Multiple Pathways Are Co-regulated by the Protein Kinase Snf1 and the Transcription Factors Adr1 and Cat8
2003; Elsevier BV; Volume: 278; Issue: 28 Linguagem: Inglês
10.1074/jbc.m301981200
ISSN1083-351X
AutoresElton T. Young, Kenneth M. Dombek, C Y Tachibana, Trey Ideker,
Tópico(s)Fermentation and Sensory Analysis
ResumoADR1 and CAT8 encode carbon source-responsive transcriptional regulators that cooperatively control expression of genes involved in ethanol utilization. These transcription factors are active only after the diauxic transition, when glucose is depleted and energy-generating metabolism has shifted to the aerobic oxidation of non-fermentable carbon sources. The Snf1 protein kinase complex is required for activation of their downstream target genes described previously. Using DNA microarrays, we determined the extent to which these three factors collaborate in regulating the expression of the yeast genome after glucose depletion. The expression of 108 genes is significantly decreased in the absence of ADR1. The importance of ADR1 during the diauxic transition is illustrated by the observation that expression of almost one-half of the 40 most highly glucose-repressed genes is ADR1-dependent. ADR1-dependent genes fall into a variety of functional classes with carbon metabolism containing the largest number of members. Most of the genes in this class are involved in the oxidation of different non-fermentable carbon sources. These microarray data show that ADR1 coordinates the biochemical pathways that generate acetyl-CoA and NADH from non-fermentable substrates. Only a small number of ADR1-dependent genes are also CAT8-dependent. However, nearly one-half of the ADR1-dependent genes are also dependent on the Snf1 protein kinase for derepression. Many more genes are SNF1-dependent than are either ADR1- or CAT8-dependent suggesting that SNF1 plays a broader role in gene expression than either ADR1 or CAT8. The largest class of SNF1-dependent genes encodes regulatory proteins that could extend SNF1 dependence to additional pathways. ADR1 and CAT8 encode carbon source-responsive transcriptional regulators that cooperatively control expression of genes involved in ethanol utilization. These transcription factors are active only after the diauxic transition, when glucose is depleted and energy-generating metabolism has shifted to the aerobic oxidation of non-fermentable carbon sources. The Snf1 protein kinase complex is required for activation of their downstream target genes described previously. Using DNA microarrays, we determined the extent to which these three factors collaborate in regulating the expression of the yeast genome after glucose depletion. The expression of 108 genes is significantly decreased in the absence of ADR1. The importance of ADR1 during the diauxic transition is illustrated by the observation that expression of almost one-half of the 40 most highly glucose-repressed genes is ADR1-dependent. ADR1-dependent genes fall into a variety of functional classes with carbon metabolism containing the largest number of members. Most of the genes in this class are involved in the oxidation of different non-fermentable carbon sources. These microarray data show that ADR1 coordinates the biochemical pathways that generate acetyl-CoA and NADH from non-fermentable substrates. Only a small number of ADR1-dependent genes are also CAT8-dependent. However, nearly one-half of the ADR1-dependent genes are also dependent on the Snf1 protein kinase for derepression. Many more genes are SNF1-dependent than are either ADR1- or CAT8-dependent suggesting that SNF1 plays a broader role in gene expression than either ADR1 or CAT8. The largest class of SNF1-dependent genes encodes regulatory proteins that could extend SNF1 dependence to additional pathways. During the diauxic shift, when yeast cells deplete the glucose in the medium, the flow of metabolites changes dramatically to adapt to the use of alternative energy and carbon sources, primarily ethanol produced during fermentation. The changes in gene expression that are required for this metabolic reorganization were dramatically revealed by DNA microarray analysis (1DeRisi J.L. Iyer V.R. Brown P.O. Science. 1997; 278: 680-686Crossref PubMed Scopus (3673) Google Scholar). The expression of more than one-quarter of the yeast genome is altered when glucose is exhausted. These alterations allow the cell to utilize carbon sources other than glucose to channel metabolites into the tricarboxylic acid and glyoxylate cycles to generate energy and synthesize intermediates for gluconeogenesis. In addition, these changes allow the cell to alter its growth rate and prepare for growth arrest and stationary phase. Numerous transcription factors and regulatory proteins are responsible for the altered transcriptional program that accompanies the depletion of glucose. Two transcription factors that play important roles during the diauxic transition are Adr1 1Proteins are indicated by gene names with a capital letter followed by lowercase letters (Adr1), and genotypes and gene names are capitalized in italics (ADR1).1Proteins are indicated by gene names with a capital letter followed by lowercase letters (Adr1), and genotypes and gene names are capitalized in italics (ADR1). and Cat8. ADR1 was discovered as a regulatory gene that is required for expression of the glucose-repressed ADH2 2The abbreviations used are: ADH, alcohol dehydrogenase; ChIP, chromatin immunoprecipitation; ORF, open reading frame; UAS, upstream activation sequence.2The abbreviations used are: ADH, alcohol dehydrogenase; ChIP, chromatin immunoprecipitation; ORF, open reading frame; UAS, upstream activation sequence. gene (2Ciriacy M. Mol. Gen. Genet. 1975; 138: 157-164Crossref PubMed Scopus (130) Google Scholar, 3Ciriacy M. Mol. Gen. Genet. 1979; 176: 427-431Crossref PubMed Scopus (89) Google Scholar). ADH2 encodes an alcohol dehydrogenase isozyme that is required for the first step in ethanol oxidation. Expression of eight other genes is dependent on ADR1. These include other genes important for ethanol and glycerol utilization (4Kratzer S. Schuller H.J. Mol. Microbiol. 1997; 26: 631-641Crossref PubMed Scopus (70) Google Scholar, 5Grauslund M. Lopes J.M. Ronnow B. Nucleic Acids Res. 1999; 27: 4391-4398Crossref PubMed Scopus (61) Google Scholar, 6Pavlik P. Simon M. Schuster T. Ruis H. Curr. Genet. 1993; 24: 21-25Crossref PubMed Scopus (97) Google Scholar). Six of the eight genes encode components of the peroxisomal pathway for β-oxidation of fatty acids (7Simon M. Adam G. Rapatz W. Spevak W. Ruis H. Mol. Cell. Biol. 1991; 11: 699-704Crossref PubMed Scopus (128) Google Scholar, 8Gurvitz A. Wabnegger L. Rottensteiner H. Dawes I.W. Hartig A. Ruis H. Hamilton B. Mol. Cell. Biol. Res. Commun. 2000; 4: 81-89Crossref PubMed Scopus (23) Google Scholar, 9Gurvitz A. Hiltunen J.K. Erdmann R. Hamilton B. Hartig A. Ruis H. Rottensteiner H. J. Biol. Chem. 2001; 276: 31825-31830Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). CAT8 is an essential gene for growth on non-fermentable carbon sources (10Hedges D. Proft M. Entian K.D. Mol. Cell. Biol. 1995; 15: 1915-1922Crossref PubMed Scopus (153) Google Scholar). Most of the genes encoding enzymes of the glyoxylate cycle and two key enzymes of gluconeogenesis are well characterized targets of Cat8. DNA miniarray and proteomic analyses extended to 34 the number of CAT8-dependent genes (11Haurie V. Perrot M. Mini T. Jeno P. Sagliocco F. Boucherie H. J. Biol. Chem. 2001; 276: 76-85Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar). Many of the newly identified genes have roles in ethanol and lactate utilization and intracellular transport of glyoxylate and tricarboxylic acid cycle intermediates. Many of the transcriptional changes occurring at the diauxic transition require an active Snf1 protein kinase complex (12Carlson M. Curr. Opin. Microbiol. 1999; 2: 202-207Crossref PubMed Scopus (458) Google Scholar). Snf1 is the yeast homolog of the AMP-activated protein kinase found in higher eukaryotes. Although Snf1 is present in glucose-grown cells, its activity is regulated, perhaps via phosphorylation by a Snf1 kinase kinase or by the accumulation of AMP that occurs in the cell when glucose is depleted. Its importance for aerobic growth is illustrated by the fact that snf1 mutant cells are unable to grow on any non-fermentable carbon source, while having apparently wild-type growth properties in the presence of glucose. SNF1 is also essential for growth on alternative sugars such as sucrose, galactose, and maltose. Snf1-dependent phosphorylation regulates several glucose-responsive transcription factors. Phosphorylation of the repressor Mig1 leads to its export from the nucleus and derepression of glucose-repressed genes encoding enzymes required for metabolism of alternative sugars (13DeVit M.J. Waddle J.A. Johnston M. Mol. Biol. Cell. 1997; 8: 1603-1618Crossref PubMed Scopus (281) Google Scholar, 14Treitel M.A. Kuchin S. Carlson M. Mol. Cell. Biol. 1998; 18: 6273-6280Crossref PubMed Scopus (258) Google Scholar). SNF1 regulates both the activity of the Cat8 and Sip4 transcription factors and the transcription of their genes (15Hiesinger M. Roth S. Meissner E. Schuller H.J. Curr. Genet. 2001; 39: 68-76Crossref PubMed Scopus (42) Google Scholar, 16Lesage P. Yang X. Carlson M. Mol. Cell. Biol. 1996; 16: 1921-1928Crossref PubMed Scopus (104) Google Scholar, 17Vincent O. Carlson M. EMBO J. 1998; 17: 7002-7008Crossref PubMed Scopus (105) Google Scholar, 18Randez-Gil F. Bojunga N. Proft M. Entian K.D. Mol. Cell. Biol. 1997; 17: 2502-2510Crossref PubMed Scopus (104) Google Scholar). The interaction between Snf1- and ADR1-dependent gene activation is unclear. The expression of Adr1 is not SNF1-dependent (19Dombek K.M. Camier S. Young E.T. Mol. Cell. Biol. 1993; 13: 4391-4399Crossref PubMed Google Scholar), but Adr1 is unable to bind the ADH2 promoter in the absence of Snf1 (20Young E.T. Kacherovsky N. Van Riper K. J. Biol. Chem. 2002; 277: 38095-38103Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar). Snf1 could phosphorylate and activate Adr1 for binding, or it could act on ADR1-dependent promoters in some other manner to allow Adr1 to bind. For example, Snf1 phosphorylates Ser-10 on histone H3, leading to acetylation of Lys-14 and activation of the INO1 promoter (21Lo W.S. Duggan L. Tolga Emre N.C. Belotserkovskya R. Lane W.S. Shiekhattar R. Berger S.L. Science. 2001; 293: 1142-1146Crossref PubMed Scopus (294) Google Scholar). If Snf1 and Adr1 act in a concerted but independent manner on target genes, some ADR1-dependent genes might be activated independently of Snf1. Such genes could be identified by examining the overlap between ADR1- and SNF1-dependent genes using DNA microarrays. Microarrays could help identify other transcription factors that play a role in the activation of Adr1 target genes. For example, two genes involved in ethanol metabolism, ADH2 and ACS1, are co-regulated by Adr1 and Cat8 in a synergistic manner (22Walther K. Schuller H.J. Microbiology. 2001; 147: 2037-2044Crossref PubMed Scopus (51) Google Scholar). The genes of β-oxidation that are ADR1-dependent are also regulated by oleate induction through the transcription factor Oaf1 ·Pip2 (8Gurvitz A. Wabnegger L. Rottensteiner H. Dawes I.W. Hartig A. Ruis H. Hamilton B. Mol. Cell. Biol. Res. Commun. 2000; 4: 81-89Crossref PubMed Scopus (23) Google Scholar, 9Gurvitz A. Hiltunen J.K. Erdmann R. Hamilton B. Hartig A. Ruis H. Rottensteiner H. J. Biol. Chem. 2001; 276: 31825-31830Abstract Full Text Full Text PDF PubMed Scopus (50) Google Scholar). GUT1 is regulated by Ino2 and Ino4 as well as by Adr1 (5Grauslund M. Lopes J.M. Ronnow B. Nucleic Acids Res. 1999; 27: 4391-4398Crossref PubMed Scopus (61) Google Scholar). Thus, Adr1 acts in concert with at least three other transcription factors. To uncover the full repertoire of ADR1-dependent genes and to identify those that are also CAT8- and SNF1-dependent, we performed whole genome transcriptome profiling of derepressed cells using DNA microarrays. Strains carrying deletions of ADR1, CAT8, SNF1, and the double mutants adr1 cat8 and adr1 snf1 were compared with a wild-type strain using a two-color hybridization assay. One hundred eight genes are ADR1-dependent in the microarray assay. Additional genes in pathways already known to be ADR1-dependent as well as genes encoding enzymes and proteins performing cellular functions not previously suspected of being ADR1-dependent were identified. Many of these genes have potential Adr1-binding sites in their promoters, and chromatin immunoprecipitation showed nine of them to be direct targets of Adr1. The ADR1-dependent genes encode proteins having diverse cellular functions, ranging from formate metabolism to meiosis. In a related microarray experiment in which repressed and derepressed mRNAs were compared, most of the ADR1-dependent genes were highly glucose-repressed. Many, but not all, ADR1-dependent genes are dependent for expression on SNF1. Despite their diversity, the ADR1-dependent genes have several unifying features. They are all non-essential in rich glucose medium, and many are important for growth or survival when yeasts are grown on a non-fermentable carbon source. Strains—Yeast strains are listed in Table I. To create isogenic strains that differ only at the ADR1 locus, we replaced the wild-type ADR1 gene with LEU2 (23Hartshorne T.A. Blumberg H. Young E.T. Nature. 1986; 320: 283-287Crossref PubMed Scopus (122) Google Scholar) and introduced a low copy plasmid with or without ADR1. This was done because we observed that an adr1::LEU2 strain grows better on glucose minimal media plates than its ADR1 leu2 isogenic parent, suggesting that leucine biosynthesis may be limiting under these conditions. By using this strategy both the ADR1 wild-type and the adr1 mutant strains are Leu+ and will not differ in gene expression because of unknown indirect effects of leucine auxotrophy. The plasmid copy of ADR1 activates ADH2 expression to a level similar to the level activated by the endogenous ADR1 gene (24Sloan J.S. Dombek K.M. Young E.T. J. Biol. Chem. 1999; 274: 37575-37582Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar). Strain TYY390 was created from TYY202 by disrupting SNF1 with a snf1::URA3 fragment derived from pST70 (25Thompson-Jaeger S. Francois J. Gaughran J.P. Tatchell K. Genetics. 1991; 129: 697-706Crossref PubMed Google Scholar). Strains TYY458 and TYY459 were created by disrupting CAT8 in strains TYY203 (ADR1) and TYY204 (adr1▵1::LEU2) with a PCR fragment generated from a KANR disruption cassette present in plasmid pUG6 (26Guldener U. Heck S. Fielder T. Beinhauer J. Hegemann J.H. Nucleic Acids Res. 1996; 24: 2519-2524Crossref PubMed Scopus (1335) Google Scholar).Table IS. cerevisiae strains used in this studyStrainGenotypeSourceW303-1A/akaTYY201MAT a ade2 can1-100 his3-11, 15 leu2-13, 112 trp1-1 ura3-1W303-1BMATα ade2 can1-100 his3-11, 15 leu2-13, 112 trp1-1 ura3-1TYY202W303-1A with adr1Δ1::LEU220TYY203W303-1A ADH2::YIpADH2/lacZ::TRP1This studyTYY204W303-1A with adr1Δ1::LEU2 ADH2::YIpADH2/lacZ::TRP1This studyTYY458W303-1A with cat8D::Kan R ADH2::YIpADH2/lacZ::TRP1This studyTYY459TYY458 with adr1Δ1::LEU2This studyTYY320TYY202 with YCpADR1-s (TRP1) (wild-type ADR1)20TYY324TYY202 with pRS314 (TRP1) (no ADR1)20TYY390W303-1A with adr1Δ1::LEU2 snf1Δ::URA320TYY391TYY390 with pRS314 (TRP1) (no ADR1)20TYY393TYY390 with YCpADR1-s (TRP1)(wild-type ADR1)20TYY461TYY201 with pWL21 (IME1/lacZ, URA3)This studyTYY462TYY202 with pWL21 (IME1/lacZ, URA3)This studyTGY25MAT a/MATα adr1Δ1::LEU2/adr1Δ1::LEU2 ade2/ade2 can1-100/can1-100 his3-11,15/his3-11,15 leu2-13,112/leu2-13, 112 URA3/ura3-1 trp1-1/TRP1This studyTGY27MAT a/MATα ADR1/adr1Δ1::LEU2 ADE2/ade2 can1-100/can1-100 HIS3/his3-11,15 leu2-13,112/leu2-13, 112 ura3-1/ura3-1 trp1-1/TRP1This study Open table in a new tab Growth of Yeast Cultures—Yeast cultures were grown in synthetic (SM) or rich (YP) medium (27Guthrie C. Fink J.R. Methods Enzymol. 1991; 194: 1-863PubMed Google Scholar). To maintain selection for plasmids containing TRP1, 0.1% casamino acids was added rather than trp- drop-out solution. For repressed growth conditions, glucose was present at 2.5%. For growth in derepressing conditions, the cells were pelleted by centrifugation when they reached an A 600 of 1.0 and resuspended in medium containing 0.05% glucose. The cells were harvested 6 h later by centrifugation and washed with 1/10 volume of cold H2O. DNA Microarray Analyses—Cy3- and Cy5-labeled probes were prepared from poly(A) RNA and hybridized to yeast open reading frame (ORF) microarrays as described previously (28Ideker T. Thorsson V. Siegel A.F. Hood L.E. J. Comput. Biol. 2000; 7: 805-817Crossref PubMed Scopus (241) Google Scholar) with minor modifications. Total yeast RNA was isolated by hot phenol/SDS-glass bead breakage, followed by ethanol precipitation (24Sloan J.S. Dombek K.M. Young E.T. J. Biol. Chem. 1999; 274: 37575-37582Abstract Full Text Full Text PDF PubMed Scopus (28) Google Scholar), and treated with DNase I that had been incubated with iodoacetate to inactivate RNase activity. Poly(A)-enriched RNA was isolated using an Oligotex mRNA Purification kit (Qiagen Corp., Valencia, CA). Forward (Cy3-wild-type/Cy5-mutant) and reverse labeling (Cy5-wild-type/Cy3-mutant) comparisons were performed for each experiment using 15 pmol of Cy3 and 25 pmol of Cy5-labeled cDNAs. Three biological replicates of the ADR1 wild-type to ▵adr1 mutant comparison and one each of the wild-type to ▵snf1, ▵snf1▵adr1, ▵cat8, and ▵cat8▵adr1 comparisons were done. The first two ▵adr1 experiments utilized low density yeast ORF microarrays prepared as described previously (28Ideker T. Thorsson V. Siegel A.F. Hood L.E. J. Comput. Biol. 2000; 7: 805-817Crossref PubMed Scopus (241) Google Scholar). PCR products from 4608 of the smaller ORFs were spotted in duplicate onto one slide and the remaining 1536 larger ORFs were spotted in duplicate onto a second slide. The third replicate of the ▵adr1 experiment and all subsequent experiments utilized high density microarrays that contained 6129 yeast ORFs spotted in duplicate on each slide. These first generation high density microarrays were produced by the Center for Expression Arrays in the Department of Microbiology at the University of Washington (Seattle, WA, ra.microslu.washington.edu/aboutus/about_us.html). Hybridized microarray slides from the first wild-type to ▵adr1 comparison experiment were scanned using a Generation II Scanner from Amersham Biosciences. Slides from all subsequent experiments were scanned using a Generation III Scanner (Amersham Biosciences) at the Center for Expression Arrays. Raw spot intensities, local background, and spot quality estimates were extracted from microarray images using the image analysis program Dapple (29Buhler J. Ideker T. Haynor D. University of Washington CSE Technical Report. University of Washington, Seattle2000: 1-12Google Scholar). The raw data were processed using the DNA Microarray Data Processing Pipeline Perl scripts and software developed at the Institute for Systems Biology (Seattle, WA, db.systemsbiology.net/software/ArrayProcess/index.html). Preprocessing of the data to normalize background-subtracted intensities and to calculate log10 ratios was performed as described previously (28Ideker T. Thorsson V. Siegel A.F. Hood L.E. J. Comput. Biol. 2000; 7: 805-817Crossref PubMed Scopus (241) Google Scholar). In brief, preprocessed data were subjected to several quality control checks before proceeding further with the analysis. First, the log10 ratio for each spot was plotted against its duplicate on the same array slide. Only data from slides having points clustered about a straight line with a positive slope of approximately 1 were further analyzed. Next, log10 ratios from forward and reverse labeling experiments were plotted against each other. Only data from experiments having points clustered about a straight line with a negative slope of approximately 1 were further analyzed. The data from each experiment were then merged and filtered using Dixon's outlier rejection test. Data from all three biological replicates of the wild-type to ▵adr1 mutant comparison were merged into one data set. To identify genes showing significantly different levels of mRNA, a maximum likelihood error model was produced for each data set using the program VERA, and this model was applied to that data set using the program SAM (www.systemsbiology.org/VERAandSAM). SAM uses the model produced by VERA to calculate a maximum likelihood ratio statistic (λ) and a log10 ratio (muRatio) that is corrected for additive and multiplicative errors that arose during the course of the experiment. Each expression ratio presented in this paper is the anti-log of the muRatio. A muRatio of 0.3 or greater, representing a 2-fold or greater difference in mRNA level, was arbitrarily chosen as one criterion of significance. A λ cutoff value of 44.0 for the low density microarrays and a cutoff value of 27.5 for the high density arrays were chosen as second criteria of significance. These λ cutoff values were determined from the distribution of λ values across all genes in a self-hybridization experiment for each type of microarray and represent the λ value below which 95% of the genes fall in the control experiment. Higher λ values eliminated many genes already known to be ADR1-dependent. For these control experiments, Cy3-labeled cDNA was prepared from the poly(A) RNA of derepressed wild-type cells and hybridized against Cy5-labeled cDNA prepared from the same poly(A) RNA. Promoter-binding Site Analyses—Upstream sequences up to 600 bp from the translational start site of each gene were retrieved and analyzed using the Web-based RSA Tools (30van Helden J. Andre B. Collado-Vides J. Yeast. 2000; 16: 177-187Crossref PubMed Scopus (150) Google Scholar) (embnet.cifn.unam.mx/rsa-tools). Transcription factor-binding sites were identified using the dna-pattern tool with the following sequence patterns: Adr1-binding site, DYGGRG, and exact matches to the UAS1-like sequence, CYCCRHN{2,36}DYGGRG (31Cheng C. Kacherovsky N. Dombek K.M. Camier S. Thukral S.K. Rhim E. Young E.T. Mol. Cell. Biol. 1994; 14: 3842-3852Crossref PubMed Scopus (64) Google Scholar); Cat8-binding site, CSRE, CCDNHN{3}CCG (11Haurie V. Perrot M. Mini T. Jeno P. Sagliocco F. Boucherie H. J. Biol. Chem. 2001; 276: 76-85Abstract Full Text Full Text PDF PubMed Scopus (124) Google Scholar); Oaf1-Pip2-binding site, ORE, CGGN{15,18}CCG (32Rottensteiner H. Kal A.J. Filipits M. Binder M. Hamilton B. Tabak H.F. Ruis H. EMBO J. 1996; 15: 2924-2934Crossref PubMed Scopus (95) Google Scholar). The DNA-pattern tool was unable to search for UAS1-like sequences with random mismatches. Therefore, we developed the Perl script seekw and used it to search for UAS1-like sequences allowing for one mismatch anywhere in the consensus. Both strands of all upstream sequences were searched for matches to each of these patterns. Construction of Reporter Plasmids—The base plasmids were pLG669Z (33Guarente L. Methods Enzymol. 1983; 101: 181-191Crossref PubMed Scopus (871) Google Scholar) and pLG669Z(K) (34Thukral S.K. Morrison M.L. Young E.T. Mol. Cell. Biol. 1992; 12: 2784-2792Crossref PubMed Scopus (56) Google Scholar). Promoter fragments from the desired ORFs were amplified from genomic DNA using primers flanking the region of interest. The PCR products were either cloned as XhoI-XmaI or XhoI-KpnI restriction fragments directly into pLG669Z or pLG669Z(K), respectively. Chromatin Immunoprecipitation Analysis—Chromatin immunoprecipitation was performed as described (20Young E.T. Kacherovsky N. Van Riper K. J. Biol. Chem. 2002; 277: 38095-38103Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar), except that cross-linking was with dimethyl adipimidate for 45 min (35Kurdistani S. Robyr D. Tavozoie S. Grunstein M. Nat. Genet. 2002; 31: 248-254Crossref PubMed Scopus (220) Google Scholar) followed by formaldehyde treatment for 15 min. Primers used for PCR are listed in Table II.Table IIPrimers used for PCR analysis of Adr1 ChIP DNAPrimersSequenceCTO-ACT1AGTATGTTCTAGCGCTTGCACCATCCTO-ACT1BTACCGACGATAGATGGGAAGACAGCTO-ADH2ATATGATCCGTCTCTCCGGTTACAGCTO-ADH2BGTGAACCCCATTTCTATGCTCTCCADY2-F-2TGACCCGGGCTCCAAGCTACGGTTTTTACGADY2-R-2ATACTCGAGGGTGTCTCTGACAGGTTGGAGALD4-FCATTCTCCCGGGCCTTTAGTACCGTCCGCACALD4-RGGAAAGCTCGAGGAAGTCAATTGTCACCGACCIT3-FGCTTATGGTACCGTTGAAGCCGTTATGTTGTCGCIT3-RTCTCCTCTCGAGGGTCATCCATTTTCAGTGTGGICL2-FAGCTAAGGTACCTGATGTAGAAAGAGGTGTCAGICL2-RTTATATCTCGAGCGTATGATATCGCTAGCATGAGIP2-FTGAAGGGGTACCGAAGTTCGTTCTATCCAACAGGGIP2-RAATAAACTCGAGACAAGATATCTGTACCCGTTCCMRF1-FTATGTTGGTACCATTTAGTTATAAAGCAAGTGGTAGCMRF1-RTGACAACTCGAGTTTGTAATCTCCAATTTCTTTAGCCPOX1-FATTATTCCCGGGCTTTAAAACCTATAATGTGCPOX1-RAAATAACTCGAGCAGGTTAGACCTTTATAAACACTCYIL057-FCGTAATCCCGGGCAATGCACCCACGACTCGCGGTGYIL057-FTGCAAGCTCGAGGGGCAATAGAGGATTGGGACAGCYPL276-F-1TTACCCGGGTTGTCACCATTTGAGAATAAGCYPL276-R-1TCTCTCGAGCAATGATTGACAGTGCAGAGTG Open table in a new tab One Hundred and Eight Genes Are ADR1-dependent for Expression— Fig. 1 illustrates the functional categories of 108 genes whose expression is reduced 2-fold or more in the absence of ADR1 when cells are grown in depressing medium. The major categories of functionally annotated genes are non-fermentative carbon metabolism, peroxisome biogenesis and β-oxidation, amino acid transport and metabolism, sporulation and meiosis, and transcriptional regulation and signal transduction. Thirteen genes are annotated but do not fall into any one of these categories. Thirty percent of the ADR1-dependent genes have unknown functions, similar to the fraction of genes of unknown function in the whole yeast genome. None of the ADR1-dependent genes are essential for growth in rich medium containing glucose as a carbon source, and none of the annotated genes are involved in cell cycle control. This was expected because adr1 mutants have no obvious growth defect in rich medium. ADR1-dependent Genes Channel Metabolites into Acetyl-CoA and NADH Production— Fig. 2 illustrates the major pathways of carbon utilization in yeast. The genes that are expressed in an ADR1-dependent manner in derepressed conditions are shown in red boxes with their positive expression ratio, wildtype/adr1. Negative values in green boxes indicate genes whose expression is higher in the absence of ADR1 than in its presence. ADR1-dependent genes figure prominently in pathways leading from ethanol, glycerol, lactate, and the oxidation of fatty acids to the formation of acetyl-CoA, generating NADH in the process. Formate metabolism is also apparently ADR1-dependent based on the strong dependence of FDH1 and FDH2 on ADR1 for expression. Formate and propionate, although unable to support yeast growth, can be co-metabolized with growth-limiting amounts of glucose to produce NADH, thus providing reducing equivalents that increase the growth potential of the cell (36Overkamp K.M. Kotter P. van der Hoek R. Schoondermark-Stolk S. Luttik M.A. van Dijken J.P. Pronk J.T. Yeast. 2002; 19: 509-520Crossref PubMed Scopus (51) Google Scholar, 37Pronk J.T. van der Linden-Beuman A. Verduyn C. Scheffers W.A. van Dijken J.P. Microbiology. 1994; 140: 717-722Crossref PubMed Scopus (61) Google Scholar). Table III lists the ADR1-dependent genes of known function, their degree of ADR1 dependence, and their biochemical or physiological function based on information available on the SGD or the MIPS web sites. Regulation of the genes in Table III is described in more detail under “Discussion.”Table IIIFunctional classification of ADR1-dependent genesGeneRatioFunctionNon-fermentative carbon metabolismFDH2 aPromoters were shown to bind Adr1 in this study64Formate dehydrogenaseFDH152Formate dehydrogenaseCIT3 aPromoters were shown to bind Adr1 in this study10Citrate synthaseYML131W9.1Quinone oxidoreductase homologACS1 bThese are genes whose promoters are known by ChIP analysis to bind Adr1 in vivo (20)9.0Acetate-CoA ligaseADH58.2Alcohol dehydrogenaseGLO47.7Hydroxyacylglutathione hydrolaseICL2 aPromoters were shown to bind Adr1 in this study7.02-Methylisocitrate lyaseADH2 bThese are genes whose promoters are known by ChIP analysis to bind Adr1 in vivo (20)(6.8)Alcohol dehydrogenaseDIC16.3Dicarboxylate transportALD4 aPromoters were shown to bind Adr1 in this study5.5Aldehyde dehydrogenaseCYB24.6L-Lactate dehydrogenaseYPL201C4.5Glycerol metabolism?YPL113C4.4Lactate dehydrogenase homologALD53.5Aldehyde dehydrogenaseYCP43.4FlavodoxinOAC13.0Oxaloacetate transportYGR043C2.6Transaldolase homologYHL008C2.6Formate/nitrite transportGUT1 bThese are genes whose promoters are known by ChIP analysis to bind Adr1 in vivo (20)2.4Glycerol kinaseMSS22.4Cox1 pre-mRNA splicing factorGUT22.3Glycerol-3-P dehydrogenaseCTP12.1Citrate transportPeroxisome biogenesis and β-oxidationPOX1 aPromoters were shown to bind Adr1 in this study55Acyl-CoA oxidaseSPS19172,4-Dienoyl-CoA reductase (NADPH)CTA1 bThese are genes whose promoters are known by ChIP analysis to bind Adr1 in vivo (20)16CatalaseFOX2163-Hydroxyacyl-CoA dehydrogenase, enoyl-CoA hydratasePOT1 bThese are genes whose promoters are known by ChIP analysis to bind Adr1 in vivo (20)14Acetyl-CoA C-acyltransferaseYMR018W6.5Pex5 homolog; putative pts1 receptorPXA15.7Peroxisome ABC transporterIDP35.0Isocitrate dehydrogenase (NADP+)DCI14.7Dodecenoyl-CoA δ-isomerasePEX113.3Peroxisomal membrane proteinYOR389W2.7Pex21 interaction by two-hybridPCD12.2Peroxisomal nudix hydrolaseMeiosis and sporulationADY2 aPromoters were shown to bind Adr1 in this study20Transporter, nitrogen utilizationYPL033C15MeiosisDMC17.0Meiotic recombinationATO36.4Ammonia transport/Ady2 homologSPO205.6Pro-spore membrane γ-SNAREBNS13.2MeiosisSPS43.1MeiosisCSM43.0Chromosome segregation meiosisSPR62.1SporulationAmino acid transport and metabolismYLR126C7.7Gln amidotransferase motifLEU16.3Leucine metabolismALP14.9Amino acid transportBAG73.8General amino acid permeaseC
Referência(s)