Estrogen-related receptor alpha directs peroxisome proliferator-activated receptor alpha signaling in the transcriptional control of energy metabolism in cardiac and skeletal muscle

Abstract

Estrogen-related receptors (ERRs) are orphan nuclear receptors activated by the transcriptional coactivator peroxisome proliferator-activated receptor gamma (PPARgamma) coactivator 1alpha (PGC-1alpha), a critical regulator of cellular energy metabolism. However, metabolic target genes downstream of ERRalpha have not been well defined. To identify ERRalpha-regulated pathways in tissues with high energy demand such as the heart, gene expression profiling was performed with primary neonatal cardiac myocytes overexpressing ERRalpha. ERRalpha upregulated a subset of PGC-1alpha target genes involved in multiple energy production pathways, including cellular fatty acid transport, mitochondrial and peroxisomal fatty acid oxidation, and mitochondrial respiration. These results were validated by independent analyses in cardiac myocytes, C2C12 myotubes, and cardiac and skeletal muscle of ERRalpha-/- mice. Consistent with the gene expression results, ERRalpha increased myocyte lipid accumulation and fatty acid oxidation rates. Many of the genes regulated by ERRalpha are known targets for the nuclear receptor PPARalpha, and therefore, the interaction between these regulatory pathways was explored. ERRalpha activated PPARalpha gene expression via direct binding of ERRalpha to the PPARalpha gene promoter. Furthermore, in fibroblasts null for PPARalpha and ERRalpha, the ability of ERRalpha to activate several PPARalpha targets and to increase cellular fatty acid oxidation rates was abolished. PGC-1alpha was also shown to activate ERRalpha gene expression. We conclude that ERRalpha serves as a critical nodal point in the regulatory circuitry downstream of PGC-1alpha to direct the transcription of genes involved in mitochondrial energy-producing pathways in cardiac and skeletal muscle.

FIG. 2.

ERRα expression causes lipid accumulation in primary cardiac myocytes. Oil Red O staining of primary rat neonatal cardiac myocytes expressing either GFP (Ad-GFP) or ERRα (Ad-ERRα) and cultured in the presence of 35 μM bovine serum albumin-complexed oleic acid. The red droplets (inset) represent accumulated neutral lipid. Magnification, ×400.

FIG. 1.

Validation of putative ERRα target genes involved in cellular fatty acid utilization and mitochondrial respiratory pathways. (A) Expression analysis of fatty acid uptake enzyme genes. Northern blotting was performed with 15 μg of total RNA isolated from Ad-GFP (GFP) or Ad-ERRα (ERRα)-infected cardiac myocytes. Blots were sequentially hybridized with probes specific for ERRα, fatty acid transporter (FAT)/CD36, FABP3, and lipoprotein lipase (LPL). (B) Induced expression of mitochondrial enzymes by ERRα or PGC-1α. Northern analysis was performed as above with probes against PGC-1α, MCAD, acyl coenzyme A oxidase (ACO), M-CPT I, cytochrome oxidase IV (COXIV), cytochrome c (Cyt. c), and ATP synthase β, with RNA from cardiac myocytes overexpressing either GFP, ERRα, or PGC-1α, as indicated. (C) Western analysis of 30 μg of whole-cell extract (WCE) prepared from rat neonatal cardiac myocytes infected with adenovirus vectors expressing GFP or ERRα. FACS-1, fatty acyl coenzyme A synthetase 1.

FIG. 3.

Deletion of the ERRα gene has differential effects on expression of fatty acid utilization enzyme genes mouse heart and skeletal muscle. (A) Northern blot studies performed with 15 μg of total RNA isolated from the hearts of wild-type (WT) or ERRα^−/− mice. Blots were hybridized sequentially with probes corresponding to MCAD, PGC-1α, PPARα, and ERRγ. Phosphorimage quantification of Northern signal intensities is shown on the right. Data represent mean intensity values (± standard error) normalized to wild-type values (= 1.0). (B) Northern analysis of total RNA comparing expression of ERR isoforms, PGC-1α, PPARα, and MCAD in the vastus lateralis muscle, comprised predominantly of fast-twitch glycolytic fibers, versus the soleus muscle, comprised of slow-twitch oxidative fibers. (C) (Left) Northern analysis of total RNA isolated from the soleus of wild-type and ERRα^−/− mice. Representative pairs of samples from each genotype are shown. (Right) Real-time PCR (Taqman) analysis of soleus gene expression in wild-type (n = 6) and ERRα^−/− (n = 6) mice. In addition to the transcripts detected in the Northern panel, quantitative analysis of mRNA encoding the cellular fatty acid utilization enzyme M-CPT I and the PPARβ isoform is shown. Data represent mean arbitrary units (± standard error) corrected to the β-actin transcript and normalized to values in the wild type (= 1.0). Asterisks indicate significant differences (P < 0.05) compared to the control.

FIG. 4.

ERRα induces endogenous PPARα expression. (A) Northern analyses to characterize the expression of regulators of mitochondrial fatty acid oxidation, PPARα, PGC-1α, and ERRα in cardiac myocytes expressing GFP, ERRα, or PGC-1α. (B) Quantification of PPARα and PPARβ mRNA levels in response to ERRα overexpression in cardiac myocytes and C₂C₁₂ myotubes by real-time PCR. Data represent mean arbitrary units (± standard error) corrected to the glyceraldehhyde-3-phosphate dehydrogenase (GAPDH) transcript and normalized to the values in GFP-expressing myocytes (= 1.0).

FIG. 5.

ERR isoforms activate the PPARα gene promoter through a conserved nuclear receptor binding site. (A) Transient cotransfection studies in CV1 cells to analyze ERR regulation of the PPARα promoter. (Left) The reporter construct pα(H-H)-pGL3, containing the −1664 to +83 region of the human PPARα gene promoter, was cotransfected with empty expression vector (−ERRα) or pcDNA3.1-hERRα (+ERRα) in the presence or absence of the coactivator PGC-1α. (Right) Activation by ERRγ was analyzed with the same conditions as for ERRα. A β-galactosidase (β-gal) expression construct was cotransfected to control for transfection efficiency. (B) Transient transfections were performed with either the wild-type PPARα promoter-reporter construct or with constructs mutated at the HNF-4-responsive element (HNF4-REmut) or farnesoid X-responsive element (FXREmut) sites. Experiments were performed as described for panel A. All bars represent mean (± standard error) corrected relative light units (RLU) for β-galactosidase normalized to the activity of the reporter cotransfected with pcDNA3.1(−) (= 1.0). Data represent at least three independent trials performed in triplicate. (C) Electrophoretic mobility shift assays were performed with ³²P-labeled probes corresponding to the wild-type HNF-4-responsive element contained in the human PPARα promoter or a mutated HNF-4-responsive element (Mut). The Mut probe contains the same nucleotide substitutions as the HNF4-REmut promoter-reporter analyzed in B. Probes were incubated with 1 or 3 μl of recombinant human ERRα (left) or mouse ERRγ (right) synthesized in a rabbit reticulocyte lysate. Control reactions included probe alone (−) and probe incubated with unprogrammed reticulocyte lysate (rl). (D) (Top) L6 myoblasts were infected with Ad-ERRα for 48 h. Cross-linked chromatin was immunoprecipitated with nonspecific antibody (immunoglobulin G) or anti-human ERRα (ERRα) antibody. Amplicons corresponding to the 310-bp region of the PPARα promoter containing the HNF-4-responsive element (PPARα) or a 210-bp nonspecific region of the TFIID promoter (control, Cont) were amplified by PCR. Input represents 0.2% of the total chromatin used in the immunoprecipitation reactions. A representative trial from multiple experiments is shown. (Bottom) Band intensities from chromatin immunoprecipitation experiments were quantified by densitometry. Data represent mean band intensity in arbitrary units (± standard error) from three independent trials with PCR performed in triplicate normalized to immunoglobulin G (= 1.0). The asterisk indicates a significant difference compared to the control (immunoglobulin G).

FIG. 6.

ERRα induction of β-oxidation enzymes genes is dependent on the presence of PPARα. Primary fibroblasts isolated from ERRα^−/− PPARα^−/− (double-knockout, DKO) or ERRα^−/− (ERR knockout, ERRKO) mice were infected with an adenoviral construct expressing GFP (−) or ERRα (+) as indicated. PGC-1α was included in some conditions to enhance ERRα activity. Real-time PCR was used to quantify the expression of endogenous MCAD and acyl coenzyme A oxidase (ACO) (A) or M-CPT I (B) in total RNA isolated from these cells. Data are reported as mean (± standard error) arbitrary units normalized to the GFP condition (= 1.0) for three independent trials performed in triplicate. (C) Palmitate oxidation rates were measured in the same primary fibroblasts as above expressingGFP or ERRα. Data were normalized to GFP (= 1.0), and values represent means (± standard error) for three overexpression trials performed with cells from two independent isolations. Asterisks indicate a significant difference compared to the controls (minus ERRα).

FIG. 7.

Role of ERRα in regulating cellular oxidative capacity. PGC-1α coactivates and regulates the expression of a number of transcription factors, including ERRα, PPARα, and NRFs, involved in mediating PGC-1α effects on cellular metabolism. PGC-1α regulation of ERRα and NRF-2 expression involves both cross- and auto-regulatory mechanisms (25, 34). Data presented in the current study demonstrated that ERRα likely directs PGC-1α upregulation of PPARα. In response to PGC-1α, activation of the NRF cascade regulates genes involved in mitochondrial respiration and biogenesis, whereas activation of the PPARα pathway regulates fatty acid uptake and mitochondrial oxidation enzymes. ERRα may also directly regulate metabolic target genes in both pathways. FAO, fatty acid oxidation; ACO, acyl coenzyme A oxidase; mtTFA, mitochondrial transcription factor A; oxid. phos., oxidative phosphorylation.