Genomic approaches to connecting the transcriptional regulatory network of Saccharomyces cerevisiae stress responses

The yeast Saccharomyces cerevisiae is the leading model organism for study by genomic methods. Whole-genome transcript profile data are now being generated routinely, and these data give important insights into the differences in transcriptional responses to many conditions. Grouping and clustering microarray data by various methods helps us to discern the dominant transcriptional groupings of all genes across some, or all, tested conditions. In Chapter II, I describe whole transcript profile data I have generated for multiple stress-inducing conditions in S. cerevisiae . Conditionally co-expressed groups of genes can be separated from other genes by certain grouping and clustering approaches, and these “enriched” groups of genes often contain upstream binding sites for site-specific transcription factors.; Motif searches can help us to find these cis elements among genes in a group or cluster. This can be done for any or all clusters in a genome. Trans factors that belong to one cluster regulate transcription of genes in other clusters by binding to cognate cis elements upstream of the genes in these other clusters. Whole-genome site matrix searches for these cognate cis elements allow us to infer a regulatory connection between the cluster to which the trans factor belongs, and the cluster(s) whose genes possess cognate upstream cis elements. These connections between trans factors and their cognate cis-acting binding sites allow us to describe a connectivity map for site specific regulatory elements. These inferred connections are relatively easy to make when cognate cis and trans regulatory elements that control expression of gene clusters have been previously described and experimentally tested; however, previously undescribed upstream cis-acting sites, such as those found by motif searches of clustered genes, need to be linked to their cognate trans factors.; In Chapter III I describe the discovery of a motif whose association with proteases is extremely statistically significant. It is found upstream of all 31 members of the yeast proteasome, and for this reason is called the p&barbelow;roteasome a&barbelow;ssociated c&barbelow;ontrol e&barbelow;lement (PACE). Using a one-hybrid approach I cloned two trans-acting factors that bind to this motif, RPT2 and RPN4. RPT2 is a proteasome subunit, and RPN4 is a zinc finger protein and disputed member of the proteasome.; In Chapter IV, I show that RPN4 regulates expression of the proteasome regulon by performing whole-genome transcript profiling of both over-expressed RPN4, and rpn4Δ strains with and without treatment with H₂O ₂. Two important stress response transcription factors, YAP1 and HSF1, the heat shock factor, appear to be part of this regulon. These factors are very important to a model of RPN4 mediated transcriptional regulation of the stress response. Not only do these factors have PACE sites upstream, and are likely members of the RPN4 regulon, but they also appear to regulate expression of RPN4. This model is supported by a variety of data, and the promoter region of RPN4 has consensus binding sites for both YAP1 and HSF1.; Three related publications I co-authored are featured in the appendices. A fourth publication on which I am a co-author is listed in the appendices, but it is not reprinted there.