Jump to
Abstract
Antagonists of the mineralocorticoid receptor improve morbidity and mortality in patients with severe heart failure. However, the cell types involved in these beneficial effects are only partially known. The aim of this work was to evaluate whether genetic deletion of mineralocorticoid receptors in mouse cardiomyocytes or fibroblasts in vivo is cardioprotective after chronic left ventricular pressure overload. After transverse aortic constriction, mice deficient in myocyte mineralocorticoid receptors but not those deficient in fibroblast mineralocorticoid receptors were protected from left ventricular dilatation and dysfunction. After pressure overload, left ventricular ejection fraction was significantly higher in mice lacking myocyte mineralocorticoid receptors (70.2±4.4%) as compared with control mice (54.3±2.5%; P<0.01). Myocyte mineralocorticoid receptor-deficient mice showed mild cardiac hypertrophy at baseline, contributing to reduced left ventricular wall tension at baseline and after pressure overload. Cardiac levels of phospho-extracellular signal–regulated kinase 1/2 were higher in myocyte mineralocorticoid receptor-deficient mice than in control mice after pressure overload. Neither fibroblast nor myocyte mineralocorticoid receptor ablation altered the development of cardiac hypertrophy or fibrosis after pressure overload. Both mineralocorticoid receptor mutant mouse strains developed similar degrees of myocyte apoptosis, proinflammatory gene expression, and macrophage infiltration after pressure overload. Thus, mineralocorticoid receptors in cardiac myocytes but not in fibroblasts protect from cardiac dilatation and failure after chronic pressure overload.
- mineralocorticoid receptor
- transgenic mouse model
- cardiac pressure overload
- cardiac hypertrophy
- fibrosis
See Editorial Commentary, pp 679–680
Pharmacological inhibition of the mineralocorticoid receptor (MR) by antagonists like spironolactone or eplerenone has improved the clinical outcome in human heart failure.1,–,3 Activation of the renin-angiotensin-aldosterone-system has been shown to induce cardiac remodeling in numerous animal studies.4,5 MR antagonists have been demonstrated to reduce interstitial and perivascular fibrosis in the heart.6,–,8 However, additional actions have been proposed to contribute to the cardioprotective effects of MR antagonists, including antihypertrophic, anti-inflammatory, and antiarrhythmic effects.9 Despite the clinical benefit of MR antagonists, their cellular target(s) and the precise mechanism of action have not yet been fully uncovered. Several mouse models have been generated to address these questions. For example, overexpression of MRs under control of the human MR promoter10 or conditional tetracycline-regulated cardiac myocyte-specific MR overexpression11 resulted in detrimental effects, including cardiac hypertrophy or severe arrhythmia, respectively. Enhanced MR signaling by transgenic expression of the 11β-hydroxysteroid dehydrogenase type 2 caused severe cardiomyopathy, fibrosis, and mortality, all of which could be partially improved by treatment with the MR antagonist eplerenone.12 Transgenic overexpression of aldosterone synthase in cardiac myocytes in vivo caused coronary dysfunction but prevented the detrimental effects of experimental diabetes mellitus on cardiac capillary density.13,14 Immune cells may play an important role for cardiac pathophysiology, because MR ablation in monocytes and macrophages prevented the development of hypertension and cardiac fibrosis after chronic deoxycorticosterone or NG-nitro-l-arginine methyl ester/angiotensin II treatment.15,16
MR is expressed in both major cardiac cell types, cardiac myocytes and fibroblasts. To evaluate the functional relevance and to determine its importance for the development of cardiac hypertrophy, fibrosis, and dysfunction, we have generated mouse models with cell type–specific deletion of the MR gene in cardiac myocytes and fibroblasts, respectively. Ablation of myocyte but not fibroblast MRs protected from left ventricular (LV) dilatation and functional deterioration after chronic pressure overload without affecting cardiac fibrosis.
Methods
Additional materials and methods are mentioned in the online Data Supplement (please see http://hyper.ahajournals.org).
Generation of MR Mutant Mice
Cardiomyocyte-specific inactivation of the MR gene was achieved using a conditional MR allele (MRflox)17 and mice expressing Cre recombinase under control of the atrial myosin light chain gene promoter (MLCCre)18,19 or the collagen 1α2 promoter (COLCre),20 respectively. All of the animal procedures were approved by the responsible animal care committees (Regierungspräsidium Freiburg, Karlsruhe/Germany), and they conformed to the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (publication No. 85-23, revised 1996). The “control” genotype includes mice with 1 or 2 floxed MR alleles that do not carry a Cre transgene. “Wild-type” refers to C57BL/6N mice.
Transverse Aortic Constriction
Male mice (8 to 10 weeks old) were anesthetized with 2% (vol) isoflurane in oxygen. After thoracotomy, a 7.0 silk suture was placed around a 27-G hypodermic needle to constrict the transverse aorta (TAC).21,22 TAC mice were compared with nonoperated mice at baseline (“basal”).
Hemodynamic Measurements and Echocardiography
For LV catheterization with a 1.4-F pressure-volume catheter, mice were anesthetized with isoflurane (2% vol/vol in O2) and placed on a 37°C table.21,22 Echocardiography was performed at baseline and every 2 weeks for 20 weeks after transverse aortic constriction using a Vivid 7 Dimension (GE Healthcare) echocardiograph equipped with a 14-MHz transducer. Ejection fraction was calculated as described.23 Wall stress was determined from echocardiography and pressure-volume catheterization data.24
Statistical Analysis
Data were analyzed using 2-way ANOVA followed by Bonferroni post hoc tests or Student t test, respectively. A P value of <0.05 was considered statistically significant. Results are displayed as mean±SEM.
Results
Generation and Validation of MR Deletion in Cardiac Myocytes
To inactivate the MR gene specifically in cardiac myocytes in vivo, mice harboring a conditional MR allele (MRflox)17,25 were crossed with myosin light chain gene promoter Cre mice (MLCCre)18,19 (Figure S1A, available in the online Data Supplement at http://hyper.ahajournals.org). Cre-mediated recombination of the conditional MRflox allele resulted in deletion of exon 3 of the MR gene, leading to ablation of MR protein expression in MRMLCCre mice.
In cardiac myocyte fractions isolated from MRMLCCre mice, MR mRNA expression was reduced to 11.6±7.7% of the level in control myocyte preparations (Figure S1B and S1C). In contrast, MR expression in cardiac nonmyocyte fractions did not differ between both genotypes (Figure S1C). To validate cell-type and organ specificity of the gene targeting strategy, MR expression was determined in the hippocampus and kidney, which contained similar or higher MR mRNA levels, respectively (Figure S2). Immunohistochemical analysis of brain and kidney sections did not reveal any alteration of MR protein expression in these tissues in MRMLCCre mice (Figure S1D and S1E). However, MR protein expression in control hearts was too low for specific detection by immunohistochemistry (data not shown).
Cardiac Function of MRMLCCre Mice
Blood pressure and LV function of MRMLCCre mice were assessed by direct aortic and cardiac catheterization and by echocardiography (Figure 1 and Tables S1 and S2). Systolic and diastolic pressures, heart rate, and LV contractility and relaxation did not differ among untreated MRMLCre, MLCCre, and control animals (Table S1).
Cardiac function of MRMLCCre mice. LV function was assessed by microtip catheterization and by transthoracic echocardiography during isoflurane anesthesia in untreated mice (basal, n=6 to 12 per genotype) or after TAC (n=6 to 7 per genotype). A, Mean pressure gradient via the transverse aorta (insert, Doppler image, bar 1 mm) was determined by pulsed-wave color Doppler imaging. B through E, LV ejection fraction (D) and inner diastolic diameter (E) were followed during 20 weeks after TAC by 2D guided M-mode echocardiography from a parasternal short axis view (B). Bars 1 mm. C, LV end-diastolic wall stress was determined by microtip catheterization and echocardiography. **P<0.01, ***P<0.001 vs untreated mice; #P<0.05, ##P<0.01, ###P<0.001 MRMLCCre vs control mice.
After TAC, mortality did not differ between genotypes (data not shown). Mean trans-stenotic pressure gradient, as determined by pulsed-wave Doppler investigation, was 32.5±2.9 mm Hg in control mice and 31.8±2.6 mm Hg in MRMLCCre mice (P=0.86; Figure 1A). At baseline, no differences in ejection fraction and LV ID in diastole were observed between genotypes (Figure 1B, 1D, and 1E). A significant decline in ejection fraction after TAC was observed in control mice but not in MRMLCCre mice (Figure 1D). LV inner diastolic diameter and wall tension increased in control mice after TAC but remained unchanged in MRMLCCre mice (Figure 1B, 1C, and 1E and Table S2).
Cardiac Hypertrophy Induced by TAC in MRMLCCre Mice
Ventricular weight and cardiac weight to body weight or tibia length indexes were higher in MRMLCCre mice at baseline as compared with control mice (Figure 2A and 2B; Table S1). Transgenic expression of Cre recombinase alone (MLCCre) did not affect the ventricular weight indexes (Table S1). After TAC ventricular weight and ventricular weight indexes increased to similar values in both genotypes (Figure 2A and 2B; Table S1). A small but significant increase in LV cardiac myocyte cross-sectional area was already apparent at baseline in MRMLCCre mice (Figure 2C through 2F). After TAC, myocyte cross-sectional areas increased to similar levels in MRMLCCre and control mice (Figure 2E and 2F). Myocyte apoptosis did not differ between genotypes at baseline or after TAC (Figure 2G and 2H).
Cardiac hypertrophy in MRMLCCre mice after TAC. A and B, Ventricular weight (A) and ventricular weight/tibia length (B) were determined in untreated mice (basal, n=24 per genotype) or after TAC (n=6 to 7 per genotype). C and D, Cardiac cross-sections stained with hematoxylin-eosin. Bars: 1 mm (C), 30 μm (D). E and F, Cross-sectional areas of LV cardiac myocytes were determined by wheat germ agglutinin staining. G and H, TUNEL of apoptotic cells in cardiac cross-sections from MRMLCCre and control mice. Red fluorescence, nuclei stained by propidium iodide; green fluorescence, TUNEL-positive nuclei (arrow). Bars 30 μm (n=6 to 7 per group). ***P<0.001 vs untreated mice; #P<0.05, ##P<0.01 MRMLCCre vs control mice.
Transcriptome Analysis and Increased Extracellular Signal–Regulated Kinase 1/2 Signaling in MRMLCCre Mice
Microarray expression profiling of mRNA isolated from total ventricular tissue revealed significant differences in gene expression between MRMLCCre and control mice at baseline and after TAC (Figure 3A and Table S3). Expression of the MR target gene serum glucocorticoid-regulated kinase (Sgk1) in ventricular tissue was induced by TAC but did not differ between genotypes (Figure S3A). However, in cell type–specific analyses, Sgk1 expression was reduced in untreated myocytes but not in nonmyocytes isolated from MRMLCCre mice compared with control mice (Figure 3B and Figure S3B). Aldosterone stimulated Sgk1 expression in control but not in MR-deficient myocytes, providing additional evidence for specific and efficient MR ablation in cardiac myocytes (Figure 3B). Similarly, in MR-deficient myocytes, mRNA expression of the MR target gene Adamts1 (a disintegrin-like and metallopeptidase with thrombospondin type 1 motif 1) was reduced to 28.5±7.4% of the level in control myocytes.
Molecular profile of cardiac hypertrophy in MRMLCCre mice. A, Altered gene expression (>1.5-fold; P<0.05) in total ventricular tissue from MRMLCCre vs control mice at baseline or after TAC was assessed by microarray analysis (n=4 to 6 per group; numbers indicate genes that are differentially expressed between groups). B, Expression of Sgk1 in cardiac myocytes isolated from adult control and MRMLCCre mice. Stimulation (24 hours) with aldosterone (100 nmol/L)-induced Sgk1 in control myocytes but not in MRMLCCre cells. C and D, Expression of the atrial natriuretic peptide (C) and ratio of α- and β-myosin heavy chain (D) gene was determined by quantitative RT-PCR at baseline (n=4 to 6 per group) or after TAC (n=6 per group). E and F, Western blot analysis of phosphorylated p70S6 kinase (p70S6K), Akt, ERK1, ERK2, and p38 at baseline (E; n=7 to 8 per group) and after TAC (F; n=4 per group). *P<0.05, **P<0.01, ***P<0.001 vs untreated cells or mice; #P<0.05, ##P<0.01, ###P<0.001 vs control mice.
Expressions of the atrial natriuretic peptide precursor gene (Nppa) and the β-isoform myosin heavy chain gene (β-MHC) were increased in MRMLCCre versus control hearts at baseline and after TAC (Figure 3C and 3D). Protein levels of phosphorylated p70S6 kinase, Akt, extracellular signal–regulated kinase (ERK) 1/2, p38 mitogen-activated protein kinase (Figure 3E), and Stat3 and JNK1,2,3 (not shown) did not differ between genotypes at baseline. However, after TAC, levels of phosphorylated ERK1 and ERK2 were significantly increased in MRMLCCre versus control mice (Figure 3F). To identify potential mechanisms of ERK1/2 activation in MR-deficient mice, cardiac myocyte expression of upstream activators of ERK1/2 was assessed. Protein kinase C-α (Prkca) mRNA levels were 1.73-fold higher (P<0.05) in MR-deficient myocytes than in control cells. However, aldosterone treatment (24 hours; 100 nmol/L) did not significantly alter Prkca mRNA levels in myocytes from both genotypes (aldosterone versus untreated: control +16.9%, P=0.82; MRMLCCre +9.8%, P=0.89).
Cardiac Fibrosis Induced by TAC in MRMLCCre Mice
Sirius Red staining of ventricular cross-sections did not reveal any significant fibrosis in MRMLCCre mice at baseline (Figure 4A through 4D). Chronic pressure overload led to similar levels of interstitial and perivascular fibrosis and increased expression of the fibrosis marker connective tissue growth factor (Ctgf) in MRMLCCre and control hearts (Figure 4A through 4E). Expression analysis of the proinflammatory gene NADPH oxidase Nox2 and the macrophage marker F4/80 (Emr1)15,16 did not reveal differences between genotypes (Figure 4F and 4G).
Cardiac fibrosis after TAC. A and B, The degree of LV interstitial (A) and perivascular (B) fibrosis was determined by Sirius Red staining in untreated mice (basal; n=4 to 6 per genotype) or after TAC (n=5 to 8 per genotype). Bars, 30 μm. C through E, Quantitative evaluation of interstitial (C) and perivascular fibrosis (D) and expression of the fibrosis marker gene connective tissue growth factor as determined by quantitative RT-PCR (Ctgf; E). **P<0.01, ***P<0.001 vs untreated mice.
Ablation of MR Expression in Fibroblasts
To investigate whether MRs in cardiac fibroblasts participate in cardiac remodeling after chronic pressure overload, the MR gene was also selectively deleted in fibroblasts using the regulatory elements of the collagen 1α2 gene to drive Cre recombinase expression20 (MRCOLCre, Figure 5 and Figure S1). Ablation of MR mRNA expression in cardiac MRCOLCre fibroblasts was confirmed by quantitative RT-PCR (Figure 5A). Ventricular weight indexes and cardiac function as determined by echocardiography did not differ between MRCOLCre and control mice at baseline and after TAC (Figure 5B and 5C and Tables S4 and S5). Importantly, pressure overload induced similar levels of interstitial and perivascular fibrosis in MRCOLCre and in control mice (Figure 5D through 5F). Expression of Nox2 and Emr1 mRNA did not differ between genotypes (Table S6). Myocyte apoptosis induced by TAC did not differ significantly between genotypes (MRCOLCre 82±25% of control; P>0.05; n=3 to 4 per genotype).
Fibroblast-specific ablation of MR expression in MRCOLCre mice. A, MR mRNA expression in fibroblasts isolated from ventricles of control or MRCOLCre mice (n=4 to 5 per genotype; *P<0.05 vs control). B and C, Ventricular weight:body weight ratio (B) and LV ejection fraction (C) in control and MRCOLCre mice at baseline and after chronic cardiac pressure overload. D through F, Interstitial and perivascular fibrosis as detected by Sirius Red staining in ventricular sections of control and MRCOLCre mice (D, bars, 30 μm). n=5 to 8 per genotype and group; *P<0.05, **P<0.01, ***P<0.001 vs untreated mice.
Discussion
The major finding of this study is that cell-specific ablation of the MR in cardiac myocytes but not in fibroblasts protects from ventricular dilatation and dysfunction after chronic pressure overload (Table).
Comparison of Cardiac Phenotypes Observed in Mice After Treatment With the MR Antagonist Eplerenone and Cell-Type–Specific MR Deletion
Validation of MR Ablation in Cardiac Myocytes
Cell-specific deletion of the exon 3 of the murine MR gene was verified by quantitative RT-PCR in myocyte and fibroblast cell fractions isolated from control, MRMLCCre, and MRCOLCre mouse hearts, respectively. Cre-mediated recombination resulted in a reduction of MR mRNA levels in MRMLCCre myocytes by 89% and in MRCOLCre fibroblasts by 98% as compared with the respective cell types isolated from control mice. The reduction in MR expression correlated well with the relative purity of the myocyte and fibroblast cell fractions as assessed by marker gene expression analysis for α-MHC or Ddr2, respectively (data not shown). Previous experiments have shown that Cre-mediated deletion of exon 3 of the MR gene leads to complete ablation of MR protein expression.17,25 In the murine heart, MR protein expression was too low for reliable detection by immunohistochemistry (data not shown). However, we demonstrated that targeted deletion of the MR gene in cardiac myocytes did not alter MR protein expression in the brain or kidney.
Cardiac Effects of Cell-Specific MR Ablation
In previous clinical and experimental studies, pharmacological blockade of MR has resulted in a significant protection of the heart from hypertrophy, fibrosis, and failure.1,2,6,–,8 In the Randomized Aldosterone Evaluation Study and Eplerenone Post-Acute Myocardial Infarction Heart Failure Efficacy and Survival Study, the MR antagonists spironolactone and eplerenone were associated with a significant reduction in overall mortality of patients with chronic heart failure, in spite of the presence of inhibitors of the renin-angiotensin system.1,2 These clinical findings have sparked a great research interest to identify mechanisms of cardioprotection, in particular for the antifibrotic effect conferred by MR antagonists.9,26
Cardiac myocyte-specific deletion of the MR gene prevented LV dilatation and failure after 20 weeks of chronic pressure overload. This finding is consistent with previous experimental evidence, including a recent report demonstrating that treatment with the MR antagonist eplerenone delayed the transition from compensated cardiac hypertrophy to dilatation and failure in mice7 (Table). We hypothesize that differences in the cardiac phenotype between MRMLCCre and wild-type mice at baseline may contribute to the protective effect of MR ablation during TAC. Loss of MR expression was accompanied by cardiac myocyte hypertrophy and distinct changes in gene expression in isolated myocytes at baseline. Myocyte hypertrophy in MRMLCCre mice at baseline may contribute to reduced wall stress during increased afterload. Reduced wall stress may thus prevent or delay the transition to failure and dilatation on chronic pressure overload.27 After TAC, we found evidence for increased activation of ERK1/2 signaling in MRMLCCre hearts. Previous studies have linked activation of cardiac mitogen-activated protein kinase kinase 1-ERK1/2 signaling with the development of physiological hypertrophy and cardioprotection.28,29 Thus, ERK1/2 activation after TAC may prevent cardiac dilatation and failure after pressure overload. However, the mechanisms leading to ERK1/2 activation after MR deletion in cardiac myocytes still need to be uncovered in future studies.
Identification of the Cell Type Involved in Cardioprotection by MR Antagonists
Most recently 2 studies have reported the effects of specific MR ablation in macrophages on cardiac remodeling in mice.15,16 Together with the results from the present study, the relative contribution of cardiac cell types to the protective effects of MR antagonists may be assessed (Table). For comparison, the study by Kuster et al7 investigating the effects of eplerenone after LV pressure overload induced by constriction of the ascending aorta in mice may be used. In this model, eplerenone improved LV function and geometry and prevented dilatation, fibrosis, myocyte apoptosis, oxidative stress, macrophage infiltration, and inflammation.7 In agreement with these results, we observed improved LV function and geometry and prevention of LV dilatation in MRMLCCre mice. Neither eplerenone7 nor myocyte-specific MR ablation in the present study prevented cardiac hypertrophy after pressure overload. Interestingly, eplerenone-treated mice also showed a trend toward cardiac hypertrophy, which did not reach statistical significance.7
In contrast to the study by Kuster et al,7 interstitial and perivascular fibrosis as induced by pressure overload did not depend on cardiac myocyte or fibroblast MR in the present study. However, this observation is entirely consistent with findings from macrophage-specific MR deletion studies.15,16 MR ablation in macrophages significantly reduced cardiac fibrosis as induced by NG-nitro-l-arginine methyl ester/angiotensin II16 or deoxycorticosterone acetate/salt treatment.15
A classic MR target gene in epithelial cells, serum glucocorticoid-inducible kinase (Sgk1)30 was downregulated to <15% of the expression level in control myocytes. Constitutive germline knockout of Sgk in mice led to reduced hypertrophy and fibrosis after deoxycorticosterone acetate/salt stimulation.8 Although these experiments demonstrate that Sgk is essential for fibrosis development after chronic MR stimulation, they do not allow any conclusion with respect to the organ or cell type involved in mediating the hypertrophic and fibrotic effects of MRs. However, in light of the current study, it would be interesting to selectively delete Sgk1 in cardiac myocytes and other cardiac cells to determine whether Sgk1 is involved in protective effects of MR ablation. Further studies are required to identify the MR target genes that are important for the cardioprotective effects during the development of hypertrophy and failure.
Conclusions
Taken together, the present study demonstrates that ablation of MRs in cardiac myocytes but not in fibroblasts protects from dilatation and failure after chronic pressure overload. Selective ablation of MR expression in other cardiac cell types, including endothelial and vascular smooth muscle cells, will help to uncover the cellular and molecular mechanisms of cardioprotection conferred by MR antagonists.
Perspectives
Antagonists of the MR, like spironolactone and eplerenone, improve morbidity and mortality in patients with severe heart failure. Recently, ablation of MR in macrophages was reported to reduce the development of cardiac inflammation, hypertrophy, and fibrosis. To identify whether additional cell types in the heart are involved in the cardioprotective effects of MR inhibition after chronic cardiac pressure overload, the MR gene was selectively deleted in cardiac myocytes or fibroblasts of transgenic mice. Ablation of MR expression in myocytes but not in fibroblasts protected from LV dilatation and dysfunction. Thus, inhibition of MR function in cardiac myocytes and in macrophages may be essential for the cardioprotective effects of MR antagonists. These findings provide an essential basis for future studies to unravel the intracellular pathways involved in the beneficial effects of MR inhibition.
Sources of Funding
This work was supported by the Deutsche Forschungsgemeinschaft (He 2073/2-1, SFB488, and SFB636), by the “Fonds der Chemischen Industrie,” and by the European Union through grant LSHM-CT-2005-018652 (CRESCENDO).
Disclosures
None.
Acknowledgments
We thank Jan Rodriguez Parkitna for his analysis of microarray data. We thank the EMBL GeneCore (Heidelberg, Germany) staff, especially Vladimir Benes and Tomi Ivacevic, for performing the Affymetrix microarray experiments.
- Received September 23, 2010.
- Revision received October 12, 2010.
- Accepted January 21, 2011.
- © 2011 American Heart Association, Inc.
References
This Issue
Jump to
Article Tools
Share this Article
- Ablation of Mineralocorticoid Receptors in Myocytes But Not in Fibroblasts Preserves Cardiac Function
