Preventive and chronic mineralocorticoid receptor antagonism is highly beneficial in obese SHHF rats

Abstract

Background and Purpose

Mineralocorticoid receptor (MR) activation contributes to heart failure (HF) progression. Its overactivity in obesity is thought to accelerate cardiac remodelling and HF development. Given that MR antagonists (MRA) are beneficial in chronic HF patients, we hypothesized that early MRA treatment may target obesity‐related disorders and consequently delay the development of HF.

Experimental Approach

Twenty spontaneously hypertensive HF dyslipidaemic obese SHHF^cp/cp rats and 18 non‐dyslipidaemic lean SHHF^+/+ controls underwent regular monitoring for their metabolic and cardiovascular phenotypes with or without MRA treatment [eplerenone (eple), 100 mg∙kg⁻¹∙day⁻¹] from 1.5 to 12.5 months of age.

Key Results

Eleven months of eple treatment in obese rats (SHHF^cp/cpeple) reduced the obesity‐related metabolic disorders observed in untreated SHHF^cp/cp rats by reducing weight gain, triglycerides and total cholesterol levels and by preserving adiponectinaemia. The MRA treatment predominantly preserved diastolic and systolic functions in obese rats by alleviating the eccentric cardiac hypertrophy observed in untreated SHHF^cp/cp animals and preserving ejection fraction (70 ± 1 vs. 59 ± 1%). The MRA also improved survival independently of these pressure effects.

Conclusion and Implications

Early chronic eple treatment resulted in a delay in cardiac remodelling and HF onset in both SHHF^+/+ and SHHF^cp/cp rats, whereas SHHF^cp/cp rats further benefited from the MRA treatment through a reduction in their obesity and dyslipidaemia. These findings suggest that preventive MRA therapy may provide greater benefits in obese patients with additional risk factors of developing cardiovascular complications.

Abbreviations

aldo

aldosterone

BMI

body mass index

BNP

brain natriuretic peptide

^cp

mutant ^cp allele of the leptin receptor

CSAA

cross‐sectional adipocyte area

EDT

E wave deceleration time

EDV

end‐diastolic volume

EF

ejection fraction

Einc

elastic incremental modulus

eple

eplerenone

ESV

end‐systolic volume

Fn1

fibronectin 1

HDL

high‐density lipoproteins

HF

heart failure

HW

heart weight

IR

insulin resistance

IVRT

isovolumetric relaxation time

LDL

low‐density lipoproteins

Lepr

leptin receptor

LV

left ventricle

LVID

Left ventricle internal diameter

MR

mineralocorticoid receptor

MRA

mineralocorticoid receptor antagonist

Nox4

NADPH oxidase 4

PWT

posterior wall thickness

RAAS

renin–angiotensin–aldosterone system

SHHF

spontaneously hypertensive heart failure

SWT

septum wall thickness

TG

triglycerides

TL

tibia length

VAT

visceral adipose tissue

Vim

vimentin

Tables of Links

TARGETS
Nuclear hormone receptors a
Mineralocorticoid receptor (MR; NR3C2)
Catalytic receptors b
Leptin receptor

LIGANDS
Adiponectin	Fibronectin
Aldosterone	Insulin
BNP	TGFβ2
COL3A1	TGFβ3
Corticosterone

These Tables list key protein targets and ligands in this article which are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY (Pawson et al., 2014) and are permanently archived in the Concise Guide to PHARMACOLOGY 2015/16 (^a,bAlexander et al., 2015a, 2015b).

Introduction

Over 23 million patients are diagnosed with heart failure (HF) worldwide (McMurray et al., 1998) among whom 32 to 49% are obese (body mass index, BMI ≥ 30 kg⋅m⁻²) and 31 to 40% are overweight (25 ≤ BMI < 30 kg⋅m⁻²) (Clark et al., 2014). Obesity, along with the ageing of the population and the increased prevalence of various underlying pathophysiological mechanisms, also referred to as HF modifier conditions (van Deursen et al., 2014), are known to precipitate individuals towards fully developed HF phenotypes.

Adverse adaptive mechanisms such as overactivation of the renin–angiotensin–aldosterone system (RAAS) are recognized to contribute to HF. Extensive study of the RAAS has resulted in the development of several pharmacological strategies aimed at inhibiting its unbalanced signalling pathways in HF. Accordingly, the addition of mineralocorticoid receptor (MR) antagonists (MRA) to standard treatments has been proven to alleviate HF symptoms and clearly reduce mortality in HF patients (Greenberg et al., 2006; Iraqi et al., 2009). Although initially believed to mainly regulate body fluid volumes via its interaction with the renal MR, extrarenal pathophysiological effects of its ligands, namely, aldosterone and corticosterone, have been substantiated by the finding that MRs are expressed in the myocardium of the failing heart and in non‐epithelial target tissues such as adipose tissue (Caprio et al., 2007; Zennaro et al., 2009; Caprio et al., 2011; Marzolla et al., 2012. Excessive MR activation and increased aldosterone concentration have been implicated in the development of co‐morbidities that are highly prevalent in HF patients. Whether alone (obesity, insulin resistance, diabetes and hypertension) or in combination (metabolic syndrome, MetS), they participate in a rapid deterioration of myocardial structure (remodelling) and function. Because increased aldosterone concentrations have also been reported in MetS, increased MR stimulation is thought to accelerate the pathophysiological conditions that lead to the development of HF (De Keulenaer and Brutsaert, 2011). Thus, when endeavouring to prevent or delay the development of HF, abnormal MR activity in non‐cardiac co‐morbidities appears as a potentially valuable target for new personalized therapeutic strategies. Although it is clear that obesity is an important risk factor for HF development, data remain relatively scarce regarding the contribution of MRAs in the prevention of the development of HF in obese patients. Given the proven role of MR activation in adipogenesis (Guo et al., 2008; Feraco et al., 2013), we hypothesized that early MRA treatment with eple, one of the most selective MRAs to date, may delay or attenuate the adverse cardiac remodelling and subsequently prevent HF progression especially in obese rats.

Our previous characterization of spontaneously hypertensive heart failure obese rats (SHHF^cp/cp, homozygous for the defective mutant ^cp allele of the leptin receptor gene Lepr) (Youcef et al., 2014) demonstrated that the onset of metabolic disorders (dyslipidaemia, overweight and insulin resistance) as well as pre‐hypertension occurred within the first 3 months of the life of these animals. These alterations appeared in the absence of any sign of cardiac dysfunction but nonetheless contributed to the earlier onset of HF in SHHF^cp/cp rats as compared with their lean controls SHHF^+/+. Considering that SHHF^cp/cp rats also develop an exacerbated hyperaldosteronism compared with SHHF^+/+ rats, the present study investigated whether early antagonism of the MR could sustainably improve the cardiovascular function of ageing SHHF rats and whether the effects of this treatment were modulated by the presence of metabolic alterations in SHHF^cp/cp rats. To this end, SHHF rats were treated for 11 months (initiated at 1.5 months of age) with either the MRA eple as a mono‐therapy or its placebo. Our conclusions with regard to the beneficial effects of chronic MRA are strengthened by the results observed with this treatment of both dyslipidaemic obese SHHF^cp/cp rats and their non‐dyslipidemic lean SHHF^+/+ controls.

Methods

Animal model

Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath and Lilley, 2015). One‐month‐old SHHF male rat littermates (SHHF/MccGmiCrl‐Lepr^cp, Charles River Laboratories, USA) were genotyped (Ishizuka et al., 1998) to determine their homozygocity for the wild‐type ‘+’ allele or the mutant ‘cp’ allele that encodes a defective leptin receptor, thus identifying the animals as the SHHF^+/+ or SHHF^cp/cp genotype respectively. The experimental protocols were carried out in our laboratory after a 2‐week acclimatization period within the animal care facility where rats were allowed ad libitum access to fluid and food (Purina Formulab chow 5008, Charles River, USA). This protocol was designed according to the ARRIVE guidelines (http://onlinelibrary.wiley.com/doi/10.1111/j.1476‐5381.2010.00872.x/pdf). It was approved by the ‘Comité d'Éthique Lorrain en Matière d'Expérimentation Animale’ (CELMEA) under agreement no. 0.1886 and was performed in an authorized animal facility (agreement no. C 54–547‐17). This agreement defined appropriate endpoints, which limited the amount of pain an animal suffered during the development of the HF phenotype; end points (survival study) were set as being lack of mobility of the animal and the development of congestion. Changes in the behaviour of the animals were assessed daily and supported by their regular phenotyping. Animals with the most noticeable phenotype were intensely monitored but none of them needed to be killed.

Study design

The effect of MR activation during the development of obesity and related disorders was evaluated for its involvement in the transition towards HF. Animals of both genotypes (SHHF^cp/cp and SHHF^+/+) were given the selective MRA eple (pure active molecule provided by Pfizer) from 1.5 months onwards (Figure 1A). This time point corresponds to the onset of metabolic disorders (dyslipidaemia, insulin resistance and overweight) and the pre‐hypertensive stage in the absence of any apparent signs of cardiac dysfunction to date (Youcef et al., 2014). Eple treatment was provided in drinking water at a final concentration of 100 mg⋅kg⁻¹⋅day⁻¹. The choice of this dosage was based on data from several reports (Lacolley et al., 2002; Bayorh et al., 2006; Rigsby et al., 2007; Susic et al., 2007; Baldo et al., 2011; Fraccarollo et al., 2011; Miana et al., 2011; Watson et al., 2013). Although much higher than that prescribed to patients with HF, the MRA concentration used herein is justified by the differences in drug metabolism and efficacy of the compounds between rats and humans. In order to minimize the effects of subjective bias when assessing results, each experimenter was blinded to the assignment of animals to the different treatment groups especially for echocardiography and haemodynamic evaluations.

Figure 1.

Study design and survival rate. (A) Thirty SHHF^cp/cp and 28 SHHF^+/+rats were randomized either to the ^1.5SHHF group (1.5‐month follow‐up, n = 10 per genotype) or to the long‐term follow‐up group to receive either the selective MRA eplerenone (eple 100 mg⋅kg⁻¹⋅day⁻¹) (^12.5SHHF^+/+eple, n = 8 and ^12.5SHHF^cp/cpeple, n = 11) or placebo (^12.5SHHF^+/+, n = 10 and ^12.5SHHF^cp/cp, n = 9). Animals underwent metabolic (M), cardiac (C), vascular (V) and histological (H) phenotyping at the beginning and at the end of the study. Cardiac parameters were regularly monitored from 1.5 to 12.5 months of age. (B) Kaplan–Meier curves indicating the survival rate for each studied group.

In order to facilitate the description of the various experimental groups, the following nomenclature is used throughout the paper: the time (expressed in months) at which the observations/assays were performed is indicated as a number in front of the SHHF strain (for example, ^1.5SHHF refers to observations made at 1.5 months of age); the genotype is indicated by a sign after SHHF (as SHHF^+/+ or SHHF^cp/cp), while the treatment status with eple is indicated with the suffix ‘eple’. The absence of eple indicates that these rats are part of a control placebo group.

Ten SHHF^cp/cp and 10 SHHF^+/+ rats were allocated to the ^1.5SHHF group. From 1.5 to 12.5 months of age, the regular monitoring of the metabolic and cardiovascular parameters of the four experimental groups (SHHF^+/+eple, n = 8; SHHF^cp/cpeple, n = 11; SHHF^+/+, n = 10 and SHHF^cp/cp, n = 8) allowed the comparison of their respective phenotypes at different time points up to 11 months of follow‐up (Figure 1a).

Echocardiography

Transthoracic echocardiographies were performed at 1.5, 3, 6, 9.5, 11.5 and 12.5 months of age on anesthetized rats (isoflurane, 5% at induction and 3% for maintenance, in 1.5 L⋅min⁻¹ O₂) using a 12 MHz transducer (Sonos 5500 Ultrasound System, Philips) (Youcef et al., 2014). Animals were positioned in left decubitus to acquire a short axis view of the left ventricle (LV) and an apical four‐chamber view.

Morphological parameters of the LV were assessed using the two‐dimensional (2D) M mode: LV internal diameter at the end diastole (LVIDd), at the end systole (LVIDs), posterior wall and septum thicknesses (PWT and SWT, respectively) were recorded to enable the calculation of LV mass according to the following equation, LVmass = 1.04*[(PWT + SWT + LVIDd)³ − (LVIDs)³], where 1.04 is the estimated specific gravity of the myocardium.

LV systolic function was assessed by (i) 2D M mode to calculate the ejection fraction (EF) as follows: EF (%) = [(EDV − ESV)/EDV × 100], where EDV and ESV correspond to the end‐diastolic and end‐systolic volumes respectively, and (ii) offline tissue Doppler imaging at the lateral and septal mitral annulus in the apical four‐chamber view which assesses longitudinal LV function and measures Sa wave velocity.

LV diastolic function and LV filling pressures were assessed by pulse wave Doppler performed on the apical four‐chamber view. This allowed the measurement and calculation of the ratio of the early (E) to late (A) ventricular filing waves, the isovolumetric relaxation time (IVRT), the E wave deceleration time (EDT), while the e′ mitral wave by tissue Doppler imaging allowed the evaluation of diastolic function of the myocardium. The LV filling pressures were determined by the E/e′ ratio. All echocardiographic examinations were based on the calculation of the mean of three consecutive cardiac cycles and were performed by a single experienced sonographer. The group sizes varied along the follow‐up due to loss of animals in untreated groups and to the restricted access to the echocardiographic transducer.

Fasting glycaemia, blood biochemical and hormonal assays

Blood samples were drawn from the carotid artery of killed 16 h‐fasted animals. Sera were obtained after a 20‐min incubation period at room temperature and centrifugation at 1000x g for 15 min. Serum lipid profiles [total cholesterol, triglycerides (TG), high‐density and low‐density lipoproteins (HDL and LDL respectively) and free fatty acids] were assessed by an automatic biochemistry analyser using enzymatic methods. Sodium and potassium (Na⁺, K⁺) were measured by a standardized indirect potentiometry technique. Adiponectin and insulin levels were determined by elisa (adiponectin Rat elisa kit #ab108784 and insulin Human elisa kit, #ab100578, respectively).

Insulin resistance (IR) was estimated by the HOMA‐IR index:

$$\text{HOMA}‐\mathrm{IR}=\left[\text{fasting plasma glucose}\left(\text{mmol}\cdot {\mathrm{L}}^{-1}\right)\times \text{fasting plasma insulin}\left(\mathrm{\mu IU}\cdot {\mathrm{mL}}^{-1}\right)\right]/22.5.$$

Plasma samples were obtained by adding sodium citrate anticoagulant to a subset of the blood samples immediately followed by centrifugation at 1700x g for 10 min and subsequent retrieval of the supernatant. Circulating levels of brain natriuretic peptide (BNP) were measured by elisa (BNP 45 Rat elisa Kit #ab108816). Aldosterone and corticosterone levels were measured on 24‐h urine samples with commercial kits according to the manufacturer's recommendations (Siemens 06615154 Coat‐A‐Count RIA Aldosterone Kit and AssayMax Corticosterone elisa Kit, Assaypro, St. Charles, MO, USA, respectively). Fasting glycaemia was measured on a drop of venous blood from conscious rats using a glucometer (Freestyle Papillon Easy, Abbott).

Total RNA isolation and RT qPCR assays

Total RNAs were isolated from a transverse section of the myocardium including portions of both ventricles (RNA Now, Ozyme) and their purity and integrity verified by spectrophotometry and capillary electrophoresis (NanoDrop^® ND‐1000 spectrophotometer and Agilent 2100 Bioanalyzer with RNA 6000 Nano assay kit respectively). Only RNAs with no sign of high levels of DNA contamination or marked degradation (RNA Integrity Number > 8) were considered of good quality and used for further analysis.

Briefly, reverse transcriptions (RT) of 400 ng of total RNA into cDNA were performed using the Qiagen RT² First Strand Kit according to the manufacturer's protocol. Real‐time quantitative PCR (qPCR) was subsequently performed with Qiagen Custom RT² Profiler PCR Arrays on 2.27 ng of cDNA per well using Qiagen RT² SYBR Green Mastermixes. Triplicate assays were run for each gene of interest [fibronectin 1 (fn1; vimentin (vim); tissue growth factor 2, tgb2; tissue growth factor 3 (tgb3); collagen 3a1 (col3a1); NADPH oxidase 4 (Nox4)] in the ViiA7 PCR system (Life Technologies) using the default settings. Comparative threshold cycles (C_T) data were used to calculate relative gene expression values, by applying the RT² Profiler PCR Array Data Analysis software v3.5, available online at http://pcrdataanalysis.sabiosciences.com/pcr/arrayanalysis.php. C_T for Sdha, Actb and B2m genes, considered as housekeeping genes, were used for qPCR data normalization. Fold changes in the gene expressions studied (Qiagen reference upon request) were calculated as 2^(−ΔCT). Results are expressed as the ratio of 2^{(−ΔCTSHHFeple)}/2^{(−ΔCTSHHF)} for each transcript in each genotype studied. The P‐values were calculated using Student's t‐test.

Histology

Immediately after the animals had been killed, ^12.5SHHF right carotid, heart and perirenal visceral adipose tissue (VAT) were rapidly dissected, rinsed in saline solution (NaCl 0.9%) and formalin fixed for further histological analysis. Perirenal location was preferred over other locations because it was always present even in the ^12.5SHHF^+/+ rats where very few small VAT pads developed. Serial 5‐μm sections of paraffin embedded tissues were prepared for heart and carotid while 10‐μm sections were prepared for VAT. Carotids were stained using the Weigert's orcein‐fuchsin method to determine the medial cross‐sectional area by manually delineating the carotid edges of several serial carotid slides for each rat using the NIS‐element software (Nikon). Hearts and perirenal VAT were stained with Sirius red to determine the degree of myocardial fibrosis and distribution of cross‐sectional adipocyte area (CSAA) respectively using imagej software. Myocardial fibrotic area was analysed by measuring the % of fibrotic area (demonstrated as magenta staining) of whole heart sections. CSAA was measured for each adipocyte representing the VAT sections from each rat.

Statistical analysis

The data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology (Curtis et al., 2015). However, some limitations apply: randomization was not used for in vivo experiments as the littermate status (genetic link) of the purchased rats was not known, and the group sizes varied in the follow‐up due to the restricted access to the echocardiographic transducer and biochemistry laboratory.

All analyses were performed using sas R9.3 software (SAS Institute, Cary, NC, USA). The overall two‐tailed significance level was set at P < 0.05. Two‐way ANOVA with genotype (SHHF^+/+ vs. SHHF^cp/cp) and treatment (placebo vs. eple) along with their interaction as fixed effects were performed, followed by four post hoc pairwise comparisons of interest (when the overall probability of the model was significant) effect of treatment according to genotype (placebo vs. eple for each genotype) and effect of genotype according to treatment (SHHF^+/+ vs. SHHF^cp/cp in each treatment group). The significance level for comparisons was adjusted for multiplicity in order to preserve the overall 5% error rate using the formula α′ = 1 − (1 − α)^1/k, where α and α′ are the overall and adjusted significance levels respectively, and k is the number of comparisons (four comparisons; *P < 0.05). Results are presented for each parameter and each time point as global results (sample size, P‐values of the model and the fixed effects), adjusted means and SEM (mean values adjusted for the effect of other factors) and the four pairwise comparisons of interest for which the significance level has to be set to P < 0.0127 in order to preserve the overall 5% α error rate. Of note, pairwise comparison results were only considered in the cases of significant interaction (P < 0.05).

Statistical analysis of the CSAA measured in adipose tissue sections was performed using a Mann–Whitney test to compare the difference between ^12.5SHHF^cp/cp and ^12.5SHHF^cp/cp eple rats (*P < 0.05). All results are expressed as adjusted means ± SEM.

To evaluate how variations in metabolic parameters may be associated with cardiac phenotypes, a statistical analysis of the raw data obtained from each group of animals were combined and computed in the coexpress software (Nazarov et al., 2013). Pearson's test was conducted to calculate the correlation coefficients (r) and a P‐value <0.05 was considered to identify the significant parameters that may associate together. Correlation coefficients rank from +1 to −1, where +1 means a strict positive correlation and −1 denotes a strict negative correlation.

Results

The present study design allowed the evaluation of the preventive effects of long‐term eple treatment observed in the presence of metabolic disorders in obese SHHF^cp/cp rats and to compare these effects with those observed in lean SHHF^+/+ rats (Figure 1A). Preventive effects were expected given that eple treatment was started before any cardiac symptoms were observed (1.5 months of age) in control lean SHHF^+/+ rats. The end point of the study (month 12.5) was chosen so as to maintain the number of surviving untreated SHHF^cp/cp rats to an appropriate level to allow statistical comparison with the other groups of rats (Figure 1B). The effective delivery of eple was confirmed by the measurement of urinary excretions of aldosterone and corticosterone, which were significantly increased and decreased respectively, upon treatment in both genotypes (Supporting Information Table S1).

Haemodynamic and vascular parameters are not affected by eplerenone monotherapy

Mean blood pressure and pulse pressure calculations were based on the invasive measurement of central blood pressures (diastolic blood pressure and systolic blood pressure) of anaesthetized ^12.5SHHFeple and their genotype‐matched untreated controls. The results demonstrated the presence of severe hypertension whose severity was independent of both genotype and of MRA treatment (Supporting Information Table S2). The distensibility‐pressure curve as well as analysis of compliance, elastic incremental modulus (Einc) and wall stress values demonstrated that the intrinsic mechanical behaviour of the carotid wall was similar across all groups at 12.5 months of age (Supporting Information Table S2 and Fig. S1A and B). The lack of effect of eple on SBP (systolic blood pressure) was confirmed on conscious animals during plethysmographic evaluation of their peripheral blood pressure (Supporting Information Table S3).

Metabolic disorders accelerate adverse cardiac remodelling in SHHF^cp/cp rats compared with SHHF^+/+ controls

Echocardiographic follow‐up between 1.5 and 12.5 months of age revealed that SHHF^+/+‐untreated rats developed LV concentric hypertrophic remodelling as evidenced by the stability of the LVIDd values (Figure 2A) and the diastolic thickening of the septal wall (SWT) (Figure 2B) resulting in an increased LV mass (Figure 2C).

Figure 2.

Chronic eple treatment alleviates cardiac remodelling. Cardiac remodelling of SHHF^+/+ (black), SHHF^+/+eple (dark grey), SHHF^cp/cp (white) and SHHF^cp/cpeple (light grey) rats was evaluated by echocardiographic follow‐up. At 1.5 months of age n = 9, SHHF^+/+ rats were evaluated then n = 8 of them at the other time points; n = 8 SHHF^+/+eple were evaluated at each time point; n = 8 SHHF^cp/cp were evaluated at 1.5, 3 and 6 months, n = 7 of them at 9 months and n = 6 of them at 11.5 and 12.5 time points. Finally, n = 10 SHHF^cp/cpeple were evaluated from 1.5 to 11. 5 months of age and n = 11 of them at 12.5 months allowing the determination of (A) left ventricular internal diameter at end diastole (LVIDd), (B) septum wall thickness (SWT) and (C) calculation of left ventricular mass (LV mass). Two‐way ANOVA with genotype (SHHF^+/+ vs. SHHF^cp/cp) and treatment (placebo vs. eple) was performed, and the significance level was set to P < 0.0127. *P < 0.0127 for the comparison between the genotypes at the same age, #P < 0.0127 for the comparison of the untreated and treated animals from the same genotype at the same age.

When compared with the lean rats, SHHF^cp/cp animals exhibited worsened phenotypes. Only 60% of untreated SHHF^cp/cp rats survived until 12.5 months of age, while 90% of SHHF^+/+ rats completed the protocol (Figure 1B). This observation was explained in part by sudden cardiac death due to the development of end‐stage systolic dysfunction observed with ageing in this genotype. Compared with SHHF^+/+ rats, ^11.5SHHF^cp/cp rats developed LV eccentric remodelling characterized by a decrease in SWT and an increase in both LVIDd and LVmass (Figure 2A–C). In addition, in ^12.5SHHF^cp/cp LV dilation was exacerbated (Figure 2A), while SWT (Figure 2B) decreased. LV remodelling was accompanied by a greater interstitial fibrosis in ^12.5SHHF^cp/cp rats than in ^12.5SHHF^+/+ rats (Figure 3A and B).

Figure 3.

Chronic eple treatment reduces myocardial fibrosis. Myocardial fibrosis and the anti‐fibrotic effects of eple were measured at the histological and molecular levels on rat myocardium. (A) Sections of paraffin‐embedded hearts obtained from untreated (upper panels) and MRA‐treated (lower panels) ^12.5SHHF^+/+ and ^12.5SHHF^cp/cp animals (^12.5SHHF^+/+, n = 5; ^12.5SHHF^+/+eple, n = 6; ^12.5SHHF^cp/cp, n = 4; ^12.5SHHF^cp/cpeple, n = 6) were stained with Sirius red enabling the identification of fibrotic areas (evidenced in magenta). (B) Fibrotic areas in each experimental group were quantified using imagej software and expressed as percentage of total section area. (C) The impact of eple on mRNA expression levels of genes involved in cardiac fibrosis and remodelling [fibronectin (Fn1), vimentin (Vim), transforming growth factor β2 and β3 (Tgfb2 and Tgfb3), collagen 3 type α 1 (Col3a1) and NADPH oxidase 4 (Nox4)] was assessed in cardiac tissue from ^12.5SHHF^+/+eple rats compared with ^12.5SHHF^+/+ rats (light grey, n = 4) and in ^12.5SHHF^cp/cpeple rats compared with ^12.5SHHF^cp/cp rats by RT‐qPCR (dark grey, n = 4). Comparative threshold cycles (C_T) for Sdha, Actb and B2m genes, considered as housekeeping genes, were used for qPCR data normalization. Gene fold regulation is expressed as the ratio of ΔCT of the treated group over the ΔCT of the placebo group of the same genotype at 12.5 months of age. #P < 0.05 for the comparison of the untreated and treated animals from the same genotype at the same age by Student's t‐test.

From 1.5 to 11.5 months of age, SHHF^cp/cp rats exhibited a progressive alteration of their diastolic function (increased IVRT and EDT values over time, Figure 4A and B). By 12.5 months, SHHF^cp/cp LV remodelling was exacerbated by concomitant functional alterations. ^12.5SHHF^cp/cp rats exhibited altered diastolic function and increased filling pressures (reduced IVRT and EDT values, Figure 4A and B) as well as a decline in systolic function (reduced EF values, Figure 4D) compared with ^12.5SHHF^+/+ rats. The reduction in Sa (marker of an altered longitudinal shortening and lengthening of the myocardium) observed in ^12.5SHHF^cp/cp rats further characterized the decline in their systolic function (^12.5SHHF^cp/cp Sa = 2.64 ± 0.10 cm⋅s⁻¹ vs. ^12.5SHHF^+/+ Sa = 3.41 ± 0.07 cm⋅s⁻¹; Figure 4C).

Figure 4.

Chronic eple treatment improves SHHF^cp/cp cardiac functional parameters. Echocardiographic monitoring of cardiac functional parameters where at 1.5 months of age n = 9 SHHF^+/+ rats were evaluated then n = 8 of them at the other time points; n = 8 SHHF^+/+eple were evaluated at each time point; n = 8 SHHF^cp/cp were evaluated at 1.5, 3 and 6 months, n = 7 of them at 9 months and n = 6 of them at 11.5 and 12.5 time points. Finally, n = 10 SHHF^cp/cpeple were evaluated from 1.5 to 11. 5 months of age and n = 11 of them at 12.5 months. (A) Isovolumic relaxation time (IVRT), (B) E wave deceleration time (EDT), (C) Sa duration (Sa) and (D) ejection fraction (EF). Two‐way ANOVA with genotype (SHHF^+/+ vs. SHHF^cp/cp) and treatment (placebo vs. eple) was performed, and the significance level was set to P < 0.0127. *P < 0.0127 for the comparison between the genotypes at the same age, #P < 0.0127 for the comparison of the untreated and treated animals from the same genotype at the same age.

The accelerated decline in cardiac function in SHHF^cp/cp rats in comparison with SHHF^+/+ rats arose in part from the early onset of drastic metabolic disorders. Already overweight at 1.5 months of age (^1.5SHHF^cp/cp = 150 ± 4 g vs. ^1.5SHHF^+/+ = 133 ± 2 g; Figure 5A), ^1.5SHHF^cp/cp rats exhibited dyslipidaemia that dramatically and significantly worsened with ageing when compared with SHHF^+/+ rats (Table 1). While the genotype did not influence fasting glycaemia, ^12.5SHHF^cp/cp insulin and adiponectin levels were increased by approximately fourfold and decreased by approximately twofold, respectively, when compared with ^12.5SHHF^+/+ animals (Table 1). This paralleled the development of massive visceral obesity in SHHF^cp/cp rats [body weight (BW), ¹²SHHF^cp/cpBW = 787 ± 17 g vs. ¹²SHHF^+/+BW = 456 ± 14 g] (Figure 5A). Characterization of ^12.5SHHF^cp/cp perirenal adipocyte size (mean cross‐sectional area) and distribution (median) (inserts within Figure 5B panels) revealed significantly larger adipocyte size and broader distribution than that observed in ^12.5SHHF^+/+ rats (Figure 5B left panels).

Figure 5.

Chronic eple treatment reduces body weight gain and visceral adipocyte hypertrophy. Rats received either MRA (eple, 100 mg⋅kg⁻¹⋅day⁻¹) or placebo from the age of 1.5 months up to 12.5 months. (A) Follow‐up of animal body weight (BW). At 1.5 months of age n = 9, SHHF^+/+ rats were evaluated then n = 8 of them at the other time points; n = 8 SHHF^+/+eple were evaluated at each time point; n = 8 SHHF^cp/cp were evaluated at 1.5, 3 and 6 months, n = 7 of them at 9 months and n = 6 of them at 11.5 and 12.5 time points. Finally, n = 11 SHHF^cp/cpeple were evaluated at each time point. *P < 0.05 SHHF^cp/cp compared with SHHF^+/+ at the same age; #P < 0.05 treated group compared with the untreated group of the same genotype at the same age by Student's t‐test. (B) Morphological analysis of perirenal VAT dissected at month 12.5. Analysis of VAT sections stained with Sirius red allowing the measurement of the cross‐sectional adipocyte area (CSAA, mm²) and cell size distribution (%) of the adipocyte population from each experimental group (SHHF^+/+ n = 5; SHHF^+/+eple n = 4; SHHF^cp/cp n = 5; SHHF^cp/cp eple n = 8).

Table 1.

Mineralocorticoid receptor blockade reduces dyslipidemia and preserves adiponectinemia

Experimental group n	1.5SHHF+/+ 7	12.5SHHF+/+ 10	12.5SHHF+/+eple 8	1.5SHHFcp/cp 7	12.5SHHFcp/cp 6	12.5SHHFcp/cpeple 11	ANOVA
							Genotype	Treatment	Interaction
Lipid profile, g⋅L−1	0.69 ± 0.03	0.90 ± 0.25	0.75 ± 0.28	0.93 ± 0.03	4.25 ± 0.33 *	2.67 ± 0.24 #	<0.05	<0.05	<0.05
Total chol.
HDL	0.24 ± 0.01	0.25 ± 0.02	0.27 ± 0.02	0.30 ± 0.01	0.56 ± 0.02 *	0.50 ± 0.02	<0.05	NS	NS
LDL	0.38 ± 0.03	0.08 ± 0.00	0.08 ± 0.01	0.48 ± 0.03	1.48 ± 0.40 *	0.55 ± 0.08 #	<0.05	<0.05	<0.05
TG	0.37 ± 0.01	0.61 ± 2.02	0.58 ± 2.26	0.77 ± 0.14	20.5 ± 2.6 *	8.26 ± 1.93 #	<0.05	<0.05	<0.05
FFA	1.2 ± 0.1	0.4 ± 0.0	0.3 ± 0.1	1.6 ± 0.2	1.0 ± 0.2	1.5 ± 0.3	<0.05	NS	NS
Fasting glycemia, mmol⋅L−1	4.6 ± 0.4	5.7 ± 0.4	6.1 ± 0.4	5.7 ± 0.5	5.4 ± 0.6	6.4 ± 0.4	NS	NS	NS
Fasting insulin, μIU⋅mL−1	11 ± 2	12 ± 5	10 ± 5	16 ± 4	52 ± 6 *	69 ± 5	<0.05	NS	NS
HOMA‐IR index	0.8 ± 0.2	0.9 ± 0.5	0.9 ± 0.5	2.6 ± 0.8	3.6 ± 0.6 *	5.4 ± 0.6	<0.05	NS	NS
Ionogram, mmol⋅L−1 Na+	141 ± 0	143 ± 3	139 ± 3	140 ± 1	146 ± 3	140 ± 2	NS	NS	NS
K+	4.8 ± 0.2	4.9 ± 0.2	5.2 ± 0.2	4.9 ± 0.1	4.6 ± 0.3	5.0 ± 0.2	NS	NS	NS
Na+/K+ ratio	29.9 ± 1.4	30.2 ± 1.9	27.1 ± 2.0	28.9 ± 0.6	33.1 ± 2.3	28.4 ± 1.7	NS	NS	NS
Adiponectin, μg⋅mL−1	6.3 ± 0.4	6.3 ± 1.3	7.4 ± 1.2	19.7 ± 1.4	11.9 ± 0.7 *	22.7 ± 1.6 #	<0.05	<0.05	<0.05
BNP, pg⋅mL−1	135 ± 18	219 ± 25	208 ± 25	102 ± 15	225 ± 29	203 ± 22	NS	NS	NS

HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; FFA, free fatty acids; TG, triglyceride; Na⁺, sodium, K⁺, potassium; BNP, brain natriuretic peptide. Values are the mean ± SEM. Two‐factor non‐parametric ANOVAs analysis with two factors allowed the evaluation of interaction between genotype and treatment at 12.5 months of age. Pairwise comparison for adjusted multiple testing set to P < 0.0127 in order to preserve the 5% overall α error rate.

^* To compare SHHF cp/cp versus SHHF+/+ at the same time point.

^# To compare eple versus placebo for a same genotype; NS, non‐significant; n represents for the number of samples.

Altogether, the above results indicate that the presence of metabolic disorders in SHHF^cp/cp rats are associated with adverse cardiac remodelling in these obese rats as compared with lean SHHF^+/+ rats, resulting in an impairment of ^12.5SHHF^cp/cp diastolic and systolic functions.

Preventive chronic eplerenone monotherapy mainly preserves SHHF^cp/cp myocardial structure and function

Preventive effects of eplerenone in SHHF^+/+eple rats

Initiation of the 11‐month eple treatment at 1.5 months of age in lean SHHF^+/+ rats had preventive effects on their cardiac phenotypes. Chronic MRA treatment significantly and sustainably reduced the hypertrophic remodelling by preventing SW thickening as early as 9.5 months of age (Figure 2B). Although myocardial fibrosis was detectable in ^12.5SHHF^+/+ rats (Figure 3A, upper left panel), its quantification (Figure 3B) did not reveal any significant effect of eple treatment. Molecular characterization (Figure 3C) of myocardial fibrosis was also assessed by the relative quantification of several transcripts encoding for structural and extracellular matrix proteins [fibronectin (Fn1), vimentin (Vim), transforming growth factor 2 and 3 (Tgfb2 and Tgfb3 respectively), collagen type 3a1 (Col3a1) and NADPH oxidase 4 (Nox4)]. None of these transcripts exhibited a significant reduction in their expression levels. The assessment of SHHF^+/+ diastolic (IVRT and EDT) and systolic (Sa and EF) cardiac function also demonstrated fairly stable parameters with ageing, none of which were significantly affected by eple treatment (Figure 4A–D, black and dark grey columns).

Hence, the prevention of LV concentric remodelling observed in lean SHHF^+/+eple rats suggests that they modestly but significantly benefited from the early initiation of eple treatment.

Greater benefits of eplerenone treatment in obese SHHF^cp/cp rats

Chronic MRA alleviated the cardiac eccentric remodelling of SHHF^cp/cpeple rats by significantly reducing their LV mass at 11.5 months of age (Figure 2C) and heart weight/tibia length ratio at age 12.5 months (^12.5SHHF^cp/cpeple HW/TL = 49.5 ± 1.3 mg⋅mm⁻¹ vs. ^12.5SHHF^cp/cp HW/TL = 57 ± 1.7 mg⋅mm⁻¹). Eple treatment also prevented the LV remodelling of ^12.5SHHF^cp/cpeple myocardium by limiting its dilation and septal thickening (Figure 2A and B). These structural improvements in ^12.5SHHF^cp/cpeple LV were accompanied by a significant reduction in myocardial fibrosis, initially evaluated and quantified at the histological level (Figure 3A and B, #P < 0.05) and subsequently confirmed at the molecular level using RT‐qPCR (Figure 3C). Significant down‐regulations of Fn1, Vim, Tgfb2, Tgfb3, Col3a1 and Nox4 were demonstrated upon treatment in the ^12.5SHHF^cp/cpeple myocardium when compared with their ^12.5SHHF^cp/cp untreated controls (Figure 3C).

At 12.5 months of age, chronic eple treatment had no significant effect on myocardial MR transcript expression in either genotype (not shown) but significantly decreased the expression of the Hsd11β1 transcript (encoding for the enzyme converting cortisone into corticosterone) only in ^12.5SHHF^cp/cpeple rats (Hsd11β1 transcript 2^{−ΔCT12.5SHHFcp/cpeple}/2^{−ΔCT12.5SHHFcp/cp} = −1.466).

At 12.5 months of age, chronic eple treatment had no significant effect on perirenal adipose tissue Hsd11β1 transcript expression in either genotype (not shown). This did not exclude the possible participation of this enzyme to the observed VAT phenotypes at earlier time points than the end of the study, but suggested that the effects on phenotype were mostly as a result of MR antagonism in adipose tissue.

Diastolic function preservation

In HF pathophysiology, the link between structure and function is relatively straightforward in particular for diastolic function, which is adversely affected by LV hypertrophy. In addition to the reduction in cardiac hypertrophy, results relative to cardiac functional parameters (Figure 4A–D) showed that eple prevented the myocardial relaxation impairment already observed in ³SHHF^cp/cp rats. At the age of 3 months, diastolic parameters were less altered in the eple group, as demonstrated by a significant increase in EDT upon treatment (³SHHF^cp/cpeple EDT = 26.5 ± 2.3 ms vs. ³SHHF^cp/cp EDT = 35.8 ± 2.6 ms) and a concomitant trend for a smaller E wave velocity ((³SHHF^cp/cpeple E = 114 mm⋅s⁻¹ vs. ³SHHF^cp/cp E = 106 mm⋅s⁻¹). In the ^12.5SHHF^cp/cpeple group, treatment also prevented the increase in E/A (not shown) and the decrease in IVRT (Figure 4A).

Systolic function preservation

Eple further improved cardiac systolic function in the ^12.5SHHF^cp/cpeple group by preventing the drop in Sa wave observed in ^12.5SHHF^cp/cp controls (^12.5SHHF^cp/cpeple Sa = 3.29 ± 0.09 cm⋅s⁻¹ vs. ^12.5SHHF^cp/cp Sa = 2.64 ± 0.10 cm⋅s⁻¹, P < 0.0001) as well as the decrease in EF (Figure 4C and D). Concomitant with the observed alteration in systolic function in obese placebo ^12.5SHHF^cp/cp rats, an increase in LV filling pressure was also noted, as shown not only by the decrease in IVRT and EDT (Figure 4A and B) but also by an elevation in E/e′ and E/A ratios, which were prevented by eple treatment (not shown).

Altogether, these results demonstrate that SHHF^cp/cpeple animals greatly benefited from early eple treatment by reducing the adverse cardiac remodelling observed in their untreated controls. The treatment positively affected both diastolic as well as systolic functions and delayed the onset of HF in the obese rats by preserving the decline in LV function.

Chronic eplerenone further benefits obese SHHF^cp/cp rats by greatly improving their metabolic status

While long‐term MR antagonism had no effect on SHHF^+/+eple body weight (BW), eple significantly and sustainably reduced SHHF^cp/cpeple BW reaching a maximal value at 12 months of age (¹²SHHF^cp/cpepleBW = 718 ± 13 g vs. ¹²SHHF^cp/cpBW = 787 ± 17 g) (Figure 5A). Eple treatment significantly minimized weight gain of ¹²SHHF^cp/cp‐treated rats (563 ± 43 g gained) when compared with ¹²SHHF^cp/cp‐untreated rats (634 ± 27 g gained) and this was achieved without significantly modifying their food and water intake or their 24 h urinary excretion (data not shown). Eple had no impact on the size and distribution of ^12.5SHHF^+/+ adipocytes (Figure 5B, upper panels) although chronic MRA treatment resulted in significantly less hypertrophy of ^12.5SHHF^cp/cpeple adipocytes than that observed in untreated controls (Figure 5B, lower panels).

Eple monotherapy partially prevented the respective increases in total cholesterol (−40%), LDL (−65%) and TG (−60%) concentrations as well as the decrease in adiponectin blood concentration observed in ^12.5SHHF^cp/cp controls (Table 1). MRA treatment had no significant effects on either the Na⁺/K⁺ ratio, excluding any potential toxic effect on salt and water balance due to the long‐term use of the MRA (Table 1), or on BNP plasma levels in surviving animals at the end of the study.

Thus, SHHF^cp/cpeple rats benefited from eple treatment at both the cardiac and metabolic level. Together, these phenotypic improvements participated in a greater survival rate as indicated by the Kaplan–Meier representation where 100% of SHHF^cp/cpeple rats completed the protocol (Figure 1B).

Correlation of metabolic and cardiac physiological profiles of all experimental groups

The statistical correlations of several metabolic and cardiac physiological parameters of the six experimental groups (^1.5SHHF^+/+; ^1.5SHHF^cp/cp; ^12.5SHHF^+/+; ^12.5SHHF^+/+eple; ^12.5SHHF^cp/cp and ^12.5SHHF^cp/cpeple) are presented as a correlogram in Figure 6. This approach allowed a better understanding of the evolution of selected cardiac parameters (vertical columns) using the probability of a significant relationship with metabolic values independently of the age, the genotype and the treatment (horizontal lines). Correlation coefficients (r) are systematically ranked between −1 to +1, where −1 value represents a perfect negative correlation between two variables, while + 1 represents a perfect positive correlation. The darker blue and red squares indicate a significant strong positive (from 0.6 to 1) and negative (from −1 to −0.6) correlation, respectively, while squares in light blue and pink denote a significant intermediate correlation (from 0.2 to 0.6 and from −0.6 to −0.2 respectively). The grey squares show the non‐significant‐matched parameters.

Figure 6.

Correlation of metabolic and cardiac physiological profiles of all experimental groups. The CoExpress software was used for calculating correlation coefficients (r) based on Pearson's test between physiological parameters and the corrplot function to elaborate the depicted correlogram. Grey coloured rectangles are poorly correlated and not significant (pVal > 0.05) while the coloured scale bar ranges from red (r from −1 to −0.6) to dark blue (r from 0.6 to 1) and intermediate values (ar from −0.6 to −0.2 in pink and from 0.2 to 0.6 in light blue). Cardiac parameters are represented horizontally with EF denoting ejection fraction; IVRT, isovolumetric relaxation time; SWT, septum wall thickness; LV mass, left ventricle mass; HW, heart weight. Metabolic parameters are represented vertically with LDL, HDL denoting low‐density and high‐density lipoprotein respectively; Tot Chol, total cholesterol; TG, triglycerides and FFA, free fatty acids.

In the present series, regardless of the genotype, the age and the treatment, the high values of the metabolic parameters (Ins, LDL, HDL, Tot Chol and TG serum levels) were systematically and significantly correlated with high heart weight (HW) values and their lower values were statistically correlated with lower HW values. Together, these above‐cited parameters exhibit a statistically positive correlation between each other (blue squares). Conversely, Ins, LDL, HDL, total cholesterol and TG serum levels showed a very systematic and significant negative correlation (red squares) with the cardiac ejection fraction (EF), suggesting that when those metabolic parameters rise, the systolic function is depressed.

Discussion and conclusion

The efficacy of eple in preventing the obesity‐accelerated progression towards HF was assessed in the present study over a course of the longest treatment duration reported to date in the literature. The design of the present study allowed us to demonstrate long‐term sustained effects of eple and their modulation by the presence (SHHF^cp/cp) or absence (SHHF^+/+) of metabolic disorders. Using this specific approach, we demonstrated that chronic MR antagonism initiated when young pre‐hypertensive SHHF rats are at risk of developing HF but not yet symptomatic was able to sustainably prevent cardiac remodelling in a haemodynamically‐independent manner. Incremental benefits were further observed when the preventive treatment was given to pre‐hypertensive, overweight and dyslipidaemic SHHF^cp/cp rats. Indeed and as suggested by the correlation data, eple had the most striking effects on traits related to obesity by reducing weight gain, adipocyte hypertrophy and dyslipidaemia, all of which contributed to the prevention of cardiac remodelling and alterations in function in SHHF^cp/cpeple rats. While clinical data are limited regarding the ability of MRA to prevent the risk of developing HF in obese patients, our results strongly suggest that in the context of obesity, the targeted antagonism of MR overactivation appears as a promising therapeutic approach in this setting. Our study provides evidence that will potentially open new avenues for drug repositioning of MRA for the prevention of HF, particularly in high‐risk patients presenting obesity‐related disorders.

Compared with other studies (Rocha et al., 1998; Bender et al., 2013; Cezar et al., 2013), our data further characterized the anti‐fibrotic properties of eple in the context of obesity by assessing the sustained long‐term effects of this MRA. The decrease in Tgfb2 (mostly expressed in epithelial cells) and Tgfb3 (primarily expressed in mesenchymal cells), Col3a1, Fn 1, Nox4 and Vim transcripts upon eple treatment strongly suggests that MR activation is also involved in the deleterious differentiation of cardiac fibroblasts into myofibroblasts. Such differentiation is known to occur during HF development (Cucoranu et al., 2005; Heymans et al., 2015) and during the endothelial‐to‐mesenchymal transition that is reactivated in the diseased adult heart (Welch‐Reardon et al., 2015). The reduced expression of the Nox4 transcript in ^12.5SHHF^cp/cpeple rats also suggests that the beneficial effect of MRA is partly the result of sustained anti‐oxidative stress effects previously reported shortly after the initiation of the treatment (Kuster et al., 2005; Bender et al., 2015).

The eple‐induced down‐regulation of these transcripts furthermore alleviated the alterations in cardiac diastolic relaxation, LV stiffness, hypertrophy and subsequent dilated remodelling observed in SHHF^cp/cpeple rats, thus confirming preclinical data obtained when MRA treatment was given for a shorter duration (Rocha et al., 1998; Cezar et al., 2013; Bender et al., 2015). The unique therapeutic opportunity provided by early preventive MRA administration enabled the SHHF^cp/cpeple rats to be protected from developing diastolic dysfunction as the increase in filling pressures and early impairment of myocardial relaxation were prevented. While the sudden death of the sickest animals in the control SHHF^cp/cp‐untreated group prevented the detection of significant differences in systolic function prior to the last echocardiographic time point, eple treatment nonetheless significantly preserved the systolic function of these SHHF^cp/cpeple rats. Furthermore, given that the severity of myocardial fibrosis has been associated with long‐term mortality in HF patients (Azevedo et al., 2010; Aoki et al., 2011; Fraccarollo et al., 2011), it is likely that myocardial fibrosis caused the lower survival rate observed in untreated animals, while the anti‐fibrotic property of eple contributed to the improved survival rate of the SHHF^cp/cp eple rats.

Given the pleiotropic activity of the MR, it is likely that the observed beneficial effects may also be related to MR antagonism in other MR target organs in addition to the myocardium and adipose tissues. Because metabolic disorders are well‐established risk factors for developing cardiovascular diseases, the authors cannot refute the possibility that the prevention of hepatocellular damage mediated by MRA treatment, demonstrated in non‐alcoholic fatty liver disease experimental models by others (Wada et al., 2010, 2013; Pizarro et al., 2015), is also involved in the global improvement of the health of treated rats. Of interest, the efficacy of eple at ameliorating histological steatosis and hepatic fibrosis in mice even allows us to consider the MR as a novel potential therapeutic target for insulin resistance and non‐alcoholic liver diseases (Pizarro et al., 2015). Likewise, an improvement in kidney function (Kang et al., 2009; Lian et al., 2012) as well as decreased sympathetic drive and improved baroreflex functions in HF are likely to contribute to the reported beneficial effects of MRA treatment. In addition to this pleiotropic effect of MR, the down‐regulation of the Hsd11b1 enzyme in the myocardium upon MRA treatment might also participate in the cardiac phenotype improvement by locally reducing the inflammatory response. Further analyses are required to better comprehend certain discrepancies in the reported metabolic effects of MRA that are difficult to reconcile or inconsistent with previous results reported in the literature, as outlined in Supporting Information Table S4 (Bender et al., 2015; Bostick et al., 2015). Antihypertensive and glucose homeostasis properties of the MRA could have been masked by the complexity of the SHHF strain and/or the 11‐month duration of the treatment, highlighting long‐term chronic effects rather than an acute response to MRA, as mostly described in the literature. In the SHHF strain, the development of metabolic and hypertensive disorders is genetically determined rather than diet‐induced (such as in high‐fat/high‐fructose and high‐salt diets), and several concomitant neuroendocrine alterations (Heyen et al., 2002; Radin et al., 2003; Radin et al., 2008; Przybylski et al., 2010) have been reported in addition to hyperaldosteronism. Moreover, MRAs vary in their effects and some of the differences reported between the effects of spironolactone and eple may result from the fact that eple is devoid of the anti‐progesterone, anti‐androgenic and anti‐glucogenic side effects of spironolactone. Notwithstanding the latter, the eple‐induced overall protection reported in the present study is predominantly the result of MR antagonism because eple is only marginally antagonistic to other receptors. The absence of expected insulin‐sensitivity properties of the preserved levels of circulating adiponectin could be explained by the fact that one of the adiponectin target organs, the liver, was shown to develop severe non‐alcoholic steatosis in the SHHF^cp/cp rat (Youcef et al., 2014) that may preclude its favourable response to adiponectin. Of note, such impairments in the benefits of adiponectin have previously been reported in obese experimental models (Hui et al., 2012). Likewise, the absence of BNP regulation upon treatment is likely to be a reflection of the complex relationship between the amount of visceral fat distribution and BNP levels (Clerico et al., 2012).

While the above issues remain to be investigated in the SHHF model and in tissues other than the myocardium and visceral adipocyte pads, eple treatment had conversely striking effects on preventing the development of dyslipidaemia and the decrease in adiponectin concentration in ^12.5SHHF^cp/cpeple rats, both of which are considered to be good prognostic factors in HF patients. The fact that adiponectin is reported to act directly on cardiomyocytes, alleviating the development of hypertrophy, cardiomyopathy and systolic dysfunction (Goldstein et al., 2009), is also consistent with its participation in the preservation of cardiac function in the present model (Guo et al., 2008; Hirata et al., 2009; Armani et al., 2014).

There is also accumulating evidence suggesting that MR overactivation mediates pathophysiological changes that participate in the development of MetS and the associated decline in cardiovascular function. Clinical reports support the notion that hypertensive and primary hyperaldosteronism patients exhibit more frequent impairment of metabolic signalling, dyslipidaemia and obesity (Manrique et al., 2005; Van Gaal et al., 2006; Sowers, 2007). Corticosterone is an upstream precursor molecule of the mineralocorticoid aldo. Upon eple treatment, the myocardial transcript of the enzyme responsible for its production, Hsd11β1and its urinary levels were decreased. This suggests that eple treatment interrupted the positive feedback loop in which MR activation participates in the up‐regulation of Hsd11β1 gene expression (Nagata et al., 2006). Because aldo is synthesized from an upper precursor, that is, cholesterol, one could also speculate that a similar positive feedback loop is also present between MR activation and regulation of cholesterol levels, which may explain the anti‐cholesterol effect of MRA treatment in our experimental model. To the best of our knowledge, the effect of MRA on dyslipidaemia in humans has only been described in the ASCOT clinical trial where MRA treatment reduced hyperlipidaemia (reduction in total and LDL cholesterol levels) in individuals with resistant hypertension (Chapman et al., 2007). This purported MRA effect on dyslipidaemia could potentially benefit from post hoc analyses of data collected in large trials evaluating the effect of MRA on HF patients in order to reinforce the data obtained in the small pilot trials by Kosmala et al. (2012, 2013). In agreement with in vitro as well as the sparse in vivo findings, our results indicate that MR activation is involved as a crucial regulator of adipocyte function and hypertrophy (Caprio et al., 2007; Hirata et al., 2009; Wada et al., 2013). More importantly, our findings further demonstrate that the reported reduced adipocyte hypertrophy observed after shorter treatment durations in vivo (Caprio et al., 2007; Hirata et al., 2009) was sustained after 11 months and may account in part for the rarely described reduced weight gain previously observed in ³SHHF^cp/cpeple rats. The fact that the BW of SHHF^+/+eple rats was not affected by eple treatment indicates that it has very minor effect, if any, on lean body mass.

Obesity and its associated metabolic abnormalities are major risk factors for the development of HF (Rimbaud et al., 2009; Lai et al., 2014). The correlogram of metabolic parameters with the cardiac parameters clearly demonstrate the physiological association between metabolic and cardiac parameters in our experimental model. The increased metabolic values paralleled the increases in cardiac remodelling and diastolic dysfunction (correlation of LV mass and HW with metabolic parameters), while the increased metabolic values paralleled the decrease in systolic function (negative correlation of EF with metabolic parameters). Our results further demonstrated that MRA not only delayed LV remodelling and dysfunction but also improved dyslipidaemia. Together with the improvement in adipocyte hypertrophy in SHHF^cp/cp rats, eple was found to have a beneficial effect on lipid homeostasis, which most likely contributed to preserving the quality of cardiac energy sources and the delay in HF development. Eple thus appears to be a promising drug that could prevent ventricular remodelling and cardiac function by preserving cardiac energy metabolism.

Given that late eple administration either rarely (Bender et al., 2015) or mostly fails to reverse cardiac outcomes (Susic et al., 2007; Cezar et al., 2013), the use of eple as a preventive rather than curative drug should be considered and further investigated in HF with preserved EF. Such an early treatment approach was assessed clinically in the treatment of HF with preserved EF using the MRA spironolactone (Pfeffer et al., 2015) but overall only neutral results were obtained. A post hoc analysis of the regional variation of this study concluded that the patients from America were the only patients benefiting from spironolactone treatment (Pfeffer et al., 2015). However, the baseline characteristics of these American patients displayed higher body mass index (32.9 vs. 29.4 kg⋅m⁻²) and dyslipidaemia (71 vs. 49%) than their Russian/Georgian counterparts. In the present experimental model, SHHF^cp/cp and SHHF^+/+ closely mimicked these patient group differences at baseline, and, likewise, our findings revealed a greater benefit of MRA in the metabolically‐affected subgroups of rats.

In summary, the present findings demonstrate that long‐term chronic MRA treatment started before the onset of cardiac alterations was beneficial to both lean and obese SHHF rats and contributed in delaying progression to the development of HF symptoms. These latter rats further benefited from the treatment through an improvement in their metabolic parameters. Given the converging evidence of the major role of excessive MR activation as an independent risk factor for metabolic and subsequent cardiovascular disease, our results support the use of MRA as a promising therapeutic strategy for the prevention of cardiac diseases especially in HF patients with altered metabolic parameters.

Author contributions

G.Y., A.O., N.N., A.M., C.D., R.M.R.G., C.L. and A.P. performed the research. A.P., F.Z. and L.V. designed the research study. R.M.R.G., C.L., R.F. and A.M. contributed essential tools. A.P., G.Y., A.O., C.L., R.F., F.Z. and P.L. analysed the data. A.P., G.Y., A.O., F.J., F.Z., P.L. and L.V. wrote the paper.

Conflict of interest

The authors disclose no conflicts of interest.

Declaration of transparency and scientific rigour

This Declaration acknowledges that this paper adheres to the principles for transparent reporting and scientific rigour of preclinical research recommended by funding agencies, publishers and other organizations engaged with supporting research.

Supporting information

Figure S1 Carotid hemodynamic assessment.

Table S1 Influence of the eplerenone treatment on urinary aldosterone and corticosterone excretion

Table S2 Lack of impact of eplerenone treatment on carotic hemodynamic properties

Table S3 Chronic eplerenone treatment does not impact hemodynamic parameters in SHHF rats

Table S4 Influence of eplerenone treatment applied as monotherapy in murine models

Supporting info item

Supporting info item

Supporting info item

Supporting info item

Supporting info item

Youcef, G. , Olivier, A. , Nicot, N. , Muller, A. , Deng, C. , Labat, C. , Fay, R. , Rodriguez‐Guéant, R.‐M. . , Leroy, C. , Jaisser, F. , Zannad, F. , Lacolley, P. , Vallar, L. , and Pizard, A. (2016) Preventive and chronic mineralocorticoid receptor antagonism is highly beneficial in obese SHHF rats. British Journal of Pharmacology, 173: 1805–1819. doi: 10.1111/bph.13479.