FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2068079504> ?p ?o ?g. }

• W2068079504 endingPage "36087" @default.
• W2068079504 startingPage "36079" @default.
• W2068079504 abstract "In bacterial adaptation to the dynamic environment, metabolic genes are typically thought to be the executors, whereas global transcription regulators are regarded as the decision makers. Although the feedback from metabolic consequence is believed to be important, much less is understood. This work demonstrates that the gluconeogenic genes in Escherichia coli, ppsA, sfcA, and maeB, provide a feedback loop to the global regulator, cAMP receptor protein (CRP), in carbon source transition. Disruption of one of the gluconeogenic pathways has no phenotype in balanced growth, but causes a significant delay in the diauxic transition from glucose to acetate. To investigate the underlying mechanism, we measured the transcriptome profiles during the transition using DNA microarray, and network component analysis was employed to obtain the transcription factor activities. Results showed that one of the global regulators, CRP, was insufficiently activated during the transition in the ppsA deletion mutant. Indeed, addition of cAMP partially rescued the delay in transition. These results suggest that the gluconeogenic flux to phosphoenolpyruvate is important for full activation of adenylate cyclase through the phosphorylated enzyme IIAglu of the phosphotransferase system. Reduction of this flux causes insufficient activation of CRP and a global metabolic deficiency, which exemplifies a significant feedback interaction from metabolism to the a global regulatory system. In bacterial adaptation to the dynamic environment, metabolic genes are typically thought to be the executors, whereas global transcription regulators are regarded as the decision makers. Although the feedback from metabolic consequence is believed to be important, much less is understood. This work demonstrates that the gluconeogenic genes in Escherichia coli, ppsA, sfcA, and maeB, provide a feedback loop to the global regulator, cAMP receptor protein (CRP), in carbon source transition. Disruption of one of the gluconeogenic pathways has no phenotype in balanced growth, but causes a significant delay in the diauxic transition from glucose to acetate. To investigate the underlying mechanism, we measured the transcriptome profiles during the transition using DNA microarray, and network component analysis was employed to obtain the transcription factor activities. Results showed that one of the global regulators, CRP, was insufficiently activated during the transition in the ppsA deletion mutant. Indeed, addition of cAMP partially rescued the delay in transition. These results suggest that the gluconeogenic flux to phosphoenolpyruvate is important for full activation of adenylate cyclase through the phosphorylated enzyme IIAglu of the phosphotransferase system. Reduction of this flux causes insufficient activation of CRP and a global metabolic deficiency, which exemplifies a significant feedback interaction from metabolism to the a global regulatory system. Bacteria adapt to environmental changes through various signaling pathways and alter the expression of metabolic genes to meet the growth requirements in the new conditions. In this paradigm, regulatory genes, particularly global regulatory genes, are the decision makers that provide unidirectional regulation to the metabolic genes. The feedback from metabolic consequence is much less understood, despite its presumed significance. This work demonstrates an example of such a feedback loop from metabolism to global regulation. When Escherichia coli is grown on excess glucose, acetate is produced and excreted as excess carbon (1Holms H. FEMS Microbiol. Rev. 1996; 19: 85-116Crossref PubMed Google Scholar), the excreted acetate can be consumed as carbon source when glucose level drops (2Shin S. Song S.G. Lee D.S. Pan J.G. Park C. FEMS Microbiol. Lett. 1997; 146: 103-108Crossref PubMed Google Scholar, 3Wolfe A.J. Microbiol. Mol. Biol. Rev. 2005; 69: 12-50Crossref PubMed Scopus (905) Google Scholar). During the metabolic switch from glucose to acetate, E. coli induces genes involved in acetate uptake, the glyoxylate shunt, the TCA cycle, as well as the gluconeogenic genes pckA and ppsA in different time scales (4Kao K.C. Yang Y.L. Boscolo R. Sabatti C. Roychowdhury V. Liao J.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 641-646Crossref PubMed Scopus (123) Google Scholar). The genes pckA and ppsA belong to two parallel pathways for the gluconeogenic conversion of the TCA cycle intermediates to phosphoenolpyruvate (PEP) 2The abbreviations used are:PEPphosphoenolpyruvateCRPcAMP receptor proteinNCAnetwork component analysisTFAtranscription factor activitiesCScontrol strength with the malic enzymes (sfcA and maeB) and ppsA forming one path for the conversion of malate to PEP, and pckA forming the other for the conversion of oxaloacetate to PEP (Fig. 1). phosphoenolpyruvate cAMP receptor protein network component analysis transcription factor activities control strength The switch from glucose to acetate metabolism involves several regulators, such as CRP and IclR. In the absence of glucose, CRP is activated by cAMP, which is produced by adenylate cyclase. The activated cAMP-CRP adduct positively regulates a large number of uptake and metabolic genes to prepare for less favorable carbon sources. The induction of the glyoxylate shunt genes in acetate is mediated directly by IclR and IHF (5Resnik E. Pan B. Ramani N. Freundlich M. LaPorte D.C. J. Bacteriol. 1996; 178: 2715-2717Crossref PubMed Google Scholar). The ligand for IclR activation has not been confirmed. Although phosphoenolpyruvate has been suggested as a ligand for the deactivation of IclR (6Cortay J.C. Negre D. Galinier A. Duclos B. Perriere G. Cozzone A.J. EMBO J. 1991; 10: 675-679Crossref PubMed Scopus (67) Google Scholar), another group was not able to show the same result (7Yamamoto K. Ishihama A. Mol. Microbiol. 2003; 47: 183-194Crossref PubMed Scopus (93) Google Scholar). This study originated from an interesting microarray result comparing E. coli transcript levels in acetate versus glucose minimal media during balanced growth (8Oh M.K. Liao J.C. Biotechnol. Prog. 2000; 16: 278-286Crossref PubMed Scopus (122) Google Scholar). The gluconeogenic gene, ppsA, which is known to be non-essential for growth in acetate, is one of the most up-regulated genes in acetate compared with glucose growth. Indeed, disruption of ppsA has no observable phenotype in acetate or glucose minimal media under balanced growth. Then, why does E. coli need to induce a non-essential gene to such a high level? In this work, we first found that ppsA is important during the transition state between glucose and acetate, although not essential in balanced growths in either carbon source. The ppsA mutant caused a significant lag during the diauxic transition. Through whole genome transcriptome profiling and network component analysis (NCA) (9Liao J.C. Boscolo R. Yang Y.L. Tran L.M. Sabatti C. Roychowdhury V.P. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 15522-15527Crossref PubMed Scopus (499) Google Scholar, 10Tran L.M. Brynildsen M.P. Kao K.C. Suen J.K. Liao J.C. Metab. Eng. 2005; 7: 128-141Crossref PubMed Scopus (81) Google Scholar), we identified that CRP is insufficiently activated without ppsA during glucose to acetate transition. Disruption of other pathway genes to PEP also showed the same phenotype, suggesting that the PEP supply during the transition state is the regulatory return loop from metabolism to modulate cAMP production. Insufficient supply of PEP reduces the phosphorylated crr gene product, EIIAglu, which is the activator of adenylate cyclase. This result demonstrates that the metabolic feedback of the gluconeogenic genes is important in determining transition time regulation, which is closely linked to bacterial survival in a dynamic environment. Strains and Culture Conditions—E. coli BW25113 (F-(araD-araB) lacZ4787 lacIp-4000 LAM-rph-1 (rhaD-rhaB) hsdR514) was obtained from the Yale E. coli stock center and is used for all physiological studies and for monitoring transcript levels unless specified otherwise. Cells were grown in M9 minimum media (11Miller J.H. A Short Course in Bacterial Genetics: A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1992Google Scholar) containing either 0.5% glucose or 0.25% acetate. In the abrupt carbon source transition experiments, cells were grown to mid-log phase (A600 of 0.5–0.6), chilled quickly in ethanol/dry ice bath, harvested by centrifugation at 8,000 × g at 4 °C, then rinsed once with 4 °C 0.25% M9 acetate. The cells were reinocuated into 0.25% M9 acetate media pre-warmed to 37 °C with starting A600 of ∼0.2. For gene expression studies, a portion of the cells was harvested as reference samples immersed in RNAlater (Qiagen, Valencia, CA), and the remaining cells were reinoculated into 0.25% M9 acetate media pre-warmed to 37 °C with starting A600 of ∼0.2. Time course samples were collected at specified times by quickly chilling in a ethanol/dry ice bath and harvested by centrifugation at 8,000 × g at 4 °C. All samples for time-dependent gene expression studies were kept in RNAlater (Qiagen, Valencia, CA) at –80 °C for RNA purification at a later time. For cAMP addition experiments, 10 mm cAMP (Sigma) was added to the pre-warmed 0.25% M9 acetate cultures. Deletion Mutation and Plasmid Construction—Each gene was disrupted using the PCR-based method developed by Datsenko and Wanner (12Datsenko K.A. Wanner B.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 6640-6645Crossref PubMed Scopus (11304) Google Scholar). Primer sequences for deletion and verification are described previously (13Oh M.K. Rohlin L. Kao K.C. Liao J.C. J. Biol. Chem. 2002; 277: 13175-13183Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). Briefly, the chloramphenicol resistance gene was PCR amplified from pKD3 using primers containing sequences homologous to the target gene locus. The linear PCR product was transformed into BW25113 containing the λ helper plasmid pKD46 to disrupt the target gene. The chloramphenicol cassette was removed by transforming pCP20 into the resulting deletion mutant strain and selected for transformants with loss of chloramphenicol resistance. Multiple deletion mutants were constructed sequentially. The plasmid pTB108 was constructed from pJF118EH (14Furste J.P. Pansegrau W. Frank R. Blocker H. Scholz P. Bagdasarian M. Lanka E. Gene (Amst.). 1986; 48: 119-131Crossref PubMed Scopus (813) Google Scholar) by replacing part of the lactose repressor gene, lacI, and the Ptac promoter with the arabinose repressor (araC) coding sequence including its promoter and the arabinose promoter (ParaBAD) from pBAD (Invitrogen) and fusing the ParaBAD to gfpmut3.1, which was amplified from pGFPmut3.1 (15Andersen J.B. Sternberg C. Poulsen L.K. Bjorn S.P. Givskov M. Molin S. Appl. Environ. Microbiol. 1998; 64: 2240-2246Crossref PubMed Google Scholar) by PCR. The gfpmut3.1 open reading frame is flanked by the unique restriction sites, XbaI on the 5′ end and HindIII on the 3′ end. Each gene was amplified by PCR using the primers listed in TABLE ONE and inserted into pTB108 between XbaI and HindIII to generate pTB108::ppsA, pTB108::pckA, and pTB108::maeB.TABLE ONEPrimers used for cloningPrimerSequence (5′ to 3′)ppsA-for-XbaIaaaaaaatctagaatgtccaacaatggctcgtcppsA-rev-HindIIIaaaaaaaaagcttttatttcttcagttcagccapckA-for-XbaIaaaaaaaaagcttttatttcttcagttcagccapckA-rev-HindIIIaaaaaaaaagcttttacagtttcggaccagccgmaeB-for-NcoIaaaaaaaccatggattaaagaggagaaatctagaatggatgaccagttaaaacaamaeB-rev-HindIIIaaaaaaaaagcttttacagcggttgggtttgcg Open table in a new tab DNA Microarray Design, Procedures, and Data Analysis—The array design, RNA purification, and hybridization procedures were described previously (13Oh M.K. Rohlin L. Kao K.C. Liao J.C. J. Biol. Chem. 2002; 277: 13175-13183Abstract Full Text Full Text PDF PubMed Scopus (248) Google Scholar). Each slide contained two spots of the same probe, each sample was hybridized to two slides, and each time course experiment was replicated up to three times. The resulting images were scanned using a VersArray ChipReader High Resolution (5 um) (Bio-Rad) scanner at two excitation wavelengths (532 nm for Cy3 and 635 nm for Cy5). The two scanned images were analyzed using the image analysis software, Imagene (Biodiscovery, Marina Del Rey, CA). The median intensities for signal and background were obtained from the software and used for downstream analysis. The software package, lcDNA (16Hyduke D.R. Rohlin L. Kao K.C. Liao J.C. Omics. 2003; 7: 227-234Crossref PubMed Scopus (16) Google Scholar), developed by our laboratory based on methods described in Tseng et al. (17Tseng G.C. Oh M.K. Rohlin L. Liao J.C. Wong W.H. Nucleic Acids Res. 2001; 29: 2549-2557Crossref PubMed Scopus (484) Google Scholar), was used to normalize the data and calculate the 97.5 and 2.5 quantiles of gene expression, and to assess the significance of expression for each gene. The resulting averaged value of the 97.5 and 2.5 quantiles for each normalized log ratio was used for subsequent analysis. Transcription Factor Activity Analysis—NCA was used to reconstruct the transcription factor activities (TFA) and control strengths (CS) from DNA microarray data (4Kao K.C. Yang Y.L. Boscolo R. Sabatti C. Roychowdhury V. Liao J.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 641-646Crossref PubMed Scopus (123) Google Scholar, 9Liao J.C. Boscolo R. Yang Y.L. Tran L.M. Sabatti C. Roychowdhury V.P. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 15522-15527Crossref PubMed Scopus (499) Google Scholar, 10Tran L.M. Brynildsen M.P. Kao K.C. Suen J.K. Liao J.C. Metab. Eng. 2005; 7: 128-141Crossref PubMed Scopus (81) Google Scholar). Briefly, the gene expression is modeled by power law (18Almeida J.S. Voit E.O. Genome Informatics. 2003; 14: 114-123PubMed Google Scholar, 19Savageau M.A. Biochemical Systems Analysis: A Study of Function and Design in Molecular Biology. Addison-Wesley, Reading, MA1976Google Scholar), [E]=[CS]·[TFA](Eq. 1) where [E] is the matrix of log (base 10) gene expression ratio, [TFA]is the log ratio of transcription factor activities, and [CS] is the matrix of control strengths. The NCA decomposition of [E] into [CS] and [TFA] is based on the topology of [CS], which is the existence of connection between transcription factors and their regulated genes. These connectivity relationships in E. coli were obtained from RegulonDB (20Salgado H. Santos-Zavaleta A. Gama-Castro S. Millan-Zarate D. Diaz-Peredo E. Sanchez-Solano F. Perez-Rueda E. Bonavides-Martinez C. Collado-Vides J. Nucleic Acids Res. 2001; 29: 72-74Crossref PubMed Scopus (182) Google Scholar) with additional modifications obtained from literature and experiments (the connectivity information used can be found at www.seas.ucla.edu/∼liaoj/), which includes 834 genes regulated by 124 transcription factors. However, not all the genes showed differential transcript abundance, and thus the transcription factors whose regulons were deemed unperturbed were eliminated from the analysis. The final NCA network that satisfied the first 2 criteria was composed of 302 genes regulated by 34 transcription factors. The network was analyzed independently for different datasets. Because we scaled the solution sets such that absolute averages of the control strengths were equal to 1 (4Kao K.C. Yang Y.L. Boscolo R. Sabatti C. Roychowdhury V. Liao J.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 641-646Crossref PubMed Scopus (123) Google Scholar, 9Liao J.C. Boscolo R. Yang Y.L. Tran L.M. Sabatti C. Roychowdhury V.P. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 15522-15527Crossref PubMed Scopus (499) Google Scholar), the TFAs of the same transcription factor were in the same scale and comparable. NCA software are available from the web site (www.seas.ucla.edu/∼liaoj/). A ppsA deletion mutant in BW25113 was generated using a PCR-based method developed by Datsenko et al. (12Datsenko K.A. Wanner B.L. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 6640-6645Crossref PubMed Scopus (11304) Google Scholar). We first compared the growth kinetics of the ppsA mutant strain with its otherwise isogenic parent in glucose and acetate minimum media. No growth rate difference was detected between the two strains (Fig. 2, a and b). However, if both glucose and acetate were present simultaneously, we observed a significantly longer diauxic lag in the ppsA strain than the parent (Fig. 2c). Thus, ppsA may be involved in the transition time regulation. To focus on the carbon source transition, we designed an abrupt carbon source transition experiment as described under “Materials and Methods.” Such experiments yield better reproducible transition time (Fig. 2d). To eliminate the possibility that this is a strain-dependent phenomenon, we also deleted the ppsA gene in another E. coli K-12 strain, MC4100, and found that the ppsA mutant strain in MC4100 also exhibited a longer growth lag during the transient state between glucose and acetate (data not shown). Note that the ppsA mutant was constructed by removing the whole open reading frame of the ppsA gene, thus the possibility of a revertant is practically eliminated. Another explanation for why the ppsA mutant strain was able to resume growth after a long lag is the possibility that it acquired a second-site suppressor mutation that allowed it to exhibit the wild-type phenotype. Therefore, we took samples of the ppsA mutant strain after the glucose to acetate transition experiment and conducted a second round of carbon source transition. Again, the ppsA mutant strain exhibits the same phenotype after a second passage, suggesting the phenomenon is not a result of a second-site mutation. To confirm that the decreased ability of the ppsA mutant strain to adapt to acetate is because of lack of a functional ppsA gene and not an artifact of the strain construction, a vector was constructed to express a functional copy the ppsA gene in trans from a plasmid (pTB108::ppsA). The presence of pTB108::ppsA in the ppsA mutant strain was able to complement the longer growth lag exhibited by the ppsA mutant strain (data not shown). Because pckA and ppsA form parallel pathways from the TCA cycle to PEP, we tested whether the pckA deletion mutation caused a similar phenotype as the ppsA mutant during the glucose to acetate transition. The pckA deletion mutant exhibited no growth phenotype during balanced growth in either glucose or acetate. However, unlike ppsA, the pckA deletion mutant exhibited no significant growth phenotype from wild-type during the glucose to acetate transition (Fig. 3). The dependence of pckA appears to be strain dependent as a prolonged growth lag was observed in a pckA deletion mutant in an MC4100 strain during the carbon source transition (Fig. 3d). Note that it has been suggested that pckA is the main gluconeogenic gene for growth in acetate metabolism (1Holms H. FEMS Microbiol. Rev. 1996; 19: 85-116Crossref PubMed Google Scholar). Apparently, the relative importance of the pckA and ppsA pathways are strain-dependent. Similarity between the Wild-type and the ppsA Mutant—To probe the transcriptional regulations involved in the ppsA mutant phenotype during the transition from glucose to acetate, the time course expression profiles of BW25113 wild-type and pps strains were monitored using cDNA microarrays spotted with ∼96% of open reading frames in the E. coli genome. The microarray chip design and experimental procedures are described under “Materials and Methods.” Time course experiments were performed for each strain where samples were harvested at times 5, 15, 30, 60, 120, 180, 240, 300, and 360 min after transition. Each time course sample was cohybridized with a reference sample, which was harvested immediately prior to transition into acetate media (designated 0th time sample). Each sample was hybridized to duplicate slides. Each experiment was repeated up to three times. The expression ratio, as well as confidence intervals of expression of the time course sample over reference sample, for each gene was determined using the software package lcDNA (16Hyduke D.R. Rohlin L. Kao K.C. Liao J.C. Omics. 2003; 7: 227-234Crossref PubMed Scopus (16) Google Scholar, 17Tseng G.C. Oh M.K. Rohlin L. Liao J.C. Wong W.H. Nucleic Acids Res. 2001; 29: 2549-2557Crossref PubMed Scopus (484) Google Scholar). We previously reported the transcriptional gene expression dynamics for the same wild-type strain (4Kao K.C. Yang Y.L. Boscolo R. Sabatti C. Roychowdhury V. Liao J.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 641-646Crossref PubMed Scopus (123) Google Scholar). In the wild-type strain, a growth lag of ∼2–3 h is observed during the transient state between glucose and acetate. During this growth lag, transcriptional regulation was highly active with 331 genes induced and 437 genes repressed by at least 2-fold in at least 2 of the 4 time points within the first hour. Even though the ppsA strain exhibited a much longer growth lag, highly active transcriptional regulation was observed in the ppsA strain with a total of 362 gene transcripts increased at least 2-fold and 392 gene transcript decreased at least 2-fold in at least 2 of the 4 time points within the first hour. Several genes known to be involved in acetate metabolism, such as the glyoxylate shunt enzymes (aceBAK) (21Cozzone A.J. Annu. Rev. Microbiol. 1998; 52: 127-164Crossref PubMed Scopus (158) Google Scholar), acs (2Shin S. Song S.G. Lee D.S. Pan J.G. Park C. FEMS Microbiol. Lett. 1997; 146: 103-108Crossref PubMed Google Scholar), pckA (8Oh M.K. Liao J.C. Biotechnol. Prog. 2000; 16: 278-286Crossref PubMed Scopus (122) Google Scholar), and glcB (24Peellicer M.T. Fernandez C. Badia J. Aguilar J. Lin E.C.C. Baldoma L. J. Biol. Chem. 1999; 274: 1745-1752Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar) showed increased transcript levels within 1–2 h of transition for both strains. In fact, acs and pckA were highly induced (>10-fold) within the first 5 min into the transient state in both strains. We reported earlier (4Kao K.C. Yang Y.L. Boscolo R. Sabatti C. Roychowdhury V. Liao J.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 641-646Crossref PubMed Scopus (123) Google Scholar) that several of the TCA cycle genes were initially down-regulated in the wild-type strain within the first hour of transition. The same trend in gene expression were also observed in the ppsA strain. As a control, we performed a similar transition experiment from glucose to glucose instead of glucose to acetate. Time points were taken at 5, 15, and 30 min after the glucose to glucose transition and gene expression profiles were measured using DNA microarrays. These TCA cycle genes were not affected in the sham transition experiment, verifying that the down-regulation of these genes were not a result of the transition treatment. Roughly 110 genes were perturbed total in the sham transition experiment. Surprisingly, several genes involved in the translational machinery (e.g. rplKA and rplCDW) showed increased expression, and the stress-related genes dps showed decreased transcript levels during the sham experiment, suggesting the transitional treatment did not cause significant stress to the cell. Comparison between the post-transition (t > 0) and pre-transition samples (t = 0) revealed an overall down-regulation of the translational machinery (e.g. rplNXFRO, rpsE, rpmOJ, and rplKA). Both the wild-type and the ppsA strains showed reduced transcript levels in the amino acid biosynthesis pathway genes (e.g. thrAC, hisIHDC), as well as the nucleotide biosynthesis genes (e.g. purA, purB, purF, guaBA), during the initial stages of transition from glucose to acetate. Exceptions include the cysteine biosynthesis genes (e.g. cysE, cysM, cysK). On the contrary, several carbohydrate transport and metabolism genes (e.g. rbsABD, malGF) were induced in both strains, likely as a means to scavenge other carbon sources in the absence of glucose. Among the genes that were up-regulated in both strains within the initial hour of transition, roughly 45% code for hypothetical proteins or proteins of putative functions. Transition from glucose to acetate resulted in the induction of several genes known to be involved in acid resistance, such as xasA, yhiE, and hdeAB. The recently identified acetate permease, yjcG, which is cotranscribed with the acetyl-coA synthase gene (acs) (25Gimenez R. Nunez M.F. Badia J. Aguilar J. Baldoma L. J. Bacteriol. 2003; 185: 6448-6455Crossref PubMed Scopus (107) Google Scholar) is induced in both strains during the carbon source transition. Two proteins, CobB and Pat, were recently identified in Salmonella enterica to regulate acs activity post-translationally (26Starai V.J. Celic I. Cole R.N. Boeke J.D. Escalante-Semerena J.C. Science. 2002; 298: 2390-2392Crossref PubMed Scopus (478) Google Scholar, 27Starai V.J. Escalante-Semerena J.C. J. Mol. Biol. 2004; 340: 1005-1012Crossref PubMed Scopus (216) Google Scholar). The CobB protein is the deacetylase that activates Acs activity, whereas Pat deactivates the ATP-dependent adenylation of acetate by acetylating Acs. In the E. coli glucose to acetate transition, the transcript level of the Pat homolog, b2584, was increased, whereas the transcript of CobB was not differentially regulated during the glucose to acetate transition. Difference between the Wild-type and the ppsA Mutant—From our wild-type gene expression data (4Kao K.C. Yang Y.L. Boscolo R. Sabatti C. Roychowdhury V. Liao J.C. Proc. Natl. Acad. Sci. U. S. A. 2004; 101: 641-646Crossref PubMed Scopus (123) Google Scholar) 3The gene expression data has been submitted to Gene Expression Omnibus accession number GSE3250. , we found that the ppsA gene was not induced until ∼2–3 h after the cells were shifted to acetate. Thus, we would expect the effect of the ppsA mutation on overall gene expression to be minimal within the first hour after transition and increase as the cells respond to the lack of this gluconeogenic pathway. Indeed, in the first hour within the transient state, the expression profiles of ppsA and wild-type strains were very similar, with the Pearson correlation coefficients around 0.9. Between 2 and 6 h during the carbon source switch, the Pearson correlation coefficient between wild-type and ppsA strains decreases from 0.7 to 0.5. This decrease in correlation coefficient between the two strains is partly because of the growth related differences between them. For example, the transcript levels of genes involved in the translational machinery (e.g. rplNXFRO, rpsE, rpmDJ, rplKA), which were initially reduced in both strains, stayed reduced longer in the ppsA mutant strain. The same trend is observed in some genes involved in biosynthesis such as the threonine biosynthesis genes (e.g. thrABC). Most TCA cycle genes were up-regulated with in 2 h of transition for both strains (Fig. 4). However, one notable exception, the malate dehydrogenase gene, mdh, whose induction is observed with in 4 h of transition in the wild-type strain but not until within 10 h of transition in the ppsA mutant strain. In addition, the gene expression of the NADP-linked malic enzyme (maeB) was significantly differentially regulated between the wild-type and ppsA strains (Fig. 4a). The DNA-binding protein from starved cells, Dps, is a nucleoid protein involved in the survival of E. coli under various environmental conditions including long term stationary phase survival, oxidative stress, acidic and alkaline pH conditions (28Nair S. Finkel S.E. J. Bacteriol. 2004; 186: 4192-4198Crossref PubMed Scopus (281) Google Scholar). The gene expression of dps is regulated by the stationary σ factor, OxyR, and IHF (29Altuvia S. Almiron M. Huisman G. Kolter R. Storz G. Mol. Microbiol. 1994; 13: 265-272Crossref PubMed Scopus (355) Google Scholar). During the glucose to acetate transition, the dps transcript level is increased in both wild-type and ppsA mutant strains. However, the induction of dps transcription in the wild-type strain decreased as the cells resumed growth by time = 4 h after the carbon source transition, whereas it remained induced in the ppsA strain. This prolonged induction of the dps transcript level may be indicative of a prolonged stress in the ppsA mutant strain during the glucose to acetate transition. The genes involved in acetate metabolism are under the control of various transcriptional regulators, such as CRP, FruR (Cra), IclR, FIS, IHF, RpoS, FNR, and ArcA. Because these regulators respond to different signals, it is difficult to decipher which ones malfunctioned because of ppsA deletion. Thus, NCA (9Liao J.C. Boscolo R. Yang Y.L. Tran L.M. Sabatti C. Roychowdhury V.P. Proc. Natl. Acad. Sci. U. S. A. 2003; 100: 15522-15527Crossref PubMed Scopus (499) Google Scholar, 10Tran L.M. Brynildsen M.P. Kao K.C. Suen J.K. Liao J.C. Metab. Eng. 2005; 7: 128-141Crossref PubMed Scopus (81) Google Scholar) was used to determine the activities of these transcription factors in the wild-type and ppsA mutant strains during the glucose to acetate transition. NCA combines microarray data with regulator-promoter binding data from RegulonDB (20Salgado H. Santos-Zavaleta A. Gama-Castro S. Millan-Zarate D. Diaz-Peredo E. Sanchez-Solano F. Perez-Rueda E. Bonavides-Martinez C. Collado-Vides J. Nucleic Acids Res. 2001; 29: 72-74Crossref PubMed Scopus (182) Google Scholar) to deduce the regulatory activity of transcription factors of interest. It was able to deconvolute the relative contributions of each regulator from a complex interaction network. In addition to the protein regulators that directly bind to DNA, a small molecule, guanosine tetraphosphate (ppGpp), is known to participate in the response to stringent conditions and g" @default.
• W2068079504 created "2016-06-24" @default.
• W2068079504 creator A5007847950 @default.
• W2068079504 creator A5020006358 @default.
• W2068079504 creator A5053537520 @default.
• W2068079504 date "2005-10-01" @default.
• W2068079504 modified "2023-10-17" @default.
• W2068079504 title "A Global Regulatory Role of Gluconeogenic Genes in Escherichia coli Revealed by Transcriptome Network Analysis" @default.
• W2068079504 cites W1519685185 @default.
• W2068079504 cites W1540505601 @default.
• W2068079504 cites W1566565565 @default.
• W2068079504 cites W1827772617 @default.
• W2068079504 cites W1890358491 @default.
• W2068079504 cites W1961858939 @default.
• W2068079504 cites W1971470015 @default.
• W2068079504 cites W1994782227 @default.
• W2068079504 cites W2002144817 @default.
• W2068079504 cites W2008687368 @default.
• W2068079504 cites W2008818443 @default.
• W2068079504 cites W2017702083 @default.
• W2068079504 cites W2018073151 @default.
• W2068079504 cites W2025086931 @default.
• W2068079504 cites W2041715942 @default.
• W2068079504 cites W2050648928 @default.
• W2068079504 cites W2061700743 @default.
• W2068079504 cites W2068165068 @default.
• W2068079504 cites W2069645361 @default.
• W2068079504 cites W2076375071 @default.
• W2068079504 cites W2082921030 @default.
• W2068079504 cites W2099114361 @default.
• W2068079504 cites W2116137883 @default.
• W2068079504 cites W2126702940 @default.
• W2068079504 cites W2129870891 @default.
• W2068079504 cites W2131603432 @default.
• W2068079504 cites W2134270815 @default.
• W2068079504 cites W2139004845 @default.
• W2068079504 cites W2140809061 @default.
• W2068079504 cites W2144495216 @default.
• W2068079504 cites W2147006435 @default.
• W2068079504 cites W2150688068 @default.
• W2068079504 cites W2152790231 @default.
• W2068079504 cites W2152875113 @default.
• W2068079504 cites W2154199509 @default.
• W2068079504 cites W2160915428 @default.
• W2068079504 cites W2163326745 @default.
• W2068079504 cites W2166516091 @default.
• W2068079504 cites W2168076793 @default.
• W2068079504 cites W2175416669 @default.
• W2068079504 doi "https://doi.org/10.1074/jbc.m508202200" @default.
• W2068079504 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/16141204" @default.
• W2068079504 hasPublicationYear "2005" @default.
• W2068079504 type Work @default.
• W2068079504 sameAs 2068079504 @default.
• W2068079504 citedByCount "75" @default.
• W2068079504 countsByYear W20680795042012 @default.
• W2068079504 countsByYear W20680795042013 @default.
• W2068079504 countsByYear W20680795042014 @default.
• W2068079504 countsByYear W20680795042015 @default.
• W2068079504 countsByYear W20680795042016 @default.
• W2068079504 countsByYear W20680795042017 @default.
• W2068079504 countsByYear W20680795042018 @default.
• W2068079504 countsByYear W20680795042020 @default.
• W2068079504 countsByYear W20680795042021 @default.
• W2068079504 countsByYear W20680795042022 @default.
• W2068079504 countsByYear W20680795042023 @default.
• W2068079504 crossrefType "journal-article" @default.
• W2068079504 hasAuthorship W2068079504A5007847950 @default.
• W2068079504 hasAuthorship W2068079504A5020006358 @default.
• W2068079504 hasAuthorship W2068079504A5053537520 @default.
• W2068079504 hasBestOaLocation W20680795041 @default.
• W2068079504 hasConcept C104317684 @default.
• W2068079504 hasConcept C150194340 @default.
• W2068079504 hasConcept C162317418 @default.
• W2068079504 hasConcept C165864922 @default.
• W2068079504 hasConcept C2908821069 @default.
• W2068079504 hasConcept C33498276 @default.
• W2068079504 hasConcept C54355233 @default.
• W2068079504 hasConcept C547475151 @default.
• W2068079504 hasConcept C67339327 @default.
• W2068079504 hasConcept C70721500 @default.
• W2068079504 hasConcept C86803240 @default.
• W2068079504 hasConceptScore W2068079504C104317684 @default.
• W2068079504 hasConceptScore W2068079504C150194340 @default.
• W2068079504 hasConceptScore W2068079504C162317418 @default.
• W2068079504 hasConceptScore W2068079504C165864922 @default.
• W2068079504 hasConceptScore W2068079504C2908821069 @default.
• W2068079504 hasConceptScore W2068079504C33498276 @default.
• W2068079504 hasConceptScore W2068079504C54355233 @default.
• W2068079504 hasConceptScore W2068079504C547475151 @default.
• W2068079504 hasConceptScore W2068079504C67339327 @default.
• W2068079504 hasConceptScore W2068079504C70721500 @default.
• W2068079504 hasConceptScore W2068079504C86803240 @default.
• W2068079504 hasIssue "43" @default.
• W2068079504 hasLocation W20680795041 @default.
• W2068079504 hasLocation W20680795042 @default.
• W2068079504 hasOpenAccess W2068079504 @default.
• W2068079504 hasPrimaryLocation W20680795041 @default.
• W2068079504 hasRelatedWork W1930517486 @default.

• next