M, Fullerton M, Dilley R, Funder J. Mineralocorticoids, hypertension, and cardiac fibrosis. J Clin Invest.. 1994;93:2578-2583.
Robert V, Thiem NV, Cheav SL, Mouas C, Swynghedauw B, Delcayre C. Increased cardiac types I and III collagen mRNAs in aldosterone-salt hypertension. Hypertension.. 1994;24:30-36.
Brilla CG, Weber KT. Reactive and reparative myocardial fibrosis in arterial hypertension in the rat. Cardiovasc Res.. 1992;26:671-677.
Fullerton MJ, Funder JW. Aldosterone and cardiac fibrosis: in vitro studies. Cardiovasc Res.. 1994;28:1863-1867.
Ikeda U, Hyman R, Smith TW, Medford RM. Aldosterone-mediated regulation of Na+,K+-ATPase gene expression in adult and neonatal rat cardiocytes. J Biol Chem.. 1991;266:12058-12066.
Charlemagne D, Orlowski J, Oliviero P, Rannou F, Sainte Beuve C, Swynghedauw B, Lois KL. Alteration of Na,K-ATPase subunit mRNA and protein levels in hypertrophied rat heart. J Biol Chem.. 1994;269:1541-1547.
Martin AF, Paul RJ, McMahon EG. Isomyosin transitions in ventricles of aldosterone-salt hypertensive rats. Hypertension.. 1986;8:128-132.
Besse S, Robert V, Assayag P, Delcayre C, Swynghedauw B. Non synchronous changes in myocardial collagen mRNA and protein during aging: effect of DOCA-salt hypertension. Am J Physiol.. 1994;267:H2237-H2244.
Bonnin CM, Sparrow MP, Taylor RR. Collagen synthesis and content in right ventricular hypertrophy in the dog. Am J Physiol.. 1981;241:H708-H713.
Chapman D, Weber KT, Eghbali M. Regulation of fibrillar collagen types I and III and basement membrane type IV collagen gene expression in pressure overloaded rat myocardium. Circ Res.. 1990;67:787-794.
Villarreal FJ, Dillmann WH. Cardiac hypertrophy-induced changes in mRNA levels for TGF-β1, fibronectin and collagen. Am J Physiol.. 1992;262:H1861-H1866.
Crawford DC, Chobanian AV, Brecher P. Angiotensin II induces fibronectin expression associated with cardiac fibrosis in the rat. Circ Res.. 1994;74:727-739.
Young M, Head G, Funder J. Determinants of cardiac fibrosis in experimental hypermineralocorticoid states. Am J Physiol. In press.
Yao J, Eghbali M. Decreased collagen gene expression and absence of fibrosis in thyroid hormone-induced myocardial hypertrophy. Circ Res.. 1992;71:831-839.
Lombès M, Oblin ME, Gasc JM, Baulieu EE, Farman N, Bonvalet JP. Immuno-histochemical and biochemical evidence for a cardiovascular mineralocorticoid receptor. Circ Res.. 1992;71:503-510.
Horisberger JD, Rossier BC. Aldosterone regulation of gene transcription leading to control of ion transport. Hypertension.. 1992;19:221-227.
Orlowski J, Lingrel JB. Tissue-specific and developmental regulation of Na,K-ATPase catalytic α isoform and β subunit mRNAs. J Biol Chem. 1988;263:10436-10442.
Herrera VLM, Chobanian AV, Ruiz-Opazo N. Isoform-specific modulation of Na+,K+-ATPase α-subunit gene expression in hypertension. Science. 1988;241:221-223.
Darrow D, Miller HC. The production of cardiac lesions by repeated injections of desoxycorticosterone acetate. J Clin Invest.. 1942;21:601-611.