Research ArticleMOLECULAR BIOLOGY

Thyroid hormone receptor and ERRα coordinately regulate mitochondrial fission, mitophagy, biogenesis, and function

See allHide authors and affiliations

Science Signaling  26 Jun 2018:
Vol. 11, Issue 536, eaam5855
DOI: 10.1126/scisignal.aam5855
  • Fig. 1 Transcriptome and ChIP-seq analyses of THRB1 and ESRRA showing co-regulation of mitochondrial pathways.

    (A) Gene Set Enrichment Analysis (GSEA) pathway analysis of transcriptomics performed in the livers from euthyroid control, TH-treated (10 μg/100 g body weight per day for 3 days), and TH + XCT790 (10 mg/day per 100 g body weight for 3 days)–treated mice (n = 3 mice per group). (B) Venn diagram representing overlapping genes between groups from (A) under various pathways. (C) THRB1 and ESRRA binding on the genes from (A) and (B) (n = 4 livers per group). (D to G) Mitochondrial oxygen consumption rate (OCR) using a Seahorse extracellular flux analyzer under TH treatment (100 nm for 48 hours) with or without ESRRA siRNA in THRB1-HepG2 cells (n = 5 replicates per group). Mitochondrial inhibitors [1 μM oligomycin, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP), rotenone (R), or antimycin A (A)] from Seahorse XF Cell Mito Stress Test kit were used. Data are representative of three or more independent experiments. Statistical significance for the data presented in (A) to (C) was calculated as described in Materials and Methods, and that in (E) and (G) was determined by using Kruskal-Wallis nonparametric analysis of variance (ANOVA), followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment (*P ≤ 0.05).

  • Fig. 2 TH-mediated induction of mitochondrial FAO and fat clearance depended on ESRRA.

    (A) Reverse transcription quantitative polymerase chain reaction (RT-qPCR) analysis of hepatic Cpt1a transcript expression (n = 5 mice per group). (B) Measurement of serum β-hydroxybutyrate/ketone bodies (n = 5 mice per group). (C) Metabolomics analysis of short-chain acylcarnitine C4-OH (d-3-hydroxybutyrylcarnitine, a derivative of ketone bodies) in euthyroid, TH-treated (10 μg/100 g body weight per day for 3 days), XCT790-treated (10 mg/day per 100 g body weight for 3 days), and XCT790 + TH–treated mice (n = 4 mice per group). (D) RT-qPCR analysis of CPT1A transcript expression in THRB1-HepG2 cells treated with control or ESRRA siRNA in the presence or absence of TH (100 nM for 48 hours) (n = 5 biological replicates per group). Data are representative of three or more independent experiments. (E and F) BODIPY 493/503 staining (E) and its quantification by flow cytometry (F) for neutral lipids and lipid droplets in fat-loaded THRB1-HepG2 cells treated with control siRNA or ESRRA siRNA in the presence or absence of TH (20,000 counts per acquisition). Quantification of BODIPY 493/503 fluorescence by flow cytometer is plotted as mean fluorescence intensity (MFI) (n = 5 biological replicates per group). Images and data are representative of three fields per group and three independent experiments. Scale bars, 200 μm. (G) Measurement of mitochondrial fuel oxidation (fatty acids; FAO) in fat-loaded (oleate/palmitate) THRB1-HepG2 cells treated with control or ESRRA siRNA in the presence or absence of TH (n = 5 biological replicates per group). Data are representative of three independent experiments. Seahorse XF Mito Fuel Flex Test kit was used with inhibitors of the glucose oxidation pathway (UK5099), glutamine oxidation pathway (BPTES), and CPT1A (long-chain FAO pathway; etoximir). Various parameters such as the measurement of cells’ reliance on a particular fuel pathway to maintain baseline respiration (dependency), cells’ ability to increase oxidation of a particular fuel to compensate for inhibition of alternative fuel pathway(s) (flexibility), and the cells’ total capability to use a fuel pathway to meet energy demand when other fuel pathways are inhibited (capacity) were calculated. Statistical significance was determined by using Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment (*P ≤ 0.05). ns, not significant.

  • Fig. 3 TH increased ESRRA gene expression by stimulating PPARGC1A expression.

    (A) RT-qPCR analysis of Esrra transcript expression in the livers from euthyroid control and TH-treated (10 μg/100 g body weight per day for 3 days) mice (n = 15 mice per group). (B) Western blot analysis of hepatic ESRRA protein expression in euthyroid and TH-treated mice (n = 20 mice per group). Blots are representative of three or more independent experiments. (C) Western blot analysis of hepatic ESRRA protein expression in euthyroid and TH-treated mice treated for the indicated periods of time (n = 5 mice per group). Blots are representative of three or more independent experiments. (D) RT-qPCR analysis of hepatic Esrra transcript expression in Thrb wild-type (Thrb+/+) mice and liver-specific Thrb knockout (Thrb−/−) treated with TH or propylthiouracil (PTU; to make mice hypothyroid) (n = 5 mice per group). (E) Western blot analysis of ESRRA protein expression in liver tissues of Thrb+/+ and liver-specific Thrb−/− mice treated with TH or PTU (n = 5 mice per group). Blots are representative of three or more independent experiments. (F) RT-qPCR analysis of hepatic Ppargc1a transcript expression in euthyroid and TH-treated mice (n = 15 mice per group). (G) RT-qPCR analysis of PPARGC1A and ESRRA transcript expression in untreated control and TH-treated (100 nM for 48 hours) THRB1-HepG2 cells with or without PPARGC1A knockdown (n = 5 biological replicates per group). Data are representative of three independent experiments. (H) RT-qPCR analysis of PPARGC1A and ESRRA transcript expression in untreated control and TH-treated THRB1-HepG2 cells with or without PPARGC1A overexpression (n = 5 biological replicates per group). Data are representative of three independent experiments. (I) Luciferase reporter analysis of ESRRA promoter activity in THRB1-HepG2 cells treated with TH by the conditions indicated in the panel below the bar graph (n = 6 biological replicates per group). Data are representative of three independent experiments. RLU, relative light units. (J) ESRRA and POLR2A ChIP-qPCR analysis on Esrra gene promoter in mouse liver tissue from euthyroid control, TH-treated, and TH + XCT790 (10 mg/day per 100 g body weight for 3 days)–treated mice (n = 3 mice per group). Statistical significance for (A), (B), and (F) was calculated using the nonparametric Mann-Whitney test, whereas for other panels, Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment, was performed (*P ≤ 0.05).

  • Fig. 4 TH induced mitochondrial biogenesis and fission through ESRRA.

    (A) Mitochondrial DNA copy number analysis in the livers from euthyroid control, TH-treated (10 μg/100 g body weight per day for 3 days), and TH + XCT790 (10 mg/day per 100 g body weight for 3 days)–treated mice (n = 5 mice per group). (B and C) Mitochondrial DNA copy number analysis in THRB1-HepG2 cells treated with ESRRA siRNA or XCT790 (2 μM) in the presence of absence of TH (100 nM) for 48 hours (n = 5 biological replicates per group). Data are representative of three independent experiments. (D to H) Transmission electron microscopy of liver tissues from euthyroid (D), TH-treated (E to G) and XCT790 + TH–treated mice (H) (n = 5 mice per group). Scale bars, 0.5 μm (D, E, and H) and 100 nm (F and G). N, nucleus; A, autophagic vesicles (white arrows). Asterisk (*) indicates mitochondria or mitochondria-like structures inside autophagic vesicles. (I and J) Mitochondrial numbers and size (I) and autophagic vesicles containing mitochondria (J) were counted. Data are representative of 10 fields per group and three independent analyses. (K) MitoTracker Red CMXRos staining was performed in THRB1-HepG2 cells to analyze mitochondrial structure. These THRB1-HepG2 cells were treated with control or ESRRA siRNA in the presence or absence of TH (100 nM for 48 hours) (n = 5 biological replicates per group). Enlarged panels represent selected digitally enlarged portions of parent images to enhance the visibility of mitochondrial structure. Images are representative of 10 fields per group and three independent experiments. Scale bars, 5 μm. Statistical significance for (I) and (J) was calculated using one-way ANOVA, followed by Tukey’s post hoc test; otherwise, Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment, was performed (*P ≤ 0.05 and #P ≤ 0.05). In (I), asterisk (*) represents the comparison between euthyroid and TH-treated mice or as indicated by the horizontal line, and number sign (#) represents the comparison between TH-treated and XCT790 + TH–treated mice.

  • Fig. 5 Induction of ULK1 expression by TH was mediated by ESRRA.

    (A) Western blot analysis and densitometric quantification of hepatic ULK1 and ESRRA protein expression in the livers from euthyroid control, TH-treated (10 μg/100 g body weight per day for 3 days), and TH + XCT790 (10 mg/day per 100 g body weight for 3 days)–treated mice (n = 5 mice per group). Blots are representative of three independent experiments. (B) RT-qPCR analysis of hepatic ULK1 transcript expression in euthyroid control, TH-treated, and TH + XCT790–treated mice (n = 5 mice per group). Data are representative of three independent experiments. (C) Western blot analysis and densitometric quantification of ESRRA and ULK1 expression in THRB1-HepG2 cells treated with control or ESRRA siRNA with or without TH (100 nM for 48 hours) (n = 5 biological replicates per group). Blots are representative of three or more independent experiments. (D) RT-qPCR analysis of ULK1 transcript expression in THRB1-HepG2 cells treated with control or ESRRA siRNA in the presence or absence of TH (n = 5 biological replicates per group). Data are representative of three independent experiments. (E) ESRRA and POLR2A ChIP-qPCR analysis on the Ulk1 gene promoter in euthyroid, TH-treated, and XCT790 + TH–treated mouse livers (n = 3 mice per group). Statistical significance was calculated using Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment (*P ≤ 0.05).

  • Fig. 6 Coordinated fission and mitophagy mediated by TH depended on ESRRA.

    (A) Western blot analysis and densitometric quantification of protein abundance in the livers from euthyroid control, TH-treated (10 μg/100 g body weight per day for 3 days), and TH + XCT790 (10 mg/day per 100 g body weight for 3 days)–treated mice (n = 5 mice per group). Blots are representative of three independent experiments. (B) Western blot analysis and densitometric quantification of cytosolic (Cyto) and mitochondrial (Mito) fractions from the livers of euthyroid, TH-treated, and XCT790 + TH–treated mice livers (n = 5 mice per group). Blots are representative of three independent experiments. (C) Western blot analysis and densitometric quantification of DRP1 phosphorylation in whole-cell lysates of THRB1-HepG2 cells treated with or without TH (100 nM) and the ULK1 inhibitor (iULK1) MRT0068921 (1 μM) for 48 hours (n = 5 biological replicates per group). Blots are representative of three independent experiments. (D) Western blot analysis and densitometric quantifications of cytosolic and mitochondrial fractions from THRB1-HepG2 cells treated with or without TH and the ULK1 inhibitor MRT0068921 to observe mitochondrial translocation of proteins (n = 5 biological replicates per group). Blots are representative of three independent experiments. (E) Coimmunoprecipitation analysis was performed to detect the interaction of FUNDC1 with ULK1 and MAP1LC3B-II (n = 5 biological replicates per group). Blots are representative of three independent experiments. Statistical significance was calculated using Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment (*P ≤ 0.05). IP, immunoprecipitation.

  • Fig. 7 Induction of DRP1 and MAP1LC3B mitochondrial localization and mitophagic flux by TH was dependent on ESRRA.

    THRB1-HepG2 cells were treated with control siRNA with or without TH (100 nM for 48 hours) or with ESRRA siRNA, FUNDC1 siRNA, or inhibitors of ULK1 (iULK1; 1 μM) or DRP1 (iDRP1; 10 μM). MitoTracker Red CMXRos was used to stain mitochondria. (A) Immunofluorescence analysis for DRP1 translocation to mitochondria. DRP1 antibody (#8570, Cell Signaling Technology) with Alexa Fluor 488 was used to detect DRP1 by confocal microscopy (n = 5 biological replicates per group). Enlarged panels represent selected digitally enlarged portions of parent images to enhance detection of DRP1 translocation to mitochondria. Images are representative of three fields per group and three independent experiments. Scale bars, 5 μm. (B) Quantification of DRP1 mitochondrial translocation (as % overlay). (C) Immunofluorescence analysis for MAP1LC3B and mitochondrial colocalization. MAP1LC3B antibody (#2775, Cell Signaling Technology) with Alexa Fluor 488 was used to detect MAP1LC3B under a confocal microscope (n = 5 biological replicates per group). Enlarged panels represent selected digitally enlarged portions of parent images to enhance the visibility of MAP1LC3B translocation to mitochondria. Images are representative of three fields per group and three independent experiments. Scale bars, 5 μm. (D) Colocalization of MAP1LC3B and mitochondria (as % overlay) was quantified. (E) Analysis of mitophagic flux (red puncta compared to yellow). Cells were transfected with siRNAs and then with the plasmid pAT016 (p-mito-mRFP-EGFP) and treated with TH (n = 5 biological replicates per group). Enlarged panels represent selected digitally enlarged portions of parent images to enhance the visibility of mitochondria inside autolysosome. Images are representative of three fields per group and three independent experiments. Scale bars, 5 μm. (F) Mitophagic flux was quantified. Statistical significance was calculated using Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment (*P ≤ 0.05).

  • Fig. 8 TH-induced mitophagy and mitochondrial turnover depended on ESRRA, ULK1, DRP1, and FUNDC1.

    (A) Western blot analysis of ESRRA and mitochondrial proteins in bafilomycin A1 (BAF A1)–treated THRB1-HepG2 cells transfected with control siRNA or ESRRA siRNA in the presence or absence of TH (100 nM for 48 hours). Blots are representative of three independent experiments. (B) Measurement of relative mitophagic flux or of mitochondrial protein abundance in THRB1-HepG2 cells after BAF A1 treatment to block autophagy. The ratios of band densities from BAF A1–treated samples to untreated samples are shown. The relative mitophagic flux as determined by the accumulation of COX-IV,VDAC, SDHA, and PDH. after BAF A1 treatment is shown for control siRNA (Control siRNA + BAF A1)/(Control siRNA), Control siRNA + TH (Control siRNA + TH + BAF A1)/(Control siRNA + TH), and ESRRA siRNA + TH (ESRRA siRNA + TH + BAF A1)/(ESRRA siRNA + TH) (n = 5 biological replicates per group). COX-IV, cytochrome c oxidase subunit IV; VDAC, voltage-dependent anion channel; SDHA, succinate dehydrogenase complex flavoprotein subunit A; PDH, pyruvate dehydrogenase. (C and D) Representative Western blot for the mitochondrial protein SDHA (C) and relative mitophagy flux (D) based on SDHA abundance in THRB1-HepG2 cells subjected to ULK1 knockdown after BAF A1 treatment to block autophagy (n = 5 biological replicates per group). Blots are representative of three independent experiments. (E and F) Representative Western blot for SDHA (E) and relative mitophagy flux (F) based on SDHA abundance in THRB1-HepG2 cells subjected to DRP1 knockdown after BAF A1 treatment to block autophagy (n = 5 biological replicates per group). Blots are representative of three independent experiments. (G and H) Representative Western blot for SDHA (G) and relative mitophagy flux (H) based on SDHA abundance in THRB1-HepG2 cells subjected to FUNDC1 knockdown after BAF A1 treatment to block autophagy (n = 5 biological replicates per group). Blots are representative of three independent experiments. (I and J) Measurement of mitochondrial OCR using a Seahorse extracellular flux analyzer (I) in HepG2 cells treated with TH and ULK1, DRP1, or FUNDC1 siRNA in THRB1-HepG2 cells (n = 5 biological replicates per group). Data are representative of three independent experiments. Functional parameters of OXPHOS (basal OCR, ATP turnover, maximum respiratory capacity, and spare respiratory capacity) were calculated in the presence of mitochondrial inhibitors [1 μM oligomycin, 1 μM FCCP, and 1 μM rotenone and 1 μM antimycin A (R + A)] (J). (K and L) Mitochondrial turnover analysis using mitochondrial targeted fluorescent TIMER protein (mitoTIMER). mitoTIMER plasmid was transfected in THRB1-HepG2 cells with control or ESRRA siRNA with or without TH (100 nM for 48 hours). Representative images are shown (K), and fluorescence was quantified (L) (n = 5 biological replicates per group). Images are representative of five fields per group and two independent experiments. Scale bars, 5 μm. Green fluorescence shows newly synthesized mitochondria, whereas red fluorescence indicates damaged mitochondria. Statistical significance for (J) was calculated using one-way ANOVA, followed by Tukey’s post hoc test; otherwise, Kruskal-Wallis nonparametric ANOVA, followed by post hoc pairwise comparisons using the Mann-Whitney test with Bonferroni adjustment, was performed (*P ≤ 0.05).

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/11/536/eaam5855/DC1

    Fig. S1. ESRRA co-regulated pathways induced by TH.

    Fig. S2. TH-ESRRA co-regulation of TCA and FAO.

    Fig. S3. TH activated ESRRA and OXPHOS in a PPARGC1A-dependent manner.

    Fig. S4. TH-ESRRA co-regulated the expression of genes encoding mitochondrial biogenesis factors but not those encoding mitochondrial fission, fusion, and mitophagy factors.

    Fig. S5. TH increased ESRRA recruitment to the ULK1 gene promoter.

    Fig. S6. DRP1 inhibition blocked the translocation of autophagic proteins but not ULK1 translocation to mitochondria.

    Fig. S7. TH increased lysosome and mitochondria colocalization in an ESRRA-dependent manner.

    Fig. S8. ESRRA inhibition dysregulated TH-induced mitochondrial membrane potential and ROS.

    Fig. S9. Proposed model for TH-ESRRA co-regulation of mitochondrial biogenesis, fission, mitophagy, and activity.

    Data file S1. THRB1 and ESRRA ChIP-seq analysis.

  • Supplementary Materials for:

    Thyroid hormone receptor and ERRα coordinately regulate mitochondrial fission, mitophagy, biogenesis, and function

    Brijesh K. Singh,* Rohit A. Sinha, Madhulika Tripathi, Arturo Mendoza, Kenji Ohba, Jann A. C. Sy, Sherwin Y. Xie, Jin Zhou, Jia Pei Ho, Ching-yi Chang, Yajun Wu, Vincent Giguère, Boon-Huat Bay, Jean-Marc Vanacker, Sujoy Ghosh, Karine Gauthier, Anthony N. Hollenberg, Donald P. McDonnell, Paul M. Yen*

    *Corresponding author. Email: paul.yen{at}duke-nus.edu.sg (P.M.Y.); singhbrijeshk{at}duke-nus.edu.sg (B.K.S.)

    This PDF file includes:

    • Fig. S1. ESRRA co-regulated pathways induced by TH.
    • Fig. S2. TH-ESRRA co-regulation of TCA and FAO.
    • Fig. S3. TH activated ESRRA and OXPHOS in a PPARGC1A-dependent manner.
    • Fig. S4. TH-ESRRA co-regulated the expression of genes encoding mitochondrial biogenesis factors but not those encoding mitochondrial fission, fusion, and mitophagy factors.
    • Fig. S5. TH increased ESRRA recruitment to the ULK1 gene promoter.
    • Fig. S6. DRP1 inhibition blocked the translocation of autophagic proteins but not ULK1 translocation to mitochondria.
    • Fig. S7. TH increased lysosome and mitochondria colocalization in an ESRRAdependent manner.
    • Fig. S8. ESRRA inhibition dysregulated TH-induced mitochondrial membrane potential and ROS.
    • Fig. S9. Proposed model for TH-ESRRA co-regulation of mitochondrial biogenesis, fission, mitophagy, and activity.

    [Download PDF]

    Other Supplementary Material for this manuscript includes the following:

    • Data file S1 (Microsoft Excel format). THRB1 and ESRRA ChIP-seq analysis.

    © 2018 American Association for the Advancement of Science

Navigate This Article