Yeast zinc cluster proteins Dal81 and Uga3 cooperate by targeting common coactivators for transcriptional activation of γ-aminobutyrate responsive genes.

In Saccharomyces cerevisiae, optimal utilization of various compounds as a nitrogen source is mediated by a complex transcriptional network. The zinc cluster protein Dal81 is a general activator of nitrogen metabolic genes, including those for γ-aminobutyrate (GABA). In contrast, Uga3 (another zinc cluster protein) is an activator restricted to the control of genes involved in utilization of GABA. Uga3 binds to DNA elements found in the promoters of target genes and increases their expression in the presence of GABA. Dal81 appears to act as a coactivator since the DNA-binding activity of this factor is dispensable but its mode of action is not known. In this study, we have mapped a regulatory, as well as an activating, region for Uga3. A LexA-Uga3 chimeric protein activates a lexA reporter in a GABA- and Dal81-dependent manner. Activation by Uga3 requires the SAGA complex as well as Gal11, a component of mediator. ChIP analysis revealed that Uga3 is weakly bound to target promoters. The presence of GABA enhances binding of Uga3 and allows recruitment of Dal81 and Gal11 to target genes. Recruitment of Gal11 is prevented in the absence of Dal81. Importantly, Dal81 by itself is a potent activator when tethered to DNA and its activity depends on SAGA and Gal11 but not Uga3. Overexpression of Uga3 bypasses the requirement for Dal81 but not for SAGA or Gal11. Thus, under artificial conditions, both Dal81 and Uga3 can activate transcription independently of each other. However, under physiological conditions, both factors cooperate by targeting common coactivators.

Large-scale analysis of genes that alter sensitivity to the anticancer drug tirapazamine in Saccharomyces cerevisiae.

Tirapazamine (TPZ) is an anticancer drug that targets topoisomerase II. TPZ is preferentially active under hypoxic conditions. The drug itself is not harmful to cells; rather, it is reduced to a toxic radical species by an NADPH cytochrome P450 oxidoreductase. Under aerobic conditions, the toxic compound reacts with oxygen to revert back to TPZ and a much less toxic radical species. We have used yeast (Saccharomyces cerevisiae) as a model to better understand the mechanism of action of TPZ. Overexpression of NCP1, encoding the yeast ortholog of the human P450 oxidoreductase, results in greatly increased sensitivity to TPZ. Likewise, overexpression of TOP2 (encoding topoisomerase II) leads to hypersensitivity to TPZ, suggesting that topoisomerase II is also a target of TPZ in yeast. Thus, our data show that yeast mimics human cells in terms of TPZ sensitivity. We have performed robot-aided screens for altered sensitivity to TPZ using a collection of approximately 4600 haploid yeast deletion strains. We have identified 117 and 73 genes whose deletion results in increased or decreased resistance to TPZ, respectively. For example, cells lacking various DNA repair genes are hypersensitive to TPZ. In contrast, deletion of genes encoding some amino acid permeases results in cells that are resistant to TPZ. This suggests that permeases may be involved in intracellular uptake of TPZ. Our discoveries in yeast may lead to a better understanding of TPZ biology in humans.

Zinc cluster protein Rdr1p is a transcriptional repressor of the PDR5 gene encoding a multidrug transporter.

The yeast PDR5 gene encodes an efflux pump that confers multidrug resistance. Expression of PDR5 is positively regulated by the transcription factors Pdr1p and Pdr3p that recognize the same pleiotropic drug resistance elements (PDREs) in the PDR5 promoter. Pdr1p and Pdr3p belong to the Gal4p family of zinc cluster proteins. The function of RDR1 (YOR380W), which also encodes a member of this family, is unknown. To identify target genes for Rdr1p, we have performed whole-genome analysis of gene expression with DNA microarrays. Our results show that Rdr1p is a transcriptional repressor of five genes, including PDR5. A Deltardr1 strain has increased resistance to cycloheximide, as expected from the overexpression of PDR5. In addition, the activity of a PDR5-lacZ reporter is increased in a Deltardr1 strain. All (but one) genes affected by removal of Rdr1p contain PDREs in their promoters. We tested if the effect of Rdr1p is mediated through PDREs by inserting this DNA element in front of a minimal promoter. Activity of this reporter was increased in a Deltardr1 strain. Moreover, mutations known to reduce binding of Pdr1/Pdr3p abolished the induction observed in the Deltardr1 strain. Thus, we have identified a transcriptional repressor involved in the control of multidrug resistance.