OUP user menu

Cardiac peroxisome proliferator-activated receptor δ is essential in protecting cardiomyocytes from oxidative damage

Guoliang Ding, Mingui Fu, Qianhong Qin, William Lewis, Ha Won Kim, Tohru Fukai, Methode Bacanamwo, Yuqing Eugene Chen, Michael D. Schneider, David J. Mangelsdorf, Ronald M. Evans, Qinglin Yang
DOI: http://dx.doi.org/10.1016/j.cardiores.2007.06.027 269-279 First published online: 1 February 2008


Objectives Peroxisome proliferator-activated receptors (PPAR) α and β/δ are essential transcriptional regulators of fatty acid oxidation in the heart. However, little is known about the roles of PPARδ in the heart. The present study is to investigate in vivo role(s) of PPARδ in the heart.

Methods A Cre–loxP mediated cardiomyocyte-restricted PPARδ knockout line was investigated. In these mice, exon 1 and 2 of PPARδ were targeted to eliminate PPARδ from cardiomyocytes.

Results PPARδ null mice exhibited pathological changes around 3 months of age, featuring progressive cardiac hypertrophy with mitochondrial oxidative damage. Most mice died from dilated cardiomyopathy. Cardiac expression of Sod2 (encoding manganese superoxide dismutase; MnSOD), a mitochondrial antioxidant enzyme was downregulated both in transcript and protein levels in cardiac samples in PPARδ knockout mice independent of pathological changes. Promoter analyses revealed that Sod2 is a target gene of PPARδ. Consequently, myocardial superoxide content in PPARδ knockout mice was increased, leading to extensive oxidative damage. Treatment with a SOD mimetic compound, MnTBAP, prevented superoxide-induced cardiac pathological changes in PPARδ knockout mice.

Conclusions The present study demonstrates that PPARδ is critical to myocardial redox homeostasis. These findings should provide new insights into understanding the roles of PPARδ in the heart.

  • PPARgamma
  • Sod2
  • Oxidative stress
  • Cardiac hypertrophy
  • Heart failure

1 Introduction

Peroxisome proliferator-activated receptors (PPARs) are members of the nuclear receptor superfamily. Cardiomyocytes express all three PPAR subtypes (α, β/δ, and δ). Of the two isoforms of PPARδ (PPARδ1 and δ2), PPARδ1 is predominantly expressed in the heart [1]. It is now clear that PPARα and PPARδ each plays a crucial role in myocardial fatty acid oxidation (FAO). Recently, important roles of PPARδ in the heart are also emerging. Studies using conventional gene targeting approach demonstrated that homozygous PPARδ null mice show embryonic lethality with documented cardiac defects [2]. Cre–loxp mediated tissue-specific PPARδ knockout mouse lines have been generated to explore the roles of PPARδ in various specific tissues. This molecular genetic approach pinpoints functional roles of PPARδ in liver [3], fat [4,5], skeletal muscle [6,7] and heart [8]. These novel observations together indicate that PPARδ is integral to structure and function in various tissues including the heart. Nevertheless, the subcellular mechanism by which PPARδ exerts its functions in the heart remains elusive.

We document here that PPARδ governs cardiac expression of Sod2, which encodes manganese superoxide dismutase (MnSOD), an essential mitochondrial antioxidant in the heart. Cardiomyocyte-restricted knockout of PPARδ in mice results in downregulated Sod2 and superoxide accumulation in the heart. The consequent oxidative damages in the heart lead to progressive cardiac hypertrophy, heart failure, and premature death.

2 Methods

2.1 Breeding of cardiomyocyte-restricted PPAR-δ knockout mice

Transgenic Cre (α-MyHC–Cre) mice (C57/BL6) that overexpress Cre specifically in cardiomyocytes [9] were crossed with PPARδflox/flox mice (C57/BL6) [7] to generate homozygous cardiomyocyte-restricted PPARδ knockout mice (CR-PPARδ−/−).

Animals received food and water on an ad libitum basis, and lighting was maintained on a 12-hour cycle. All experimental procedures were conducted in accordance with the Guide for Care and Use of Laboratory Animals of the National Institutes of Health, and were approved by the Institutional Animal Care and Use Committee of Atlanta University Center.

2.2 Transcript analyses

Total RNA samples were extracted using a RNA extraction kit (Qiagen). QPCR (Quantitative real-time RT-PCR) analyses (Roche LightCycler PCR system) were carried out to determine transcript levels of target genes. QPCR results from each gene/primer pair were normalized to 18 S or β-actin, and compared across conditions.

2.3 Protein analysis

Nuclear and cytosolic proteins were extracted using a nuclear protein extraction kit (Pierce). Mitochondria were extracted using a mitochondria isolation kit (SIGMA) as described previously [10]. Samples were subjected to SDS-PAGE gels and Immunoblotting was performed. Antibodies were obtained from commercial sources: PPARδ, Complex I and pan-actin (Santa Cruz Biotechnology); MnSOD (Upstate Biotechnology). Protein samples from adenoviral mediated PPARδ overexpressed cardiomyocytes were used as positive control for anti-PPARδ.

2.4 Pathological examinations

Five-μm paraffin embedded sections were stained with Masson's trichrome. To obtain tissues for transmission electron microscopy (TEM), hearts of anesthetized mice were perfused under gravity with 3.5% glutaraldehyde in cardioplegic solution (25 mM KCl, 5% dextrose in PBS, pH 7.4) for 2 min followed by perfusion with 3.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.3 for another 2 min. Samples were taken from the left ventricles.

2.5 In vivo hemodynamic measurement

The in vivo measurement of cardiac function was performed as previously described with modification [11]. In brief, mice were anesthetized with Avertin (375 mg/kg, i.p.) and ventilated. A 1.4 French Millar catheter-tip micromanometer catheter was inserted through the right carotid artery into the left ventricle, where pressure and volume were recorded by a computerized data-acquisition system (Millar Instrument).

2.6 Echocardiographic measurement

Mice were lightly anesthetized with Avertin (250 mg/kg, i.p.). An echocardiographic machine (Hewlett-Packard HP5500) with a 15 MHz linear transducer was used. Two-dimensional (2-D) guided M-mode in the mouse were performed. Left ventricle diastolic dimensions (EDD), and LV end-systolic chamber dimensions (ESD) were measured. All measurements were performed from leading edge to leading edge according to the American Society of Echocardiography guidelines [12]. The percentage of LV fractional shortening (LV%FS) were calculated as [(EDD−ESD)/EDD×100].

2.7 Cytochrome C reduction assay

The cytochrome C reduction assay was based on the measurement of NADPH-dependent superoxide generation. Left ventricular tissues were homogenized and centrifuged at 800 g for 10 min. The supernatant was incubated in the presence of 30 μmol/l succinylated ferricytochrome c and 1 mmol/l NADPH (Sigma). The change in absorbance at 550 nm was measured. The difference of reduced succinylated ferricytochrome c in the presence or absence of 0.3 mg/ml superoxide dismutase (SOD) (Sigma) was used to estimate the amount of superoxide generation by using an absorbance coefficient 21.1 mmol/l−1 cm−1 [13].

2.8 Cardiac aconitase activity

Aconitase activity was measured spectrophotometrically by monitoring the formation of cis-aconitate from isocitrate at 240 nm as previously described [14].

2.9 Mitochondrial potential measurement

Mitochondria from CR-PPARδ−/− and controlled hearts were isolated using a Mitochondria isolation kit (SIGMA) as described previously [10]. We used the JC-1 assay kit (SIGMA) to measure mitochondrial potential from the above mitochondria according to the manufacturer's instruction. Thirty μg of mitochondria protein in 30 μl were used. The relative fluorescence of the sample was measured in a spectrofluorometer (GENios plus) with 490 (excitation)/590 (emission) nm.

2.10 Adult mouse cardiomyocyte isolation and culture

Adult mouse cardiomyocytes were isolated from the hearts of adult mice and cultured as previously described [10].

2.11 Luciferase transfection assays

We subcloned the indicated mouse Sod2 promoter fragments with or without the predicted PPRE element into a luciferase reporter vector (pGL3-Basic). They were co-transfected with β-gal plasmid for normalization. H9C2 cells were seeded into 24-well plates the day before transfection to give a confluence of 50–80% when transfection was performed. After co-transfection of the above vectors into cultured H9C2 cells treated with rosiglitazone (10 μM) or DMSO, the relative luciferase activity were tested and compared in different vectors to see whether the PPRE elements are still functionally active.

2.12 Quantitative Chromatin Immunoprecipitation (ChIP) Assay

ChIP assays were performed as previously described [15,16]. H9C2 cells were treated with 10 μM rosiglitazone or DMSO for 8 h. An anti-PPARδ polyclonal antibody (Santa Cruz Biotechnology) was used in chromatin immunoprecipitation with IgG as a negative control. QPCR was performed using the target primer that recognizes a fragment (−956 to −1236) containing the putative PPRE consensus sequence and the other two primers (upstream and downstream) that recognize up-and downstream sequences (−490 to −592 and −1185 to −1490) apart from the putative PPRE site in the rat Sod2 promoter. The QPCR results from samples treated with PPARδ antibody were normalized to results from samples treated with IgG. Target primers: 5′-GGATTGTCGTTCATTCCAGA-3′ (forward) and 5′-ACCCCCTTTGCCTATTAAGGA-3′(reverse). upstream primers 5′-TTCGAGAGATCTCCTGGCATT-3′ (forward) and 5′-TAGTGGCTGTTCCTGGTTGT-3′ (reverse), downstream primer: 5′-TGCTAGTTAATTGCGAGGCTG-3′ (forward) and 5′-ATGGCGTAATCAGGGTCCTT-3′ (reverse). They were located at upstream (−1185 to −1490) and downstream (−490 to −592) of targeted regions.

2.13 MnTBAP treatment

The SOD mimetic drug, MnTBAP (OxisResearch) was dissolved in phosphate-buffered saline (PBS). CR-PPARδ−/− and α-MyHC–Cre mice at their ages of 2.5 months were given one bolus dose of MnTBAP (5 mg/kg/day, IP) daily for the periods of 1 month. Control groups of CR-PPARδ−/− and α-MyHC–Cre mice were given the same amount of PBS. At the end of the experiment, hearts were harvested after echocardiographic assessment.

2.14 Statistics analyses

All data were analyzed by one factor or mixed, two-factor analysis of Variance (ANOVA) followed by post-hoc analysis with the Student–Newman–Keuls test to detect differences between experimental groups using Statview 4.01 software (Abacus Concepts Inc.).

3 Results

3.1 Cardiomyocyte-restricted PPARδ knockout leads to cardiac hypertrophy and heart failure

We conducted RT-PCR on samples extracted from left ventricular tissues of the cardiomyocyte-restricted PPARδ knockout line (CR-PPARδ−/−) to confirm the cardiomyocyte-specific knockout of PPARδ. As reported previously [4,7], the recombined mRNA with the deletion of exons 1 and 2 in CR-PPARδ−/− hearts produced 300 and 400 bp products and the wild-type mRNA yielded a product of 700 base pairs (bp) (Fig. 1A). Only a 700 bp product could be detected on samples of other tissues such as lung, skeletal (sk) muscle, liver and adipose from CR-PPARδ−/− mice (Fig. 1A). Western blot analysis revealed that PPARδ protein contents in heterozygous and homozygous CR-PPARδ−/− cardiomyocytes were ∼40% and 90% of those in controls (α-MyHC–Cre) (Fig. 1B), respectively. Adult CR-PPARδ−/− mice showed no changes in cardiac expression of PPARα and PPARδ (data not shown).

Fig. 1

Mice with cardiomyocyte-restricted knockout of PPARδ develop cardiac hypertrophy and heart failure: A) Transcript products in heart tissues from CR-PPARδ−/− mice. cDNA samples were converted from RNA samples isolated from left ventricle, lung, skeletal muscle, liver and adipose tissues of the indicated mouse line, and subjected to RT-PCR analyses. Arrows indicate the corresponding wild-type (700 bp) and mutant (300 and 400 bp) PPARδ PCR products as previously shown [7]. B). Western blot analyses: Western blot analysis of PPARδ on nuclear protein samples extracted from isolated adult cardiomyocytes of CR-PPARδ+/−, CR-PPARδ−/− and α-MyHC–Cre mice. Protein samples extracted from adenoviral mediated PPARδ overexpressed cardiomyocytes were used as positive control to indicate the specificity of the anti-PPARδ antibody.

CR-PPARδ−/− mice displayed no gross phenotype after birth till 3–4 months of their age. CR-PPARδ−/− mice exhibited greater heart weight to body weight ratios compared to age- and gender-matched controls (Fig. 2A). Quantitative real-time PCR (QPCR) revealed that the transcript levels of skeletal α-actin and atrial natriuretic factor (ANF), two molecular markers of cardiac hypertrophy, rose over 10- and 40-fold of control levels in CR-PPARδ−/− hearts, respectively (Fig. 2B). Food intake, body weight, blood glucose concentration, glucose tolerance and body fat distribution were not altered in CR-PPARδ−/− mice compared with controlled (PPARδlox/lox and α-MyHC–Cre) (data not shown). We assessed cardiac performance using in vivo catheterization in 2-month-old CR-PPARδ−/− and control mice. CR-PPARδ−/− hearts exhibited a 13% decrease of maximal dP/dt and an 18% decrease of cardiac output relative to control hearts (Table 1). As CR-PPARδ−/− mice aging, they displayed dilated cardiomyopathy (Fig. 2C) with evidence of heart failure, including ascites, pleural effusions, and dyspnea. Masson's Trichrome staining revealed dramatically increased fibrosis and enlarged cardiomyocytes on CR-PPARδ−/− heart sections compared to those of controls (Fig. 2D). Transmission electron microscope (TEM) assessments of the adult CR-PPARδ−/− hearts revealed mitochondrial abnormalities in >90 % of cardiomyocytes, including variation in size and shape, and random distribution (Fig. 2E, middle and right panels). Some cardiomyocytes showed striking mitochondrial depletion with abnormal sarcomeres with distorted Z disk, I band, and M band structures, and radical focal myofibrillar breakdown (Fig. 2E, right panel). Other CR-PPARδ−/− cardiomyocytes exhibited increased mitochondrial density, smaller mitochondria with less well organized sarcomeric and subsarcolemmal distribution (Fig. 2E, middle panel). Echocardiographic assessment revealed that CR-PPARδ−/− mice (6-month-old, age- and gender-matched) displayed a substantially reduced fractional shortening compared to those of control mice (Fig. 2F). CR-PPARδ−/− mice showed markedly increased mortality at their first year of life (Fig. 2G). Contrastingly, there was no overt phenotype in α-MyHC–Cre and PPARδflox/flox mice over their life time.

Fig. 2

Cardiomyocyte-restricted PPARδ knockout leads to heart failure and premature death: A). Ratios of heart weight to body weight (HW/BW) from 3–4 month-old CR-PPARδ−/− mice were compared to those from age- and gender-matched α-MyHC–Cre mice. Data were expressed as mean±SEM, n=8, *P<0.01. B). QPCR measurement of mRNA abundance of skeletal α-actin and ANF normalized to mRNA levels of β-actin in samples from 3–4 month-old CR-PPARδ−/− and α-MyHC–Cre hearts. Data are expressed as mean±SEM, n=4, *P<0.01. C). Images of hearts from an end stage CR-PPARδ−/− mouse and from an age- and gender-matched α-MyHC–Cre mouse. D). Trichrome blue staining: Heart sections from 4-month-old, sex-matched CR-PPARδ−/− and α-MyHC–Cre mice were stained with Masson's Trichrome and photomicrographed. Representative histopathological images (×200, Masson's trichrome stain) revealed interstitial fibrosis in CR-PPARδ−/− hearts with extensive blue stained connective tissue. E). TEM images of heart sections from 4-month-old CR-PPARδ−/− and α-MyHC–Cre mice. Note the coexistence of sparse and dense of mitochondrial number in CR-PPARδ−/− heart compared to that of α-MyHC–Cre. Sarcomeres were disrupted and degenerated. (image magnification: 10,000×). F). Echocardiographic assessment of cardiac function in CR-PPARδ−/− and α-MyHC–Cre mice: fractional shortening in 4–6 month old CR-PPARδ−/− and α-MyHC–Cre mice were shown. Data are means ±SEM, n=11 (α-MyHC–Cre) and 14 (CR-PPARδ−/−), *P<0.01. G). Kaplan–Meier survival curves for CR-PPARδ−/− (n=26) and α-MyHC–Cre (n=24) mice at 12 month of their ages. The survival curves are statistically different (P<0.001) by Log-Rank test.

View this table:
Table 1

In vivo hemodynamic measurement

HR (beat/min)393±35428±29
SV (μl)17±318±2
CO (μl/min)6684±302*7707±223
LVPsys (mm Hg)98±5104±7
LVEDP (mm Hg)9.5±2.48.7±3.1
LVPdia (mm Hg)1.7±3.72±1.6
LVDP (mm Hg)97±3.3102±1.6
dLVPdtmax (mm Hg/S)6003±239*6783±182
dLVPdtmin (mm Hg/S)4955±4526097±555
RT50 (ms)17.9±519.2±3.2
TAUlog (ms)15±2.516±2.4
  • In vivo hemodynamic measurement: CR-PPARδ

  • −/− and α-MyHC–Cre mice at their ages of 2 months with mixed gender were anesthetized and a 1.4 French Millar catheter was placed through the carotid artery into the left ventricle as described in Methods. Values of heart rate (HR), stroke volume (SV), cardiac output (CO), left ventricular systolic pressure (LVPsys), left ventricular end-diastolic pressure (LVEDP), Left ventricular diastolic pressure (LVPdia), Left ventricular developed pressure (LVDP), 50% relaxation time (RT50) and time constant of relaxation (Taulog) were showed, respectively. Values represent mean±SEM (n=8 for each group,

  • *P<0.05).

3.2 PPARδ regulates Sod2 expression in cardiomyocytes

To investigate the mechanisms underlying pathological changes in CR-PPARδ−/− hearts, we evaluated expression levels of many potential PPARδ target genes using QPCR on RNA samples from CR-PPARδ−/− and α-MyHC–Cre mice before overt phenotypic changes. Interestingly, there was no substantial change on the expression levels of many key lipid metabolic genes (such as muscle carnitine palmitoyl transferase-1 and PGC-1α) and nuclear factor-κB (NF-κB) target genes (data not shown). Conversely, Sod2 was decreased ∼40% in both transcript and protein levels in CR-PPARδ−/− hearts compared with controls (Fig. 3A and B). Transcript abundances of other key antioxidants such as copper/zinc superoxide dismutase (Sod1), extracellular superoxide dismutase (Sod3), glutathione peroxidase type I (Gpx1), catalase (Cat) and thioredoxin were not changed (data not shown). Furthermore, a PPARδ-selective activator, Rosiglitazone (10 μM) induced over 3-folds more Sod2 transcript than DMSO (Fig. 3C) in cultured cardiomyocytes from healthy control adult mice. We identified a putative PPRE consensus sequence in the −1 kb region within the Sod2 promoter (∼2.3 kb). Three fragments of the mouse Sod2 promoter of various lengths were cloned into the pGL3-Basic luciferase reporter vector. Each of these vectors was transfected into H9C2 myocytes with or without rosiglitazone (10 μM). Sod2-985 but not Sod2-935 and Sod2-650 contained the putative PPRE consensus (Fig. 3D). Rosiglitazone treatment induced nearly 2-fold increase in Luciferase signal in those cells transfected with the Sod2 promoter fragment containing the putative PPRE (Fig. 3D). Conversely, rosiglitazone treatment did not increase luciferase activity in cells transfected with the two Sod2 promoter fragments without PPRE (Fig. 3D). Quantitative Chromatin Immunoprecipitation (ChIP) Assays were used to confirm that the activated PPARδ indeed bound to this PPRE sequence in the rat Sod2 promoter in cellular context. QPCR on immunoprecipitated samples from ChIP assays revealed an over 15-fold increased product from promoter region with the putative PPRE compared with those from DNA fragments matching regions up or downstream of the putative PPRE (Fig. 3E). Together, these data provide strong evidence to support Sod2 as a PPARδ target gene in heart.

Fig. 3

PPARδ regulates Sod2 expression: A). QPCR measurement of mRNA abundance of Sod2 normalized to mRNA levels of 18S in samples from 1-month-old CR-PPARδ−/− and α-MyHC–Cre heart. Data are expressed as mean±SEM, n=4, *P<0.01. B). Western blot analyses: Total protein samples were extracted from the above CR-PPARδ−/− and control hearts. MnSOD was detected by a polyclonal antibody (SIGMA). The relative amount of MnSOD was normalized to actin protein. Data are expressed as mean±SEM, n=4, *P<0.05. C). QPCR measurement of mRNA abundance of Sod2 normalized to mRNA levels of 18S in samples from cultured adult cardiomyocytes treated with rosiglitazone. Data are expressed as mean±SEM, n=4, *P<0.01. D). Luciferase assays: A schematic diagram of the Sod2 promoter fragments cloned into the pGL3-Basic luciferase reporter vector. Numbers in the names of plasmids represent the deletion points of the Sod2 promoter fragments. Sod2-985 contains the PPRE site, but Sod2-935 and Sod2-650 does not. Numbers are also given with reference to the translation start site of luciferase (Luc). Cultured H9C2 cells were transfected the above reporter constructs followed by treatment with vehicle (DMSO) alone or 10 μM rosiglitazone. The results of Luciferase assays are the mean of 4 independent experiments and expressed as mean±SEM, n=4, *P<0.05. E) ChIP assays: ChIP assays were carried out by chromatin immunoprecipitation in rosiglitazone (5 μM) or DMSO treated H9C2 cells with an antibody against PPARδ or preimmune serum (PI). QPCR revealed the relative abundances of fragments of promoter sequence flanking the putative PPRE sequence or up- and downstream of the PPRE region in the immunoprecipitated chromatin. Input represents 1% of total chromatin. The data are the means of three independent experiments and expressed as mean±SEM, n=3, *P<0.01.

3.3 Increased superoxide leads to myocardial oxidative stress in CR-PPARδ−/− hearts

Cytochrome C reduction assay was used to assess the relative superoxide content. Myocardial superoxide content in 2-month-old CR-PPARδ−/− hearts was increased ∼3-folds compared with controls (Fig. 4A). Markers for superoxide-induced mitochondrial oxidative damage were further examined. Aconitase activity decreased after 5 min of reaction and continued to fall at 10 min and 15 min on mitochondrial samples from CR-PPARδ−/− hearts compared with those of control n hearts (Fig. 4B). In parallel, quantitative analysis of immunoblots revealed ∼35% less abundance of the 17 KD subunit of the complex I in CR-PPARδ−/− hearts than in comparable control samples (Fig. 4C). Furthermore, we detected attenuated mitochondrial membrane potential in mitochondrial samples isolated from CR-PPARδ−/− hearts compared with those from control hearts (Fig. 4D). Therefore, these results support that the progressive pathological development of CR-PPARδ−/− hearts is the consequence of oxidative damage.

Fig. 4

Assessments of myocardial superoxide content and oxidative stress: A). Cytochrome C reduction assays: Superoxide content were measured using cytochrome C assays in heart samples of CR-PPARδ−/− and α-MyHC–Cre mice at their ages of 2 months. Data are expressed as mean±SEM, n=5 (CR-PPARδ−/−) and 7 (α-MyHC–Cre), P<0.01. B). Aconitase activities: Cardiac aconitase activities on samples from 3-month-old CR-PPARδ−/− and controlled hearts were measured using an Aconitase activities kit (OxisResearch) at various time points. Data are expressed as mean±SEM, n=4*P<0.01. C). Western blot analyses of protein content of complex I 17 KD subunit: Total protein samples were extracted from 3-month-old CR-PPARδ−/− and control hearts. Complex I was detected by a polyclonal antibody (Molecular Probe). The relative amount of complex I 17 KD subunit was normalized to actin protein. Data are expressed as mean±SEM, n=4, *P<0.01. D). Mitochondria membrane potential: Mitochondria membrane potential on freshly isolated mitochondria from 3-month-old CR-PPARδ−/− and controlled hearts were detected by using a JC-1 Mitochondrial Membrane Potential Detection Kit (SIGMA). Data are expressed as mean±SEM, n=4, *P<0.01.

3.4 MnTBAP treatment largely prevents phenotypic changes in adult CR-PPARδ−/− mice

To investigate if treatment of an SOD mimetic compound MnTBAP can block the progressive pathological development, we treated CR-PPARδ−/− mice at their ages of 2.5-month-old with MnTBAP (5 mg/kg/day, IP) for 1 month. Sod2 mRNA remained low in CR-PPARδ−/− hearts after MnTBAP treatment (Fig. 5A), but intracellular superoxide content in CR-PPARδ−/− hearts was reduced to a normal level (Fig. 5B). MnTBAP treatment prevented >90% of ultrastructural changes in CR-PPARδ−/− hearts (Fig. 5C). Echocardiographic measurement revealed improved cardiac performance in CR-PPARδ−/− mice treated with MnTBAP (Fig. 5D). MnTBAP treatment in CR-PPARδ−/− mice largely normalized the depressed fractional shortening values seen in those without MnTBAP treatment (Fig. 5D). Heart weight to body weight ratio (Fig. 5E) and ANF expression levels (Fig. 5F) were similar between the two MnTBAP treated groups. These results indicate that early cardiac pathological development in CR-PPARδ−/− hearts is largely preventable by the SOD mimetic drug.

Fig. 5

Effect of MnTBAP treatment: A). QPCR measurement of mRNA abundance of Sod2 normalized to mRNA levels of β-actin in samples CR-PPARδ−/− and α-MyHC–Cre hearts after 1-month MnTBAP treatment. Data are expressed as mean±SEM, n=5, P<0.05. B). Relative superoxide content: Superoxide content were measured using cytochrome C reduction assay in heart samples of CR-PPARδ−/− and α-MyHC–Cre mice after 1-month of MnTBAP (5 mg/kg) treatment. Data are expressed as mean±SEM, n=6, P>0.05. C). TEM images of heart sections from CR-PPARδ−/− and α-MyHC–Cre mice after 1-month MnTBAP treatment. (image magnification: 5000×). D). Echocardiographic assessment of cardiac function in CR-PPARδ−/− and α-MyHC–Cre mice. Fractional shortening in CR-PPARδ−/− and α-MyHC–Cre mice with or without MnTBAP treatment were shown. Data are means±SEM, n=8, *P<0.05. E). Ratios of heart weight to body weight (HW/BW) from CR-PPARδ−/− mice were compared to those from age- and gender-matched α-MyHC–Cre mice after 1-month MnTBAP treatment. Data were expressed as mean±SEM, n=8, P>0.01. F). QPCR measurement of mRNA abundance of ANF normalized to mRNA levels of β-actin in samples from CR-PPARδ−/− and α-MyHC–Cre hearts 1-month MnTBAP treatment. Data are expressed as mean±SEM, n=6, P>0.05.

4 Discussion

In the present study, we demonstrated that cardiomyocyte-restricted PPARδ knockout led to depressed cardiac Sod2 expression. The consequently elevated superoxide content caused cardiac mitochondrial damage in these mice, leading to cardiac hypertrophy, dilated cardiomyopathy, and premature death.

We identify PPARδ as an important transcriptional determinant of Sod2 expression in the heart. Depressed Sod2 transcript and protein expression in CR-PPARδ−/− hearts before any cardiac pathological change should account for the dramatically accumulated superoxide. More importantly, MnTBAP treatment in young CR-PPARδ−/− mice prior to overt phenotypic changes effectively prevented the progressive pathological development seen in aged-matched CR-PPARδ−/− mice. Since existing evidence indicates that Sod2 mRNA level is upregulated in failing hearts [17], it is very unlikely that the decreased Sod2 in transcript and protein levels in CR-PPARδ−/− hearts is due to cardiac hypertrophy and heart failure. MnSOD is a mitochondrial antioxidant protecting cells from oxidative damage (See review [18]). Mitochondrial dysfunction and damage is linked to dilated cardiomyopathy in patients with mutations in mtDNA [19,20]. Polymorphisms in the Sod2 gene are associated with nonfamilial idiopathic dilated cardiomyopathy in Japanese population [21]. In addition, MnSOD is involved in the pathogenic processes of ischemic heart disease [22] and doxorubicin-induced cardiomyopathy [23]. Cardiac tissues become susceptible to superoxide-related damage when the expression or activity of MnSOD in mitochondria is reduced. MnSOD-deficient mice (Sod2 knockout mice) show severe mitochondrial damage in the heart and die early from dilated cardiomyopathy [24,25]. Heterozygous Sod2 knockout mice show impaired cardiac mitochondrial function and increased apoptosis [26,27]. Therefore, Sod2 deficiency should substantially contribute to the pathological development in CR-PPARδ−/− hearts.

Phenotypes in this CR-PPARδ−/− mouse line appear more severe than those described in another cardiomyocyte-restricted PPARδ knockout line [8], in which exon 2 of the PPARδ gene is deleted through the use of similar Cre–loxP technology [6]. The present CR-PPARδ−/− mouse line is derived from a floxed PPARδ line that has loxP sites flanking both exons 1 and 2 [4,7]. While it is not immediately clear why these targeted deletions lead to different degrees of phenotypic changes in the heart, our long-term observations were consistent. The striated muscle-specific PPARδ knockout mice from the same floxed PPARδ line as the one in the present study also appear to have cardiac hypertrophy [7]. Systemic phenotypic effects, such as fatty liver, are quite striking in this mouse line [7]. In contrast, it is interesting that PPARδ knockout from cardiomyocytes does not exert major systemic effects. Minor differences in genetic background among different PPARδ knockout models could contribute to the observed phenotypic differences. Despite these differences, it is in accordance that cardiomyocyte-PPARδ deficiency impaired cardiac structure/function. We show that mitochondria abnormalities were key features. Secondary mitochondrial damage due to elevated oxidative stress is a reasonable consideration for the cause of the cardiac pathological changes. The dramatic increase of superoxide in CR-PPARδ−/− hearts supports this notion. Increased superoxide contents suppress levels or activities of superoxide targeted mitochondrial enzymes, such as complex I complex and aconitase. Since complex I deficiency in humans leads to hypertrophic cardiomyopathy [28,29] with functional blockade of the electron transport system in the mitochondria, attenuated complex I 17 KD subunit should account for at least some of the phenotypic changes in CR-PPARδ−/− hearts. In addition, the depressed mitochondrial membrane potential in CR-PPARδ−/− hearts should play a major role in the progressive cardiomyocyte degeneration. These observations demonstrate that oxidative stress is one of the primary causes of the severe structure/function impairment in CR-PPARδ−/− hearts.

Activation of NF-κB has been suggested as the central event, which may in turn trigger cardiac growth in CR-PPARδ−/− mice [8]. However, we detect no change in the expression of common NF-κB target genes in young CR-PPARδ−/− hearts before the onset of hypertrophy or heart failure (data not shown). Furthermore, in contrast to a known effect of NF-κB in elevating Sod2 expression [30], Sod2 expression in CR-PPARδ−/− hearts were decreased. These results do not support the possibility of an early activation of cardiac NF-κB signaling in CR-PPARδ−/− hearts. Instead, increased superoxide anion should activate signaling transduction pathways that trigger cardiac growth. Therefore, the present results clearly indicate that the severe phenotypic changes in CR-PPARδ−/− mice are largely derived from mitochondrial damage.

A hybrid type of rat Sod1, which is analogous to human Cu, Zn-SOD, was previously found to have PPRE consensus sequence in its promoter region [31] and it is activated partly through the PPRE element in its promoter [32]. PPARδ ligands have been shown to reduce superoxide by stimulating both activity and expression of Cu, Zn-SOD in human umbilical vein endothelial cell and suppressing NADPH oxidase [33]. In addition, Cat has been characterized as PPARδ target gene with a PPRE consensus sequence located on its promoter region [34]. However, little is known if PPARδ exerts any transcriptional regulation on key endogenous antioxidants in cardiomyocytes. The findings that lower transcript expression of Sod2 but not other known endogenous antioxidants in CR-PPARδ−/− hearts are intriguing. While our study adds Sod2 into the list of those antioxidants regulated by PPARδ, the molecular mechanisms underlying the selective nature of PPARδ on its regulating capacity on the heart are obscure.

Therefore, our results provide strong evidence to support a role of PPARδ-dependent regulation of constitutive Sod2 under normal physiological conditions, i.e. in the absence of externally or pathologically induced oxidative stress. This study uncovers a crucial mechanism by which PPARδ constitutively regulates myocardial redox homeostasis and provides new insights into the understanding of a role of PPARδ in the heart.


This work was supported by grants from NIH (S06GM08248, 1R01HL085499 and 1R01HL084456 to QY; HL068878 and HL075397 to YQC) and from the American Heart Association national center to QY. DJM is and investigator of the Howard Hughes Medical Institute. RME is an investigator of the Howard Hughes Medical Institute and March of Dimes Chair in Molecular and Developmental Biology.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
View Abstract