Multiomic investigation on the relationship between metabolism and gene expression

Cellular metabolism and gene expression are proposed to be highly integrated but a detailed mechanistic understanding of this relationship is lacking. The central metabolite, acetyl-CoA, is the substrate for histone acetylation and increases in levels of acetyl-CoA are associated with increased histone acetylation and active gene expression. How histone acetylation leads to upregulated gene expression and how the altered patterns of gene expression subsequently influence cellular metabolism is not fully understood. This work addresses the ‘acetylation-recruitment’ hypothesis: that the upregulated histone acetylation in response to the increased acetyl-CoA can recruit a bromodomain containing chromatin remodelling complex to activate gene expression, which can further regulate cellular metabolism. To facilitate this study, a growth phenomenon in budding yeast, S.cerevisiae, known as the YMC (yeast metabolic cycle) is exploited. In the YMC, yeast cells exhibit highly synchronized and periodic oscillations between the high and low oxygen consumption states, associated with cycling levels of acetyl-CoA, histone acetylation and transcripts. To link histone acetylation to cycling transcripts, the role of various bromodomain containing proteins in maintaining the YMC is tested and the Rsc2 isoform of the RSC complex is identified as important. By comparing the temporal profiles of transcripts between the wild-type and the rsc2Δ strain, it is found that the expression of genes having histone H3 acetylation at their promoters is significantly more likely to be dependent on the RSC complex. Furthermore, ACO2 is identified and verified as a gene associated with the upregulation of oxygen consumption and also a downstream target of the acetylation-RSC-recruitment pathway. Finally, two new techniques have been developed with the aim of addressing the question about the relationship between transcription and transcripts during the YMC and how this relates to metabolism. One is ultralow background NET-seq, which reduces the percentage of background signals in a yeast NETseq library from 32% to 0.7%. The other is size-fractionated NET-seq, which can map transcription start sites for both coding and noncoding transcription units.