| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research Institute of Innovative Technology for the Earth (RITE), 9-2 Kizugawadai, Kizugawa, Kyoto, 619-0292, Japan
Correspondence
Hideaki Yukawa
mmg-lab{at}rite.or.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The GenBank/EMBL/DDBJ accession nos for the complete C. glutamicum R genome DNA and native episome sequences are AP009044 and AP009045, respectively.
The array data discussed in this publication have been deposited in GenomeNet EXPRESSION (http://www.genome.jp/kegg/expression/) and are accessible through accession number ex0001754.
Tables showing oligonucleotides used in this study, expression data of genes observed to be up- or down-regulated under oxygen deprivation conditions, and the COGs functional annotation of genes and the numbers of genes showing increased or decreased transcriptional levels during oxygen deprived reactions, are available as supplementary data with the online version of this paper.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Hermann, 2003; Inui et al., 1999a; Kelle et al., 2005; Kumagai, 2000; Vertès et al., 2005), as well as heterologous proteins (Billman-Jacobe et al., 1995; Salim et al., 1997). We previously observed that when aerobically grown C. glutamicum R (Yukawa et al., 2007) is packed to a high cell density under oxygen deprivation conditions, despite the cessation of cellular growth, the cells remain able to excrete in significant amounts several metabolites, including lactate and succinate (Inui et al., 2004b). There are two main advantages of using these oxygen deprivation conditions for the production of organic acids in corynebacteria. First, energy is primarily used for compound production, since microbial growth is essentially halted. This results in high yields and low by-product formation. Second, the absence of growth enables the use of micro-organisms at high density. This leads to high volumetric productivity.
The metabolism of C. glutamicum R for organic acid production has previously been investigated by measuring organic acid production, glucose consumption and intracellular NAD+/NADH ratios in the wild-type and various mutant strains in which the genes coding for key metabolic enzymes have been inactivated (Inui et al., 2004b). For L-lactate production, glucose is metabolized to pyruvate via the glycolytic pathway. Pyruvate is in turn converted to L-lactate by the action of NAD-dependent lactate dehydrogenase (LDH). On the other hand, for succinate production, phosphoenolpyruvate, also metabolized from glucose, is converted to oxaloacetate (OAA) by the action of phosphoenolpyruvate carboxylase (PEPC), an anaplerotic enzyme. OAA is subsequently metabolized to succinate using the reductive arm of the tricarboxylic acid (TCA) cycle (i.e. OAA
malatefumaratesuccinate). The observation that under oxygen deprivation conditions PEPC is the predominant anaplerotic enzyme for succinate production (Inui et al., 2004b) is particularly noteworthy, as under aerobic conditions it is pyruvate carboxylase (PC) which is known to be the predominant anaplerotic enzyme to replenish OAA pools for the production of amino acids such as glutamate or lysine (Peters-Wendisch et al., 2001). In addition, the metabolism of C. glutamicum R under oxygen deprivation conditions is controlled mainly by the intracellular NAD+/NADH ratio. Moreover, bicarbonate or pyruvate addition into the oxygen deprived reaction can act as a trigger to stimulate the glycolytic pathway. This phenomenon is known as the ‘pump-priming effect’ (Inui et al., 2004b). Presumably, the uptake of extracellular pyruvate enables the regeneration of NAD+ to fuel the glycolytic pathway at the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) step.
To harness oxygen deprived metabolism in corynebacteria, we have developed a new, high-productivity bioprocess for organic acid production based on these observations (Inui et al., 2004b; Okino et al., 2005). Furthermore, we have recently applied this concept to the production of ethanol from recombinant C. glutamicum (Inui et al., 2004a).
The present study aims at a better understanding of the physiological mechanisms underlying the performance of C. glutamicum for organic acid and ethanol production under process conditions of oxygen deprivation, as controlled at the level of the regulation of genetic expression. We used global DNA microarrays to establish complete transcription profiles of the C. glutamicum genome, in which cells were incubated under normal aeration conditions in A rich medium at physiological cell densities and under oxygen deprivation conditions in BT minimal medium at very high cell densities. These incubation conditions are referred to in the rest of this manuscript as ‘aerobic conditions’ and ‘oxygen deprivation conditions’, respectively.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
|
Conditions for growth and organic acid production.
For genetic manipulations, Escherichia coli strains were grown at 37 °C in Luria–Bertani (LB) medium (Sambrook et al., 1989) and C. glutamicum strains were grown at 33 °C in A medium (2 g yeast extract l–1, 7 g Casamino acids l–1, 2 g urea l–1, 7 g (NH4)2SO4 l–1, 0.5 g KH2PO4 l–1, 0.5 g K2H2PO4 l–1, 0.5 g MgSO4 . 7H2O l–1, 6 mg Fe2SO4 . 7H2O l–1, 4.2 mg Mn2SO4 . H2O l–1, 0.2 mg biotin l–1, 0.2 mg thiamine l–1) with 4 % glucose on a rotary shaker at 200 r.p.m. When appropriate, media were supplemented with antibiotics. The final antibiotic concentrations for E. coli were 50 µg tetracycline ml–1 and 50 µg chloramphenicol ml–1; for C. glutamicum, the final antibiotic concentrations were 5 µg chloramphenicol ml–1 and 50 µg kanamycin ml–1.
Corynebacteria were cultivated at 33 °C, pH 7.6, for 13 h in 500 ml A medium containing 4 % glucose in a 1 l jar fermenter (BMJ01PI, Biott). The aeration was set at 1 l min–1 and the agitation speed was 1000 r.p.m. The culture was harvested by centrifugation at 6500 g and 4 °C for 15 min. The cell pellet was subsequently washed once with minimal medium (BT medium), which differs from the A medium by the absence of yeast extract and Casamino acids. For organic acid production under oxygen deprivation conditions, washed cells were resuspended at a final cell wet cell concentration of 50 g l–1 with 80 ml BT medium containing 200 mM glucose and 400 mM bicarbonate. The cell suspension was incubated at 33 °C with constant agitation without aeration in a lidded 100 ml medium bottle. The pH of the cell reaction was maintained at 8 using a pH controller (DT-1023, Biott) to supplement the medium with NH3.
Analytical methods.
Reaction mixtures were centrifuged (1000 g, 4 °C, 1 min), and the supernatants were analysed for sugars and organic acids. Organic acids were quantified by HPLC (8020, Tosoh) equipped with an electric conductivity detector. The concentration of glucose was measured by an enzyme electrode glucose sensor (BF-4, Oji Scientific Instruments). Cell growth was monitored by measuring OD610 with a spectrophotometer (DU640, Beckman Coulter).
Detection methods for dissolved oxygen (DO) and oxidation–reduction potential (ORP).
The concentration of DO and the ORP were continuously monitored using an oxygen electrode (Type 12, Biott) and a Redox Fermprobe (Broadley-James), respectively.
Enzymic assays.
Cultures were harvested by centrifugation at 5000 g and 4 °C for 10 min. The cell pellet was washed once with extract buffer (100 mM Tris/HCl, pH 7.5, 20 mM KCl, 20 mM MgCl2, 5 mM MnSO4, 0.1 mM EDTA, 2 mM DTT). Cell suspensions were sonicated using an ultrasonic homogenizer (Astrason model XL2020) in an ice–water bath for three 2 min periods, interrupted by 2 min cooling intervals. Cell debris was removed by centrifugation (10 000 g, 4 °C, 30min). The supernatant was used as a crude extract for enzyme assays. Protein concentrations were measured with a Bio-Rad protein assay kit.
Assays for GAPDH (Omumasaba et al., 2004), phosphoglycerate kinase (PGK) (Eikmanns, 1992), triosephosphate isomerase (TPI) (Eikmanns, 1992), PEPC (Inui et al., 1997) and LDH (Bunch et al., 1997) were performed as previously described. Malate dehydrogenase (MDH) activity was determined as described by Sridhar et al. (2000), modified by using 0.1 M Tris/HCl (pH 8.0) and 0.15 mM NADH to obtain optimum enzyme activities. β-Galactosidase activities were measured as described previously (Inui et al., 1999b).
DNA techniques.
Plasmid DNA was isolated by the alkaline lysis procedure (Sambrook et al., 1989) or by using a HiSpeed Plasmid Midi kit (Qiagen) according to the manufacturer's instructions, modified, when extracting DNA from corynebacteria, by using 4 mg lysozyme ml–1 at 37 °C for 30 min. Chromosomal DNA was isolated from corynebacteria following the method of Sambrook et al. (1989), modified by using 4 mg lysozyme ml–1 at 37 °C for 30 min. Restriction endonucleases and Klenow fragment were purchased from Takara and used as per the manufacturer's instructions. PCR was performed using an Applied Biosystems GeneAmp PCR System 9700 in a total volume of 100 µl with 50 ng chromosomal DNA, 0.5 µM each primer, 0.2 mM dNTP, 2 % DMSO in ExTaq polymerase buffer with MgCl2 and 4 U ExTaq DNA polymerase (Takara) for 30 cycles at temperatures of 94 °C for denaturation (1 min), 55 °C for annealing (1 min), and 72 °C for extension (2 min). The PCR fragments were purified with a QIAquick PCR purification kit (Qiagen).
Corynebacteria were transformed by electroporation as previously described (Vertès et al., 1993a). Transformation of E. coli cells was performed by the CaCl2 procedure, as described by Sambrook et al. (1989).
Preparation of DNA microarray.
PCR primers (20-mers) were designed to amplify the full-length coding DNA segments of 3080 predicted ORFs based on the annotation of the complete C. glutamicum R genome sequence (Yukawa et al., 2007). Amino-terminus primers amplified 20 bp downstream from the initiation codon, which was included. Carboxy-terminus primers amplified 20 bp upstream from the termination codon, which also was included. The ORF-specific double-stranded DNA fragments were produced by standard PCR amplification methods, as described above, with genomic DNA of C. glutamicum R as template and the above primer sets. Each PCR product was analysed routinely by agarose gel electrophoresis to ensure its size and purity. When PCR products of the expected size or purity were not obtained, modification of PCR conditions, redesign of primer pairs, or purification of DNA fragments from gels were attempted. As a result, a library of 3076 single PCR products was obtained that contained 99.9 % of a total of 3080 predicted ORFs present in the C. glutamicum R genome. PCR products were purified by a PCR96 Cleanup kit (Millipore), desiccated and resuspended in Solution-T (Takara Bio). The PCR products thus obtained were printed in duplicate at two separate positions onto a 1x3 inch (2.5x7.5 cm approx.) Takara slide glass TX704 (Takara Bio) using an arraying robot. The DNA microarrays were rehydrated in a humidity chamber at 37 °C for 1 h. They were subsequently exposed to UV light at an energy setting of 60 mJ cm–2 in a UV cross-linker, and soaked in 0.2 % SDS for 2 min at room temperature. After being washed twice with ultrapure water, the slides were immersed in a succinic anhydride solution for 20 min at room temperature, and placed in boiling water for 2 min. The slides were then immersed in 100 % ethanol for 1 min and dried by centrifugation at 185 g for 2 min. Spotting quality was assessed by checking as controls the configuration and concentration of non-specific stained DNA spots on the DNA microarrays. In addition, 12 spots each of human transferrin receptor (1 kb), E. coli plasmid pUC19 and 
DNA were printed onto the slides as negative controls for DNA hybridization. DNA microarrays were stored at room temperature in the dark under controlled-humidity conditions until use.
RNA isolation.
Total RNA was extracted from C. glutamicum R cells using the Qiagen RNeasy Midi kit. Cultures were added to two volumes of RNAprotect Bacteria reagent (Qiagen), incubated for 5 min, and centrifuged at 3000 r.p.m. for 5 min at room temperature. The pellet was resuspended in RLT buffer containing β-mercaptoethanol (RNeasy Midi kit, Qiagen) to a final concentration of 1 g dry cell weight (DCW) per litre. A 1 ml sample of this cell suspension was subsequently mixed with 0.5 g 0.1 mm zirconia/silica beads (BioSpec Products). Cells were mechanically disrupted for eight cycles of 45 s at a speed rating of 6.5 spaced by 5 min resting intervals in a FastPrep FP120 instrument (Qbiogene). The resulting mixture was centrifuged for 5 min at 15 000 r.p.m. The supernatant was processed using an RNeasy system (Qiagen) with DNase on-column treatment according to the manufacturer's instructions for RNA extraction. Isolated RNA samples were checked for purity by both agarose gel electrophoresis and spectrophotometric analysis, and stored at –80 °C until use.
Preparation of labelled cDNA.
Fluorescently labelled cDNAs were prepared by cDNA synthesis with random nonamer primers and 30 µg total RNA of C. glutamicum R as template, with subsequent Cy3/Cy5 labelling, using a CyScribe cDNA Post-Labelling kit (Amersham Biosciences), according to the manufacturer's instructions.
Hybridization and analysis.
Fifty microlitres of each Cy3- and Cy5-labelled cDNA probe, prepared as mentioned above, was mixed and added to 100 µl 2x hybridization buffer [12x saline sodium citrate (SSC), 0.4 % SDS, 10x Denhardt's solution, 0.2 mg ml–1 denatured salmon sperm DNA]. All steps, from hybridizing to washing and drying, were automatically performed using a Lucidea Automated Slide Processor (Amersham Biosciences). The hybridization solution, which was heated at 95 °C for 2 min and cooled to room temperature, was added to hybridization buffer to a final combined volume of 200 µl. The DNA microarray was inserted into the Lucidea Automated Slide Processor and hybridization at 60 °C with the whole-genome DNA microarray was allowed to proceed for 14 h. The hybridization arrays were subsequently washed (once at 35 °C for 6 min in 2x SSC and 0.2 % SDS, once at 35 °C for 6 min in 0.2x SSC and 0.2 % SDS), rinsed (twice at 35 °C in 0.2x SSC and once at 37 °C in isopropanol) and dried. The slides were scanned at 532 nm for Cy3 and at 635 nm for Cy5 with a Fujifilm Fluorescent Image Analyser FLA-8000 (Fuji). Images were analysed with a GenePix Pro 5.0 instrument (Axon Instruments). Individual spots were located, the Cy3 and Cy5 fluorescence intensity at each spot was measured, and the background signal intensities were determined. Artefacts on slides such as dust, scratches or salts were excluded from further analysis. Data analysis was performed using GeneSpring 6.1 software (Silicon Genetics). The data generated by GenePix 5.0 were loaded to GeneSpring 6.1 and normalized by global normalization and intensity-dependent LOWESS normalization to eliminate Cy dye bias. The normalization parameter selected in the GeneSpring 6.1 software was ‘Per spot and per chip: intensity dependent (LOWESS) normalization’. The resultant raw intensity data were consecutively calculated as Cy3/Cy5 ratios that represented the average of a total of 12 hybridization signals for each gene: duplicate spots on six different slides of six replicate RNA samples from six independent biological samples. Negative controls consistently gave hybridization signals that were similar to the background. Estimated differences were considered significant if the P value of a t test was <0.05.
Quantitative RT-PCR.
Quantitative RT-PCR was performed using an ABI PRISM 7000 Sequence Detection System (Applied Biosystems) and a QuantiTect SYBR Green RT-PCR kit (Qiagen) according to the manufacturer's instructions. Specific primers (primers 1–40 and 66–81) (Supplementary Table S1) were designed by the Primer Express Software version 2.0 (Applied Biosystems). Each PCR reaction consisted of 25 µl 2x QuantiTect SYBR Green RT-PCR Master Mix, 0.5 µM forward primer, 0.5 µM reverse primer, 0.5 µl QuantiTect RT Mix and 50 ng total RNA in a total volume of 50 µl. PCR parameters were 50 °C for 30 min, 95 °C for 15 min, and 45 cycles at 95 °C for 15 s, 55 °C for 30 s, 72 °C for 30 s. The comparative CT method (Applied Biosystems) was used to quantify relative expression, with a threshold cycle (CT) being defined as the cycle at which the reporter fluorescence is distinguishable from the background in the extension phase of the PCR reaction. The CT values were computed as the mean of triplicates.
Construction of gapA–, ldhA– and mdh–lacZ fusions.
β-Galactosidase was expressed as hybrid gapA'–, ldhA'– and mdh'–'lacZ fusion proteins, using the native gapA, ldhA and mdh transcriptional and translational signals, respectively. In order to integrate the fusion genes on the C. glutamicum R chromosome by homologous recombination, a 2.0 kb DNA fragment from the strain-specific island 7 (SSI 7) (Suzuki et al., 2005) of C. glutamicum R was amplified using the two oligonucleotide primers 41 and 42 with C. glutamicum R chromosomal DNA as template. The 2.0 kb PCR product was digested with SphI, and inserted into the unique SphI site of the suicide vector pCRA725, yielding pCRA740. The E. coli lacZ gene was amplified using oligonucleotide primers 43 and 44 with the E. coli lacZ fusion vector pMC1871 as template. The 3.1 kb PCR product containing the E. coli lacZ was digested with NheI, and inserted into the unique NheI site of pCRA740, yielding pCRA741. In order to amplify the gapA promoter region, the oligonucleotide primers 45 and 46 were used in a PCR with C. glutamicum R chromosomal DNA as template. The 0.6 kb PCR product encompassing the gapA promoter region was digested with NaeI, and inserted into the unique NaeI site of pCRA741, yielding pCRA742. Similarly, the C. glutamicum ldhA and mdh promoter regions were amplified by PCR using their respective oligonucleotide primers (primers 47–50) and C. glutamicum R chromosomal DNA as template. Each amplified PCR product was cloned into pCRA741, yielding pCRA743 for the ldhA–lacZ fusion and pCRA744 for the mdh–lacZ fusion (Table 1
). The plasmids pCRA742, pCRA743 and pCRA744 were isolated as non-methylated DNA from E. coli JM110 for efficient gene introduction into C. glutamicum R (Vertès et al., 1993a), and were subsequently integrated on the C. glutamicum R chromosome by markerless gene insertion methods, as described previously (Inui et al., 2004a).
Primer extension.
IRD700-labelled primers (primers 51–61) were designed as shown in Supplementary Table S1. Total RNA (100 µg) and 1.5 pmol primer were mixed and annealed at 80 °C for 90 min, at 60 °C for 90 min and at 30 °C for 90 min, using a GeneAmp PCR System 9700 instrument (Applied Biosystems). cDNA was synthesized at 42 °C for 1 h using AMV Reverse Transcriptase (Promega). The reaction was terminated by adding EDTA (pH 8.0) at a final concentration of 250 mM and DNase-free RNase A at a final concentration of 3 ng µl–1 to trigger degradation of the RNA templates. The resulting cDNA was treated by phenol/chloroform extraction and ethanol precipitation. Upon centrifugation, the precipitated DNA pellet was resuspended in IR2 Stop Solution (LI-COR). The primer extension products were treated at 95 °C for 2 min and placed on ice for 5 min, and separated on 3.7 % KBPlus Gel Matrix (LI-COR) using a LI-COR 4300 DNA analyser. The migration position of each primer extension product was determined by comparing a sequencing ladder generated from a DNA fragment corresponding to the same chromosomal region, using the same primers and a DYEnamic Direct Cycle Sequencing kit with 7-deaza-dGTP (Amersham Biosciences).
Gene disruption and replacement of the mdh and sdhA genes.
Based on the C. glutamicum R whole genome sequence (Yukawa et al., 2007), two oligonucleotide primers were designed to amplify a portion of the mdh gene (Supplementary Table S1). The oligonucleotide primers 62 and 63 were used in a PCR with C. glutamicum R chromosomal DNA as template. The 1.0 kb PCR product bearing the mdh gene was digested with BamHI and SphI, and inserted into the unique BamHI–SphI site of the E. coli vector pHSG398, which cannot replicate in C. glutamicum, yielding pCRA745. A 1.2 kb PstI kanamycin cassette from pUC4K was blunt-ended with the Takara DNA blunting kit according to the manufacturer's instructions and subsequently inserted into the blunt-ended unique BglII site that lies within the mdh gene-coding region borne by plasmid pCRA745. The resulting plasmid pCRA746 was electroporated into C. glutamicum R according to the method that uses non-methylated DNA isolated from E. coli, as reported previously (Vertès et al., 1993b). Similarly, the C. glutamicum sdhA gene was amplified by PCR using the oligonucleotide primers 64 and 65, and C. glutamicum R chromosomal DNA as template (Supplementary Table S1). The amplified PCR product was cloned into the E. coli vector pHSG398, yielding pCRA747. A 1.2 kb blunt-ended kanamycin cassette was inserted into the unique StuI site of the sdhA gene on pCRA747. The resulting plasmid pCRA748 was electroporated into C. glutamicum R to generate an sdhA-deficient mutant. Disruptions of the two genes were confirmed by Southern hybridization and PCR analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
shows the time-course of the aerobic cultivation of C. glutamicum in a 1 l jar fermenter followed by its incubation under conditions of oxygen deprivation accompanied by the production of organic acid. We have previously reported that C. glutamicum cells produce organic acids from glucose at high yields in mineral medium, even though their proliferation is arrested (Inui et al., 2004b; Okino et al., 2005). Here, we confirmed that the glucose consumption rate during an oxygen deprived reaction [8.1 mmol (g DCW)–1 h–1] (Fig. 1b) was 1.4-fold higher than that attained during aerobic cultivation [6.0 mmol (g DCW)–1 h–1] (Fig. 1a). This phenomenon, which occurs in numerous organisms subjected to anaerobic conditions, is known as the ‘Pasteur effect’ (Emmerling et al., 2000; Gottschalk, 1985).
|
), and after 2 h incubation under oxygen deprivation conditions at a DO concentration of <0.01 p.p.m (ORP –451 mV) (Fig. 1b) for DNA microarray analysis.
The ratios of mRNA levels in the cells under oxygen deprivation conditions relative to those attained under aerobic conditions were calculated by comparing normalized signal intensities between the two samples. The scatter plot of the resulting data showed a correlation coefficient of 0.37, demonstrating that significant changes in the gene expression profiles were taking place between the two conditions (Fig. 2). A total of 161 genes showed a greater than twofold increase in expression, and 221 genes were transcribed at levels less than half those observed under aerobic conditions (Supplementary Table S2). Using the nomenclature based on gene function as defined by the NCBI COG (Clusters of Orthologous Groups) database, 62 of the observed 161 upregulated genes and 64 of the observed 221 down-regulated genes belonged to the metabolism category, suggesting a profound change had occurred when cells were shifted from aerobic conditions to oxygen deprivation conditions (Supplementary Table S3). We focused our detailed analysis on the expression of genes coding for proteins involved in glycolysis, the anaplerotic pathway and the TCA cycle that are related to organic acid production under oxygen deprivation conditions. Several key enzyme-encoding genes in the glycolytic and anaplerotic pathways, and the reductive arm of the TCA cycle, were more than twofold upregulated (Table 2, Figs 2 and 3). In contrast, most genes in the oxidative arm of the TCA cycle (pyruvatecitrateisocitrateoxoglutaratesuccinyl-CoAsuccinate) were down-regulated (Table 2, Figs 2 and 3).
|
|
|
, b; Omumasaba et al., 2004). Moreover, under oxygen deprivation conditions for succinate production by C. glutamicum, the major anaplerotic reaction is driven by the ppc gene product PEPC, rather than by the pyc gene product PC (Inui et al., 2004b). The gapA and ppc genes form, with the pgk and tpi genes, a cluster in the C. glutamicum genome (Eikmanns, 1992). Consistent with these previous observations, we observed during the present study that these four genes were highly upregulated at the transcriptional level under oxygen deprivation conditions (Table 2, Figs 2 and 3). Furthermore, the ldhA gene, which codes for the enzyme LDH, was 8.6-fold induced at the transcriptional level under oxygen deprivation conditions (Table 2, Figs 2 and 3). Likewise, the mdh gene coding for MDH, an enzyme of the reductive arm of the TCA cycle, was 2.7-fold upregulated at the transcriptional level (Table 2, Figs 2 and 3). We have previously reported that C. glutamicum fum (fumarase; FUM) gene disruptants are unable to produce succinate (Inui et al., 2004b). Similarly, we observed during the course of this work that two C. glutamicum mutants in which the MDH-encoding gene (mdh) or the succinate dehydrogenase (SDH) encoding gene (sdhA) had been disrupted were unable to produce succinate. These data indicate that MDH, FUM and SDH are all necessary for succinate production. Related to this finding, the fum, sdhA, sdhB and sdhC genes coding for enzymes of the reductive arm of the TCA cycle were all down-regulated under oxygen deprivation conditions (Table 2, Fig. 3). In addition, we have previously reported that overexpression in a C. glutamicum ldhA mutant of the E. coli frdABCD genes, encoding the enzyme fumarate reductase, which catalyses the conversion of fumarate to succinate, leads to a marginal albeit consistent increase in succinate production when cells are incubated under oxygen deprivation conditions (Murakami et al., 2005). Thus, it was interesting that the malE gene, coding for the malic enzyme, was twofold overexpressed under oxygen deprivation conditions compared to aerobic conditions (Table 2, Figs 2 and 3). It was also noted that the gltA, acn, icd, sucA, sucB, sucC and sucD genes, which encode enzymes involved in the oxidative arm of the TCA cycle, were all down-regulated to approximately half to two-thirds of their expression levels under standard aeration conditions (Table 2, Figs 2 and 3). Moreover, the transcriptional expression of the aceA and aceB genes, respectively coding for isocitrate lyase and malate synthase, two enzymes of the glyoxylate shunt, were not significantly changed under oxygen deprivation compared to aerobic conditions (Table 2, Fig. 3). We have previously reported the absence of gas release by C. glutamicum during organic acid production under oxygen deprivation conditions (Inui et al., 2004b) and that similar amounts of succinate are produced by aceA mutants compared to the wild-type (Inui et al., 2004b). Together, these data corroborated the view that under oxygen deprivation conditions, C. glutamicum scarcely used either the forward direction of the TCA cycle, which is known to release CO2 at several enzymic steps, or the glyoxylate shunt.
The DNA microarray data collected in this work were confirmed by quantitative RT-PCR analyses (Table 2). Furthermore, the increased gene expression at the transcriptional level was reflected in the corresponding specific enzyme activities when these were measured in parallel (Table 3). The ratios of the activities of the enzymes TPI and MDH when shifted from aerobic to oxygen deprivation conditions were notably higher than the ratios of their corresponding gene expression by the DNA microarray analysis (Tables 2 and 3). To understand the difference, additional analyses under oxygen deprivation conditions of mRNA stability, the efficiency and control of translation, and protease activity towards the enzymes, are needed.
|
). These data suggest the existence of a correlation between reduced H+-ATPase activity in C. glutamicum under oxygen deprivation conditions and enhanced glycolytic metabolism, as seen in H+-ATPase mutants of E. coli (Jensen & Michelsen, 1992; Jensen et al., 1995), Bacillus subtilis (Santana et al., 1994) and C. glutamicum (Sekine et al., 2001).
Effect of oxygen tension on expression of the gapA gene cluster, and the ldhA and mdh genes
The differences in transcriptional profiles between aerobic cultivation in A medium and the oxygen deprived reaction in BT medium could alternatively be explained by nutrient limitation, not simply oxygen tension. To verify whether the expression of the gapA gene cluster, and the ldhA and mdh genes, which were more than twofold upregulated in the above DNA microarray and quantitative RT-PCR analyses, were regulated via oxygen tension, we observed the changes of the expression of gapA, pgk, tpi, ppc, ldhA and mdh genes by supplying nitrogen gas during aerobic growth.
After 4 h aerobic growth (DO 6.9 p.p.m), a 99.9 % nitrogen gas supply was substituted for aeration (Fig. 4). DO was immediately reduced to 0.01 p.p.m. and the growth rate was drastically reduced. Cells after 5, 6 and 7 h (sampling points I, II and III in Fig. 4) of incubation were harvested from cultures supplied by air or nitrogen gas. Total RNAs were subsequently isolated from these samples and used for quantitative RT-PCR analysis.
|
). In particular, the ldhA gene was more highly upregulated than the others, and we observed that L-lactate production was immediately started after nitrogen was supplied (data not shown). These data confirmed that these genes were upregulated by decreased oxygen tension.
|
When C. glutamicum that integrated the gapA–, ldhA–, and mdh–lacZ fusions was grown in A medium under aerobic conditions, the expression of the three genes was strongly induced, and there was a reduction of the DO tension at the onset of the stationary phase (Fig. 5a, c, e). High-level expression of the gapA–, ldhA– and mdh–lacZ fusions was maintained under oxygen deprivation conditions (Fig. 5b, d, f).
|
) and oxygen deprivation conditions (at point II in Fig. 1b).
Primer-extension analyses of the gapA, ldhA and mdh genes revealed that, in all cases, a single major extension product was generated both under oxygen deprivation conditions (Fig. 6, lane 1) and aerobic conditions (Fig. 6, lane 2) with all three/four primers (Supplementary Table S1), corresponding to a 5'-end located 185, 93 or 34 bp upstream of the gapA, ldhA or mdh translation initiation codon, respectively. The results obtained with primer 51 for the gapA gene (Fig. 6a), primer 55 for the ldhA gene (Fig. 6b) and primer 58 for the mdh gene (Fig. 6c) are shown. The transcriptional signals of the gapA, ldhA and mdh genes under oxygen deprivation conditions were much stronger than those of the same genes under aerobic conditions, as observed in DNA microarray (Table 2, Fig. 3) and lacZ fusion (Fig. 5) experiments. Upstream of the transcriptional start sites of the gapA, ldhA and mdh genes, putative –10 and –35 promoter sequences, which share some similarities with the E. coli 
70 promoter consensus (Hawley & McClure, 1983), were observed. The transcriptional start site of the gapA gene in C. glutamicum R under oxygen deprivation conditions, as studied in this report, was identical to that in C. glutamicum ATCC 13032 grown under aerobic conditions and reported elsewhere (Schwinde et al., 1993).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). However, the respective contributions of each of these enzymes to the increase of the glycolytic flux in C. glutamicum under oxygen deprivation conditions remain unclear. Studies with a gapA–lacZ fusion suggest that the gapA gene is strongly induced at the onset of the stationary phase under aerobic growth conditions. Moreover, increased expression of the gapA gene was sustained under oxygen deprivation conditions. The maintenance of the upregulation of gapA gene expression under oxygen deprivation conditions was consistently observed with three different methods: lacZ fusion, DNA microarray and quantitative RT-PCR.
Notwithstanding this observation, GAPDH activity has been demonstrated not to exert a significant control over the glycolytic flux in Lactococcus lactis, since the wild-type levels of this enzyme observed in this organism represent a three- to fourfold excess (Solem et al., 2003). In addition, moderate overexpression of phosphofructokinase (pfk) and pyruvate kinase (pyk) has been shown not to dramatically modify flux ratios in E. coli (Sauer et al., 1999). Perhaps the physiological significance of the latter observed transcriptional increases is limited, as recent studies in E. coli have demonstrated that, in this organism, the glycolytic flux is chiefly controlled by the intracellular demand for ATP, rather than by the overexpression of the enzymes of the pathway (Koebmann et al., 2002a, b). On the other hand, PEPC, encoded by ppc, which belongs to the same gene cluster as gapA, is threefold upregulated in C. glutamicum R under oxygen deprivation conditions. Under these conditions, the OAA pool is replenished via conversion of phosphoenolpyruvate catalysed by PEPC, rather than via conversion of pyruvate through the action of the enzyme PC; this observation raises the question whether the pyruvate kinase step could be rate limiting (Inui et al., 2004a, b), due either to redox imbalances or to insufficient pyruvate kinase activity. Pyruvate kinase has been shown elsewhere to be a controlling node of the corynebacterial glycolytic pathway, as it is subject to allosteric activation by AMP (Jetten et al., 1994). Similarly, in E. coli under anaerobic conditions, the activity of the TCA cycle is reduced and anaplerosis via PEPC action predominantly generates OAA that feeds the reductive arm of the TCA cycle (Sauer et al., 1999). PEPC overexpression enables succinate production by C. glutamicum under oxygen deprivation conditions, involving the subsequent exclusive action of MDH, as demonstrated by the inability of C. glutamicum MDH mutants to produce succinate. C. glutamicum possesses two types of L-malate dehydrogenase, MDH, a cytoplasmic, NAD-dependent malate dehydrogenase, and a highly active membrane-associated malate : quinone oxidoreductase (MQO) (Molenaar et al., 1998). Under aerobic conditions, MQO, not MDH, is responsible for the net flux from malate to OAA within the TCA cycle (Eikmanns, 2005), and an MQO-deficient mutant is unable to grow on minimal medium, whereas an MDH-negative mutant has no obvious phenotype (Molenaar et al., 2000). In addition, this enzyme may contribute to the backflux from OAA to fumarate, as detected by NMR flux quantification (Eikmanns, 2005). We therefore inferred that under oxygen deprivation, mdh gene expression is induced and MDH mainly functions for the backflux from OAA to fumarate with subsequent succinate production. The physiological value to the C. glutamicum cell under oxygen deprivation conditions of upregulating both ppc and mdh, which serve as key nodes in the metabolic network, is perhaps to commit the carbon chains to the reductive arm of the TCA cycle. A similar response is observed at the level of the expression of ldhA, which is present in a single copy in the C. glutamicum genome (Inui et al., 2004b). The NAD-dependent LDH was 8.6-fold upregulated under oxygen deprivation conditions, perhaps to enable the cell to more efficiently regenerate its NAD+ pool, and thus compensate, among other things, for the down-regulation of the oxidative arm of the TCA cycle and the cessation of energy-generating respiration.
Most genes in the oxidative arm of the TCA cycle, namely gltA, acn, icd, sucA, sucB, sucC and sucD, are down-regulated in C. glutamicum under oxygen deprivation conditions, which is consistent with the observation made in other organisms that under anaerobic conditions the primary function of the TCA cycle is to replenish biosynthetic precursors and not to generate energy. Likewise, the citrate synthase encoded by the gltA gene is down-regulated in oxygen-deprived C. glutamicum, similar to what is observed in B. subtilis (Nakano et al., 1998). The physiological impact of this phenomenon remains to be elucidated.
The ubiquitous changes occasioned particularly by a sharp decrease in oxygen tension are apparent at the level of other energy components of the corynebacterial cell. For example, it is worth noting that transcription of the corynebacterial H+-ATPase operon under oxygen deprivation conditions is sharply decreased to one-quarter to one-half of its level under aerobic conditions. This is in sharp contrast to what is observed in the facultative anaerobe E. coli, in which anaerobic cell growth on glucose minimal salts medium leads to H+-ATPase expression that is 30 % higher than that attained under aerobic growth (Kasimoglu et al., 1996). The physiological function of the ATPase complex is to generate ATP from ADP and Pi by electron transport-linked phosphorylation under aerobic conditions. Under non-respiratory conditions, the ATPase complex hydrolyses ATP to maintain an electrochemical gradient across the bacterial membrane. E. coli ATPase-deficient mutants exhibit increased flow through the TCA and glycolytic pathways (Jensen & Michelsen, 1992; Jensen et al., 1995), with ATP being generated by an increased rate of glucose consumption and acetate generation. The increased flow through glycolysis and the TCA cycle results in an increased generation of NADH, which is compensated by an increased respiration rate without the generation of energy. Similarly, B. subtilis mutants grown aerobically consume twice the amount of glucose for the same increase in biomass, and exhibit increased synthesis of terminal oxidases, including ubiquinol oxidase and a-type haem, leading to an increased respiration rate, albeit also uncoupled from the generation of ATP (Santana et al., 1994). H+-ATPase mutant C. glutamicum also exhibits an increased respiration rate, an increased glucose consumption rate per cell, increased glycolysis, and decreased 2-oxoglutarate production, suggesting a decreased flow through the oxidative arm of the TCA cycle, increased production and intracellular accumulation of pyruvate, and increased lactate production (Sekine et al., 2001). Combined, these various observations suggest that, like E. coli and B. subtilis, C. glutamicum adjusts its metabolism in the absence of oxidative phosphorylation to maintain its energy level as high as possible, through mechanisms that, overall, have been highly conserved under different selective forces during two billion years of evolution since the divergence of Gram-negative and Gram-positive bacteria.
In C. glutamicum, and again similar to what is observed in B. subtilis (Ye et al., 2000) and E. coli (Kang et al., 2005), numerous other genes are induced or repressed with changes in environmental conditions such as oxygen tension. Among the genes observed to be highly inducible in the last two species, beyond those genes involved in anaerobic respiration, carbon metabolism and electron transport, iron uptake, stress response and several hypothetical genes are also upregulated. In B. subtilis and E. coli, the major regulator of the physiological switch between aerobic and anaerobic growth conditions is the DNA binding protein FNR (Gunsalus & Park, 1994; Kang et al., 2005; Pao et al., 1994; Ye et al., 2000). For example, in E. coli, 297 genes constituting 184 different operons are regulated by FNR and/or oxygen levels (Kang et al., 2005). However, other global transcriptional regulators have been shown to play important roles, such as ArcA, which impacts redox regulation, and whose knockout increases the activity of the TCA cycle by 60 % in E. coli (Perrenoud & Sauer, 2005). Similarly, down-regulation of the transcription of citrate synthase and aconitase in B. subtilis under anaerobiosis is driven by a mechanism that is different from those involving CcpA, CcpB, FNR and ResDE (Nakano et al., 1998).
The overall physiological consequence of the observed increased transcription in corynebacteria under oxygen deprivation process conditions is an increased rate of carbon-source consumption, and presumably an increased flow of carbon through the glycolytic pathway and the reductive arm of the TCA cycle, accompanied by a concomitant increase in organic acid production. The numerous transcriptional changes observed under varying environmental conditions suggest the presence in C. glutamicum, as in other organisms, of one or more pleiotropic transcriptional regulators that are activated when cells are subjected to oxygen deprivation conditions and perhaps mediated by sensing-effector cascades. In addition to its fundamental scientific interest, the understanding of the molecular basis of the biological response of C. glutamicum to oxygen deprivation conditions will help design more efficient processes for industrial biotechnology.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: C. W. Chen
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bunch, P. K., Mat-Jan, F., Lee, N. & Clark, D. P. (1997). The ldhA gene encoding the fermentative lactate dehydrogenase of Escherichia coli. Microbiology 143, 187–195.[Abstract]
Eggeling, L. & Sahm, H. (1999). Amino acid production: principles of metabolic engineering. In Metabolic Engineering, pp. 153–176. Edited by S. Y. Lee & E. T. Papoutsakis. New York: Marcel Dekker.
Eikmanns, B. J. (1992). Identification, sequence analysis, and expression of a Corynebacterium glutamicum gene cluster encoding the three glycolytic enzymes glyceraldehyde-3-phosphate dehydrogenase, 3-phosphoglycerate kinase, and triosephosphate isomerase. J Bacteriol 174, 6076–6086.
Eikmanns, B. (2005). Central metabolism: tricarboxylic acid cycle and anaplerotic reactions. In Handbook of Corynebacterium glutamicum, pp. 241–276. Edited by L. Eggeling & M. Bott. Boca Raton, FL: CRC Press.
Emmerling, M., Bailey, J. E. & Sauer, U. (2000). Altered regulation of pyruvate kinase or co-overexpression of phosphofructokinase increases glycolytic fluxes in resting Escherichia coli. Biotechnol Bioeng 67, 623–627.[CrossRef][Medline]
Gottschalk, G. (1985). Bacterial Metabolism, 2nd edn. New York: Springer.
Gunsalus, R. P. & Park, S. J. (1994). Aerobic-anaerobic gene regulation in Escherichia coli: control by the ArcAB and Fnr regulons. Res Microbiol 145, 437–450.[Medline]
Hawley, D. K. & McClure, W. R. (1983). Compilation and analysis of Escherichia coli promoter DNA sequences. Nucleic Acids Res 11, 2237–2255.
Hermann, T. (2003). Industrial production of amino acids by coryneform bacteria. J Biotechnol 104, 155–172.[CrossRef][Medline]
Inui, M., Dumay, V., Zahn, K., Yamagata, H. & Yukawa, H. (1997). Structural and functional analysis of the phosphoenolpyruvate carboxylase gene from the purple nonsulfur bacterium Rhodopseudomonas palustris no. 7. J Bacteriol 179, 4942–4945.
Inui, M., Terasawa, M. & Yukawa, H. (1999a). L-Isoleucine. In Encyclopedia of Bioprocess Technology: Fermentation, Biocatalysis, and Bioseparation, pp. 1498–1503. Edited by M. C. Flickinger & S. W. Drew. New York: Wiley.
Inui, M., Nakata, K., Roh, J. H., Zahn, K. & Yukawa, H. (1999b). Molecular and functional characterization of the Rhodopseudomonas palustris no. 7 phosphoenolpyruvate carboxykinase gene. J Bacteriol 181, 2689–2696.
Inui, M., Kawaguchi, H., Murakami, S., Vertès, A. A. & Yukawa, H. (2004a). Metabolic engineering of Corynebacterium glutamicum for fuel ethanol production under oxygen deprivation conditions. J Mol Microbiol Biotechnol 8, 243–254.[CrossRef][Medline]
Inui, M., Murakami, S., Okino, S., Kawaguchi, H., Vertès, A. A. & Yukawa, H. (2004b). Metabolic analysis of Corynebacterium glutamicum during lactate and succinate productions under oxygen deprivation conditions. J Mol Microbiol Biotechnol 7, 182–196.[CrossRef][Medline]
Jensen, P. R. & Michelsen, O. (1992). Carbon and energy metabolism of atp mutants of Escherichia coli. J Bacteriol 174, 7635–7641.
Jensen, P. R., Michelsen, O. & Westerhoff, H. V. (1995). Experimental determination of control by the H+-ATPase in Escherichia coli. J Bioenerg Biomembr 27, 543–554.[CrossRef][Medline]
Jetten, M. S., Gubler, M. E., Lee, S. H. & Sinskey, A. J. (1994). Structural and functional analysis of pyruvate kinase from Corynebacterium glutamicum. Appl Environ Microbiol 60, 2501–2507.
Kang, Y., Weber, K. D., Qiu, Y., Kiley, P. J. & Blattner, F. R. (2005). Genome-wide expression analysis indicates that FNR of Escherichia coli K-12 regulates a large number of genes of unknown function. J Bacteriol 187, 1135–1160.
Kasimoglu, E., Park, S. J., Malek, J., Tseng, C. P. & Gunsalus, R. P. (1996). Transcriptional regulation of the proton-translocating ATPase (atpIBEFHAGDC) operon of Escherichia coli: control by cell growth rate. J Bacteriol 178, 5563–5567.
Kelle, R., Hermann, T. & Bathe, B. (2005). L-Lysine production. In Handbook of Corynebacterium glutamicum, pp. 465–488. Edited by L. Eggeling & M. Bott. Boca Raton, FL: CRC Press.
Koebmann, B. J., Westerhoff, H. V., Snoep, J. L., Nilsson, D. & Jensen, P. R. (2002a). The glycolytic flux in Escherichia coli is controlled by the demand for ATP. J Bacteriol 184, 3909–3916.
Koebmann, B. J., Westerhoff, H. V., Snoep, J. L., Solem, C., Pedersen, M. B., Nilsson, D., Michelsen, O. & Jensen, P. R. (2002b). The extent to which ATP demand controls the glycolytic flux depends strongly on the organism and conditions for growth. Mol Biol Rep 29, 41–45.[CrossRef][Medline]
Kumagai, H. (2000). Microbial production of amino acids in Japan. Adv Biochem Eng Biotechnol 69, 71–85.[Medline]
), DO (
) and ORP (
), and the concentration of glucose (
), lactate (
) and succinate (
) during aerobic cultivation (a) and oxygen deprived conditions (b) of C. glutamicum R. The arrows indicate the sampling points of C. glutamicum cells under aerobic (I) and oxygen deprivation conditions (II) for transcriptional analyses. Under oxygen deprivation conditions, the DO value was always maintained at a level lower than 0.01 p.p.m. A glucose spike was performed at 9.5 h after the beginning of the experiment by adding 20 g glucose to the reaction vessel containing 500 ml culture.
