RGD1, encoding a RhoGAP involved in low-pH survival, is an Msn2p/Msn4p regulated gene in Saccharomyces cerevisiae

Gene 351 (2005) 159 – 169 www.elsevier.com/locate/gene RGD1, encoding a RhoGAP involved in low-pH survival, is an Msn2p/Msn4p regulated gene in Saccharomyces cerevisiae Xavier Gatti, Geoffroy de Bettignies, Sandra Claret, François Doignon, Marc Crouzet, Didier Thoraval* Laboratoire de Biologie Moléculaire et de Séquençage, Institut de Biochimie et Génétique Cellulaires, UMR Université Victor Segalen Bordeaux 2-CNRS 5095, Box 64, 146 rue Léo Saignat, Bordeaux Cedex 33076, France Received 12 July 2004; received in revised form 14 February 2005; accepted 22 March 2005 Received by B. Dujon Abstract The RhoGAP Rgd1p is involved in different signal transduction pathways in Saccharomyces cerevisiae through its regulatory activity upon the Rho3 and Rho4 GTPases. The rgd1D mutant, which presents a mortality at the entry into the stationary phase in minimal medium, is sensitive to medium acidification caused by biomass augmentation. We showed that low-pH shock leads to abnormal intracellular acidification of the rgd1D mutant. Transcriptional regulation of RGD1 was studied in several stress conditions and we observed an activation of RGD1 transcription at low pH and after heat and oxidative shocks. The transcription level at low pH and after heat shock was demonstrated to depend on the STRE box located in the RGD1 promoter. The general stress-activated transcription factors Msn2p and Msn4p as well as the HOG pathway were shown to mainly act on the basal RGD1 transcriptional level in normal and stress conditions. D 2005 Elsevier B.V. All rights reserved. Keywords: Stress; STRE; Transcription; Acidic shock 1. Introduction All living cells encounter environmental changes that require adaptation of metabolism and physiology. The ability to respond rapidly to fluctuations in temperature, nutrient availability, and other medium variations is important for competitive fitness and cell survival. Significant clues to the mechanisms involved in adaptation to environment have come from studies of genes that are expressed in response to specific stresses. In the yeast Abbreviations: cAMP, cyclic AMP; HOG, high osmolarity glycerol; HSE, heat shock element; NLS, nuclear localisation sequence; MAP, mitogen-activating protein; ONPG, o-nitrophenyl-hd-galactopyranoside; PCR, polymerase chain reaction; pHi, intracellular pH; PKC, protein kinase C; Rho, Ras homologous; RhoGAP, Rho GTPase-activating protein; STRE, stress response element; TOR, target of rapamycin. * Corresponding author. Tel.: +33 5 57 57 47 20; fax: +33 5 57 57 47 19. E-mail address: thoraval@u-bordeaux2.fr (D. Thoraval). 0378-1119/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2005.03.034 Saccharomyces cerevisiae, study of the temporal expression profiles in various stress conditions demonstrated that more than half of the genome was regulated by environmental changes and allowed the identification of a global set of genes regulated by each stress condition (Gasch et al., 2000; Causton et al., 2001). Transcriptional activators and repressors contributing to coordinate remodelling of genome expression were identified. For example, the yeast transcriptional factor Hsf1p and the cis-acting sequence HSE (heat shock element) mediate the response to heat shock (Sorger, 1991), and the Yap1p factor mediates the response to oxidative stress (Toone and Jones, 1999). Another pathway responding to a wide variety of stress conditions is the so-called general stress response pathway (Martinez-Pastor et al., 1996; Schmitt and McEntee, 1996). Msn2p and Msn4p transcription factors are known to mediate this response through binding to the cis element STRE (STress Response Element), of which the core consensus sequence is CCCCT or AGGGG (Marchler et 160 X. Gatti et al. / Gene 351 (2005) 159 – 169 al., 1993; Martinez-Pastor et al., 1996; Schmitt and McEntee, 1996). These two proteins, normally present in the cytoplasm, are transported into the nucleus upon stress exposure (Gorner et al., 1998). The activity of Msn2p/ Msn4p during stress is regulated by at least two independent pathways: the cAMP – PKA pathway downregulates the nuclear localisation of Msn2p and Msn4p by phosphorylating the NLS sequence (Gorner et al., 2002), and the TOR pathway downregulates these transcription factors by allowing sequestration of Msn2/Msn4 in the cytosol via Bmh2p (Beck and Hall, 1999). In addition, Reiser et al. (1999) suggested that the HOG pathway may act positively by sequestrating Msn2p and Msn4p in the nucleus. The general stress response pathway is activated under heat shock, osmotic stress, acidic shock, oxidative stress, glucose starvation, ethanol, DNA damage, and others (Schuller et al., 1994). In our laboratory, we focused our research on the RhoGAP Rgd1p and associated biological processes in S. cerevisiae (Barthe et al., 1998; de Bettignies et al., 1999). We have demonstrated its negative regulatory action on the small GTPases Rho3p and Rho4p (Doignon et al., 1999), implicated in the actin cytoskeleton organization (Matsui and Toh, 1992; Imai et al., 1996) and in exocytosis (Robinson et al., 1999). Moreover, using different genetic screens, two pathways have shown close relationships with the RhoGAP Rgd1p encoding gene: the cell integrity PKC pathway (de Bettignies et al., 2001) and the actin assembly machinery (Roumanie et al., 2000; Roumanie et al., 2002). Physical and genetic partners of Rgd1p have also been identified by global approaches. Proteins implicated in cell polarity, such as Arp2p and Bni4p, were isolated by a synthetic lethality screening (Tong et al., 2004), and Mkk2p, one of the MAP kinase kinases of the PKC pathway, was identified by protein complexes purification (Ho et al., 2002). In S. cerevisiae, Bem2p, one RhoGAP of the GTPase Rho1p, has been involved in stress response: cells lacking Bem2p are hypersensitive to heat shock and ethanol (Takahashi et al., 2001). In this study, we wonder whether the RGD1 gene is also involved in stress response, since an STRE sequence has been found in its promoter. To confirm this hypothesis, we present here a phenotypical study of the rgd1D mutant in various stress conditions, showing the sensitivity of rgd1D cells to acidic conditions. Then, using a lacZ fusion reporter gene, we analyzed the transcription profile of RGD1 in stress conditions, and we identified cis and trans elements responsible for the regulation of RGD1 transcription in these conditions. 2. Materials and methods 2.1. Materials and strains The XL1Blue Escherichia coli strain (Stratagene, CA) was used for cloning. Experiments were mainly performed in the yeast BY4742 background used by the EUROSCARF project (http://www.uni-frankfurt.de/fb15/mikro/ euroscarf/), or in the congenic X2180 background (see Table 1). The msn2D msn4D double mutant has been obtained after crossing msn2D with msn4D strain. The msn2D msn4D hog1D triple mutant was constructed from the msn2D msn4D double mutant using PCR-generated HIS3MX6 disruption cassette (Wach et al., 1997). To this aim, the HIS3MX6 gene was flanked with DNA sequences homologous to the HOG1 promoter ( 27 to + 3 from the ATG) and terminator ( 3 to + 27 from the STOP codon) regions. HOG1 disruption was confirmed by PCR and by sensitivity to high osmolarity, due to HOG1 inactivation (Brewster et al., 1993). Growth temperature was 30 -C for yeast strains except for heat shock experiments. Liquid culture was performed in YPD (1% Bacto yeast extract, 2% Bacto peptone, and 2% glucose) or in synthetic minimal YNB (0.67% yeast nitrogen base without a.a. (Difco), 2% glucose) supplemented with bases and amino acids when necessary. The buffered YNB was prepared by adding to YNB, 1% succinic acid, and 0.6% NaOH; the final pH of this medium is 5.6 after autoclaving. Strain lethality was determined by methylene blue staining and dead blue cells were counted by microscopic examination from at least 300 cells (Rose, 1975; de Bettignies et al., 2001). Yeasts were transformed by the method of Gietz et al. (1995). Table 1 List of yeast strains used in this study Strain Relevant genotype Source X2180-1A LBG3-1B BY4742 msn2D msn4D msn2D msn4D hog1D pbs2D msn2D msn4D hog1D rgd1D MATa MATa MATa MATa MATa MATa MATa MATa MATa MATa YGSC de Bettignies et al., (1999) Euroscarf Euroscarf Euroscarf This work Euroscarf Euroscarf This work Euroscarf SUC2 SUC2 his3D his3D his3D his3D his3D his3D his3D his3D mal mel gal2 CUP1 mal mel gal2 CUP1 his3-11,15 rgd10HIS3 leu2D lys2D ura3D leu2D lys2D ura3D msn20KanMX4 leu2D lys2D ura3D msn40KanMX4 leu2D lys2D ura3D msn20KanMX4 msn40KanMX4 leu2D lys2D ura3D hog10KanMX4 leu2D lys2D ura3D pbs20KanMX4 leu2D lys2D ura3D msn20KanMX4 msn40KanMX4 hog10HIS3 leu2D lys2D ura3D rgd10KanMX4 X. Gatti et al. / Gene 351 (2005) 159 – 169 161 2.2. Stress conditions 2.5. Mutagenesis of the RGD1 promoter All stresses were performed on a yeast culture at 0.5 OD600 nm with a medium pH of 3.7. The culture in minimal medium (150 ml) was split: one half was maintained in standard growth condition as control, and the other half was submitted to stress. Acidic shock was performed by adding hydrochloric acid until the pH of the liquid medium was 2.8. Heat shock was performed by transferring the culture from 21 -C to 39 -C. Oxidative shock was performed by adding hydrogen peroxide to a final concentration of 0.3 mM, which was just below the growth-inhibitory concentration. Osmotic shock was performed by adding sodium chloride to a final concentration of 0.5 M. Alkali shock was done by adding sodium hydroxide in the medium to reach a final pH of 7.5. In order to replace the STRE (CCCCT) sequence in the RGD1 promoter by a neutral sequence (AAGCA), we performed a directed mutagenesis by a two-step PCR approach. In a first step, two PCRs were done using complementary oligonucleotides containing the mutated STRE sequence and two oligonucleotides complementary of the extremities of the promoter region as primers. The two PCR products were used as matrix in the second step PCR, as well as the two outside primers in order to amplify the full promoter containing the neutral sequence. This mutation was further verified by sequencing. The mutated promoter was introduced into the YEp357R vector as described above and the resulting vector was named YEp/ PRGD1*-LacZ. 2.3. Estimation of the intracellular pH 2.6. b-galactosidase liquid assay About 8  109 cells were harvested by centrifugation and resuspended into 200 Al of MilliQ water. Glass beads (diameter 0.45 mm; Biospec Products) were added up to the top of the liquid phase. The tube was shaken three times during 30 s with Mini-Beadbeater (Biospec Products) and the cell lysate was recovered. The beads were then washed with water and the liquid was collected; this last step was repeated three times to reach a final volume of 2 ml. Cell debris were removed by centrifugation and the pH of the supernatant was determined with a pH meter (Hanna Instruments). 2.4. Plasmids General molecular biological protocols were from Sambrook et al. (1989). DNA amplifications were performed using the Pwo DNA polymerase (Roche Molecular Biochemicals), according to the manufacturer’s instructions, using genomic yeast DNA as template. The plasmid YEp357R (Myers et al., 1986) was used to follow gene expression: it contains the lacZ sequence without its ATG nor its promoter sequence. RGD1 promoter (262 nucleotides upstream from the ATG) and the first 45 nucleotides of RGD1 coding sequence were first amplified by PCR from genomic DNA, with addition of one endonuclease cleavage site at each end. The EcoRI site was used to fuse the RGD1 coding sequence in-frame with the lacZ gene encoding the hgalactosidase. The resulting plasmid was named YEp/ PRGD1-LacZ. In the same way, for control experiments, ADE1 promoter ( 500 nt to + 45 nt from ATG) was amplified from genomic DNA and inserted into YEp357R to give the YEp-PADE1-lacZ plasmid. The ADE1 housekeeping gene encodes an enzyme implicated in purine synthesis. pGR213 is a low copy plasmid containing the MSN2 gene fused to yEGFP3 at its 5V end, under the control of MET25 promoter, and pGR247 is a low copy plasmid containing the MSN4 gene fused to EGFP at its 3V end, under the control of ADH1 promoter (Jacquet et al., 2003). Cells containing the plasmid with lacZ reporter gene were grown in selective synthetic medium overnight to an OD600 nm of 0.5, and aliquots of the culture were harvested when notified. For each assay, 20 ml of culture medium was collected. After a 10,000  g centrifugation for 5 min, cell pellets washed with water were stored at 20 -C until transcriptional activity measurement. The pellets were resuspended with 100 Al of 0.1 M phosphate buffer (Na2HPO4/NaH2PO4, pH 7.0), and cells were disrupted by mechanical shaking (Mini-Beadbeater; Biospec Products). After a 10,000  g centrifugation for 3 min, the protein concentration in the supernatant was determined by a Bradford assay, and 10 Ag of total protein was incubated in 1 ml of 0.1 M phosphate buffer containing 0.8 mg of ONPG (o-nitrophenyl-hd-galactopyranoside). After 20 min of incubation, the reaction was stopped by addition of 500 Al of 1 M Na2CO3, and h-galactosidase activity was measured with the spectrophotometer at 420 nm. Results are presented in nanomoles per minute per milligram of proteins, with 4.5  10 3 OD420 nm units representing 1 nmol of ONPG cleaved (Casadaban et al., 1983). We verified that with 10 Ag of protein extract, ortho-nitrophenol production by h-galactosidase was linear during 20 min. hGalactosidase activity was measured for each point in triplicate. For each condition, three independent transformants were tested and variation was always <20%. Figures present the results obtained from one representative clone. 3. Results 3.1. Acidification of synthetic medium induces lethality of rgd1D cells When we explored the role of the RGD1 gene, we first observed a lethality of the rgd1D mutant at the entry of the 162 X. Gatti et al. / Gene 351 (2005) 159 – 169 stationary growth phase. This phenomenon is specific to synthetic medium and does not occur in YPD-rich medium (Barthe et al., 1998; de Bettignies et al., 1999). We then tried to determine which element of the minimal medium was responsible for this peculiar lethality. Considering the differences of composition between the rich and synthetic media, the amounts of nitrogen and carbon sources, vitamins, salts, and oligo-elements were tested for their implication in the mortality of rgd1D cells. After testing these various components, we found that the component concentrations could act on the rgd1D cell mortality through their buffering effect. Indeed, the pH of the medium was a critical factor for the viability of rgd1D cells at the entry of stationary phase. Whereas rich YPD medium had a stable pH value of approximately 6 during yeast growth, the pH of synthetic medium was about 5.7 at the beginning of the culture, and this value continually decreased to reach 2.6 with biomass augmentation. In Fig. 1, the growth curves and percentages of lethality of wild type X2180 and rgd1D cells cultivated in synthetic medium and in pH 5.6-buffered synthetic medium are shown. Whereas up to 20% of rgd1D cells died in synthetic medium, only 5% of mutant cells died A X2180 synthetic medium rgd1∆ synthetic medium X2180 buffered medium rgd1∆ buffered medium Growth X2180 synthetic medium rgd1∆ synthetic medium X2180 buffered medium rgd1∆ buffered medium Lethality 30 10 25 OD600nm 20 15 0.1 10 Lethality (%) 1 0.01 5 0.001 0 0 10 20 30 40 50 60 70 Time (h) B 6 pH 5 4 synthetic medium buffered medium 3 2 0 10 20 30 40 Time (h) 50 60 70 Fig. 1. Growth and lethality of wild type and rgd1D strains. (A) Wild type (X2180-1A) and rgd1D (LBG3-1B) strains were cultivated at 30 -C either in synthetic medium or in synthetic medium buffered at pH 5.6 with succinate and sodium hydroxide. Growth was followed by measuring optical density at 600 nm and the percentage of dead cells was determined after methylene blue staining. (B) Follow up of the pH value of synthetic medium buffered or not during growth. in buffered medium, which was comparable to the value obtained with wild type cells. In the BY4742 background, the same phenomenon, although less pronounced, was also observed. Using several synthetic media at different initial pH values (from 6.0 to 3.0) and by following cell mortality and pH variation, we have demonstrated that the rgd1D lethality began when pH was below 3.2/3.1 (data not shown). 3.2. Low-pH shock is detrimental to rgd1D mutant and acidifies the intracellular medium Lethality of the rgd1D mutant cells in synthetic medium appeared at the entry of the stationary phase. This growth phase is characterized by a lot of physiological changes (Werner-Washburne et al., 1996), like modification of the expression pattern of a great majority of genes, or remodelling of cell wall components. We wanted to study the effect of medium acidification, irrespective of yeast physiological modifications associated to the entry into the stationary growth phase. For this purpose, we developed a test consisting in a rapid acidification (acidic shock) of the medium containing exponentially growing cells. The cells (OD600 nm of 0.5) were submitted to an acidic shock by adding hydrochloric acid in the culture medium to reach a pH of 2.8, a pH value beyond the threshold of 3.1, for which rgd1D cells began to die. Cell lethality was followed during several hours after the shock by methylene blue staining. As shown in Fig. 2, lethality appeared in the mutant rgd1D cells within the hour following the shock to reach about 45%, 4 h after the shock. In the same experiment, the acidic shock was without effect on cell lethality of wild type strain. In standard growth conditions, no rgd1D cell lethality was observed. In this case, the pH of the synthetic medium did not reach the pH threshold for cell mortality. In addition, growing rgd1D cells in rich YPD medium adjusted to pH 2.8 also led to cell mortality, indicating that the effect of low pH is independent of medium composition. Therefore a too low pH of the medium was detrimental to rgd1D cells. Medium acidification induces cell wall modifications (Kapteyn et al., 2001) and, in rgd1D cells, this acidification could lead to an intracellular acidification due to defective cell wall and/or plasma membrane, which is detrimental for cell survival. Thus, we investigated whether the intracellular pH of rgd1D cells was affected by extracellular acidification. We therefore estimated the intracellular pH of mutant and wild type cells submitted to acidic shock or not (see Materials and Methods for details). Results are presented in Table 2. Whereas the pH of acellular extract of wild type and rgd1D cells without shock was comparable, it appeared that after acidic shock, the pH measured with this method was significantly lower (about 0.2 pH unit) in rgd1D cells than in the wild type cells treated in the same conditions. These data argue in favour of a default in the maintenance of appropriate intracellular pH in rgd1D cells. 163 X. Gatti et al. / Gene 351 (2005) 159 – 169 A BY4742 with acidic shock rgd1∆ without acidic shock rgd1∆ with acidic shock Growth BY4742 with acidic shock rgd1∆ without acidic shock rgd1∆ with acidic shock Lethality 10 100 75 OD600nm 50 0.1 Lethality (%) 1 25 0.01 0 -1 0 1 2 3 4 Time (h) after acidic shock B pH 4 with acidic shock without acidic shock 3 2 -1 0 1 2 3 Time (h) after acidic shock 4 Fig. 2. Growth and lethality of wild type and rgd1D strains after an acidic shock. (A) Wild type and rgd1D strains (BY4742 background) were cultivated in synthetic medium. At an OD600 nm of 0.5, half of the culture was submitted to acidic shock by adding hydrochloric acid to reach a pH of 2.8; the remaining half was kept in normal culture conditions. Growth and percentage of lethality were followed as described in Materials and methods. For clarity of figure, data from unshocked wild type cells were not shown; wild type cells responded in the same way after acidic shock or not. (B) Follow-up of the pH value of cultures submitted to acidic shock or not. The plasma membrane H+ ATPase Pma1p is the major intracellular pH regulator of yeast (McCusker et al., 1987). In order to explore further this intracellular pH variation, we overexpressed the PMA1 gene in rgd1D cells to compensate the default in maintaining a proper pHi. Unfortunately, this overexpression remained without effect neither on the lethality nor in the pHi of rgd1D cells during acidic shock, suggesting the existence of other altered mechanisms involved in intracellular pH maintenance. Acidification of the medium as performed is a stress signal for the yeast cell, and we wondered whether the rgd1D mutant cells responded in the same way to other stresses, or whether the rgd1D behaviour was specific to acidic conditions. Therefore, heat, oxidative, osmotic, and alkaline shocks were also tested on rgd1D cells. All these stress conditions remained without apparent lethal effect on mutant cells (data not shown). Nevertheless, unlike the BY4742 background used in these experiments, we noticed that rgd1D in the X2180 background was sensitive to heat shock, with appearance of a slight mortality (de Bettignies et al., 1999). These phenotypical data in acidic stress show the important role of the RhoGAP Rgd1p in the low-pH response and the maintenance of intracellular pH during acidification of the medium, which is essential for cell viability. The role of Rgd1p seems to be shock-specific in BY4742 background considering the absence of other stress effects on the viability of rgd1D mutant cells. We then wondered if, in the culture conditions where the presence of Rgd1p was decisive (i.e., at the end of exponentially growing phase and after acidic shock), the expression level of the RGD1 gene was affected by pH variation in the medium. 3.3. RGD1 is a stress-induced gene In order to evaluate the RGD1 transcriptional level, we inserted into the plasmid YEp357R the RGD1 promoter region (262 bp upstream of the translation initiation codon) and the first 15 codons in frame with the lacZ coding sequence. More precisely, this promoter region covers the intergenic region between the ATG of RGD1 and the STOP codon of the contiguous YBR261C ORF (193 pb) and upstream. This construct was introduced into the BY4742 strain and the transcription level of RGD1 was monitored during growth in synthetic medium and in buffered synthetic medium by measuring the h-galactosidase activity (Fig. 3). We observed an increase of h-galactosidase activity by about 2.5-fold at the entry into stationary phase in the strain grown in synthetic medium, compared to values obtained from the buffered medium, indicating an activation of the RGD1-driven lacZ expression specific to nonbuffered medium. This particular transcriptional activation was not due to a general activation of transcription at this growth phase, as the transcription of an ADE1-lacZ reporter construct did not show such activation (data not shown). Activation of RGD1-driven lacZ expression followed the lethality profile of rgd1D cells and this expression was activated within the period where cell mortality appeared. As this cell lethality was pH-dependent, RGD1-driven lacZ Table 2 pH of acellular extract of wild type (BY4742) and rgd1D strains treated with or without hydrochloric acid pH of acellular extract Wild type without acidic shock Wild type with acidic shock rgd1D without acidic shock rgd1D with acidic shock 5.57 T 0.05 5.51 T 0.02 5.60 T 0.02 5.33 T 0.01 Cells were submitted to acidic shock or kept in standard culture conditions (without acidic shock), harvested after 1 h, and disrupted in Milli-Q water. Acellular extract was obtained after centrifugation of the lysate, and pH was measured. Data represent results from three independent experiments. 164 X. Gatti et al. / Gene 351 (2005) 159 – 169 0.1 0 0 10 20 30 40 50 60 Time (h) Growth in synthetic medium Growth in buffered medium β-galactosidase activity in synthetic medium β-galactosidase activity in buffered medium Fig. 3. RGD1 transcriptional activity during growth in buffered and nonbuffered media. The promoter region of RGD1 was PCR-amplified and introduced into the YEp357R plasmid, in frame with lacZ reporter gene. This YEp-PRGD1-LacZ plasmid was introduced into BY4742 strain. The transformed strain was cultivated either in synthetic medium or in synthetic medium buffered at pH 5.6 with succinate and sodium hydroxide. Growth was followed as described in Fig. 1. At each time point, an aliquot of the culture was taken to perform h-galactosidase activity assay (see Materials and methods for details). expression should also be linked to the medium pH variation. So RGD1 transcriptional activity was measured in synthetic medium after acidic shock. As shown in Fig. 4A, the RGD1-driven lacZ expression was activated up to fourfold within 1 h of treatment with hydrochloric acid, compared to cells at t 0, and this activation was enhanced 4 h after the shock. A slight and continuous increase of the RGD1-driven lacZ expression in the nonshocked cells, due to natural acidification of the synthetic medium, was observable. In parallel, the BY4742 strain with the reporter gene was also used to follow the RGD1driven lacZ expression after a heat shock and an oxidative shock (Fig. 4B and C, respectively). The lacZ expression was also activated in these two stress conditions. Unlike heat and acidic shock activations, the oxidative shock led to a response mainly visible during the first 2 h of treatment. After 4 h, we can observe that the expression level was similar in shocked and non-shocked cells. The decrease observed in treated cells might be the consequence of hgalactosidase enzyme destabilization caused by the oxidative agent in medium. During heat shock, a very slight activation of transcription was visible in the non-shocked control cells, probably due to a weaker natural acidification of medium linked to a slower yeast growth at 21 -C. 3.4. STRE box in RGD1 promoter is involved in transcriptional activity during normal and stress conditions In order to identify a cis element involved in these stress regulations, the RGD1 promoter sequence was first analysed in silico using the appropriate tools in S. cerevisiae β-galactosidase activity (nmoles/min/mg) 100 300 with acidic shock without acidic shock 250 200 150 100 50 0 0 B β-galactosidase activity (nmoles/min/mg) 200 A 1 2 4 Time (h) after acidic shock 350 with heat shock without heat shock 300 250 200 150 100 50 0 0 1 2 4 Time (h) after heat shock C β-galactosidase activity (nmoles/min/mg) 1 β-galactosidase activity (nmoles/min/mg) OD600nm 300 250 with oxidative shock without oxidative shock 200 150 100 50 0 0 1 2 4 Time (h) after oxidative shock Fig. 4. RGD1-driven lacZ expression after different stresses. The BY4742 strain transformed with YEp-PRGD1-LacZ was cultivated in synthetic medium at 30 -C (acidic shock, oxidative shock) or at 21 -C (before heat shock). At t 0, when optical density was of 0.5 and the pH was 3.7, the culture was divided in two, one of which was submitted to stress. Acidic stress was performed as described in Materials and methods, heat shock was performed by switching the culture from 21 -C to 39 -C, and oxidative stress was performed by adding hydrogen peroxide to a final concentration of 0.3 mM. Follow up of h-galactosidase activity during 4 h in treated and untreated cells after acidic shock (A), after heat shock (B), and after oxidative shock (C). At each time point, an aliquot of the culture was taken to perform h-galactosidase activity assay. 165 X. Gatti et al. / Gene 351 (2005) 159 – 169 Taken together, these results showed that the STRE box is an important cis element necessary for RGD1-driven lacZ expression by acting on its basal level. Consequently STRE mutation affects the transcriptional activity in stressed conditions even if it does not affect the induction factor. A yet to be identified cis element could be responsible for the specific stress-induced activation of RGD1 expression. promoter database (http://cgsigma.cshl.org/jian/). Beyond the TATA boxes, a STRE (STress Response Element) sequence was found at 161 bp upstream of the ATG (Fig. 5A). As this sequence is known to mediate the global stress response in yeast (Marchler et al., 1993), we have focused the analysis on this cis element. To investigate the role of the STRE sequence in the transcriptional activity of RGD1 during stress, we constructed a mutated promoter by replacing the core CCCCT sequence (STRE consensus) by a Fneutral_ sequence (AAGCA). This mutated promoter was inserted in frame with the lacZ gene into the YEp357R plasmid, leading to the YEp-PRGD1*/LacZ plasmid. This molecular construct was used to follow up h-galactosidase activity in the BY4742 strain after 2 h of treatment with acidic and heat shocks. As illustrated in Fig. 5B, we observed as before the strong effect of stress conditions on the RGD1 wild type promoter. Mutation of the STRE box resulted in a severe impairment in the level of RGD1-driven lacZ expression in standard and stress conditions. Indeed, when using the mutated RGD1 promoter, both basal and acid-induced expression fell to 16% of wild type level. Likewise, the level of RGD1-driven lacZ expression under control of the mutated promoter compared to that of wild type promoter was reduced to 27% upon heat shock and to 13% without heat shock. Similar results were obtained using a promoter region deleted for the STRE box (data not shown). A 3.5. Msn2p and Msn4p transcription factors are necessary for RGD1-driven lacZ expression Two transcription factors, Msn2p and Msn4p, are known to bind to STRE sequence and trigger activation during stress (Martinez-Pastor et al., 1996; Schmitt and McEntee, 1996) and have partially overlapping functions (Estruch and Carlson, 1993). To investigate the involvement of these transcription factors in the level of RGD1 expression during low pH stress, we have determined the expression of hgalactosidase under the control of RGD1 promoter in cells lacking Msn2p, Msn4p, or both proteins. As presented in Fig. 6A, basal transcription level in the msn2D, msn4D, and msn2D msn4D strains was altered in the absence of stress, showing a role of these factors in RGD1 expression. When cells were submitted to acidic shock, we also observed a lower activation, but it still remained a stress induction of the RGD1-driven lacZ expression, like that observed with the promoter mutated for the STRE box. RGD1-driven lacZ expression was Promoter region -196 YBR261c -99 -161 TATA STRE TATA +1 RGD1 TATA ATG STOP B β-galactosidase activity (nmoles/min/mg) RGD1 promoter 0 20 40 60 80 100 120 140 160 180 200 STRE ACGCCCCT TTT lacZ ACGAAGCATTT lacZ ACGCCCCT TTT lacZ ACGAAGCATTT lacZ Acidic shock with shock without shock Heat shock Fig. 5. Implication of the STRE box in activation of RGD1-driven lacZ expression during stress. (A) Localisation of TATA and STRE boxes in RGD1 promoter region, as revealed by using the appropriate tools at S. cerevisiae promoter database (http://cgsigma.cshl.org/jian/). (B) The STRE sequence (CCCCT) was replaced by a neutral sequence (AAGCA) in the RGD1 promoter by two-step PCR mutagenesis and the mutated promoter was inserted in YEp357R. The resulting YEp-PRGD1*-LacZ plasmid was introduced into strain BY4742. The resulting strain was cultivated in synthetic medium, and acidic and heat shocks were performed as described in Materials and methods. h-Galactosidase activity was measured 2 h after the shocks. 166 β-galactosidase activity (nmoles/min/mg) A X. Gatti et al. / Gene 351 (2005) 159 – 169 600 with acidic shock without acidic shock 500 400 300 200 100 0 BY4742 msn2∆ msn4∆ msn2∆msn4∆ β-galactosidase activity (nmoles/min/mg) B 2200 with acidic shock without acidic shock 2000 1800 600 400 200 0 BY4742 β-galactosidase activity (nmoles/min/mg) C BY4742 + MSN2 BY4742 + MSN4 600 500 with acidic shock without acidic shock 400 300 200 100 0 BY4742 pbs2∆ hog1∆ msn2∆msn4∆ msn2∆msn4∆hog1∆ Fig. 6. Implication of Msn2p, Msn4p, and the HOG pathway in the activation of RGD1-driven lacZ expression during acidic shock. (A) Effect of the msn2D and msn4D mutations on RGD1 transcription activation. The msn2D, msn4D and msn2D msn4D strains were transformed with YEpPRGD1-LacZ and cultivated in synthetic medium. Acidic shock was performed by adding hydrochloric acid to half of the culture medium. After 2 h, h-galactosidase activity was determined for treated and non-treated cells. (B) Transcription level of RGD1 in BY4742 strain overproducing Msn2p or Msn4p. The strain BY4742 containing YEp-PRGD1-LacZ was transformed either with pGR213, containing the MSN2 gene or pGR247 containing the MSN4 gene. These strains were cultivated in synthetic medium and acidic-shocked as in (A) and h-galactosidase activity was measured after 2 h. (C) Effect of the HOG pathway on the transcription of RGD1. The hog1D, pbs2D, msn2D msn4D, and msn2D msn4D hog1D strains transformed with YEp-PRGD1-LacZ plasmid were cultivated in synthetic medium and acidic-shocked as above. h-Galactosidase activity was measured 2 h after the acidic shock. particularly decreased in msn2D and msn2D msn4D strains, indicating a preponderant role of Msn2p with regard to Msn4p. In order to confirm the role of Msn2p and Msn4p on the transcription of RGD1, we finally tested the effect of the overexpression of these two transcriptional activators (Treger et al., 1998). To this aim, we transformed the BY4742 wild type strain with plasmids pGR213 or pGR247 overproducing Msn2p and Msn4p, respectively (Fig. 6B). We can see that, when overproduced, Msn2p and Msn4p enhanced the RGD1-driven lacZ expression, up to eightfold and twofold, respectively. These results confirmed the action of both transcription factors, with again a preponderant role of Msn2p over Msn4p in RGD1 expression. Msn2/4p-dependent transcription activation may be positively regulated by the HOG pathway (Schuller et al., 1994). This latter regulatory pathway comprises a MAP kinase cascade, with Hog1p as the MAP kinase and Pbs2p as the MAP kinase kinase. We verified that RGD1-driven lacZ expression was positively regulated by the HOG pathway, by measuring in the same way as above the level of h-galactosidase activity in the hog1D and pbs2D mutant strains with or without acidic shock (Fig. 6C). As for msn2D mutant, lacZ expression during acidic shock was strongly impaired in the hog1D and pbs2D cells, showing the positive role of the HOG pathway in the level of RGD1driven lacZ expression. As the Msn2/4p transcription factors and the Hog1p MAP kinase could act on RGD1-driven lacZ expression by two distinct pathways, we tested the transcriptional activity of RGD1 in the msn2D msn4D hog1D triple mutant. As shown in Fig. 6C, the effects of these different mutations were not cumulative as the level of RGD1-driven lacZ expression remained close to those measured in the msn2D msn4D and hog1D mutants. This result argues in favour of a common pathway including the Hog1p MAP kinase and the Msn2/4p transcription factors, leading to activation of RGD1-driven lacZ expression. Taken together, all these results demonstrated that Msn2p and Msn4p as well as the HOG pathway act as key regulators of the basal and acid-stressed expression levels of RGD1. 4. Discussion 4.1. Transcriptional profile of RGD1 During this work, we performed the analysis of RGD1 transcription during growth and different stress conditions. We demonstrated the stress inducibility of RGD1 transcription during acidic, heat, and oxidative shocks, and that the STRE sequence in RGD1 promoter plays a major role in the basal level of transcription of this gene. Furthermore, Msn2p and Msn4p, as well as the HOG pathway, have been demonstrated to act positively on RGD1 transcription X. Gatti et al. / Gene 351 (2005) 159 – 169 through a common pathway. The stress inducibility of the RGD1 gene and the implication of the STRE sequence in RGD1 transcription make it part of the numerous genes controlled by the general stress response. In two global studies, Treger et al. (1998) and Moskvina et al. (1998) searched for putative STRE-driven genes in the genome of S. cerevisiae. Treger et al. (1998) identified 186 genes with at least two STRE consensus sequences in their promoter and Moskvina et al. (1998) identified 55 genes containing clustered STRE (i.e., at least two STRE separated by less than 30 bases). The RGD1 gene was not identified during these two analyses because of the presence of only one STRE box in its promoter. Here, we can consider that RGD1 expression is STRE-regulated according to the three characteristics described by Treger et al. (1998): (i) the expression is induced during heat shock, (ii) stress-induced expression is greatly reduced in cells lacking Msn2p and Msn4p, and (iii) basal transcription is greatly enhanced when Msn2p is overproduced. In addition, at the entry into the stationary phase, RGD1 transcription is activated in response to medium acidification, even if part of this activation is pH-independent, as shown with h-galactosidase activity in buffered medium. Such an expression profile is found in genes regulated by STRE and Msn2/4 as the entry into the stationary phase leads to a fall in cAMP concentration, which in return induces STRE-driven expression (Marchler et al., 1993; Gorner et al., 2002). These last data are in agreement with an STRE-Msn2/4-dependent regulation of RGD1 expression. Gasch et al. (2000) performed a global analysis of stressinduced genes. In this study, RGD1 was not highlighted despite its clear activation pattern observed during stress in our experimental conditions. However, Causton et al. (2001) also did a systematic analysis of stress response among which response to low pH revealed a RGD1 activation by twofold following an acidic shock consistent with our experimental data. Differences in the methods employed to perform shocks could account for the partial non-identification of RGD1 in global microarrays. The RGD1 activation level during shock could also account for this non-identification, as it can be lower than the threshold used to select activated genes in these global studies. The global studies are very interesting to isolate clusters of genes identically or similarly expressed during various growth conditions, or in different mutant backgrounds, but they remain nonexhaustive, and complementary targeted analyses may still bring new data as the expression study on YAP4 (Nevitt et al., 2004) or on GPX2 (Tsuzi et al., 2004). It is interesting to notice that another RhoGAP has been identified in global studies as being regulated by stress: the RhoGAP encoding BAG7 gene is indeed activated by heat and oxidative stresses (Gasch et al., 2000). This regulation is not surprising as five STRE boxes are found in the promoter of BAG7. We have shown the implication of Msn2p and Msn4p in RGD1 regulation during stress; however, Msn2p seemed to 167 have a much more important role than Msn4p despite the high similarity between these two transcription factors (Estruch and Carlson, 1993). Such a differential pattern is relatively common for Msn2/4p-driven genes (Nevitt et al., 2004; Treger et al., 1998). Difference in MSN2 and MSN4 expression in glucose medium (Garreau et al., 2000) might account for the stronger implication of Msn2p in RGD1 transcription. The HOG pathway has been demonstrated to act positively on stress-induced expression dependent on Msn2p and Msn4p, although the effect of HOG pathway inactivation varies with the gene studied (Schuller et al., 1994). We have shown that RGD1 expression is also dependent on Hog1p and Pbs2p, the two major components of the HOG MAP kinase pathway. Although Schuller et al. (1994) did not consider that these two proteins are activators of CTT1 transcription after acidic shock, the effect of hog1D and pbs2D in our study is as high as that of msn2D mutation, indicating the dependence on the HOG pathway for the Msn2/4 transcription activation during acidic shock. 4.2. Cellular implication of Rgd1p in low pH response We demonstrated that cells lacking RGD1 are sensitive to low pH. One explanation might involve the integrity of yeast cell wall and/or plasma membrane which could not resist a too high external proton concentration without diminishing the intracellular pH; this defect in intracellular pH maintenance leads to mortality as was shown elsewhere (Imai and Ohno, 1995). Links between the lack of Rgd1p and defects in cell integrity might be considered through the defective regulation of Rho3p and Rho4p GTPases. Loss of Rgd1p function leads to an accumulation of active forms of Rho3p and Rho4p (Roumanie et al., 2000), and consistent with these data, overexpressing constitutively active form of Rho3p leads to sensitivity to low pH, as for the rgd1D mutant (Claret et al., in preparation). As Rho3p plays a major role in actin dynamics (Imai et al., 1996) and in exocytosis particularly in the docking of vesicles to plasma membrane (Adamo et al., 1999), we propose that deregulation of these processes could alter the amount of plasma membrane components implicated in low-pH adaptation, as the H+ ATPase Pma1p (Fernandes and Sa-Correia, 2001) or alter cell wall composition and structure, thus indirectly allowing abnormal entry of protons within the cell. Defect of proton extrusion through Pma1p does not seem to be the main explanation for the sensitivity of the rgd1D cells, as overexpression of PMA1 does not compensate this default. On the contrary, rgd1D cells seem to have defects in cell wall structure, as they are sensitive to alcian blue (Conde et al., 2003). Another argument in favour of an impairment of the rgd1D cell wall is based on behaviour of rgd1D mid2D double mutant towards parietal drugs. The mid2D strain is resistant to Congo Red and Calcofluor White (Ketela et al., 1999), and we have reported that the lack of Rgd1p function in mid2D mutant decreases resistance to these drugs (de Bettignies et al., 1999). 168 X. Gatti et al. / Gene 351 (2005) 159 – 169 In silico analysis revealed that, as RGD1, other genes encoding RhoGAPs could be regulated during stress. Indeed, 5 of 11 identified RhoGAP genes of S. cerevisiae present STRE sequences in their promoter region: BAG7, BEM2, RGA2, RGD1, and SAC7, and 10 of them present a HSE motif in their promoter region: BAG7, BEM3, ECM25, LRG1, RGA1, RGA2, RGD2, SAC7, and YHR182W. The presence of potential regulatory motifs suggests activation of these genes in stress conditions. Such activation was already demonstrated for BAG7 (Gasch et al., 2000), but investigating the regulatory profile for the entire set of RhoGAPs genes would be of great interest to understand the involvement of this protein family in yeast growth in normal and unfavorable environmental conditions. Acknowledgments We are grateful to Dr. G. Renault and Pr. M. Jacquet for the pGR213 and pGR247 plasmids. We thank M.F. Peypouquet for expert technical assistance. This work was supported by grants from the Université Victor Segalen Bordeaux 2, the Centre National de la Recherche Scientifique, and the Conseil Régional d’Aquitaine. X. Gatti is the recipient of an MJENR fellowship. References Adamo, J.E., Rossi, G., Brennwald, P., 1999. The Rho GTPase Rho3 has a direct role in exocytosis that is distinct from its role in actin polarity. Mol. Biol. Cell 10, 4121 – 4133. Barthe, C., et al., 1998. First characterization of the gene RGD1 in the yeast Saccharomyces cerevisiae. C. R. Acad. Sci., III 321, 453 – 462. Beck, T., Hall, M.N., 1999. The TOR signalling pathway controls nuclear localization of nutrient-regulated transcription factors. Nature 402, 689 – 692. Brewster, J.L., de Valoir, T., Dwyer, N.D., Winter, E., Gustin, M.C., 1993. An osmosensing signal transduction pathway in yeast. Science 259, 1760 – 1763. Casadaban, M.J., Martinez-Arias, A., Shapira, S.K., Chou, J., 1983. Betagalactosidase gene fusions for analyzing gene expression in Escherichia coli and yeast. Methods Enzymol. 100, 293 – 308. Causton, H.C., et al., 2001. Remodeling of yeast genome expression in response to environmental changes. Mol. Biol. Cell 12, 323 – 337. Conde, R., Pablo, G., Cueva, R., Larriba, G., 2003. Screening for new yeast mutants affected in mannosyl phosphorylation of cell wall mannoproteins. Yeast 20, 1189 – 1211. de Bettignies, G., Barthe, C., Morel, C., Peypouquet, M.F., Doignon, F., Crouzet, M., 1999. RGD1 genetically interacts with MID2 and SLG1, encoding two putative sensors for cell integrity signalling in Saccharomyces cerevisiae. Yeast 15, 1719 – 1731. de Bettignies, G., Thoraval, D., Morel, C., Peypouquet, M.F., Crouzet, M., 2001. Overactivation of the protein kinase C-signaling pathway suppresses the defects of cells lacking the Rho3/Rho4-GAP Rgd1p in Saccharomyces cerevisiae. Genetics 159, 1435 – 1448. Doignon, F., Weinachter, C., Roumanie, O., Crouzet, M., 1999. The yeast Rgd1p is a GTPase activating protein of the Rho3 and rho4 proteins. FEBS Lett. 459, 458 – 462. Estruch, F., Carlson, M., 1993. Two homologous zinc finger genes identified by multicopy suppression in a SNF1 protein kinase mutant of Saccharomyces cerevisiae. Mol. Cell. Biol. 13, 3872 – 3881. Fernandes, A.R., Sa-Correia, I., 2001. The activity of plasma membrane H(+)-ATPase is strongly stimulated during Saccharomyces cerevisiae adaptation to growth under high copper stress, accompanying intracellular acidification. Yeast 18, 511 – 521. Garreau, H., Hasan, R.N., Renault, G., Estruch, F., Boy-Marcotte, E., Jacquet, M., 2000. Hyperphosphorylation of Msn2p and Msn4p in response to heat shock and the diauxic shift is inhibited by cAMP in Saccharomyces cerevisiae. Microbiology 146 (Pt 9), 2113 – 2120. Gasch, A.P., et al., 2000. Genomic expression programs in the response of yeast cells to environmental changes. Mol. Biol. Cell 11, 4241 – 4257. Gietz, R.D., Schiestl, R.H., Willems, A.R., Woods, R.A., 1995. Studies on the transformation of intact yeast cells by the LiAc/SS- DNA/PEG procedure. Yeast 11, 355 – 360. Gorner, W., et al., 1998. Nuclear localization of the C2H2 zinc finger protein Msn2p is regulated by stress and protein kinase A activity. Genes Dev. 12, 586 – 597. Gorner, W., et al., 2002. Acute glucose starvation activates the nuclear localization signal of a stress-specific yeast transcription factor. EMBO J. 21, 135 – 144. Ho, Y., et al., 2002. Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature 415, 180 – 183. Imai, T., Ohno, T., 1995. The relationship between viability and intracellular pH in the yeast Saccharomyces cerevisiae. Appl. Environ. Microbiol. 61, 3604 – 3608. Imai, J., Toh-e, A., Matsui, Y., 1996. Genetic analysis of the Saccharomyces cerevisiae RHO3 gene, encoding a rho-type small GTPase, provides evidence for a role in bud formation. Genetics 142, 359 – 369. Jacquet, M., Renault, G., Lallet, S., De Mey, J., Goldbeter, A., 2003. Oscillatory nucleocytoplasmic shuttling of the general stress response transcriptional activators Msn2 and Msn4 in Saccharomyces cerevisiae. J. Cell Biol. 161, 497 – 505. Kapteyn, J.C., et al., 2001. Low external pH induces HOG1-dependent changes in the organization of the Saccharomyces cerevisiae cell wall. Mol. Microbiol. 39, 469 – 479. Ketela, T., Green, R., Bussey, H., 1999. Saccharomyces cerevisiae mid2p is a potential cell wall stress sensor and upstream activator of the PKC1 – MPK1 cell integrity pathway. J. Bacteriol. 181, 3330 – 3340. Marchler, G., Schuller, C., Adam, G., Ruis, H., 1993. A Saccharomyces cerevisiae UAS element controlled by protein kinase A activates transcription in response to a variety of stress conditions. EMBO J. 12, 1997 – 2003. Martinez-Pastor, M.T., Marchler, G., Schuller, C., Marchler-Bauer, A., Ruis, H., Estruch, F., 1996. The Saccharomyces cerevisiae zinc finger proteins Msn2p and Msn4p are required for transcriptional induction through the stress response element (STRE). EMBO J. 15, 2227 – 2235. Matsui, Y., Toh, E.A., 1992. Yeast RHO3 and RHO4 ras superfamily genes are necessary for bud growth, and their defect is suppressed by a high dose of bud formation genes CDC42 and BEM1. Mol. Cell. Biol. 12, 5690 – 5699. McCusker, J.H., Perlin, D.S., Haber, J.E., 1987. Pleiotropic plasma membrane ATPase mutations of Saccharomyces cerevisiae. Mol. Cell. Biol. 7, 4082 – 4088. Moskvina, E., Schuller, C., Maurer, C.T., Mager, W.H., Ruis, H., 1998. A search in the genome of Saccharomyces cerevisiae for genes regulated via stress response elements. Yeast 14, 1041 – 1050. Myers, A.M., Tzagoloff, A., Kinney, D.M., Lusty, C.J., 1986. Yeast shuttle and integrative vectors with multiple cloning sites suitable for construction of lacZ fusions. Gene 45, 299 – 310. Nevitt, T., Pereira, J., Azevedo, D., Guerreiro, P., Rodrigues-Pousada, C., 2004. Expression of YAP4 in Saccharomyces cerevisiae under osmotic stress. Biochem. J. (Pt). Reiser, V., Ruis, H., Ammerer, G., 1999. Kinase activity-dependent nuclear export opposes stress-induced nuclear accumulation and retention of Hog1 mitogen-activated protein kinase in the budding yeast Saccharomyces cerevisiae. Mol. Biol. Cell 10, 1147 – 1161. Robinson, N.G., Guo, L., Imai, J., Toh, E.A., Matsui, Y., Tamanoi, F., 1999. Rho3 of Saccharomyces cerevisiae, which regulates the actin cytoske- X. Gatti et al. / Gene 351 (2005) 159 – 169 leton and exocytosis, is a GTPase which interacts with Myo2 and Exo70. Mol. Cell. Biol. 19, 3580 – 3587. Rose, A.H., 1975. Growth and handling of yeasts. Methods Cell Biol. 12, 1 – 16. Roumanie, O., Peypouquet, M.F., Bonneu, M., Thoraval, D., Doignon, F., Crouzet, M., 2000. Evidence for the genetic interaction between the actin-binding protein Vrp1 and the RhoGAP Rgd1 mediated through Rho3p and Rho4p in Saccharomyces cerevisiae. Mol. Microbiol. 36, 1403 – 1414. Roumanie, O., Peypouquet, M.F., Thoraval, D., Doignon, F., Crouzet, M., 2002. Functional interactions between the VRP1-LAS17 and RHO3 – RHO4 genes involved in actin cytoskeleton organization in Saccharomyces cerevisiae. Curr. Genet. 40, 317 – 325. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Schmitt, A.P., McEntee, K., 1996. Msn2p, a zinc finger DNA-binding protein, is the transcriptional activator of the multistress response in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U. S. A. 93, 5777 – 5782. Schuller, C., Brewster, J.L., Alexander, M.R., Gustin, M.C., Ruis, H., 1994. The HOG pathway controls osmotic regulation of transcription via the stress response element (STRE) of the Saccharomyces cerevisiae CTT1 gene. EMBO J. 13, 4382 – 4389. 169 Sorger, P.K., 1991. Heat shock factor and the heat shock response. Cell 65, 363 – 366. Takahashi, T., Shimoi, H., Ito, K., 2001. Identification of genes required for growth under ethanol stress using transposon mutagenesis in Saccharomyces cerevisiae. Mol. Genet. Genomics 265, 1112 – 1119. Tong, A.H., et al., 2004. Global mapping of the yeast genetic interaction network. Science 303, 808 – 813. Toone, W.M., Jones, N., 1999. AP-1 transcription factors in yeast. Curr. Opin. Genet. Dev. 9, 55 – 61. Treger, J.M., Schmitt, A.P., Simon, J.R., McEntee, K., 1998. Transcriptional factor mutations reveal regulatory complexities of heat shock and newly identified stress genes in Saccharomyces cerevisiae. J. Biol. Chem. 273, 26875 – 26879. Tsuzi, D., Maeta, K., Takatsume, Y., Izawa, S., Inoue, Y., 2004. Regulation of the yeast phospholipid hydroperoxide glutathione peroxidase GPX2 by oxidative stress is mediated by Yap1 and Skn7. FEBS Lett. 565, 148 – 154. Wach, A., Brachat, A., Alberti-Segui, C., Rebischung, C., Philippsen, P., 1997. Heterologous HIS3 marker and GFP reporter modules for PCRtargeting in Saccharomyces cerevisiae. Yeast 13, 1065 – 1075. Werner-Washburne, M., Braun, E.L., Crawford, M.E., Peck, V.M., 1996. Stationary phase in Saccharomyces cerevisiae. Mol. Microbiol. 19, 1159 – 1166.