A transcriptional regulatory mechanism of genes in the tricarboxylic acid cycle in the heart

Abstract

The tricarboxylic acid (TCA) cycle plays a crucial role in mitochondrial ATP production in the healthy heart. However, in heart failure, the TCA cycle becomes dysregulated. Understanding the mechanism by which TCA cycle genes are transcribed in the healthy heart is an important prerequisite to understanding how these genes become dysregulated in the failing heart. PPARγ coactivator 1α (PGC-1α) is a transcriptional coactivator that broadly induces genes involved in mitochondrial ATP production. PGC-1α potentiates its effects through the coactivation of coupled transcription factors, such as estrogen-related receptor (ERR), nuclear respiratory factor 1 (Nrf1), GA-binding protein-a (Gabpa), and Yin Yang 1 (YY1). We hypothesized that PGC-1α plays an essential role in the transcription of TCA cycle genes. Thus, utilizing localization peaks of PGC-1α to TCA cycle gene promoters would allow the identification of coupled transcription factors. PGC-1α potentiated the transcription of 13 out of 14 TCA cycle genes, partly through ERR, Nrf1, Gabpa, and YY1. ChIP-sequencing showed PGC-1α localization peaks in TCA cycle gene promoters. Transcription factors with binding elements that were found proximal to PGC-1α peak localization were generally essential for the transcription of the gene. These transcription factor binding elements were well conserved between mice and humans. Among the four transcription factors, ERR and Gabpa played a major role in potentiating transcription when compared to Nrf1 and YY1. These transcription factor-dependent PGC-1α recruitment was verified with Idh3a, Idh3g, and Sdha promoters with DNA binding assay. Taken together, this study clarifies the mechanism by which TCA cycle genes are transcribed, which could be useful in understanding how those genes are dysregulated in pathological conditions.

Keywords: ERR; Gabpa; Nrf1; PGC-1α; YY1; mitochondria; transcription; tricarboxylic acid (TCA) cycle.

Figure 1

Regulation of TCA cycle genes in stress conditions.A, a schematic representation of the TCA cycle. The TCA cycle consists of eight enzymes composed of 14 genes involved in sequential linear steps required for mitochondrial energy production. Abbreviations of the enzymes are as follows. Aco2, Aconitase 2; Cs, Citrate Synthase; Fh1, Fumarate Hydratase 1; Idh3, Isocitrate Dehydrogenase; Mdh2, Malate Dehydrogenase 2; Ogdh, Oxoglutarate Dehydrogenase; Sdh, Succinate Dehydrogenase; and Suclg, Succinate-CoA Ligase. B-C, the expression of TCA cycle genes in response to pressure overload (2 weeks) (B) and fasting (C). n = 8(A) and 6 to 8(B). ★p < 0.05 vs control mice.

Figure 2

Many, if not all, TCA cycle genes are PGC-1α targets.A and B, the levels of TCA cycle genes at mRNA (A) and protein (B) in PGC-1αcKO mice. C, ChIP-seq shows PGC-1α localization in the promoter regions of 13 out of 14 TCA cycle genes. These ChIP-seq images magnifying a promoter region are reused in Figure 3A. D, the levels of citrate, succinate and malate metabolites in PGC-1αcKO mice, detected with enzymatic assay (citrate) and metabolome analysis (succinate and malate). E, decrease in ATP production in PGC-1αcKO mice. n = 8(A), 7(B), 6 to 7(D) and 6(E). (F) Schematic representation of how the TCA cycle is regulated. ▼: Downregulation at mRNA or protein levels. ▼ or ▲: Trend of down or upregulation. Metabolites not detected is shown by gray. ★p < 0.05 vs control mice. PGC-1α, PPARγ coactivator 1α.

Figure 3

Possible binding elements of ERR, YY1, Nrf1, and Gabpa in TCA cycle gene promoters.A, a bioinformatics analysis with transcription factor binding profiling identifies possible binding elements of ERR, NRf1, YY1, and Gabpa in TCA cycle gene promoters. Distance (bp) from the TSS and the peak PGC-1α location relative to the TSS are shown on the top, whereas the full genomic size is shown in the right bottom corner of ChIP-seq images. The ChIP-seq images are reused in Figure 4, Figure 6 and 6D, S1, S2 and S4 for specific purposes stated in the figure legends. B, the presence or absence of transcription binding element found within 100 bp ± of PGC-1a peaks are shown. ERR, estrogen-related receptor; Gabpa, GA-binding protein-a; Nrf1, nuclear respiratory factor 1; PGC-1α, PPARγ coactivator 1α; YY1, Yin Yang 1.

Figure 4

Identification of transcription factor essential for transcription of TCA cycle enzymes. Gene expressional analyses were performed in cultured cardiomyocytes with knockdown of indicated transcription factors with indicated n. ★p < 0.05 vs control siRNA. The ChIP-seq images are reused from Figure 3A to indicate the extent to which a transcription factor whose binding sequence is found proximal to the PPARγ coactivator 1α localization peak essentially transcribes the gene. ERR, estrogen-related receptor; Gabpa, GA-binding protein-a; Nrf1, nuclear respiratory factor 1; YY1, Yin Yang 1.

Figure 5

The critical transcription factor binding elements are conserved between mice and humans. Homology search in comparison between mice and human genomes was performed with the selected binding elements. M and H indicate mice and humans, respectively. The number indicates the position from the transcription start site. The back bar indicates the identical nucleotide sequence. The homology (ratio of identical nucleotides relative to total number of nucleotides) in the genomic regions is also indicated. ERR, estrogen-related receptor; Gabpa, GA-binding protein-a; Nrf1, nuclear respiratory factor 1; YY1, Yin Yang 1.

Figure 6

The binding of ERR, Nrf1, Gabpa and YY1 to the TCA cycle gene promoters.A, biotin-labeled DNA used by this study. B and C, the binding of ERRα, Nrf1, and Gabpa to the indicated biotin-labeled DNA with nuclear extract from cultured cardiomyocytes. D, biotin-labeled DNA comprising indicated endogenous promoter sequences. The binding elements of the transcription factors are shown. The ChIP-seq images are reused from Figure 3A to indicate the promoter region with possible transcription factor binding elements used for DNA binding assays. E, the DNA binding assays were performed with indicated biotin-labeled DNA and nuclear extract from cultured cardiomyocytes with knockdown of indicated transcription factors. The results were verified with at least three independent experiments. ERR, estrogen-related receptor; Gabpa, GA-binding protein-a; Nrf1, nuclear respiratory factor 1; PGC-1α, PPARγ coactivator 1α; YY1, Yin Yang 1.

Figure 7

Transcriptional mechanism for TCA cycle genes.A, Schematic representation for the basal transcriptional mechanism for TCA cycle genes. B, the expression levels of indicated proteins in the PO model and sham-operated LV as a control. C, recruitment of PGC-1α, but not ERRα, Nrf1, Gabpa, or YY1, is inhibited in the PO model. DNA binding assays were performed with nuclear extract of the LV subjected to 4 weeks of PO. The most representative images are shown with 4 to 6 independent experiments. ERR, estrogen-related receptor; Gabpa, GA-binding protein-a; Nrf1, nuclear respiratory factor 1; PGC-1α, PPARγ coactivator 1α; YY1, Yin Yang 1.