Mineralocorticoid Receptor Blockade Reverses Obesity-Related Changes in Expression of Adiponectin, PPARγ and Pro-inflammatory Adipokines

Abstract

Background

In obesity, decreases in adiponectin and increases in pro-inflammatory adipokines are associated with heart disease. Since adipocytes express mineralocorticoid receptor (MR) and MR blockade reduces cardiovascular inflammation and injury, we tested the hypothesis that MR blockade reduces inflammation and expression of pro-inflammatory cytokines in adipose tissue and increases adiponectin expression in adipose tissue and hearts of obese mice.

Methods and Results

We determined the effect of MR blockade (eplerenone, 100 mg/kg/day for 16 weeks) on gene expression in retroperitoneal adipose and heart tissue from obese, diabetic db/db mice (n=8) as compared with untreated obese, diabetic db/db mice (n=10) and lean, non-diabetic db/+ littermates (n=11). There was increased expression of tumor necrosis factor-α (TNF-α), monocyte chemoattractant protein-1 (MCP-1), plasminogen activator inhibitor type-1 (PAI-1) and macrophage protein CD68 and decreased expression of adiponectin and peroxisome proliferator-activated receptor-γ (PPARγ) in retroperitoneal adipose tissue from obese versus lean mice. Also, adiponectin expression in heart was reduced in obese versus lean mice. MR blockade prevented these obesity-related changes in gene expression. Further, treatment of undifferentiated preadipocytes with aldosterone (10⁻⁸ M for 24 h) increased mRNA levels of TNF-α and MCP-1, and reduced mRNA and protein levels of PPARγ and adiponectin, supporting a direct aldosterone effect on gene expression.

Conclusions

MR blockade reduced expression of pro-inflammatory and pro-thrombotic factors in adipose tissue and increased expression of adiponectin in heart and adipose tissue of obese, diabetic mice. These effects on adiponectin and adipokine gene expression may represent a novel mechanism for the cardioprotective effects of MR blockade.

INTRODUCTION

Obesity is a pro-inflammatory state characterized by adipose tissue inflammation, increased adipose tissue production of pro-inflammatory cytokines (e.g. tumor necrosis factor-α (TNF-α) and monocyte chemoattractant protein-1 (MCP-1)) and pro-thrombotic factors (e.g. plasminogen activator inhibitor type-1 (PAI-1)), and decreased adipose tissue production of adiponectin and peroxisome proliferator-activated receptor-γ (PPAR-γ) ^{1, 2}. These obesity-related changes in adipose tissue are linked to the development of insulin resistance, type 2 diabetes mellitus and cardiovascular injury ^1–6.

Activation of the mineralocorticoid receptor (MR) has been implicated in mediating the inflammation observed in vessels, heart and renal cortex of rodent models of diabetes and hypertension ^7–10. Further, large-scale clinical studies demonstrate beneficial effects of MR blockade on cardiovascular morbidity and mortality in patients with heart failure ^{11, 12}. MR blockade also decreases left ventricular mass in hypertensive patients with left ventricular hypertrophy with and without type 2 diabetes ¹³, improves coronary vascular function in patients with diabetes ¹⁴, reduces markers of inflammation in patients with diabetes ^{15, 16}, and decreases proteinuria in patients with diabetic and non-diabetic renal injury ^{17, 18}.

In the present studies we used in vivo and in vitro approaches to test the hypotheses that MR activation regulates expression of adipokines and that chronic MR blockade has beneficial effects on adipose tissue inflammation and adipose tissue expression of TNF-α, MCP-1, PAI-1, PPAR-γ and adiponectin. In vivo studies were performed in obese, diabetic db/db mice ¹⁹ and lean, nondiabetic db/+ heterozygous littermates. The effects of acute MR activation on expression of TNF-α, MCP-1, PPAR-γ and adiponectin were assessed in 3T3-L1 cells.

RESEARCH DESIGN AND METHODS

Animal procedures

Male obese db/db mice (Jackson Laboratory, Bar Harbor, ME), which are homozygous for an inactivating mutation in the leptin receptor leading to hyperphagia, obesity, insulin resistance and hyperglycemia ¹⁹, received either vehicle (saline) or the MR antagonist eplerenone via chow (0.6 mg eplerenone per gm chow) from age 8 to 25 weeks. The non-diabetic control group mice were male lean db/+ heterozygous littermates (Jackson Laboratory) studied from age 8 to 25 weeks. We reported previously that eplerenone reduces renal injury in these animals ⁹. Animals were kept in a room lighted 12 h/day at an ambient temperature of 22 ± 1°C. Animals had free access to drinking water and Purina Rodent Chow (Purina, St. Louis, MO). Body weight was obtained at 8 and 25 weeks of age. At 25 weeks of age, systolic blood pressure was measured in conscious animals by tail-cuff plethysmography (Blood Pressure Analyzer, Model 179, IITC Life Science) and animals were placed in individual metabolic cages for collection of urine over 24 hours. Mice were anesthetized with isoflurane. Blood, adipose tissue from the retroperitoneum, and hearts were harvested. Tissue samples were processed for immunohistochemistry studies as previously described ⁹. The Institutional Animal Care and Use Committee at Harvard University approved our experimental procedures.

Cell culture

3T3-L1 cells (ATCC, #CL-173, Manassas, VA) were maintained in Dulbecco’s modified Eagle’s medium (DMEM containing 4.5g/L glucose) with 10% fetal calf serum (FCS). Twelve hours prior to treatment, cells were switched to DMEM without FCS. At the time of treatment, cells were washed twice with phosphate buffer saline (PBS) and incubated with vehicle, aldosterone (10⁻⁸M) or aldosterone (10⁻⁸M) plus the water-soluble MR antagonist canrenoate (10⁻⁶M) in DMEM without FCS. The cells were harvested for RNA and protein analyses after 24-hour or 72-hour incubation, respectively. 3T3-L1 cells were differentiated into adipocytes in six well plates using the 3T3-L1 Adipocyte kit (Zen bio, #KT-01, Research Triangle Park, NC) as described previously ²⁰. Differentiated 3T3-L1 cells were cultured in DMEM, 0% FCS for 24 hours and then stimulated with aldosterone (10⁻⁸M) for 24 hours.

Plasma and urine assays

Plasma insulin was measured using the LincoPlex mouse insulin assay (LINCO Research, St. Charles, MO). Plasma glucose and triglycerides were measured using Roche Cobas Integra 400 (Roche Diagnostics, Indianapolis, IN) via a hexokinase enzymatic reaction for determination of glucose and via an enzymatic and colorimetric method with glycerol phosphate oxidase and 4-aminophenazone for triglycerides. To estimate insulin resistance the homeostatic model assessment (HOMA) index was calculated by the formula: fasting plasma insulin (µIU/ml) × fasting plasma glucose (mmol/l)/22.5 ²¹. The Cytometric Bead Array System was used according to manufacturer's instructions (mouse inflammation CBA kit, Cat #552364, BD Biosciences, San Jose, CA) to measure protein concentrations of interleukin-6 (IL-6), interleukin-10 (IL-10), MCP-1, interferon-γ (IFN-γ), TNF-α and interleukin-12p70 (IL-12p70) in mouse plasma. Cytokine concentrations were determined by flow cytometry (BD FACScan, BD biosciences). Results were calculated to take into account the total protein concentration of the plasma and are expressed as pg/ml. Intra-assay variability was 2% and inter-assay variability was 5%.

Twenty-four hour urine collections were assayed for aldosterone and creatinine. Urine was extracted with ethyl acetate, and after evaporation of ethyl acetate, extract was reconstituted in buffer. Aldosterone was measured using solid-phase RIA (Diagnostic Products Corp., Los Angeles, CA). Creatinine was assayed using the DCA 2000+analyzer (Bayer, Elkhart, IN) and the COBAS Integra 400 (Roche Diagnostics, Indianapolis, IN) for values that were below the limit of detection (15 mg/dl) of the Bayer assay. The Roche and Bayer assays correlated very well (y=0.97x + 2.52, r=0.97, p<0.0001).

Quantitative Real-Time PCR

Total mRNA was extracted from adipose tissue using the RNeasy Lipid Tissue Mini Kit (Qiagen Sciences, Germantown, MD) and from heart or 3T3-L1 cells using the RNeasy Mini Kit (Qiagen Sciences). cDNA was synthesized from 5µg RNA with the First Strand cDNA Synthesis kit (Amersham, Buckinghamshire, UK). PCR amplification reactions were performed with TaqMan gene expression assays in duplicate using the ABI Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, CA). ΔΔCT method was used to determine mRNA levels. Target gene expression was normalized to 18S rRNA levels.

Western blot analysis

Protein in 3T3-L1 cells was analyzed by quantitative Western Blotting as previously described ⁹. Briefly, 3T3-L1 cells were sonicated in lysis RIPA buffer (Santa Cruz Biotechnology Inc, Santa Cruz, CA), followed by centrifugation. Equal amounts (30 µg) of the protein were subjected to electrophoresis on 10% SDS-polyacrylamide gels. In addition, detection of circulating levels of high molecular weight (HMW) adiponectin was conducted as described ²². Briefly, 1 µl of plasma was loaded to 3–15% SDS-PAGE gel under non-reducing and non-heat denaturing conditions. The gels were transferred onto a nitrocellulose membrane by electro-blotting. Membranes were immunoblotted with 1:5,000 anti-adiponectin antibody (Chemicon, Temecula, CA) or 1:5,000 anti-PPARγ antibody (Santa Cruz). The blots were scanned using the Epson Perfection 1650 scanner and densitometric analysis was performed with the Imagequant 5.2 software (Molecular Dynamics).

Immunofluorescent Microscopy

Retroperitoneal adipose tissue sections (4 µm) were fixed in ice-cold acetone for 10 minutes, and preincubated with blocking solution containing 1% preimmune serum for 10 minutes. Tissue slices were incubated overnight at 4°C with mouse primary antibody to adiponectin (Chemicon, Temecula, CA) and rabbit primary antibody to MR (Santa Cruz Biotechnology, Santa Cruz, CA). After three washes with 0.5% blocking solution in PBS, tissue slices were incubated at 37°C for 30 minutes with secondary goat anti-mouse antibody tagged with Alexa Fluor 488 and secondary goat anti-rabbit antibody tagged Alexa Fluor 594. After washing with PBS and deionized water, the slices were air-dried, mounted with Vectashield mounting medium containing 4',6-diamidino-2-phenylindole (DAPI) (Cat #H1200, Vector Laboratories Inc, Burlingame, CA) and stored under cool and dark conditions. Images were obtained with a Nikon Eclipse 90i microscope (Nikon Instruments Inc, Melville, NY) and processed using NIS-Elements Advanced Research Imaging Software AR 2.30, SP3 (Nikon Instruments Inc) according to the manufacturer's instructions. Alexa Fluor 488 yields a green color, Alexa Fluor 594 a red color and DAPI a blue color. In the absence of the primary antibody there was minimal fluorescence.

Data Analysis

Data was analyzed using One-way ANOVA followed by Tukey’s or Newman-Keuls post-hoc test for multiple comparisons. Differences in means with p values ≤0.05 were considered statistically significant. Values are expressed as mean ± standard error (SE). The authors had full access to the data and take responsibility for its integrity. All authors have read and agree to the manuscript as written.

RESULTS

Characteristics of lean db/+ and obese db/db mice

We studied three groups of animals from age 8 to 25 weeks: 1) lean, nondiabetic db/+ mice; 2) obese, diabetic db/db littermates receiving no treatment; and 3) obese, diabetic db/db mice receiving eplerenone 100 mg/kg daily. At age 25 weeks, db/db mice were heavier than db/+ mice and had elevated blood glucose levels (Table 1). Treatment of db/db mice with eplerenone did not have a significant effect on body weight, weight gain from 8 to 25 weeks, blood glucose levels or systolic blood pressure (Table 1). Urinary aldosterone to creatinine ratios in 24 hour urines collected at 25 weeks were higher in db/db mice as compared to db/+ mice and further elevated in db/db mice receiving eplerenone (Table 1).

Table 1.

Characteristics of lean db/+ and obese db/db mice at 25 weeks

	db/+	db/db	db/db
			eplerenone
	(n=11)	(n=10)	(n=8)
Weight gain from 8 to 25 weeks (g)	8.5 ± 0.6	19.9 ± 1.8*	18.3 ± 2.1*
Body weight (g)	32 ± 0.6	57 ± 2.2*	55 ± 2.0*
Systolic Blood Pressure (mm Hg)	112 ± 5	121± 5	126 ± 2
Urinary aldosterone/creatinine (ng/mg)	0.09 ± 0.02	0.24 ± 0.02†	1.18 ± 0.07*‡
Blood measurements
Triglycerides (g/l)	0.21 ± 0.04	0.41 ± 0.07†	0.22 ± 0.03§
Glucose (mg/dl)‖	159 ± 21	769 ± 41*	680 ± 70*
HOMA index‖	35 ± 11	334 ± 96#	132 ± 27§
MCP-1 (pg/ml) **	190 ± 4	286 ± 7*	224 ± 5#‡
HMW adiponectin (relative units)**	1.00 ± 0.09	0.68 ± 0.06***	0.93 ± 0.16
HMW adiponectin/total adiponectin**	0.51 ± 0.01	0.47 ± 0.02	0.48 ± 0.03

Data are mean ± SE

^*p<0.001 versus db/+

^†p<0.05 versus db/+

^‡p<0.001 versus db/db

^§p<0.05 versus db/db

^‖n = 5–7 per group

^#p<0.01 versus db/+

^**n = 4–6 per group

^***p<0.01 versus db/+ by Fisher Exact.

Effect of obesity and MR blockade on adipose tissue expression of adipokines

There was a 7- to 10-fold increase in expression of pro-inflammatory cytokines TNF-α and MCP-1 and the macrophage marker CD68 in retroperitoneal adipose tissue from 25 week-old obese db/db mice, as compared with lean db/+ animals (Fig 1 A–C). Adipose tissue from obese animals also showed elevated levels of PAI-1 mRNA as compared with adipose tissue from lean animals (Fig 1 D). Treatment with eplerenone markedly reduced mRNA levels of MCP-1, TNF-α, PAI-1 and CD68 in adipose tissue of db/db mice (Fig 1). Consistent with the gene expression studies in adipose tissue, plasma levels of MCP-1 were increased in db/db mice versus db/+ mice and treatment of db/db mice with eplerenone significantly reduced plasma MCP-1 levels (Table 1). Adipose tissue mRNA levels of MR and IL-6 and plasma levels of IL-6, IL-10, IL-12p70, TNF-α and IFN-γ were similar in the three groups (data not shown).

Figure 1. Adipose tissue expression of adipokines in db/db mice.

(A) TNF-α, (B) MCP-1, (C) CD68 and (D) PAI-1 mRNA levels in retroperitoneal adipose tissue of 25-week old lean db/+ mice, obese db/db mice, and obese db/db mice treated with the MR antagonist eplerenone (100 mg/kg/day from age 8 to 25 weeks). mRNA levels are expressed relative to 18S rRNA. n=8 per group. Data are mean ± SE.

Effect of obesity and MR blockade on cardiac and adipose tissue expression of adiponectin and PPARγ

We determined adipose tissue expression of adiponectin, PPARγ, and leptin (Fig 2 A–C). Leptin mRNA expression in fat was significantly increased in diabetic db/db mice versus db/+ mice, consistent with the loss of functional leptin receptors in db/db mice. In contrast, the levels of adiponectin and PPARγ mRNA were markedly decreased in adipose tissue of obese db/db mice compared to lean db/+ mice. MR blockade in db/db mice reduced adipose tissue expression of leptin and increased expression of both adiponectin and PPARγ to levels similar to those observed in lean animals (Fig 2 A–C).

Figure 2. Expression of adiponectin, PPARγ and leptin in adipose and heart tissue of db/db mice.

(A) Adiponectin, (B) PPARγ and (C) leptin mRNA levels in retroperitoneal adipose tissue and (D) adiponectin and (E) PPARγ mRNA levels in heart tissue of 25-week old lean db/+ mice, obese db/db mice, and obese db/db mice treated with the MR antagonist eplerenone (100 mg/kg/day from age 8 to 25 weeks). mRNA levels are expressed relative to 18S rRNA. n=8 per group. Data are mean ± SE.

Obese db/db mice had reduced levels of adiponectin mRNA in heart compared with lean db/+ mice (Fig 2 D). Treatment of db/db mice with eplerenone increased cardiac expression of adiponectin to levels observed in lean animals (Fig 2 D). In contrast, PPARγ expression in heart was similar across the three groups (Fig 2 E). Cardiac mRNA levels of MR, PAI-1 and TNF-α were similar in the three groups and there was no detectable leptin mRNA (data not shown).

Obese db/db mice, as compared with lean db/+ control mice, had lower levels of circulating HMW adiponectin, a form of adiponectin that is decreased in humans with type 2 diabetes. Plasma levels of HMW adiponectin were similar in db/+ mice and db/db mice receiving eplerenone (Table 1). The ratio of HMW to total adiponectin was similar in the three groups (Table 1).

Effect of obesity and MR blockade on HOMA index and triglyceride levels

Since our data demonstrates that MR blockade increases expression of insulin-sensitizing factors (adiponectin in heart and adipose tissue, and PPARγ in adipose tissue), and reduces expression of cytokines such as MCP-1 and TNF-α, which impair insulin sensitivity, we examined the effect of eplerenone on HOMA index and levels of triglycerides in blood samples obtained from anesthetized animals at the time of sacrifice. Triglyceride levels and HOMA index were elevated in obese diabetic db/db mice compared with lean mice, consistent with the insulin-resistant state of db/db mice, and eplerenone treatment reduced triglyceride levels and HOMA index (Table 1).

Effect of aldosterone on expression of adiponectin, PPARγ and adipokines in cultured adipocytes

Immunofluorescent staining of adipose tissue from db/db mice demonstrated expression of MR protein in adipocytes (Fig 3). To determine whether some of the effects of chronic MR blockade could be mediated through direct actions of MR on gene expression, we treated undifferentiated 3T3-L1 preadipocytes with aldosterone (10⁻⁸M) for 24 hours. Aldosterone increased mRNA levels of TNF-α six-fold and mRNA levels of IL-6 and MCP-1 approximately two-fold as compared to control treatment (Fig 4 A–C). In contrast, aldosterone treatment decreased mRNA levels of adiponectin and PPARγ to less than 50% of that observed with control treatment (Fig 4 D, E). With 72 hours of exposure to aldosterone there was a significant decrease in protein levels of adiponectin and PPARγ, consistent with the aldosterone-mediated decrease in mRNA expression (Fig 4 F, G). Furthermore, the MR antagonist, canrenoate, prevented these effects of aldosterone on gene expression (Fig 4). We also studied differentiated 3T3-L1 preadipocytes. Consistent with the observations in preadipocytes, incubation of differentiated adipocytes with 10⁻⁸ M aldosterone for 24 hours caused a 33% decrease in adiponectin mRNA levels, as compared to cells incubated with vehicle (0.67 ± 0.04 with aldosterone versus 1.00 ± 0.03 with vehicle, p<0.01, n=3 per condition).

Figure 3. Expression of MR in adipocytes of db/db mice.

Retroperitoneal adipose tissue of db/db mice was immunostained for MR - red (A) and adiponectin-green (B), which identifies adipocytes. DAPI-blue (C) staining identifies nuclei. Merged colors are shown in (D). Original magnification was 40x.

Figure 4. Expression of pro-inflammatory adipokines, adiponectin, and PPARγ in cultured 3T3-L1 preadipocytes.

(A) TNF-α, (B) IL-6, (C) MCP-1, (D) adiponectin and (E) PPARγ mRNA levels, and (F) HMW adiponectin and (G) PPARγ protein levels in 3T3-L1 preadipocytes treated with vehicle (control group), 10⁻⁸ M aldosterone (ALDO) and 10⁻⁸ M aldosterone with the MR antagonist 10⁻⁶ M canrenoate (ALDO+MRA). Treatment was 24 hours for mRNA studies and 72 hours for protein studies. mRNA levels are expressed relative to 18S rRNA. Protein levels are expressed relative to β-actin. n=8 per condition for mRNA and n=6–8 for protein per group. Data are mean ± SE.

DISCUSSION

These studies demonstrate reduced expression of adiponectin in heart and retroperitoneal adipose tissue and reduced expression of PPARγ in adipose tissue of obese, diabetic db/db mice as compared to lean, nondiabetic db/+ mice. Further, obese animals have increased adipose tissue inflammation and increased adipose tissue expression of PAI-1 and the pro-inflammatory cytokines TNF-α and MCP-1. Treatment of obese db/db mice with a MR antagonist reverses all of these obesity-related changes. Studies in 3T3-L1 preadipocytes demonstrate that aldosterone increases mRNA levels of TNF-α, MCP-1 and IL-6 and decreases mRNA and protein levels of adiponectin and PPARγ. Thus, some of the chronic effects of MR blockade in vivo may be secondary to blockade of MR’s effects on gene expression. These findings indicate that MR activation is a key factor mediating obesity-related changes in adipose tissue expression of pro-inflammatory and insulin-sensitizing factors, and in regulating adiponectin expression in heart.

Adipose tissues from obese db/db mice show obesity-related changes in adipose tissue inflammation and gene expression ²³ similar to those reported in humans ^{1, 2, 24, 25}. The current studies demonstrate that MR blockade prevents these obesity-related changes in adipose tissue, suggesting that obesity is associated with an activated MR. Multiple factors could contribute to increased MR activity in obesity, including increased levels of MR and aldosterone. While cardiac and adipose tissue expression of MR were similar in obese and lean mice, urinary aldosterone was elevated in the obese mice. Similarly, 24 hr urinary aldosterone levels are elevated in overweight as compared to lean individuals on a high salt diet, without any differences in urinary free cortisol levels ²⁶. These results suggest that increases in aldosterone contribute to increased MR activation in obesity.

Adipose tissue is the main site of adiponectin production, however, adiponectin is also expressed in heart tissue ²⁷. To our knowledge this is the first report that adiponectin expression is decreased in hearts of obese versus lean mice and that MR blockade increases cardiac adiponectin expression. This effect is specific for adiponectin, as cardiac expression of PPARγ is not regulated by obesity or eplerenone treatment. The decreased expression of adiponectin in heart and adipose tissue and reduced plasma levels of HMW adiponectin in the obese, diabetic mice are consistent with studies showing reduced adiponectin, and in particular HMW adiponectin, in individuals with insulin resistance and type 2 diabetes ^{22, 28, 29}. Our observation that MR blockade increases adiponectin expression is consistent with a report showing an increase in circulating adiponectin levels in nine diabetic individuals with poor glycemic control treated with spironolactone ¹⁵.

In db/db mice, the improvements in adipokine expression with MR blockade did not appear to be mediated by changes in weight, glycemia or blood pressure. However, we cannot rule out the possibility that our assays were not sufficiently sensitive to detect changes in glycemia and blood pressure. Our immunofluorescent studies demonstrate MR expression in retroperitoneal fat cells of db/db mice. To test whether MR activation has direct effects on adipokine gene expression, we studied aldosterone’s effects in 3T3-L1 cells, which express a functional MR ^30–32. Aldosterone increases expression of MCP-1, TNF-α, and IL-6 in cultured preadipocytes, consistent with aldosterone’s stimulatory effects on MCP-1 expression in brown adipocytes ³³, on MCP-1 and PAI-1 expression in the vasculature ^{34, 35}, and on IL-6 levels in humans ³⁶. Our studies do not determine whether these changes in gene expression are due to direct transcriptional effects of MR or are mediated through the rapid effects of MR on intracellular signaling pathways that could lead indirectly to changes in gene expression.

Two studies demonstrate a role for MR activation in the differentiation of 3T3-L1 preadipocytes into adipocytes, a process accompanied by increases in adiponectin expression ^{30, 32}. This aldosterone-mediated increase in adiponectin during adipocyte differentiation differs from the current studies showing decreases in adiponectin with aldosterone in both undifferentiated and differentiated 3T3-L1 cells, and may indicate that the ultimate effects of MR activation depend on the underlying state of the cell. However, our observation that MR activation decreases adiponectin and PPARγ expression in 3T3-L1 cells is in agreement with our observation that MR blockade increases adiponectin and PPARγ expression in adipose tissue of obese diabetic mice.

Adipose tissue contains multiple cells including preadipocytes, adipocytes, small blood vessels and macrophages. Cross-talk between the different adipose tissue cell types, in particular between macrophages, preadipocytes and adipocytes, is thought to be an important factor in promoting a pro-inflammatory state in obesity ^{25, 37}. It is possible that some of the eplerenone-mediated changes in adipose tissue gene expression may be related to changes in gene expression in these cells, as well as in adipocytes. Finally, leptin expression in adipose tissue is increased in db/db mice, which is attributed to the lack of a functional leptin receptor in these animals ¹⁹. Our observation that MR blockade reduces leptin expression in adipose tissue of db/db mice is consistent with a study showing aldosterone-stimulated increases in leptin gene expression in vitro ³³ and suggests there are MR-modified factors involved in leptin regulation. The leptin promoter contains multiple AP-1 sites ³⁸ and MR has been shown to regulate transcription through inhibition of AP-1/NFκB protein - DNA complexes ³⁹.

Obesity-related changes in adipose tissue expression (e.g. increases in cytokines such as TNF-α and MCP-1 and decreases in adiponectin and PPARγ) are thought to contribute to insulin resistance ^{4, 22, 28, 40–43}. Therefore, we examined indirect measures of insulin sensitivity (e.g. levels of triglycerides, insulin and glucose) in our studies. Obese, diabetic db/db mice are hyperglycemic and have elevated triglyceride levels and HOMA index compared with lean mice, consistent with an insulin-resistant state in db/db animals ¹⁹. Treatment with eplerenone leads to decreases in triglycerides and HOMA index, suggesting that MR blockade improves insulin sensitivity. Similarly, studies in healthy individuals on a high salt diet demonstrate a positive correlation between aldosterone and insulin resistance that is independent of age and body weight index ^{26, 44}. Further, in primary hyperaldosteronism, removal of an aldosterone-producing adenoma or treatment with a MR antagonist improves insulin sensitivity ⁴⁵. Aldosterone also decreases insulin receptor substrate-1 in rat vascular smooth muscle cells, which reduces insulin sensitivity ⁴⁶. These studies support the concept that aldosterone impairs insulin sensitivity.

Our studies suggest that MR activation plays an important role in obesity-related changes in cardiac expression of adiponectin and changes in adipose tissue function that lead to low-grade inflammation, insulin resistance and ultimately cardiovascular injury. While a number of treatments (diet and exercise ⁴⁷ and HMG-CoA reductase inhibitors ^{48, 49}) reduce obesity-related changes in adipose tissue inflammation and improve insulin sensitivity, diabetes mellitus and cardiovascular injury remain major causes of morbidity and mortality in obese individuals. Large clinical trials demonstrate that MR blockade reduces cardiovascular morbidity and mortality in patients with heart failure on maximum therapy including statins ^{11, 12}. The current findings suggest a novel mechanism for beneficial cardiovascular effects of MR blockade through reductions in adipose tissue inflammation and increases in adiponectin.