Portal de Periódicos da CAPES

Sok2p Transcription Factor Is Involved in Adaptive Program Relevant for Long Term Survival of Saccharomyces cerevisiae Colonies

2004; Elsevier BV; Volume: 279; Issue: 36 Linguagem: Inglês

10.1074/jbc.m404594200

1083-351X

Libuše Váchová, Frédéric Devaux, Helena Kučerová, Markéta Řičicová, Claude Jacq, Zdena Palková,

Plant-Microbe Interactions and Immunity

Volatile ammonia functions as a long range alarm signal important for the transition of yeast colonies to their adaptive alkali developmental phase and for their consequent long term survival. Cells of aged Saccharomyces cerevisiae sok2 colonies deleted in the gene for Sok2p transcription factor are not able to release a sufficient amount of ammonia out of the cells, they are more fragile than cells of wild type colonies, and they exhibit a survival defect. Genome-wide analysis on gene expression differences between sok2 and WT colonies revealed that sok2 colonies are not able to switch on the genes of adaptive metabolisms effectively and display unbalanced expression and activity of various enzymes involved in cell protection against oxidative damage. Impaired amino acid metabolism and insufficient activation of genes for putative ammonium exporters Ato and of those for some other membrane transporters may be responsible for observed defects in ammonia production. Thus, Sok2p appears to be an important regulator of S. cerevisiae colony development. Gene expression differences caused by its absence in colonies differ from those described previously in liquid cultures, which suggests a pleiotropic effect of Sok2p under different conditions. Volatile ammonia functions as a long range alarm signal important for the transition of yeast colonies to their adaptive alkali developmental phase and for their consequent long term survival. Cells of aged Saccharomyces cerevisiae sok2 colonies deleted in the gene for Sok2p transcription factor are not able to release a sufficient amount of ammonia out of the cells, they are more fragile than cells of wild type colonies, and they exhibit a survival defect. Genome-wide analysis on gene expression differences between sok2 and WT colonies revealed that sok2 colonies are not able to switch on the genes of adaptive metabolisms effectively and display unbalanced expression and activity of various enzymes involved in cell protection against oxidative damage. Impaired amino acid metabolism and insufficient activation of genes for putative ammonium exporters Ato and of those for some other membrane transporters may be responsible for observed defects in ammonia production. Thus, Sok2p appears to be an important regulator of S. cerevisiae colony development. Gene expression differences caused by its absence in colonies differ from those described previously in liquid cultures, which suggests a pleiotropic effect of Sok2p under different conditions. Yeast colonies growing on complex solid medium produce ammonia, which functions as a volatile signal influencing long term colony development. The pulse of ammonia production and transition of stressed aged colony (e.g. Saccharomyces cerevisiae colony 12 days old) to the alkali developmental phase are accompanied by transient growth arrest followed by the renascent colony growth during the second phase of acidification (1Palkova Z. Janderova B. Gabriel J. Zikanova B. Pospisek M. Forstova J. Nature. 1997; 390: 532-536Crossref PubMed Scopus (171) Google Scholar). In neighboring colonies, ammonia induces their ammonia production regardless of their current developmental phase, thus causing synchronization of development of all colonies in the respective territory. Ammonia appears to function as an alarm signal released by the colony, which first feels the stress induced by shortage in nutrients. Then, via the ammonia induction mechanism, this signal is spread through the whole colony population, causing its switch to the alkali phase (2Palkova Z. Forstova J. J. Cell Sci. 2000; 113: 1923-1928PubMed Google Scholar). Our previous findings revealed that the acid-to-alkali colony transition is connected with significant changes in gene expression, indicating a parallel decrease in stress factors and mitochondrial respiratory functions, a switch from the mitochondrial citrate cycle to the methylglyoxylate cycle, enabling more economical exploitation of nutrients, and mobilization of cell reserves by means of β-oxidation of fatty acids in peroxisomes. Additionally, we observed significant changes in the expression of genes coding for various membrane proteins/transporters including those that may be important for ammonia production (Ato transporters) (3Palkova Z. Devaux F. Ricicova M. Minarikova L. Le Crom S. Jacq C. Mol. Biol. Cell. 2002; 13: 3901-3914Crossref PubMed Scopus (116) Google Scholar). Monitoring the behavior of colonies of strains defective in several particular genes allowed us to identify the gene encoding transcription factor Sok2p as a regulator possibly affecting long term colony development. Sok2p is a transcriptional repressor of the basic helix-loop-helix family, which was initially characterized as a multicopy suppressor of mutations in the cAMP-dependent protein kinase pathway (4Ward M.P. Garrett S. Mol. Cell. Biol. 1994; 14: 5619-5627Crossref PubMed Scopus (42) Google Scholar). It has been proposed to counteract the function of Msn2p/Msn4p in the activation of some stationary phase-induced genes (SSA3, GAC1) in the presence of glucose, thus being an actor in glucose repression triggered by the cAMP-dependent protein kinase pathway (5Ward M.P. Gimeno C.J. Fink G.R. Garrett S. Mol. Cell. Biol. 1995; 15: 6854-6863Crossref PubMed Scopus (157) Google Scholar). It was shown that Sok2p can physically interact with the transcription factor Msn2p and is involved in the starvation-induced metabolic changes that control the decision of diploid yeast cells to switch from the mitotic growth to the meiosis and sporulation (6Shenhar G. Kassir Y. Mol. Cell. Biol. 2001; 21: 1603-1612Crossref PubMed Scopus (61) Google Scholar). Additionally, Sok2p was shown to be a negative regulator of sporulation (6Shenhar G. Kassir Y. Mol. Cell. Biol. 2001; 21: 1603-1612Crossref PubMed Scopus (61) Google Scholar) and switch to pseudohyphal growth in diploid cells under conditions of nitrogen limitation (7Pan X. Heitman J. Mol. Cell. Biol. 2000; 20: 8364-8372Crossref PubMed Scopus (103) Google Scholar). In the sok2/sok2 diploid mutant, both sporulation and switch to pseudohyphal growth occur even when a rich nitrogen source is present in the medium. The role of Sok2p in pseudohyphal growth appears to be independent of the cAMP-dependent protein kinase pathway. Microarray and Northern blot experiments indicate that Sok2p negatively regulates the expression of the PHD1, ASH1, and SWI5 genes, which encode positive regulators of the genes required for pseudohyphal growth (e.g. FLO11) (7Pan X. Heitman J. Mol. Cell. Biol. 2000; 20: 8364-8372Crossref PubMed Scopus (103) Google Scholar). As regards these data, Sok2p appears to be a key regulator involved in many processes connected with yeast growth, development, and adaptation to environmental changes, particularly changes in nutrient availability. Previously, we have shown that colonies of SOK2-deleted strains are not able to produce sufficient amount of ammonia and that they exhibit growth problems at later phases of colony development (3Palkova Z. Devaux F. Ricicova M. Minarikova L. Le Crom S. Jacq C. Mol. Biol. Cell. 2002; 13: 3901-3914Crossref PubMed Scopus (116) Google Scholar). Here, using cytological, biochemical, and microarray approaches, we observed premature dying of sok2 colonies and significant anomalies in the expression of several groups of genes previously proposed to be involved in ammonia-induced adaptation of S. cerevisiae colonies. We also found that SOK2-deleted colonies have an unbalanced activity of some of the reactive oxygen species detoxification enzymes. Our data show that the loss of SOK2 causes significant pleiotropic disorders already in young colonies. These finally result in faulty transcriptional and physiological reprogramming of aged colonies in their alkali phase of development, leading to failure in colony long term survival and adaptation. Strains and Media—Strain S. cerevisiae BY4742 (MATα, his3Δ1, leu2Δ0, lys2Δ0, ura3Δ0) and the isogenic mutant sok2 are from the EUROSCARF collection. Colonies were grown on GM agar (1% yeast extract, 3% glycerol, 2% agar, 30 mm CaCl2) or GM-BKP 1The abbreviations used are: BKP, bromcresol purple; MDR, multidrug resistance; WT, wild type; MES, 4-morpholineethanesulfonic acid; TES, 2-{[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]amino}ethanesulfonic acid. agar (GM, 0.01% bromcresol purple). Ammonia Production Measurement—Ammonia released by growing colonies was absorbed into acidic traps as described (1Palkova Z. Janderova B. Gabriel J. Zikanova B. Pospisek M. Forstova J. Nature. 1997; 390: 532-536Crossref PubMed Scopus (171) Google Scholar), and the amount of ammonia in various liquid samples was determined by the use of the Nessler reagent. Determination of the Amount of Ammonium/Ammonia in Cells and in the Agar below Colonies—Colonies were removed at particular time points (see Fig. 1B). The agar surface was thoroughly cleaned of the remaining cells, and discs (diameter of 7 mm) of the entire agar layer below the colonies were cut out. For each parallel measurement, three discs were collected into one Eppendorf tube, and 50 μl of 10% citric acid was immediately added to lower pH and to prevent premature volatilization of ammonia present in the agar. At this step, samples can be stored frozen at -80 °C. To separate ammonia/ammonium from other nitrogen forms present in the agar, NH4+ was converted to NH3 by the addition of NaOH, causing rapid increase in pH and immediate NH3 volatilization from the agar discs. The precise arrangement was as follows. The agar discs in Eppendorf tube were overlaid by 400 μl of 5 m NaOH, and the tube was immediately closed by the stopper containing a small piece of cotton wool soaked with 200 μl of 10% citric acid. A small piece of perforated nylon cloth was placed between the tube and the stopper to fix the wool in the stopper. Released ammonia was captured into the citric acid solution in the wool for the interval of 10 min. Then, the stopper with the soaked wool was transferred on an empty Eppendorf tube, the liquid was spun down by centrifugation at 1000 rpm for 3 min, and its nitrogen content was determined by using Nessler reagent. The colonies harvested from agar plates were washed and treated in the same way as the agar discs. Concentrated 5 m NaOH was used both for cell disruption and for ammonia volatilization. Detection of Cells with Permeabilized Membrane and of Disrupted Cells—The whole colony growing on GM-BKP agar was picked up and resuspended in 10 mm MES buffer, pH 6, with 1 m sorbitol, and the percentage of cells stained by BKP was counted under a fluorescence microscope (Leica; filter N2.1). Disrupted cells were photographed either with fluorescence filter N2.1 or under the Nomarski contrast by using a Hitachi camera. Detection of Proteins Released from Cells of Colonies during Phosphate-buffered Saline Washing—Whole colonies growing on GM agar were picked up at different time points (see Fig. 2D), and wet biomass was weighted and gently suspended in phosphate-buffered saline (8Sambrook J. Maniatis T. Fritsch E.F. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989: A8.41-A8.45Google Scholar). After centrifugation, proteins of the supernatant were precipitated by methanol/chloroform treatment (9Wessel D. Flugge U.I. Anal. Biochem. 1984; 138: 141-143Crossref PubMed Scopus (3108) Google Scholar). Sedimented proteins were dissolved in Laemmli sample buffer and run on SDS-PAGE (12% gel). Proteins extracted from 5 mg of wet biomass were loaded on each slot. Silver staining of proteins in gels was performed as described (10Rabilloud T. Electrophoresis. 1992; 13: 429-439Crossref PubMed Scopus (244) Google Scholar). RNA Isolation and Northern Analysis—For total RNA isolation, colonies (approximately 1010 cells) were directly suspended in TES buffer (10 mm Tris, pH 7.5, 10 mm EDTA, 0.5% SDS). The exact isolation procedure is at www.biologie.ens.fr/fr/genetiqu/puces/protocoles_puces.html. The RNA samples were quantified on an Agilent Bioanalyzer 2100 using RNA Nano assay (Agilent Technologies, Palo Alto, CA) (11Ricicova M. Palkova Z. FEMS Yeast Res. 2003; 4: 119-122Crossref PubMed Scopus (17) Google Scholar). mRNA was prepared from total RNA using the Micro Fast track mRNA purification kit (Invitrogen). For Northern blot, 10 μg of total RNA was loaded. The rRNA content was visualized by EtBr staining. The radioactive probes for ATO1, JEN1, POX1, ICL2, TPO2, MSN4, DIA1, CMK2, ARG1, and ACT1 were obtained by random priming on the complete open reading frame (amplified by PCR from the Research Genetics bank of yeast open reading frames) using the Nonaprimer kit. The individual Northern blots were quantified using the TINA (Ray-test) software with ACT1 as the reference. Microarray Analyses and Biocomputational Analyses of Microarray Data—We used homemade microarray slides containing probes for most of the yeast open reading frames (about 6000 oligonucleotides). The oligonucleotides were supplied by MWG Biotech (Yeast Oligo set). The slides used are Ultragaps from Corning. 4 μg of mRNA was used for each reverse transcription reaction. Detailed protocols are at www.biologie.ens.fr/fr/genetiqu/puces/protocoles_puces.html. The arrays were read by a Genepix 4000 scanner (Axon) and were analyzed with the Genepix 4.0 software. We excluded artifactual spots, saturated spots, and low signal spots. Assuming that most of the genes have unchanged expression, the Cy3/Cy5 ratios were normalized block per block using the LOWESS normalization tool of the Varan software, and genes with significant changes in expression were selected using the confidence interval of 0.999; other parameters were according to the default settings (www.bionet.espci.fr/varan/) (12Golfier G. Tran Dang M. Dauphinot L. Graison E. Rossier J. Potier M.C. Bioinformatics. 2004; 20: 1641-1643Crossref PubMed Scopus (13) Google Scholar). The cluster shown on Fig. 3 was generated by Treeview (13Eisen M.B. Spellman P.T. Brown P.O. Botstein D. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 14863-14868Crossref PubMed Scopus (13076) Google Scholar). Each microarray result presented here is an average of 5-6 independent biological measurements. The data were deposited in the hybridization array data repository GEO (www.ncbi.nlm.nih.gov/geo/) under the GEO accession number GSE1454. Determination of Catalase and Superoxide Dismutase Activity—Cells of 6-18 giant colonies were harvested and stored at -80 °C until use. Cell lysates were prepared in 10 mm MES buffer, pH 6, supplemented with protease inhibitor mixture Complete, EDTA-free (Roche Applied Science) and 250 μm 4-(2-aminoethyl)benzenesulfonyl fluoride (Sigma) by vortexing with glass beads under cooling. Cell debris was removed by centrifugation for 5 min at 13,000 rpm. Protein concentration in supernatant (lysate) was determined using a protein detection kit (Bio-Rad). Supernatant (5-15 μg of proteins/slot) was run on PAGE (9% gel) under non-denaturing conditions (8Sambrook J. Maniatis T. Fritsch E.F. Molecular Cloning: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1989: A8.41-A8.45Google Scholar); SDS was omitted in all steps). The activity of superoxide dismutase in the gel was visualized as described (14Beauchamp C. Fridovich I. Anal. Biochem. 1971; 44: 276-287Crossref PubMed Scopus (9521) Google Scholar). For detection of catalase activity, the method of Clare et al. (15Clare D.A. Duong M.N. Darr D. Archibald F. Fridovich I. Anal. Biochem. 1984; 140: 532-537Crossref PubMed Scopus (212) Google Scholar) was used. Enzyme activities were determined by the image analyses using the Un-Scan-It gel program (Silk Scientific Corp.). The commercial catalase from beef liver (Roche Applied Science) and superoxide dismutase from horseradish (Sigma) were used as the standards. Calibration curve using 4-5 different dilutions of the commercial enzyme of the known activity was prepared for each gel. Determination of Concentrations of Low Molecular Weight Compounds Containing Sulfhydryl Groups—The assay was performed as described (16Ellman G.L. Arch. Biochem. Biophys. 1959; 82: 70-77Crossref PubMed Scopus (20881) Google Scholar, 17Hu M.L. Methods Enzymol. 1994; 233: 380-385Crossref PubMed Scopus (772) Google Scholar) with several modifications as follows. Low molecular weight compounds were extracted from colonies by 7.5% trichloroacetic acid. The wet biomass concentration in each sample was 77.5 mg/ml. Extraction was completed by three freeze-thaw cycles; cell debris was removed by centrifugation. 900 μl of 0.4 m Tris-HCl, pH 8.9, was mixed with 100 μl of the extract to adjust the pH in the assay to 8.2-8.3. 10 μl of 10 mm 5,5′-dithiobis-2-nitrobenzoate was added. The absorbance at 412 nm was measured before and 10 min after the addition of 5,5′-dithiobis-2-nitrobenzoate. Physiological Alterations of sok2 Colonies—The first perceptible difference between colonies formed by the S. cerevisiae strain BY4742 (WT colonies) and colonies formed by the isogenic strain deleted in the SOK2 gene (sok2 colonies) appears even in the first acidic phase, when the coloring of the pH dye indicator BKP present in the GM agar indicates more efficient acidification of sok2 colonies in comparison with their WT counterparts (data not shown). As indicated before (3Palkova Z. Devaux F. Ricicova M. Minarikova L. Le Crom S. Jacq C. Mol. Biol. Cell. 2002; 13: 3901-3914Crossref PubMed Scopus (116) Google Scholar), sok2 colonies are not able to reach the fully developed alkali phase and to produce a significant level of volatile ammonia. The defect in NH3 production can be caused either by an inefficient alkalization of sok2 colony surroundings, leading to an inefficient conversion of protonated NH4+ to unprotonated NH3 (pKa = 9.25, where Ka is the dissociation constant) (18Palkova Z. Vachova L. Int. Rev. Cytol. 2003; 225: 229-272Crossref PubMed Scopus (29) Google Scholar) and therefore leading to poor NH3 volatilization, or by a defect in the formation/export of NH4+ out of the sok2 colonies or by a combination of both. As the first possibility would lead to the accumulation of ammonium outside of the cells, we measured the amounts of ammonium in agar below WT and sok2 colonies in different developmental phases (Fig. 1B). The profile of ammonium concentration in the agar below the WT colonies indicates an efficient initiation of ammonium secretion from WT cells in the interval between the 9th and 13th day. Later, the ammonium concentration in the agar transiently decreases, apparently due to the sharp increase in pH, leading to the efficient NH3 volatilization (Fig. 1A). The timing of changes in ammonium concentration in the agar below sok2 colonies is similar to that of WT colonies; however, the amount of ammonium present in agar between the 12th and 13th day (0.04 mg nitrogen/colony) is about half of that in agar below WT colonies (0.1 mg of nitrogen/colony) (Fig. 1B). The sum of ammonia volatilized from WT colonies between the 8th and 13th day is much higher (2.1 mg of nitrogen/colony) than its residual concentration in agar, and it is 11 times lowered in sok2 colonies (0.2 mg of nitrogen/colony). In summary, during the interval of transition of colonies from the acidic to the alkali/abortive alkali phase, the total amount of ammonia released in the sok2 colony (0.24 mg of nitrogen/colony) is about 10% of those released in the WT colony (2.2 mg of nitrogen/colony). There is no dramatic difference in the content of ammonium in WT and sok2 cells (Fig. 1C). At the time of the increased production of extracellular ammonia/ammonium by WT colonies, the ammonium concentration within sok2 cells (0.06 mg of nitrogen/colony) is about 75% of that within WT cells. Later, the intracellular ammonium in sok2 colonies slightly increases, reaching a higher level than that of WT colonies, from which the ammonia/ammonium is efficiently excreted. However, there is no substantial accumulation of ammonium in sok2 cells in concentrations corresponding to the amount of released ammonia in WT colonies (approximately 2 mg of nitrogen/colony). In other words, the inability of sok2 colonies to produce extracellular ammonium/ammonia is not connected with any significant increase of intracellular ammonium, which also means that the production of free ammonium within cells is lowered in sok2 colonies (see below). All these data suggest that besides an inefficient alkalization, sok2 colonies exhibit additional defects in formation and export of ammonium out of the cells. The observed defect in the expression of putative ammonium exporter Ato1p (Fig. 1D) is in agreement with this (for details, see below). The appearance of an increasing number of papillae, which initiates at the time of the transition to the abortive alkali phase (Fig. 2A), indicates that the coordinated growth of cells within the sok2 colony is impaired and results in irregular colony morphology. This can be the result of the progressive dying of the majority of cells in colonies alongside with the effort of some cells to escape the fate of the majority and to exploit nutrients released from dying cells for their own growth. To prove this hypothesis, we estimated the number of dying cells within sok2 and WT colonies in different phases of the colony development using the BKP dye, which enters only the cells with a permeabilized plasma membrane. Results summarized on Fig. 2B show that there is a significant increase in the number of sok2 cells stained by BKP dye after the switch to the abortive alkali phase (starting approximately at the 13th day). This is accompanied by an increased amount of stained cell debris within samples from sok2 colonies as compared with WT ones (Fig. 2C). Gentle washing of cells picked up from colonies of different ages by phosphate-buffered saline buffer and subsequent analysis of protein content in the washing extract (Fig. 2D) revealed an increased amount of released proteins in extracts from aged sok2 colonies. All these observations indicate that aged sok2 cells became more fragile than WT cells and a large number of them is ultimately disrupted. Gene Expression Differences between WT and SOK2-deleted Colonies—For microarray comparisons, we isolated transcriptomes from giant colonies (1Palkova Z. Janderova B. Gabriel J. Zikanova B. Pospisek M. Forstova J. Nature. 1997; 390: 532-536Crossref PubMed Scopus (171) Google Scholar) of WT strain BY4742 and the isogenic strain deleted in the SOK2 gene. We compared WT and sok2 colonies occurring in five different developmental phases: 1) the early first acidic phase (4th day), 2) the middle first acidic phase (7th day), 3) the neutral phase (11th day), 4) the fully developed alkali phase (WT) and the abortive alkali phase (sok2) (13th day), and 5) the second acidic phase (21st day). We observed differences in the expression of ∼500 genes in sok2 colonies at least in one of the five developmental phases (supplementary Table 1S and Fig. 3). Expression changes of some genes representing the main functional categories were confirmed by Northern blot (Fig. 4). The finding that even in the early first acidic phase there are groups of genes that are differently expressed in sok2 colonies when compared with WT colonies is in agreement with the observation that the sok2 versus WT diversity starts even in the first acidic phase. However, the existence of distinct groups of genes differently expressed in any of the particular developmental phases indicates divergences between sok2 and WT during the whole examined interval of colony growth. The SOK2-deleted Colonies Do Not Switch on Effectively the Genes of Plasma Membrane Transporters Important for Volatile Ammonia Production—Several groups of genes, which were found to be induced during the transition of WT colonies from the acidic phase to the ammonia-producing period of life (3Palkova Z. Devaux F. Ricicova M. Minarikova L. Le Crom S. Jacq C. Mol. Biol. Cell. 2002; 13: 3901-3914Crossref PubMed Scopus (116) Google Scholar), are down-regulated in sok2 colonies as compared with WT colonies (Fig. 3 and see also Fig. 6). They namely include genes for some transporters and metabolic genes. The genes for phosphate (Pho84p, Pho89p, Pho90p), sulfate (Sul1p), and carboxylic acid (Jen1p) transporters possibly involved in the import of protons into WT cells and initial increase of extracellular pH are down-regulated in sok2 colonies as compared with WT colonies (Fig. 3). Also, the expression of all three genes encoding putative ammonium transporters Ato1p, Ato2p, Ato3p (3Palkova Z. Devaux F. Ricicova M. Minarikova L. Le Crom S. Jacq C. Mol. Biol. Cell. 2002; 13: 3901-3914Crossref PubMed Scopus (116) Google Scholar) is significantly lowered in sok2 colonies. Northern blot analysis of time course ATO1 gene expression (Fig. 4) revealed that even sok2 colonies start to induce ATO1 gene at approximately the same developmental time as WT colonies. However, in contrast to WT colonies, the expression is much lower and is not sufficiently kept up at later developmental phases. The moderate ATO1 gene induction (Fig. 1D) correlates with the initial appearance of ammonium in the agar below sok2 colonies (Fig. 1B). The SOK2-deleted Colonies Do Not Switch on Effectively the Adaptation Program—Groups of metabolic genes possibly involved in the adaptation of WT colonies, which are switched on during their acid-to-alkali transition (3Palkova Z. Devaux F. Ricicova M. Minarikova L. Le Crom S. Jacq C. Mol. Biol. Cell. 2002; 13: 3901-3914Crossref PubMed Scopus (116) Google Scholar), are not activated during sok2 colony development (Fig. 3). They include genes for peroxisome biogenesis and fatty acid β-oxidation (e.g. FAA1, POX1, IDP3, PXA1, SPS19, CAT2, PEX11, PEX21, FOX2), genes important for non-degradative steps of the citrate cycle and methylglyoxylate cycle in mitochondria (CIT1, CIT2, CIT3, ICL2), and genes encoding various dehydrogenases (ADH2, ADH1, FDH1) of cytosol and mitochondria, which might participate in NADH regeneration during the escape of cells from oxidative stress (18Palkova Z. Vachova L. Int. Rev. Cytol. 2003; 225: 229-272Crossref PubMed Scopus (29) Google Scholar). Differences in the expression of POX1 (encoding acyl-CoA oxidase, an enzyme involved in fatty acid β-oxidation in peroxisomes) and ICL2 (encoding isocitrate lyase, a key enzyme of the methylglyoxylate cycle) detected on Northern blots of RNAs isolated from sok2 and WT colonies in different developmental phases are even more expressive (Fig. 4). These findings indicate that sok2 colonies are not able to adapt efficiently to a shortage of nutrients. They seem not to be able to switch on the economical mitochondrial methylglyoxylate pathway as well as to activate genes of metabolisms providing essential substrates, acetyl-CoA, and oxaloacetate. The later is demonstrated by the absence of activation of genes essential for utilization of fatty acids (important for acetyl-CoA production) and by the absence of activation of the gene JEN1 for the plasma membrane transporter of carboxylic acids and genes of the cytosolic glyoxylate cycle MLS1, ICL1, and MDH3 (important for the import of carboxylic acids and their conversion to oxaloacetate). The SOK2-deleted Colonies Remain Stressed—The gene encoding Msn4p, one of the major activators (in concert with Msn2p) of various stress-response genes via their stress-response element sequence (19Martinez-Pastor M.T. Marchler G. Schuller C. Marchler-Bauer A. Ruis H. Estruch F. EMBO J. 1996; 15: 2227-2235Crossref PubMed Scopus (836) Google Scholar), is strongly activated in sok2 colonies as compared with WT colonies, starting even in the 'middle' first acidic phase (Figs. 3 and 4). Various stress genes, which are induced in the sok2 mutant (Fig. 3), contain a stress-response element in their promoter and seem to be under the Msn4p control (e.g. CTT1, HOR2). The gene YER130c encoding the Msn4p homologue, a transcription factor of unknown function, is also strongly activated. The YER130c gene was shown to be activated by Haa1p, a transcription factor that also activates the TPO2 gene encoding the polyamine transport protein (see below) (20Keller G. Ray E. Brown P.O. Winge D.R. J. Biol. Chem. 2001; 276: 38697-38702Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). Expression changes indicate that sok2 colonies are not able to switch on the adaptation program and become more stressed by reactive oxygen species than WT colonies. This prediction was checked by the direct analysis of the activity of important oxidative stress defense enzymes, namely cytosolic catalase Ctt1p and superoxide dismutase Sod1p (Fig. 5A). The enzymatic assays revealed the higher activity of Ctt1p in cell lysates of sok2 colonies in comparison with WT colonies in all estimated phases. On the other hand, the activity of Sod1p in sok2 colonies is almost identical with that of WT colonies at later developmental phases (phases 3-5), whereas it is significantly lower in sok2 colonies occurring at the early and middle first acidic phases. The prediction of cumulative oxidative stress in sok2 colonies is also supported by the observation that the amount of low molecular weight compounds containing sulfhydryl groups (e.g. glutathione) is lowered in sok2 colonies during the first acidic phase (Fig. 5B) Alterations in Expression of Amino Acid Metabolic Genes and Possible Role of TPO Genes—Amino acids and their metabolism appear to be important for the proper transition of colonies from acidic to alkali phase as well as for ammonia production (1Palkova Z. Janderova B. Gabriel J. Zikanova B. Pospisek M. Forstova J. Nature. 1997; 390: 532-536Crossref PubMed Scopus (171) Google Scholar, 21Zikanova B. Kuthan M. Ricicova M. Forstova J.