In Saccharomyces cerevisiae ρ 0 Cells, UME6 Contributes to the Activation of ABC Transporter Genes and Pleiotropic Drug Resistance via RPD3 and PDR3

: In Saccharomyces cerevisiae , the Rpd3L complex includes the histone deacetylase Rpd3 and the DNA binding proteins Ume6 and Ash1 and serves as a transcriptional silencer or enhancer. In S. cerevisiae , the transcription of PDR5 , which encodes a major drug efflux pump, and pleiotropic drug resistance (PDR) are hyperactivated by the transcription factor Pdr3 in ρ 0/ − cells, which lack mitochondrial DNA. We previously showed that RPD3 and UME6 are required for the activation of PDR5 transcription and PDR in S. cerevisiae ρ 0 cells. Here, using real-time PCR analysis, we revealed that RPD3 and UME6 are responsible for the activated basal expression of the ABC transporter-encoding genes SNQ2 , PDR15, and PDR5 in S. cerevisiae ρ 0 cells. Furthermore, using real-time PCR analysis and a spot dilution assay, we found that Ume6 increases the basal expression of PDR5 and PDR15 and induces PDR in a manner dependent on RPD3 and PDR3 in ρ 0 cells. This finding may contribute to the elucidation of the relationships between the molecules required for the activation of ABC transporter genes in S. cerevisiae ρ 0/ − cells and in pathogenic Candida species.


Introduction
Multidrug resistance refers to the acquired resistance of cancer cells and microorganisms to a wide variety of chemotherapeutic drugs [1,2].Therefore, multidrug resistance is a serious concern in the treatment of cancer and microbial infections.In mammalian cells, overexpression of the MDR1 gene, which encodes P-glycoprotein, an ATP-binding cassette (ABC) transporter, is a major mechanism underlying multidrug resistance in cancer cells [3][4][5].Yeast Candida species Candida albicans and Candida glabrata are opportunistic pathogenic microorganisms [1].In pathogenic Candida species, the most prevalent mechanism of multidrug resistance involves increased activity of ABC transporters and major facilitator superfamily pumps [6][7][8].Therefore, the elucidation of multidrug resistance mechanisms in cancer cells and Candida species is needed to develop anticancer and antifungal drugs that overcome multidrug resistance [3,9,10].
The yeast Saccharomyces cerevisiae is used as a model organism for studies of multidrug resistance in pathogenic Candida species [11].The resistance phenotype of S. cerevisiae to a broad range of cytotoxic compounds is referred to as pleiotropic drug resistance (PDR) [12].PDR in S. cerevisiae is often associated with the overexpression of plasma membrane ABC transporters [13][14][15].Plasma membrane ABC transporters, such as Pdr5, Snq2, and Yor1, export a variety of functionally and structurally unrelated compounds from cells [13,16,17].PDR5 encodes a major efflux pump of structurally and functionally unrelated drugs and xenobiotics, such as fluconazole and cycloheximide [18,19].
The retrograde signalling pathway is strongly activated in S. cerevisiae and C. glabrata ρ 0/− cells, which have mitochondrial DNA defects [25,26].Deletion of mitochondrial DNA in C. glabrata cells results in increased expression of multidrug resistance genes, including CgCDR1, and increased PDR.Pdr3 but not Pdr1 is strongly activated in S. cerevisiae ρ 0/− cells; therefore, the transcription of ABC transporter genes such as PDR5 and PDR is hyperactivated [27].It has previously been reported that the Hsp70 chaperone Ssa1, the mitochondrially localised phosphatidylserine decarboxylase Psd1, the subunit of the RNA polymerase II mediator complex Med12, and the histone H2 ubiquitination enzyme Lge1 are required for the activation of ABC transporter gene transcription and PDR in S. cerevisiae ρ 0/− cells.Pdr3 is negatively regulated by the Hsp70 protein Ssa1 through a direct interaction [28].Furthermore, less Ssa1 is bound to Pdr3 in ρ 0 cells than in ρ + cells, which contain mitochondrial DNA, suggesting the release of Pdr3 from the negative regulation of Ssa1 in ρ 0 cells [28].Deletion of Med12 completely suppresses the induction of PDR5 expression in ρ 0 cells but not in ρ + cells [29].Psd1 is involved in phosphatidylethanolamine (PE) synthesis, and loss of the PSD1 gene from ρ 0 cells prevents the normal activation of PDR5 [30].In addition, the expression of a catalytically inactive form of Psd1 induces PDR5 transcription in ρ 0 cells [30].Lge1 is indispensable for the induction of PDR5 transcription in ρ 0 cells [31,32].However, the relationships among these molecules required for the activation of ABC transporter genes in ρ 0 cells are unknown.
We previously showed that RPD3 and UME6 are required for the hyperactivation of PDR5 transcription and PDR via retrograde signalling in S. cerevisiae ρ 0 cells [45].In contrast, we also previously reported that Ume6 suppresses the basal transcription of ABC transporters, including PDR5, and the PDR in ρ + cells; however, RPD3 is required for drug resistance but does not alter the basal PDR5 mRNA level [46].
Here, we show that RPD3 and UME6 are responsible for the activated basal expression of not only PDR5 but also SNQ2 and PDR15 in S. cerevisiae ρ 0 cells.We also show that UME6 increases the basal expression of PDR5 and PDR15 and increases the PDR in a manner dependent on RPD3 and PDR3 in ρ 0 cells.

Spot Dilution Assay
The relative resistance of each yeast strain to fluconazole or cycloheximide was estimated with a spot dilution assay using YPD media [45,46,49].The ρ 0 cells from each yeast strain were aerobically grown to the logarithmic phase (at an OD 600 of 0.6-0.9) at 30 • C in YPD media in triplicate.Five microlitre aliquots of 10-fold serial dilutions of cultures containing the same number of cells were spotted on YPD plates with or without 10 µg/mL of fluconazole (Nacalai Tesque, Kyoto, Japan) (or 0.3 µg/mL of cycloheximide (Wako Pure Chemicals, Osaka, Japan)) and incubated at 30 • C for 7 days.
2.3.RNA Extraction from ρ 0 Cells of Each Yeast Strain Grown to the Logarithmic Growth Phase ρ 0 cells from each yeast strain were grown to an OD 600 of 7-9 in YPD media in duplicate.The cultures were diluted to an OD 600 of 0.1 and grown to an OD 600 of 0.4-0.8[45,46,49].The cultures were recovered, and the cells in the cultures were pelleted, washed, frozen at −80 • C, and used for total RNA extraction [45,46,49].Total RNA was isolated from the yeast cells using a NucleoSpin RNA kit (TaKaRa Bio Inc., Shiga, Japan) according to the manufacturer's protocol.

Real-Time RT-PCR
Reverse transcription of total RNA was performed using FastGene Scriptase II cDNA 5× ReadyMix (NIPPON Genetics, Tokyo, Japan) and oligo dT primers (TaKaRa Bio Inc., Shiga, Japan).SYBR Green qRT-PCR for cDNA from the individual duplicate samples was performed using TB Green Premix Ex Taq II (TaKaRa Bio Inc., Shiga, Japan) in a Step One Real-time PCR system (Applied Biosystems, Foster City, CA, USA) [52].A minus reverse transcriptase control was used as the negative control.Serial dilutions of control cDNA were prepared to produce a standard curve for each primer pair.The primers used for qRT-PCR are listed in Table S1.The mRNA levels of each target gene were measured by qRT-PCR and normalised to those of the housekeeping gene ACT1, which was used as an endogenous control.The normalised mRNA levels are shown relative to samples from the wild-type strains, which were set to 1.

Statistical Analysis
An unpaired Student's t-test was used for statistical analysis.Results with p < 0.05 and p < 0.01 were considered statistically significant.We previously reported that activated PDR5 transcriptional expression by retrograde signalling was significantly reduced in ρ 0 cells of the rpd3∆ and ume6∆ strains.Thus, we investigated the mRNA levels of the major ABC transporter genes PDR15, PDR10, SNQ2, and YOR1, including PDR5, in ρ 0 cells of the wild-type, ume6∆, and rpd3∆ strains by qRT-PCR.qRT-PCR revealed that the YOR1, SNQ2, PDR15, PDR10, and PDR5 mRNA levels were statistically significantly lower in the ume6∆ strain than in the wild-type strain (p < 0.05) (Figure 1).In contrast, the expression of SNQ2, PDR15, and PDR5, but not that of YOR1 or PDR10, was statistically significantly lower in the rpd3∆ strain than in the wild-type strain (p < 0.05) (Figure 2).In both the ume6∆ and rpd3∆ mutants, although the SNQ2 mRNA levels were relatively mildly reduced, the mRNA levels of SNQ2, PDR15, and PDR5 were reduced compared with those in the wild-type strain (Figures 1 and 2).These results suggest that SNQ2, PDR15, and PDR5 are upregulated by Ume6 via Rpd3 in ρ 0 cells.

UME6 and RPD3
according to the manufacturer's protocol.

Real-Time RT-PCR
Reverse transcription of total RNA was performed using FastGene Scriptase II cDNA 5× ReadyMix (NIPPON Genetics, Tokyo, Japan) and oligo dT primers (TaKaRa Bio Inc., Shiga, Japan).SYBR Green qRT-PCR for cDNA from the individual duplicate samples was performed using TB Green Premix Ex Taq II (TaKaRa Bio Inc., Shiga, Japan) in a Step One Real-time PCR system (Applied Biosystems, Foster City, CA, USA) [52].A minus reverse transcriptase control was used as the negative control.Serial dilutions of control cDNA were prepared to produce a standard curve for each primer pair.The primers used for qRT-PCR are listed in Table S1.The mRNA levels of each target gene were measured by qRT-PCR and normalised to those of the housekeeping gene ACT1, which was used as an endogenous control.The normalised mRNA levels are shown relative to samples from the wild-type strains, which were set to 1.

Statistical Analysis
An unpaired Student's t-test was used for statistical analysis.Results with p < 0.05 and p < 0.01 were considered statistically significant.

UME6 and RPD3 Are Required for the Upregulation of the Steady-State mRNA Levels of SNQ2, PDR15, and PDR5 in ρ 0 Cells
We previously reported that activated PDR5 transcriptional expression by retrograde signalling was significantly reduced in ρ 0 cells of the rpd3∆ and ume6∆ strains.Thus, we investigated the mRNA levels of the major ABC transporter genes PDR15, PDR10, SNQ2, and YOR1, including PDR5, in ρ 0 cells of the wild-type, ume6∆, and rpd3∆ strains by qRT-PCR.qRT-PCR revealed that the YOR1, SNQ2, PDR15, PDR10, and PDR5 mRNA levels were statistically significantly lower in the ume6∆ strain than in the wild-type strain (p < 0.05) (Figure 1).In contrast, the expression of SNQ2, PDR15, and PDR5, but not that of YOR1 or PDR10, was statistically significantly lower in the rpd3∆ strain than in the wildtype strain (p < 0.05) (Figure 2).In both the ume6∆ and rpd3∆ mutants, although the SNQ2 mRNA levels were relatively mildly reduced, the mRNA levels of SNQ2, PDR15, and PDR5 were reduced compared with those in the wild-type strain (Figures 1 and 2).These results suggest that SNQ2, PDR15, and PDR5 are upregulated by Ume6 via Rpd3 in ρ 0 cells.

RPD3 and PDR3 Are Required for the Partial Rescue of the Reduction in PDR15 and PDR5 Expression in the ume6∆ Strain by Exogeneous UME6 Expression
To validate whether reduced mRNA levels of PDR15 and PDR5 in the ume6∆ mutant result from the deletion of UME6, we investigated whether decreased mRNA levels of PDR15 and PDR5 in the ume6∆ mutant are complemented with exogeneous expression of UME6.Although PDR15 and PDR5 mRNA levels were statistically significantly lower in ume6∆ pRS313 than in the wild-type strain (p < 0.01), the reductions in basal PDR15 and PDR5 mRNA levels in ume6∆ pRS313 were incompletely restored to wild-type pRS313 levels in ume6∆ pRS313-UME6 (p < 0.01) (Figure 3).The incomplete rescue by introduction of pRS313-UME6 into the ume6∆ mutant could result from nonphysiological levels of UME6 expression, because qRT-PCR revealed that UME6 is overexpressed at high levels-at least sevenfold-in ume6∆ pRS313-UME6 compared with wild-type pRS313.Regardless, this restoration of PDR15 and PDR5 mRNA levels suggests that the deletion of UME6 is a cause of reduced mRNA levels of PDR15 and PDR5 in the ume6∆ mutant.
Next, we investigated whether RPD3 and PDR3 are required for the partial rescue of reduced mRNA levels of PDR15 and PDR5 in the ume6∆ mutant by exogenous expression of UME6.Although there was no significant difference in the mRNA levels of PDR15 and PDR5 between ume6∆ pRS313 and ume6∆rpd3∆ pRS313 (p > 0.05), no upregulation of PDR15 and PDR5 mRNA levels was detected in ume6∆rpd3∆ pRS313-UME6 compared to ume6∆rpd3∆ pRS313 (p > 0.05) (Figure 3).Similarly, there was no significant difference in the mRNA levels of PDR15 and PDR5 between ume6∆ pRS313 and ume6∆pdr3∆ pRS313 (p > 0.05), while the mRNA levels of PDR15 and PDR5 in ume6∆pdr3∆ pRS313-UME6 were  To validate whether reduced mRNA levels of PDR15 and PDR5 in the ume6∆ mutant result from the deletion of UME6, we investigated whether decreased mRNA levels of PDR15 and PDR5 in the ume6∆ mutant are complemented with exogeneous expression of UME6.Although PDR15 and PDR5 mRNA levels were statistically significantly lower in ume6∆ pRS313 than in the wild-type strain (p < 0.01), the reductions in basal PDR15 and PDR5 mRNA levels in ume6∆ pRS313 were incompletely restored to wild-type pRS313 levels in ume6∆ pRS313-UME6 (p < 0.01) (Figure 3).The incomplete rescue by introduction of pRS313-UME6 into the ume6∆ mutant could result from nonphysiological levels of UME6 expression, because qRT-PCR revealed that UME6 is overexpressed at high levelsat least sevenfold-in ume6∆ pRS313-UME6 compared with wild-type pRS313.Regardless, this restoration of PDR15 and PDR5 mRNA levels suggests that the deletion of UME6 is a cause of reduced mRNA levels of PDR15 and PDR5 in the ume6∆ mutant.
Next, we investigated whether RPD3 and PDR3 are required for the partial rescue of reduced mRNA levels of PDR15 and PDR5 in the ume6∆ mutant by exogenous expression of UME6.Although there was no significant difference in the mRNA levels of PDR15 and PDR5 between ume6∆ pRS313 and ume6∆rpd3∆ pRS313 (p > 0.05), no upregulation of PDR15 and PDR5 mRNA levels was detected in ume6∆rpd3∆ pRS313-UME6 compared to ume6∆rpd3∆ pRS313 (p > 0.05) (Figure 3).Similarly, there was no significant difference in the mRNA levels of PDR15 and PDR5 between ume6∆ pRS313 and ume6∆pdr3∆ pRS313 (p > 0.05), while the mRNA levels of PDR15 and PDR5 in ume6∆pdr3∆ pRS313-UME6 were not statistically significantly higher than those in ume6∆pdr3∆ pRS313 (p > 0.05) (Figure 3).These results indicate that RPD3 and PDR3 are required for incomplete rescue of the reduction in PDR15 and PDR5 mRNA levels in the ume6∆ mutant by exogenous expression of UME6.
not statistically significantly higher than those in ume6∆pdr3∆ pRS313 (p > 0.05) (Figure 3).These results indicate that RPD3 and PDR3 are required for incomplete rescue of the reduction in PDR15 and PDR5 mRNA levels in the ume6∆ mutant by exogenous expression of UME6.

RPD3 and PDR3 Are Required for Incomplete Restoration of Susceptibility to Fluconazole and Cycloheximide in the ume6∆ Strain via Exogeneous UME6 Expression
To examine whether exogeneous UME6 expression from a plasmid with a low copy number can rescue the susceptibility of ume6∆ mutant ρ 0 cells to the PDR substrates fluconazole and cycloheximide, a spot dilution assay was carried out.We found that ume6∆ pRS313 displays greater susceptibility to fluconazole and cycloheximide than does wildtype pRS313 (Figure 4).In contrast, exogeneous UME6 expression in ume6∆ pRS313-UME6 incompletely rescued the susceptibility to fluconazole and cycloheximide in ume6∆ pRS313 (Figure 4).These results can be explained by the partial rescue of PDR15 and PDR5 reduction by exogenous UME6 expression in ume6∆ pRS313-UME6, as shown in Figure 3.This partial restoration of susceptibility to fluconazole and cycloheximide also suggests that the deletion of UME6 is a cause of high susceptibility to fluconazole and cycloheximide in the ume6∆ mutant.
We next examined whether RPD3 and PDR3 are required for incomplete restoration of high susceptibility to fluconazole and cycloheximide in the ume6∆ mutant by exogenous UME6 expression.In contrast to the case of ume6∆ pRS313 and ume6∆ pRS313-UME6, high susceptibility to fluconazole and cycloheximide in ume6∆rpd3∆ pRS313 and ume6∆pdr3∆ pRS313 was not complemented with exogenous UME6 expression in ume6∆rpd3∆

RPD3 and PDR3 Are Required for Incomplete Restoration of Susceptibility to Fluconazole and Cycloheximide in the ume6∆ Strain via Exogeneous UME6 Expression
To examine whether exogeneous UME6 expression from a plasmid with a low copy number can rescue the susceptibility of ume6∆ mutant ρ 0 cells to the PDR substrates fluconazole and cycloheximide, a spot dilution assay was carried out.We found that ume6∆ pRS313 displays greater susceptibility to fluconazole and cycloheximide than does wild-type pRS313 (Figure 4).In contrast, exogeneous UME6 expression in ume6∆ pRS313-UME6 incompletely rescued the susceptibility to fluconazole and cycloheximide in ume6∆ pRS313 (Figure 4).These results can be explained by the partial rescue of PDR15 and PDR5 reduction by exogenous UME6 expression in ume6∆ pRS313-UME6, as shown in Figure 3.This partial restoration of susceptibility to fluconazole and cycloheximide also suggests that the deletion of UME6 is a cause of high susceptibility to fluconazole and cycloheximide in the ume6∆ mutant.
We next examined whether RPD3 and PDR3 are required for incomplete restoration of high susceptibility to fluconazole and cycloheximide in the ume6∆ mutant by exogenous UME6 expression.In contrast to the case of ume6∆ pRS313 and ume6∆ pRS313-UME6, high susceptibility to fluconazole and cycloheximide in ume6∆rpd3∆ pRS313 and ume6∆pdr3∆ pRS313 was not complemented with exogenous UME6 expression in ume6∆rpd3∆ pRS313-UME6 and ume6∆pdr3∆ pRS313-UME6 (Figure 4).These results suggest that Ume6mediated activation of PDR in ρ 0 cells is dependent on RPD3 and PDR3.

Discussion
In this report, we showed that UME6 and RPD3 are required for upregulating the steady-state mRNA expression of SNQ2, PDR15, and PDR5 in S. cerevisiae ρ 0 cells.Although SNQ2, PDR15, PDR5, and PDR10 are activated by Pdr3 in ρ 0 cells [50], the mRNA levels of SNQ2, PDR15, and PDR5 but not that of PDR10 were increased by Ume6 and Rpd3 in ρ 0 cells (Figures 1 and 2).Thus, Ume6 and Rpd3 may bind to the promoter regions of SNQ2, PDR15, and PDR5 but not to the promoter region of PDR10 and increase the transcription of SNQ2, PDR15, and PDR5 via Pdr3 in ρ 0 cells.Although Ume6 is bound to the promoters of PDR5, PDR10, and YOR1 but not to that of SNQ2 or PDR15 in S. cerevisiae ρ + cells, it is unknown whether Ume6 is localised at the SNQ2, PDR15, and PDR5 promoters in ρ 0 cells [53,54].Alternatively, the upregulation of SNQ2, PDR15, and PDR5 mRNA expression mediated by Rpd3 and Ume6 in ρ 0 cells may be indirectly caused by changes in the expression of other genes.In addition, the expression of PDR10 was significantly lower in the ume6∆ strain but significantly greater in the rpd3∆ strain than in the wild-type strain (Figures 1 and 2).This may be because Rpd3 deacetylates proteins involved in the regulation of PDR10 independent of Ume6.
In an example similar to the transcriptional regulation of SNQ2, PDR15, and PDR5 by Ume6 and Rpd3 in ρ 0 cells, the histone chaperone Rtt106 specifically localises to the SNQ2, PDR15, and PDR5 promoters in a manner dependent on Pdr3 but not Pdr1 in S. cerevisiae ρ + cells [54].The histone chaperone Rtt106 is also essential for Pdr3-mediated basal expression of SNQ2, PDR15, and PDR5 in ρ + cells [54].In addition, PDR3 carrying the gain-of-function allele pdr3-7 in ρ + cells has fewer target genes than Pdr3 activated by retrograde signalling in ρ 0 cells [50].These findings suggest that other transcription factors act in concert with Pdr3 in ρ 0 cells.Therefore, the colocalisation of cooperative factors such as Ume6 and Rpd3 with Pdr3 at promoters may be required for the activation of target genes by Pdr3 in ρ 0 cells.
We also revealed that Ume6 activates PDR15 and PDR5 transcription and PDR via RPD3 and PDR3 in ρ 0 cells.Although it is currently unknown whether the histone deacetylase activity of Rpd3 is required for the activation of PDR15 and PDR5 transcription and PDR by retrograde signalling in ρ 0 cells, histone deacetylation by Rpd3 can lead to the transcriptional activation of DNA damage-inducible and osmoresponsive genes [43,55].Therefore, the histone deacetylase activity of Rpd3 recruited to the PDR15 and PDR5 promoters by Ume6 may activate the transcription of PDR15 and PDR5.Moreover, in an example similar to the dependency of UME6 on RPD3 and PDR3 for the activation

Discussion
In this report, we showed that UME6 and RPD3 are required for upregulating the steady-state mRNA expression of SNQ2, PDR15, and PDR5 in S. cerevisiae ρ 0 cells.Although SNQ2, PDR15, PDR5, and PDR10 are activated by Pdr3 in ρ 0 cells [50], the mRNA levels of SNQ2, PDR15, and PDR5 but not that of PDR10 were increased by Ume6 and Rpd3 in ρ 0 cells (Figures 1 and 2).Thus, Ume6 and Rpd3 may bind to the promoter regions of SNQ2, PDR15, and PDR5 but not to the promoter region of PDR10 and increase the transcription of SNQ2, PDR15, and PDR5 via Pdr3 in ρ 0 cells.Although Ume6 is bound to the promoters of PDR5, PDR10, and YOR1 but not to that of SNQ2 or PDR15 in S. cerevisiae ρ + cells, it is unknown whether Ume6 is localised at the SNQ2, PDR15, and PDR5 promoters in ρ 0 cells [53,54].Alternatively, the upregulation of SNQ2, PDR15, and PDR5 mRNA expression mediated by Rpd3 and Ume6 in ρ 0 cells may be indirectly caused by changes in the expression of other genes.In addition, the expression of PDR10 was significantly lower in the ume6∆ strain but significantly greater in the rpd3∆ strain than in the wild-type strain (Figures 1 and 2).This may be because Rpd3 deacetylates proteins involved in the regulation of PDR10 independent of Ume6.
In an example similar to the transcriptional regulation of SNQ2, PDR15, and PDR5 by Ume6 and Rpd3 in ρ 0 cells, the histone chaperone Rtt106 specifically localises to the SNQ2, PDR15, and PDR5 promoters in a manner dependent on Pdr3 but not Pdr1 in S. cerevisiae ρ + cells [54].The histone chaperone Rtt106 is also essential for Pdr3-mediated basal expression of SNQ2, PDR15, and PDR5 in ρ + cells [54].In addition, PDR3 carrying the gain-of-function allele pdr3-7 in ρ + cells has fewer target genes than Pdr3 activated by retrograde signalling in ρ 0 cells [50].These findings suggest that other transcription factors act in concert with Pdr3 in ρ 0 cells.Therefore, the colocalisation of cooperative factors such as Ume6 and Rpd3 with Pdr3 at promoters may be required for the activation of target genes by Pdr3 in ρ 0 cells.
We also revealed that Ume6 activates PDR15 and PDR5 transcription and PDR via RPD3 and PDR3 in ρ 0 cells.Although it is currently unknown whether the histone deacetylase activity of Rpd3 is required for the activation of PDR15 and PDR5 transcription and PDR by retrograde signalling in ρ 0 cells, histone deacetylation by Rpd3 can lead to the transcriptional activation of DNA damage-inducible and osmoresponsive genes [43,55].Therefore, the histone deacetylase activity of Rpd3 recruited to the PDR15 and PDR5 promoters by Ume6 may activate the transcription of PDR15 and PDR5.Moreover, in an example similar to the dependency of UME6 on RPD3 and PDR3 for the activation of PDR15 and PDR5 transcription and PDR in ρ 0 cells, PDR1 is required for the activation of PDR5 and YOR1 transcription by the C-terminal region of Zuo1 (Zuo1C) in ρ + cells [56].Zuo1, a ribosome-associated J protein, can positively regulate the transcription of PDR5 and YOR1, increasing the activity of the transcription factor Pdr1 when it is not bound to ribosomes [56].Exogenous expression of Zuo1C increases the expression of reporter genes driven by the PDR5 or YOR1 promoter (PDR5-lacZ and YOR1-lacZ) in the order of 10-and 4-fold, respectively, in both wild-type cells and pdr3∆ cells [56].In contrast, no activation of PDR5-lacZ or YOR1-lacZ was observed in pdr1∆ cells expressing Zuo1C [56].In addition, similar levels of drug resistance were observed in wild-type and pdr3∆ cells expressing Zuo1C, whereas no drug resistance in the presence of the drug was observed in pdr1∆ cells [56].Thus, Prunuske et al. concluded that Zuo1-mediated activation of PDR5, YOR1, and PDR is dependent on Pdr1 [56].In addition, Hallstrom et al. investigated the dependency of PDR13 on PDR1 for PDR13-mediated induction of cycloheximide resistance in ρ + cells using a spot assay [57].Hallstrom et al. concluded that PDR1 is required for the induction of cycloheximide resistance by PDR13 in ρ + cells [57].These reports support the relevance of our approach in examining the dependency of UME6 on RPD3 and PDR3 for the activation of PDR15 and PDR5 transcription and PDR.
In addition, how Ume6 and Rpd3 are involved in the activation of PDR15 and PDR5 transcription and PDR by Pdr3 in ρ 0 cells is unknown.It has been reported that the Rpd3 complex is required for the normal function of the transcriptional activator Upc2 and its stable binding to the promoter of the anaerobic gene DAN1 [44].Therefore, Ume6 and Rpd3 may also localise to the PDR15 and PDR5 promoters in ρ 0 cells and facilitate the normal function of Pdr3 and its stable binding to these promoter regions (Figure 5).
of PDR15 and PDR5 transcription and PDR in ρ 0 cells, PDR1 is required for the activation of PDR5 and YOR1 transcription by the C-terminal region of Zuo1 (Zuo1C) in ρ + cells [56].Zuo1, a ribosome-associated J protein, can positively regulate the transcription of PDR5 and YOR1, increasing the activity of the transcription factor Pdr1 when it is not bound to ribosomes [56].Exogenous expression of Zuo1C increases the expression of reporter genes driven by the PDR5 or YOR1 promoter (PDR5-lacZ and YOR1-lacZ) in the order of 10and 4-fold, respectively, in both wild-type cells and pdr3Δ cells [56].In contrast, no activation of PDR5-lacZ or YOR1-lacZ was observed in pdr1Δ cells expressing Zuo1C [56].In addition, similar levels of drug resistance were observed in wild-type and pdr3Δ cells expressing Zuo1C, whereas no drug resistance in the presence of the drug was observed in pdr1Δ cells [56].Thus, Prunuske et al. concluded that Zuo1-mediated activation of PDR5, YOR1, and PDR is dependent on Pdr1 [56].In addition, Hallstrom et al. investigated the dependency of PDR13 on PDR1 for PDR13-mediated induction of cycloheximide resistance in ρ + cells using a spot assay [57].Hallstrom et al. concluded that PDR1 is required for the induction of cycloheximide resistance by PDR13 in ρ + cells [57].These reports support the relevance of our approach in examining the dependency of UME6 on RPD3 and PDR3 for the activation of PDR15 and PDR5 transcription and PDR.
In addition, how Ume6 and Rpd3 are involved in the activation of PDR15 and PDR5 transcription and PDR by Pdr3 in ρ 0 cells is unknown.It has been reported that the Rpd3 complex is required for the normal function of the transcriptional activator Upc2 and its stable binding to the promoter of the anaerobic gene DAN1 [44].Therefore, Ume6 and Rpd3 may also localise to the PDR15 and PDR5 promoters in ρ 0 cells and facilitate the normal function of Pdr3 and its stable binding to these promoter regions (Figure 5).Currently, the relationships among the molecules required for the activation of ABC transporter genes in ρ 0 cells are unknown.We revealed that UME6 activates basal PDR15 and PDR5 transcription and PDR in a manner dependent on RPD3 and PDR3 in ρ 0 cells (Figures 3 and 4).In contrast, we previously showed that UME6 suppresses basal PDR5 expression and PDR in ρ + cells [46].Med12 in the L-Mediator complex also contributes to Currently, the relationships among the molecules required for the activation of ABC transporter genes in ρ 0 cells are unknown.We revealed that UME6 activates basal PDR15 and PDR5 transcription and PDR in a manner dependent on RPD3 and PDR3 in ρ 0 cells (Figures 3 and 4).In contrast, we previously showed that UME6 suppresses basal PDR5 expression and PDR in ρ + cells [46].Med12 in the L-Mediator complex also contributes to the induction of PDR5 expression in ρ 0 cells but not in ρ + cells.The transcriptional mediator complex serves as the interface between gene-specific transcription factors and the RNA polymerase II machinery [58].The L-Mediator complex in S. cerevisiae contains the core mediator complex and the Cdk8 subcomplex.The Cdk8 subcomplex is composed of Med12/Srb8, Med13/Srb9, the cyclin-dependent kinase Cdk8/Srb10, and cyclin C/Srb11 [59].Pdr1 and Pdr3 can bind to the KIX domain of a mediator subunit called Med15/Gal11 of the L-Mediator complex [32].Loss of Med12 from the Cdk8 complex completely suppresses the induction of PDR5 expression in ρ 0 cells but not in ρ + cells [29].In addition, Lge1 is required for proper PDR5 induction in ρ 0 cells but not in ρ + cells, independent of its role in histone H2B ubiquitination [31].These results indicate a difference in the regulatory machinery of PDR5 transcription between ρ + and ρ 0 cells.Thus, the identification of all molecules specifically required for the activation of ABC transporter genes and PDR in ρ 0/− cells is needed to reveal the relationships among these molecules.
C. albicans and C. glabrata are the two most common yeast pathogens in humans [60,61].S. cerevisiae is phylogenetically closer to C. glabrata than to C. albicans [62].Loss of the mitochondrial genome also leads to increased PDR in C. glabrata [63].However, C. albicans cannot survive the loss of mitochondrial DNA and therefore is petite negative [63].The molecules involved in PDR pathway activation in S. cerevisiae ρ 0 cells are parallel to those in C. glabrata ρ 0 cells.For example, the Pdr3 homologue CgPdr1 in C. glabrata, upon compromise of mitochondrial function, upregulates the expression of CDR1 and CDR2, the homologues of S. cerevisiae PDR5 [64].In addition, in C. glabrata ρ + cells, the loss of the Rpd3 orthologue CgRpd3 increases susceptibility to caspofungin at high concentrations [65].However, no mechanistic explanation-for example, gene targets or changes in histone modifications-has been provided for the caspofungin hypersensitive phenotype.Thus, the Ume6 orthologue Zcf11 in C. glabrata may also be responsible for multidrug resistance via transcriptional regulation of efflux genes.Therefore, identifying specific inhibitors of Zcf11 may lead to the development of drugs with activity against the multidrug-resistant pathogen C. glabrata.

Conclusions
We previously showed that RPD3 and UME6 are required for the activation of PDR5 and PDR in ρ 0 cells.This study investigated the dependence of Ume6 on Rpd3 and Pdr3 in basal transcription of the ABC transporters, including PDR5, and PDR in S. cerevisiae ρ 0 cells.Using a real-time PCR, RPD3 and UME6 were responsible for the activated basal expression of the ABC transporter-encoding genes SNQ2, PDR15, and PDR5 in S. cerevisiae ρ 0 cells.Furthermore, Ume6 increased the basal expression of PDR5 and PDR15, and induced PDR in a manner dependent on RPD3 and PDR3 in ρ 0 cells.This work may contribute to an elucidation of the relationships between molecules required for the activation of the ABC transporter genes in ρ 0/− cells.
Are Required for the Upregulation of the Steady-State mRNA Levels of SNQ2, PDR15, and PDR5 in ρ 0 Cells

Figure 1 .
Figure 1.Transcription levels of ABC transporters in ρ 0 cells of the wild-type and rpd3∆ strains in the logarithmic growth phase.Relative YOR1, SNQ2, PDR15, PDR10, and PDR5 mRNA levels in ρ 0 cells of the wild-type and ume6∆ strains in the logarithmic growth phase were determined by qRT-PCR.One asterisk (*) or two asterisks (**) indicate p values less than 0.05 or 0.01, respectively.

Figure 1 .
Figure 1.Transcription levels of ABC transporters in ρ 0 cells of the wild-type and rpd3∆ strains in the logarithmic growth phase.Relative YOR1, SNQ2, PDR15, PDR10, and PDR5 mRNA levels in ρ 0 cells of the wild-type and ume6∆ strains in the logarithmic growth phase were determined by qRT-PCR.One asterisk (*) or two asterisks (**) indicate p values less than 0.05 or 0.01, respectively.

Figure 2 .
Figure 2. Transcription levels of ABC transporters in ρ 0 cells of the wild-type and rpd3∆ strains in the logarithmic growth phase.The relative transcription levels of YOR1, SNQ2, PDR15, PDR10, and PDR5 were determined in ρ 0 cells of the wild-type and rpd3∆ strains in the logarithmic growth phase by qRT-PCR.One asterisk (*) or two asterisks (**) indicate p values less than 0.05 or 0.01, respectively.

Figure 2 .
Figure 2. Transcription levels of ABC transporters in ρ 0 cells of the wild-type and rpd3∆ strains in the logarithmic growth phase.The relative transcription levels of YOR1, SNQ2, PDR15, PDR10, and PDR5 were determined in ρ 0 cells of the wild-type and rpd3∆ strains in the logarithmic growth phase by qRT-PCR.One asterisk (*) or two asterisks (**) indicate p values less than 0.05 or 0.01, respectively.

3. 2 .
RPD3 and PDR3 Are Required for the Partial Rescue of the Reduction in PDR15 and PDR5 Expression in the ume6∆ Strain by Exogeneous UME6 Expression

Figure 4 .
Figure 4. RPD3 and PDR3 are required for partial complementation of sensitivity to fluconazole and cycloheximide by exogenous UME6 expression in ρ 0 cells of the ume6∆ mutant.Fluconazole or cycloheximide resistance in the ρ 0 cells of the wild-type, ume6∆, ume6∆rpd3∆, and ume6∆pdr3∆ strains with an empty pRS313 plasmid or a pRS313-UME6 plasmid was determined by the spot dilution assay.

Figure 4 .
Figure 4. RPD3 and PDR3 are required for partial complementation of sensitivity to fluconazole and cycloheximide by exogenous UME6 expression in ρ 0 cells of the ume6∆ mutant.Fluconazole or cycloheximide resistance in the ρ 0 cells of the wild-type, ume6∆, ume6∆rpd3∆, and ume6∆pdr3∆ strains with an empty pRS313 plasmid or a pRS313-UME6 plasmid was determined by the spot dilution assay.

Figure 5 .
Figure 5. Model of the correlation among Ume6, Rpd3, and Pdr3 and the activation of SNQ2, PDR15, and PDR5 in ρ 0 cells.In this model, Ume6 recruits Rpd3 to the SNQ2, PDR15, and PDR5 promoter regions, and the histone deacetylase activity of Rpd3 facilitates the binding of Pdr3 to these promoter regions.

Figure 5 .
Figure 5. Model of the correlation among Ume6, Rpd3, and Pdr3 and the activation of SNQ2, PDR15, and PDR5 in ρ 0 cells.In this model, Ume6 recruits Rpd3 to the SNQ2, PDR15, and PDR5 promoter regions, and the histone deacetylase activity of Rpd3 facilitates the binding of Pdr3 to these promoter regions.