|Home | About | Journals | Submit | Contact Us | Français|
Peroxisome proliferator activated receptor δ (PPARδ) is an essential determinant of basal myocardial fatty acid oxidation (FAO) and bioenergetics. We wished to determine whether increased lipid loading affects the PPARδ deficient heart in transcriptional regulation of FAO and in the development of cardiac pathology. Cardiomyocyte-restricted PPARδ knockout (CR-PPARδ−/−) and control (α-MyHC-Cre) mice were subjected to 48 hours of fasting and to a long-term maintenance on a (28 weeks) high-fat diet (HFD). The expression of key FAO proteins in heart was examined. Serum lipid profiles, cardiac pathology, and changes of various transduction signaling pathways were also examined. Mice subjected to fasting exhibited upregulated transcript expression of FAO genes in the CR-PPARδ−/− hearts. Moreover, long-term HFD in CR-PPARδ−/− mice induced a strikingly greater transcriptional response. After HFD, genes encoding key FAO enzymes were expressed remarkably more in CR-PPARδ−/− hearts than in those of control mice. Despite the marked rise of FAO gene expression, corresponding protein expression remained low in the CR-PPARδ−/− heart, accompanied by abnormalities in sarcomere structures and mitochondrial abnormalities that were similar to those CR-PPARδ−/− hearts with regular chow feeding. The CR-PPARδ−/− mice displayed increased expression of PPARγ coactivator-1α (PGC-1α) and PPARα in the heart with deactivated Akt and p42/44 MAPK signaling in response to HFD. We conclude that PPARδ is an essential determinant of myocardial FAO. Increased lipid intake activates cardiac expression of FAO genes via PPARα/PGC-1α pathway, albeit it is not sufficient to improve cardiac pathology due to PPARδ deficiency.
Obesity and its many health-related hazards have become a serious and growing problem worldwide. High-fat diet (HFD)-induced obesity is a major risk factor for the development of heart failure with features of myocardial lipid overload, which has been recently named as lipotoxic cardiomyopathy . High-fat feeding increases fat mass, elevates plasma free fatty acid (FFA) levels, and suppresses insulin signaling. HFD increased fatty acid oxidation (FAO) rates and myocardial oxygen consumption . On the other hand, impaired FAO has been linked to cardiac dysfunction in obese animal models . Under dietary stress conditions, such as fasting, the heart shows increased expression of genes encoding key enzymes that are involved in lipid uptake and FAO [4, 5]. However, little is known about the mechanisms underlying this change.
Peroxisome proliferator-activated receptors (PPARα, δ, and γ) are nuclear receptors that respond to a variety of endogenous and exogenous ligands to govern transcriptional expression of genes encoding a broad spectrum of proteins, especially, those key enzymes involved in fatty acid metabolism. It is established that PPARα and δ transcriptionally regulate fatty acid metabolism in the heart (see review ). Our previous studies demonstrated that PPARδ is an essential determinant of basal myocardial FAO [7–9]. PPARδ selective ligands and overexpression of wild type PPARδ in cultured cardiomyocytes induce upregulation of a series of genes encoding many key enzymes of FAO [7, 8, 10]. PPARα has been shown to be a transcriptional determinant for myocardial FAO in response to fasting and HFD. However, it remains unclear whether PPARδ also plays a similar role. PPARγ co-activator α (PGC-1α) is an inducible regulator of energy metabolism. PGC-1α is relatively abundant in tissues with high capacities for mitochondrial FAO, including heart and brown adipose . PGC-1α and other potential transcription co-factors are likely served as transducers of physiologic stimuli to the control of cardiac energy metabolism through PPARs. PGC-1α is a co-activator for both PPARα and PPARδ for their transcriptional activities[12, 13]. However, how PGC-1α works preferentially or equally with PPARα and/or PPARδ is not known.
To determine whether PPARδ mediates increased transcriptional activity of genes encoding key proteins in lipid metabolism of the heart under fasting and chronic high-fat feeding, wild-type and cardiomyocyte-restricted PPARδ knockout (CR-PPARδ−/−) mice fed an HFD were studied. The present data show that 1) Both PPARδ and PPARα signaling are activated in the heart by chronic high-fat feeding; 2) High fat loading in the heart leads to an upregulation of PPARα and PGC-1α accompanied by a greater expression of their target genes 3) Depressed Akt and p42/p44 MAPK signaling pathways in PPARδ deficient hearts may be responsible for the upregulation of PPARα and PGC-1α in response to high-fat feeding.
The cardiomyocyte-restricted PPAR-δ knockout (CR-PPARδ−/−) mice have been described [7, 8]. Transgenic Cre (α-MyHC-Cre) mice that overexpress Cre specifically in cardiomyocytes have been reported before . The Cre-lox controlled PPARδ gene-targeting mice have been described . In this mouse model, the first two exons were targeted by the Cre-lox methodology. Homozygous PPAR-δflox/flox mice were bred and crossed with α-MyHC-Cre mice. Animals received food and water on an ad libitum basis, and lighting was maintained on a 12-hour cycle.
All experiments were performed in mice at their ages of 3-months (20–30 g). Age- and gender-matched littermates were separated into individual cages at the beginning of each fasting experiment. Fasting was initiated at 5:00 pm and lasted for 48 hours. Control mice (fed ad libitum) and fasted mice were kept in identical light/dark cycles. The control mice were allowed free access to standard lab chow. At the time of tissue harvest, animals were killed by CO2 inhalation and tissues were rapidly dissected free, snap-frozen in liquid nitrogen, and stored at −80°C until processed for isolation of RNA or lipid extraction.
Male and female littermates at their ages of six weeks were fed with a high-fat diet or regular chow for 28 weeks. The HFD provided 60% energy in the form of lard fat (D12450B or D12451, Research Diets, New Brunswick, USA). Body weight and food intake were measured at regular intervals throughout the feeding intervention. At the time of tissue harvest, animals were killed by CO2 inhalation and tissues were rapidly dissected free, snap-frozen in liquid nitrogen, and stored at −80°C until processed for isolation of RNA or lipid extraction.
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 the University of Alabama at Birmingham.
Total RNA samples were extracted from left ventricles using a RNA extraction kit (Qiagen) according to the manufacturer’s instructions. Real-time RT-PCR measurement of transcript levels (Roche LightCycler PCR system) were carried out to determine transcript levels of target genes. Real time PCR results from each gene/primer pair were normalized to β-actin, and compared across conditions.
Both nuclear and cytosolic protein samples were extracted using a nuclear protein extraction kit (Pierce). Mitochondria extraction from left ventricular tissues was performed using a tissue mitochondria extraction kit from SIGMA. Protein samples were subjected to SDS-PAGE gels and electro-transferred to nitrocellulose membranes (Amersham). Immunoblotting was performed using the Enhanced Chemiluminescence Detection System (Santa Cruz Biotechnology) or the Chemofluorescence Detection Kit from Molecular Probe. A polyclonal antibody for malonyl-CoA Decarboxylase (MCD) is a generous gift from Dr. Lopaschuk’s group of the University of Alberta . Other antibodies were obtained from commercial sources: pan-actin, and PGC-1α (Santa Cruz Biotechnology); PPARα (Abcam); p44/42 MAP kinase, p-P44/42 MAP kinase (Thr202/Tyr204), pan-Akt and Phospho-Akt (Ser 473) (Cell Signaling Technology) and carnitine palmitoyltransferase 1b (CPT-1b) (Alpha Diagnostic International).
Myocardial triglyceride and serum content of triglyceride, free fatty acids (FFA) and Cholesterol were assayed using lipid diagnostic kits (Wako Chemicals USA, Inc.). For myocardial lipid extraction, left ventricular tissues were homogenized with ice-cold chloroform-methanol-water mixture (2:1:0.8) for 2 min. Additional chloroform and water were added to separate the organic and aqueous layers. After centrifugation, the aqueous layer was removed, and the chloroform layer was decanted and evaporated at 70°C. The residue was dissolved in 0.5 ml of isopropanol.
To obtain tissue for 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 two minutes followed by perfusion with 3.5% glutaraldehyde in 0.1 M cacodylate buffer, pH 7.3 for another 2 minutes. Samples were taken from the left ventricles.
Plasma concentrations of adiponectin and leptin were determined by enzyme-linked immunosorbent assay (ELISA). The assays were conducted according to manufacturer’s protocols (R&D Systems).
Approximately 2μl of blood sample was collected from the mouse tail vein. Blood glucose level was measured with a One Touch Ultra glucometer (LifeScan, Milpitas, CA).
All data were analyzed by one factor or mixed, two-factor analysis of Variance (ANOVA) using GraphPad Prism software (GraphPad Software Inc.) and SPSS software (SPSS Inc.). Unless indicated in figure legends, values of quantitative results were expressed as mean±SEM. Differences between groups and treatments were regarded as significant at the P<0.05 probability level.
The expression levels of PPARα and PPARδ in response to 48-hour fasting in normal mice were examined. Real time PCR measurement revealed that cardiac PPARα transcript levels were increased by about 2 folds, whereas PPARδ transcript expression was not changed (Figure 1A). On the other hand, mRNA of both PPARα and PPARδ were substantially upregulated in mice with a long-term HFD (6-month) (Figure 1C and D). These results indicate that cardiac PPARα expression is increased in response to both fasting and HFD, whereas cardiac PPARδ expression is only upregulated in the HFD condition.
To investigate if cardiac deficiency of PPARδ is required for the heart to respond to dietary stress, three-month-old CR-PPARδ−/− mice with no major pathological phenotypes mice were subjected to 48-hour fasting. CR-PPARδ−/− littermates of the above mice were fed regular chow as the control group. Age- and gender-matched α-MyHC-Cre mice were used as controls. PCR revealed that the transcript expression of genes encoding key FAO proteins, such as CPT-Ib, 3-acyl-CoA thiolase (thiolase), MCD and long chain acyl-CoA dehydrogenase (LCAD) were upregulated in both the CR-PPARδ−/− and control hearts to similar levels, suggesting a greater increased expression of the above genes in CR-PPARδ−/− hearts to match those of control hearts (Figure 2A-D). Therefore, the diminished expression of FAO genes in CR-PPARδ−/− hearts did not block the fasting induced FAO gene upregulation.
To further investigate the roles of cardiac PPARδ in response to dietary associated myocardial lipid overload, we fed the CR-PPARδ−/− and control mice with an HFD for 28 weeks. Littermates of the above CR-PPARδ−/− and control mice were fed regular chow for the same duration and served as the control groups. With comparable food intake, high fat feeding for 28 weeks in both PPARδ and α-MyHC-Cre mice led to a similar increase (~50%) in their body weights (Table 1). Serum adiponectin contents were not different and leptin contents were markedly increased after a 28-week high fat feeding in both CR-PPARδ−/− and α-MyHC-Cre mice (Table 1). After a 28-week high fat feeding, serum lipid profiling revealed similarly increased triglyceride (TG), free fatty acids (FFA) and total cholesterol contents over basal condition in both PPARδ and α-MyHC-Cre mice (Table 1). With a 28-week HFD, both experimental mice showed increased serum glucose and leptin levels but unchanged adiponectin levels (Table 1). As we showed previously , CR-PPARδ−/− mice fed regular chow exhibited diminished cardiac expression of genes encoding lipid metabolism, such as CPT-Ib, thiolase, MCD, LCAD, Acyl-CoA oxidase (ACOX1), LCAD, and very long chain acyl-CoA dehydrogenase (vLCAD) (Figure 3A-F). After 28 weeks of high fat feeding, the above genes were markedly upregulated in CR-PPARδ−/− (Figure 3A-F). On the other hand, cardiac expression of these genes in the control hearts were no change, increased or decreased (Figure 3A-F). Surprisingly, Western blot analyses revealed that at least protein levels of MCD and CPT-1 were not increased but remained decreased in high fat feeding CR-PPARδ−/− hearts (Figure 4A and B).
At the end of the high fat feeding, both CR-PPARδ−/− and control mice exhibited cardiac hypertrophy with increased heart weight to tibial length ratios (Table 1). During the course of the 28-week feeding period, two out of ten CR-PPARδ−/− mice died due to apparent heart failure as those observed in the basal condition. Despite the remarkable upregulation of PPARα target genes involved in FAO, myocardial lipid accumulation in CR-PPARδ−/− hearts remained high (Table 1). TEM revealed disrupted sarcomeric structures and abnormal mitochondria in the CR-PPARδ−/− heart with high fat feeding, similar to those CR-PPARδ−/− hearts with regular chow feeding at the same age. There were increased lipid droplets in both CR-PPARδ−/− and α-MyHC-Cre heart sections in response to HFD. Compared with the control, cardiomyocytes from CR-PPARδ−/− heart showed disrupted sarcomeric structures and swelling mitochondria with loss of cristae (Figure 5). Lipid and remnant matrix-type materials were visible in the vacuolated mitochondria (Figure 5). Therefore, it appears that the long-term high fat feeding may magnify expression of lipid metabolic proteins at the transcript level but not in protein levels, thus exerting no apparent rescue to the pathological development of the CR-PPARδ−/− heart.
To determine if the upregulation of other transcriptional regulators in the CR-PPARδ−/− hearts after high fat feeding is responsible for the magnified upregulation of FAO genes, we further examined the expression of PGC-1α and PPARα in CR-PPARδ−/− hearts after high fat feeding. Real time PCR measurement revealed that PGC-1α and PPARα expression was not altered at basal condition but markedly increased after HFD in the CR-PPARδ−/− hearts compared with that of controls (Figure 6A and B). Western blot analyses revealed that cardiac protein level of PGC-1α, but not PPARα, was correspondently increased in long-term high fat feeding CR-PPARδ−/− mice compared with controls (Figure 6C and D). These results indicate that elevated PGC-1α may be responsible for the remarkable upregulation of FAO genes.
We further investigated whether P42/44 MAPK and Akt signaling pathways are involved in the regulation of PGC-1α expression and PPARα activities. Western blot analyses revealed that cardiac PPARδ deficiency did not alter protein expression of Akt and P42/44 MAPK, and their phosphorylation compared with α-MyHC-Cre hearts (Figure 7A and B). On the other hand, phosphorylation of Akt and P42/44 MAPK was substantially less in the CR-PPARδ−/− hearts than in the control hearts (Figure 7C-D), indicating that both the Akt and p42/44 MAPK signaling pathways are suppressed in the CR-PPARδ−/− hearts with HFD.
In this study, we seek to uncover the effects of PPARδ deficiency in the heart under dietary stress and lipid overload conditions. CR-PPARδ−/− mice with an increased lipid loading exhibit greater increases of transcript expression of cardiac FAO genes but fail to recue cardiac phenotype. The overexpression of FAO genes in the CR-PPARδ−/− heart is directed by augmented PPARα/PGC-1α signaling via diminished Akt and P42/44 MAPK pathways.
Fasting in animals has been shown to increase the concentration of FFAs in plasma . The fatty acids are taken up by the heart and oxidized to supply energy. Cardiac expression of genes encoding lipid metabolism was upregulated in the fasting condition [5, 18], presumably by PPARα  and maybe PPARδ. Our current findings showed that cardiac PPARα, but not PPARδ, was upregulated in response to 48-hour fasting. It suggests that PPARα be more responsive to dietary stress than PPARδ in the heart. On the other hand, cardiac expression of both PPARα and PPARδ was markedly increased in mice subjected to long-term high fat feeding. Therefore, it appears that the expression of PPARα is more responsive to both short-term and long-term demands of lipid metabolism in the heart, whereas PPARδ seems to be more important in maintaining long-term stable myocardial lipid metabolisms.
It is now clear that both PPARδ and PPARα are involved in transcriptional activation of FAO and lipid transportation in the heart. PPARα involves in the transcriptional response to fasting and to a HFD . PPARα−/− mice develop marked hepatic and cardiac lipid accumulation after short-term fasting , accompanied by markedly increased circulating FFA. More importantly, it has been shown that PPARα governed FAO is essential for the normal response to fasting as revealed by severe hypoglycemia with early death in fasted PPARα deficient mice . Since PPARδ exhibits almost identical function on regulating FAO gene expression as PPARα does in the heart, we speculate that PPARδ inactivation in the heart may also block the upregulation FAO genes in response to fasting. To our surprise, fasting-induced transcript expression of myocardial lipid metabolic genes was not blunted in the PPARδ deficient hearts. With diminished basal expression, fasting induced eventually greater cardiac expression of these genes in the PPARδ deficient heart to catch up with the expression of the control hearts. Cardiac expression of FAO genes was eventually rose substantially more in CR-PPARδ−/− hearts fed an HFD compared to those of controls. Therefore, increased lipid loading in the CR-PPARδ−/− heart with deficient myocardial FAO appears to trigger a compensatory transcriptional upregulation, which is especially apparent in a situation with stable supply of fatty acids to the heart. It is interesting that the overexpression of these FAO genes did not translate into correspondingly increased proteins. While the mechanisms are not clear, similar observation of dissociation between mRNA and protein expression of a number of enzymes of FAO in the heart has been noted . With the impaired protein expression of key FAO enzymes, myocardial lipid accumulation in CR-PPARδ−/− hearts remained high, which presumably leads to elevated ceramide, a cardiotoxin proposed as the cause of lipotoxic cardiomyopathy . Consequently, the CR-PPARδ−/− hearts after long term HFD manifested similar pathological changes to those hearts of age- and gender-matched CR-PPARδ−/− littermate fed with regular chow.
The unexpected overexpression of lipid metabolic genes in the high-fat feeding CR-PPARδ−/− hearts may be the results of a secondary activation or inactivation of an important singling pathway(s). PGC-1α is essential for maximal and efficient cardiac mitochondrial FAO and lipid homeostasis . PGC-1α is a co-activator for both PPARα and PPARδ for their transcriptional activities [12,13], albeit it is not clear whether PGC-1α works preferentially or equally with PPARα and/or PPARδ. Similar to PPARδ/PGC-1α, PPARα/PGC-1α has been well documented as the key FAO regulator in the heart (See review[6, 24]). We suspected that PPAR α/PGC-1α signal was upregulated in the PPARδ deficient heart under lipid overloading conditions. Indeed, CR-PPARδ−/− mice fed an HFD exhibited increased PGC-1α and PPARα transcripts and correspondingly increased PGC-1α protein in the heart. The greater increase of cardiac expression of PGC-1α and FAO genes was also observed in CR-PPARδ−/− mice after shorter-term HFD (4 weeks) before apparent obesity (data not shown). The presumably augmented PPARα/PGC-1α signal should be able to drive a more pronounce overexpression of lipid metabolic genes. In an effort to explore the underlying mechanisms that direct the upregulation of PPARα/PGC-1α signal, we have assessed multiple upstream signaling pathways that determine the expression and activity of PPARα/PGC-1α. It has been shown that PPARα can be deactivated by p42/44 MAPK-mediated phosphorylation through a mechanism that reduces affinity for ligand . On the other hand, Akt activities are involved in transcriptional regulation of PGC-1α . PGC-1α gene expression is down-regulated by Akt-mediated phosphorylation in insulin stimulated skeletal muscle . Many stimuli including Ang II  and mechanical stress activate Akt , which is linked to the changes of PGC-1α expression and adaptive changes in myocardial FAO in hypertrophy. Interestingly, PGC-1α expression can be controlled by physiologic stimuli. For example, PGC-1α gene expression is markedly induced in brown adipose tissue by cold exposure  and in the heart by short-term fasting . It is possible that the excessive lipid loading in the PPARδ deficient heart deactivates both the Akt and p42/44 MAPK signaling pathways. The deactivated Akt signaling may result in increased expression of both PGC-1α transcript and protein, which should be sufficient to enhance FAO gene expression through PPARα. Furthermore, the deactivation of p42/44 MAPK may also directly enhance PPARα transcriptional activity. Even though the current study is not possible to exclude many other potential signaling events, the upregulation of PGC-1α expression, together with potentially increased PPARα activity, should induce the robust expression of FAO genes in CR-PPARδ−/− mice with an HFD.
Results from the current study strongly suggest that PPARδ plays a major role in the metabolic adaptations to the lipid overload condition. It is plausible that PPARδ should participate in protecting the heart from lipid overload in the high fat feeding condition. Therefore, we conclude that PPARδ is an essential determinant of myocardial FAO in both basal and lipid loading conditions. Whereas PPARα is important in response to both short-term and long-term demands of lipid metabolism in the heart, PPARδ plays an important role in maintaining long-term stable myocardial lipid metabolisms.
This work was supported by grants from NIH (1R01HL085499, 1R01HL084456 and R21 AT003734).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.