The molecular dissection of the regulation of PGC-1α and the genome-wide activity of its transcriptional network in skeletal muscle cells

The transcriptional coactivator peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) is a potent inducer of mitochondrial biogenesis and oxidative metabolism. Highest expression levels of PGC-1α are found in tissues with high energy demands like brain, skeletal muscle, heart, brown adipose tissue and kidney. In these tissues, PGC-1α can be induced by metabolic stress, like cold in brown fat, exercise in skeletal muscle or fasting in liver. Once activated, PGC-1α regulates the activity of transcription factors and is thus capable of inducing entire biological programs like mitochondrial biogenesis, fatty acid oxidation, angiogenesis or gluconeogenesis. PGC-1α does not possess an intrinsic histone acetylase activity (HAT), but instead binds proteins with HAT activity and recruits them to the site of transcription. Similarly, PGC-1α binds the mediator and to the SWI/SNF complexes and thereby serves as a platform to connect the transcription factors with transcription initiation complex, chromatin remodelling complex and proteins with HAT activity. Therefore, PGC-1α senses extracellular stimuli and metabolic stress and connects these events with gene transcription. In skeletal muscle, basically all pathways triggered by exercise at some point converge at PGC-1α and change the expression of Ppargc1a transcripts or stabilize the PGC-1α protein through posttranslational modifications. When expressed in skeletal muscle, PGC-1α induces mitochondrial biogenesis, glucose uptake, promotes angiogenesis, protects skeletal muscle from atrophy and leads to a muscle fibre type switch towards more oxidative fibres. Thus, PGC-1α acts as a master regulator of exercise-induced adaptations in skeletal muscle.
Some key transcription factors mediating these PGC-1α induced changes like ERRα, NRF-1 and MEF2C have been identified. However, a profound knowledge about the transcriptional network of transcription factors and other proteins mediating PGC-1α gene regulation is still missing. To reveal this transcriptional network and to be able to draw general conclusions about the role of PGC-1α as a coactivator, we have investigated the activity of PGC-1α on genome-wide scale. By combining ChIP-Seq studies with expression arrays, we have identified all interactions of PGC-1α with the genome in cultured skeletal muscle cells and gained knowledge about how PGC-1α regulates gene expression. PGC-1α induced expression of genes involved in oxidative metabolism and suppressed the expression of inflammatory response genes. Surprisingly, the induction of gene expression by PGC-1α was not only directly by binding to transcription factors in promoters, but also indirectly, without the need for PGC-1α to be present at the promoters of some induced genes. Inversely, the suppression of inflammatory genes was almost exclusively indirect because it did not require the recruitment of PGC-1α to the promoters of suppressed genes, indicating that PGC-1α does not act as a corepressor in skeletal muscle cells. We identified ERRα as a major mediator of PGC-1α induced gene expression. By performing ChIP-sequencing of ERRα, we have found that ERRα can be transcriptionally active and regulate gene expression with and without PGC-1α. In addition to ERRα, we predict several other transcription factors to cooperate with PGC-1α and directly regulate gene expression. By knocking down some of these transcription factors, we validated our predictions and showed that these transcription factors are involved in the transcription of a subset of PGC-1α target genes. These results suggest that PGC-1α coactivates the transcription factor complex AP-1 to regulate the expression of genes involved in the response to hypoxia. Last, even though the inhibition of phosphodiesterases PDE1 and PDE4 led to induction of Ppargc1a expression in cultured skeletal muscle cells, this effect could not be shown in vivo. Because the activation of β2-AR signaling strongly induced Ppargc1a expression in skeletal muscle in vivo, the involvement of cAMP in the regulation of Ppargc1a expression is very likely. Therefore the involvement of other cAMP-specific PDEs in this regulation cannot be excluded.
In conclusion, in this thesis, we describe how the transcriptional coactivator PGC-1α controls gene expression in cultured skeletal muscle cells on a genome-wide scale. We identified and validated some key transcription factors as members of the PGC-1α transcriptional network. The large amount of data generated in this study and our predictions could serve as a starting point for future projects that aim to study PGC-1α.