SGM Journals
PHYSIOLOGY AND BIOCHEMISTRY

Regulation of ldh expression during biotin-limited growth of Corynebacterium glutamicum

Christiane Dietrich, Aimé Nato, Bruno Bost, Pierre Le Maréchal, Armel Guyonvarch

  • 1Université Paris-Sud, IGM, UMR 8621, Orsay F-91405, France
  • 2Université Paris-Sud, IBBMC, UMR 8619, Orsay F 91405, France
  • 3CNRS, Orsay F-91405, France
  • Correspondence
    Armel Guyonvarch
    armel{at}igmors.u-psud.fr

    • Microbiology 2009; 155(4):1360–1375 · https://doi.org/10.1099/mic.0.022004-0

    View at publisher PubMed

    Abstract

    Corynebacterium glutamicum is a biotin-auxotrophic bacterium and some strains efficiently produce glutamic acid under biotin-limiting conditions. In an effort to understand C. glutamicum metabolism under biotin limitation, growth of the type strain ATCC 13032 was investigated in batch cultures and a time-course analysis was performed. A transient excretion of organic acids was observed and we focused our attention on lactate synthesis. Lactate synthesis was due to the ldh-encoded l-lactate dehydrogenase (Ldh). Features of Ldh activity and ldh transcription were analysed. The ldh gene was shown to be regulated at the transcriptional level by SugR, a pleiotropic transcriptional repressor also acting on most phosphotransferase system (PTS) genes. Electrophoretic mobility shift assays (EMSAs) and site-directed mutagenesis allowed the identification of the SugR-binding site. Effector studies using EMSAs and analysis of ldh expression in a ptsF mutant revealed fructose 1-phosphate as a highly efficient negative effector of SugR. Fructose 1,6-bisphosphate also affected SugR binding.

    Edited by: R. J. M. van Spanning

    INTRODUCTION

    Corynebacterium glutamicum is a Gram-positive organism, isolated by Kinoshita and coworkers as a biotin-auxotrophic bacterium, that excretes large amounts of l-glutamate when grown aerobically under biotin limitation with glucose as a carbon source (Kinoshita et al., 1957; Kikuchi & Nakao, 1986). C. glutamicum is now widely used in the industrial production of amino acids such as l-glutamate and l-lysine (Abe et al., 1967). Improvements of the fermentation strategies employed and of the bacterial strains have been achieved, leading to progressively increasing rates of production and/or yields of glutamate and lysine. The development of tools for genetic engineering (Kirchner & Tauch, 2003), the availability of the genome sequence (Kalinowski et al., 2003), genome-wide expression analyses using microarrays (Wendisch, 2003), proteome analyses (Hermann et al., 2001) and carbon-flux distribution analyses (Dominguez et al., 1998; Rollin et al., 1995) allowed a comprehensive knowledge of C. glutamicum metabolism during exponential growth and led to the rational improvement of C. glutamicum strains by metabolic design for the production of d-pantothenate, l-isoleucine, l-valine or l-threonine (Sahm et al., 2000; Hüser et al., 2005).

    Despite its described habitat, soil, which is unlikely to contain excess nutrients, only few studies have attempted to understand C. glutamicum metabolism under nutrient limitation. To our knowledge, only iron (Wennerhold et al., 2005), phosphate (Ishige et al., 2003), ammonium (Silberbach et al., 2005a) and nitrogen (Silberbach et al., 2005b) starvation, and glucose limitation (Cocaign-Bousquet et al., 1996), have been investigated. The effects of biotin limitation on C. glutamicum metabolism have not been studied to date, even though biotin was postulated to be essentially supplied by plant roots and then limiting in soils (Rovira & Harris, 1961).

    Biotin is an important cofactor for the activity of a family of enzymes that includes pyruvate carboxylases, acyl-CoA carboxylases, oxaloacetate decarboxylases and transcarboxylases (Samols et al., 1988; Toh et al., 1993). In micro-organisms, biotin is produced in four steps from l-alanine and pimeloyl-CoA (Eisenberg, 1987). In C. glutamicum, biotin auxotrophy is at least due to the lack of the bioF-encoded 7-keto-8-aminopelargonic acid synthetase (EC 2.3.1.47), the enzyme that catalyses the first step of biotin synthesis from l-alanine and pimeloyl-CoA. There is also no evidence of the presence of either the bioC and bioH genes, which are postulated to be involved in pimeloyl-CoA synthesis in Escherichia coli (Eisenberg, 1987), or bioI and bioW, involved in pimelic acid synthesis in Bacillus subtilis (Bower et al., 1996), or bioZ, involved in pimeloyl-CoA synthesis in Mesorhizobium sp. (Sullivan et al., 2001). On the other hand, the bioA, bioD and bioB genes are present and encode functional enzymes that are able to catalyse the last three steps (Hatakeyama et al., 1993a, b).

    In C. glutamicum, pyruvate carboxylase (EC 6.4.1.1), the major anaplerotic enzyme, and acetyl-CoA carboxylase (EC 6.4.1.2), which catalyses the committed step in fatty acid synthesis, have been shown to be biotinylated in vivo (Peters-Wendisch et al., 1998; Jäger et al., 1996). As the reactions catalysed by these two enzymes are closely linked to pyruvic acid, a rigid branch point of central metabolism (Vallino & Stephanopoulos, 1994) and the precursor of many important metabolites, biotin limitation is expected to lead to an important disturbance of central metabolism. In addition, this disturbance could be magnified by the expected biotin dependency of propionyl-CoA carboxylase β-chain (EC 6.4.1.3), the pccB-encoded carboxylase involved in mycolic acid synthesis (Portevin et al., 2005). Based on these assumptions of a possible metabolic disturbance at the pyruvate node during biotin-limited growth of C. glutamicum, we followed the growth features of C. glutamicum during biotin-limited batch cultures and focused our attention on the effect of biotin limitation on pyruvate redirection under these conditions. An unexpected mixed-acid fermentation occurred under aerobiosis without stress. We restricted our investigation to l-lactate synthesis and the expression and regulation of the l-lactate dehydrogenase (Ldh)-encoding ldh gene. In-depth analysis of ldh expression allowed us to identify SugR, the product of the cg2115 gene, as the transcriptional regulator of ldh. During the preparation of this manuscript, Engels & Wendisch (2007), Gaigalat et al. (2007) and Tanaka et al. (2008a) reported the identification of SugR as the transcriptional repressor of sugar PTS (phosphotransferase system) genes in C. glutamicum. None of these papers reported the role of SugR in ldh transcriptional regulation.

    METHODS

    Bacterial strains, growth conditions, metabolite analysis, plasmids and oligonucleotides.

    The bacterial strains and plasmids used in this study are listed in Table 1⇓. The sequences of oligonucleotides are listed in Table 2⇓. C. glutamicum strains were grown aerobically (250 r.p.m.) at 34 °C in Brain Heart Infusion (BHI) rich medium (Difco) or in GCXII chemically defined medium (Keilhauer et al., 1993) containing 0.02 mg biotin l−1 and either 2 % (w/v) glucose, 2 % (w/v) fructose, 4 % (w/v) sucrose, 2 % (w/v) acetate or a mixture of glucose and fructose (2 % each). Growth was followed by measuring the OD600 with a Beckman DU 7400 spectrophotometer. Time-course experiments were performed as follows. C. glutamicum strains were grown overnight in BHI medium. Cells were concentrated by centrifugation and resuspended in CGXII medium at an initial OD600 of 2. Samples were taken every 1.5 h. Experiments were done in triplicate and all the analyses were done in triplicate for each sampling time. Escherichia coli strains (DH5α and BL21(DE3)pLysS) were grown aerobically at 37 °C in Luria–Bertani medium (Difco). Antibiotics used were kanamycin (25 μg ml−1), chloramphenicol (10 μg ml−1) and ampicillin (100 μg ml−1). IPTG was used at 0.1 mM final concentration. All chemicals were purchased from Sigma-Aldrich. The dry mass was estimated by measuring the OD600 and assuming that an optical density equal to 1 corresponds to 0.34 mg dry mass ml−1 (Ruffert et al., 1997). The elemental formula of C. glutamicum biomass was taken to be C4.5H8O2.2N with a molecular mass of 121 Da when ash content (9 % total cell dry mass) was allowed for (Dominguez et al., 1998). Extracellular metabolites were measured with a Beckman System Gold HPLC. l-Lactate concentrations were determined enzymically with the lactate reagent kit from Sigma-Aldrich. d-Lactate was determined spectrophotometrically at 340 nm by combining 100 μl of the solution to be tested and 900 μl of 0.1 M sodium pyrophosphate buffer pH 8.5 containing 10 mM NAD+ and 2 U d-lactate dehydrogenase from Leuconostoc mesenteroides (Sigma-Aldrich). Glutamate concentration was determined enzymically (Sigma-Aldrich). Intracellular fructose 1,6-bisphosphate was measured as described by Dominguez et al. (1998). Emission was measured at 460 nm after exitation at 350 nm with an SFM 25A fluorescence spectrophotometer (BIO-TEK Instruments). Intracellular concentrations were expressed as molar concentrations using the intracellular water content value of 1.8 ml (mg dry mass)−1 as determined by Hoischen & Krämer (1989).

    Table 1.

    Strains and plasmids used in this study

    Table 2.

    Oligonucleotides used in this study

    DNA isolation, manipulation and transfer.

    All the molecular biology procedures used in this study were as described by Sambrook et al. (1989), Ausubel et al. (1987) and Merkamm & Guyonvarch (2001). C. glutamicum and E. coli cells were transformed by electroporation as described by Bonamy et al. (1990). Unless otherwise stated, enzymes were purchased from Promega.

    Construction of mutants.

    PCR experiments were carried out with either Taq DNA Polymerase (Promega) or High Fidelity PCR enzyme mix (Fermentas) and a GeneAmp PCR System 2700 thermocycler (Applied Biosystems). Chromosomal C. glutamicum DNA was used as template. PCR products were purified using the High Pure PCR Product Purification kit (Roche Diagnostics). All oligonucleotides were purchased from Sigma-Genosys. The genome sequence of C. glutamicum deposited in GenBank with the accession number NC_006958 was used throughout this work as the reference.

    To disrupt the chromosomal ldh gene (cg3219), an internal ldh fragment was amplified by PCR using primers Ldh-1 and Ldh-2. The purified 356 bp PCR fragment was cloned into the pCR2.1-TOPO cloning vector (Invitrogen) to create pMM29. Plasmid pMM29 was introduced into C. glutamicum by electroporation. Kanamycin-resistant transformants were the result of chromosomal integration by single cross-over events, as verified by Southern blotting. The same procedure was followed to disrupt the chromosomal ptsF gene (cg2120), with primers Fruca-1 and Fruca-2 (plasmid pAG1032) and the chromosomal fruR gene (cg2118) with primers FruR-1 and FruR-2 (plasmid pAG1035). As fruR is the first cistron of the cg2118-cg2119-cg2120 operon (Gaigalat et al., 2007), the orientation of the fruR fragment in pAG1035 was chosen in order to allow the expression of the cg2119-cg2120 genes under the control of Plac, then allowing the recombinant strain to grow on minimal media containing glucose or fructose as carbon sources.

    Defined chromosomal deletions of the four putative transcriptional-regulator-encoding genes were constructed with the pK18 mobsacB vector system (Schäfer et al., 1994) as follows. The DNA region upstream of the cg2115 gene encoding YP_226173 was amplified by PCR with oligonucleotides Regg1-1 and Regg1-2 (Table 2⇑). The DNA region downstream of cg2115 was amplified with oligonucleotides Regg1-3 and Regg1-4 (Table 2⇑). The PCR fragments were cloned independently into the pCR2.1-TOPO cloning vector. The upstream region was isolated as a KpnI/NcoI fragment and the downstream region as a NcoI/NheI fragment. These fragments were cloned into KpnI/NheI-digested vector pK18 mobsacB. The construction was transferred into C. glutamicum RES167 by electroporation. Allelic exchanges were performed as described by Schäfer et al. (1994). PCR experiments were performed to verify the chromosomal deletions. The same procedure was followed to delete the YP_227153-encoding cg3224 gene (oligonucleotides Regg2-1, Regg2-2 and Regg2-3, Regg2-4), the YP_224408-encoding cg0146 gene (Rega1-1, Rega1-2 and Rega1-3, Rega1-4) and the YP_227238-encoding cg3315 gene (Mar-1, Mar-2 and Mar-3, Mar-4).

    Construction of β-galactosidase (β-Gal) fusions.

    The promoter-probe vector pMF2 (Soual-Hoebeke et al., 1999) was used for construction of transcriptional fusions of the ldh promoter region or of the sugR-cg2116 promoter region and the promoterless lacZ gene. Plasmid pMF2 was first digested with EcoRI and BamHI and purified. The native promoter fragments were generated by PCR using primers Ldh-20 and Ldh-21 or Regg1-13 and Regg1-14 respectively, and cloned into pCR2.1-TOPO. The resulting plasmids were digested with EcoRI and BamHI, and the purified fragments were cloned in front of the lacZ gene in pMF2. Successful cloning was verified by sequencing (Genome Express). The resulting plasmids pAG1022 and pAG1030 allowed the expression of lacZ under the control of the ldh promoter and of the sugR-cg2116 promoter respectively and were introduced into C. glutamicum by electroporation. Kanamycin-resistant transformants were the result of chromosomal integration at the icd locus (cg0766) by single cross-over events, as verified by Southern blotting.

    The in vivo binding of SugR on Pldh was tested with the following construct. The 5′ part of Pldh was amplified by PCR with oligonucleotides Ldh-37 and Ldh-21. The 3′ part of Pldh was amplified by PCR with oligonucleotides Ldh-38 and Ldh-20. The PCR fragments were combined and used as templates for PCR amplification with oligonucleotides Ldh-20 and Ldh-21. The resulting fragment was digested with EcoRI and BamHI, and cloned in front of the lacZ gene in pMF2 as before. The resulting transcriptional fusion (pAG1033) was introduced into C. glutamicum by electroporation and the integrations events were confirmed as before.

    Enzyme assays and Ldh purification.

    Bacterial cells were harvested and resuspended in 50 mM sodium phosphate buffer (pH 6.6), 10 mM MgCl2 to 50 OD600 units. RNase A (1 mg ml−1) and DNase I (1 mg ml−1) were added when appropriate. Cells were disrupted in a French press (Spectronic Instruments) at 110 400 kPa. Cell debris was removed by centrifuging at 10 000 g for 15 min at 4 °C. The supernatant was ultracentrifuged at 130 000 g for 2 h at 4 °C. The resulting supernatant was used as the cytosoluble fraction. The protein concentration was measured by the Lowry method using bovine immunoglobulin G as a standard.

    Lactate dehydrogenase (EC 1.1.1.27) activity was assayed photometrically at 30 °C as described by Bergmeyer & Bernt (1974) in a slightly modified form. The reaction was performed in 50 mM sodium phosphate buffer pH 6.6, 5 mM MgCl2, 20 mM sodium pyruvate, 0.3 mM disodium NADH, 2 mM fructose 1,6-bisphosphate with a final concentration of 10 μg protein ml−1. Activity staining for lactate dehydrogenase after non-denaturing gel electrophoresis was performed according to Selander et al. (1986). β-Galactosidase (EC 3.2.1.23) activity was assayed as described by Miller (1972). The glutamate dehydrogenase (EC 1.4.1.4) and isocitrate dehydrogenase (EC 1.1.1.42) activities were assayed spectrophotometrically as described by Meers et al. (1970) and Nachlas et al. (1963) respectively. All the activities were measured in triplicate. Mean values are given with confidence interval (P=0.05).

    l-Lactate dehydrogenase purification was performed essentially according to Dartois et al. (1995). The ÄKTA FPLC device (Amersham Pharmacia) was used for chromatography. The cytosoluble fraction was fractionated with (NH4)2SO4. The enzyme precipitated between 45 and 60 % saturation. The precipitate was dialysed overnight against buffer A (50 mM Tris/HCl pH 7.5, 1 mM EDTA, 1 mM β-mercaptoethanol). The solution was loaded on a DEAE-Sephadex A-50 column pre-equilibrated with buffer A. Ldh was eluted using increasing concentrations from 0.1 to 1 M KCl in buffer A. Fractions displaying Ldh activity were pooled and concentrated with an ultrafiltration cell PM30 (Amicon). The concentrate was then loaded onto a Cibacron Blue column pre-equilibrated with 20 mM phosphate buffer pH 6.6. The column was washed and the enzyme was eluted by addition of 1 mM NADH in phosphate buffer pH 6.6. Coomassie blue staining after SDS-PAGE electrophoresis revealed a single band.

    DNA-affinity purification and identification of regulatory proteins.

    The purification of DNA-binding proteins was performed essentially as described by Gerstmeir et al. (2004). The ldh promoter/operator probe was generated by PCR using C. glutamicum chromosomal DNA and oligonucleotides Ldh-10 and Ldh-11, the former being tagged with biotin. Cytosoluble proteins from C. glutamicum cells grown on minimal medium with either acetate or glucose as a carbon source were prepared as described above and dialysed against binding buffer (Gerstmeir et al., 2004). For each condition, 10 mg samples of proteins were incubated with 1 mg coupled beads for 2 h at 25 °C. Unbound proteins were washed and DNA-bound proteins were eluted. Eluted proteins were concentrated and dialysed against 10 mM Tris/HCl pH 8.0, 1 mM EDTA buffer by ultrafiltration of the samples with Vivaspin 500 concentrators (Sartorius). SDS-PAGE was performed according to the method of Laemmli (1970) followed by Coomassie brilliant blue R-250-staining.

    Spots of interest were excised and analysed by MALDI-TOF mass spectrometry as described by Gerstmeir et al. (2004). Spectra were acquired on a Voyager DE-STR MALDI-TOF mass spectrometer (Applied Biosystems) equipped with a 337 nm nitrogen laser, in positive-ion reflector mode with delayed extraction. External calibration was applied, based on a mixture of six reference peptides covering the m/z 900–3700 Da range. Proteins were identified using the search programs PeptIdent () and Mascot (), reducing the Swiss-Prot/TrEMBL and NCBI databases to the Bacteria species.

    Construction and purification of a His-tagged SugR protein (SugRHis).

    A PCR product fusing the coding region of the C. glutamicum sugR gene (cg2115) with a nucleotide sequence encoding a C-terminal His-tag was obtained by using primers Regg1-5 and Regg1-6. The PCR product was digested with NdeI and NotI and cloned into the expression vector pET9sn1 (Trésaugues et al., 2004). The resulting plasmid pAG1025 was transferred to E. coli BL21(DE3)pLysS (Novagen) by electroporation. The His-tagged SugR protein was purified with Ni-NTA spin columns (Qiagen) according to the manufacturer's protocol. SDS-PAGE was performed to control purity of the His-tagged SugR protein (SugRHis).

    To control the in vivo functionality of SugRHis, pAG1025 was digested by SalI and the digested plasmid was self-ligated to give pAG1034. This led to the deletion of both the T7 promoter and the first 20 nucleotides of sugR in pAG1025. The plasmid pAG1034 was then introduced into C. glutamicum by electroporation. In the recombinant strain, the sugRHis gene was expressed as a single copy under the control of the native sugR promoter.

    Identification of transcriptional starts by the RACE method.

    Total RNA from glucose-grown cells was prepared as described by Merkamm & Guyonvarch (2001) and incubated in 0.1 M sodium acetate pH 3.0, 5 mM MgSO4 and 0.3 U DNase I ml−1 (Roche Diagnostics) for 30 min at 37 °C. After heat inactivation for 5 min at 75 °C and phenol/chloroform extraction, the RNA was precipitated with LiCl as described by Sambrook et al. (1989). Primers (Ldh-32 and Regg1-10) binding downstream of the putative translational starts of the ldh and sugR genes respectively along with 2 μg total RNA were used for cDNA synthesis. The cDNAs were then modified and amplified by two further PCRs with nested primers (Ldh-31 and Ldh-30 or Regg1-9 and Regg1-8) using the 5′/3′ RACE kit second generation (Roche Diagnostics) according to the manufacturer's protocol. The PCR products were purified and cloned into the pCR2.1-TOPO vector. Three different clones from each experiment were selected for plasmid preparation and DNA sequencing (Genome Express).

    Transcriptional organization of the sugR-cg2116 locus.

    cDNAs were obtained as described before with primer Regg1-12. The cDNAs were further amplified by PCR with oligonucleotides Regg1-12 and Regg1-3 or Regg1-12 and Regg1-11. PCR products were analysed by agarose gel electrophoresis.

    Electrophoretic mobility shift assays (EMSAs).

    Standard EMSA studies were performed essentially as described by Rey et al. (2005), with appropriate PCR products and purified His-tagged SugR protein (SugRHis). The samples were loaded onto a 2 % non-denaturing agarose gel prepared in gel buffer (10 mM Tris/HCl pH 7.8, 10 mM sodium acetate, 1 mM EDTA) and separated at 50 V for 2 h. DNA and DNA–protein complexes were visualized under UV light after ethidium bromide staining of the gels. Effector screening studies were performed as described by Gaigalat et al. (2007). After electrophoresis and ethidium bromide staining, the gels were scanned with a Typhoon TRIO Variable Imager (GE Healthcare).

    Bioinformatic tools for the interpretation of the C. glutamicum genomic sequences.

    Orthologues of Ldh and SugR proteins in three other Corynebacterium species, namely C. efficiens YS-314 [GenBank accession NC_004369 (Nishio et al., 2003)], C. diphtheriae NCTC 13129 [GenBank accession NC_002935 (Cerdeño-Tárraga et al., 2003)] and C. jeikeium K411 [GenBank accession NC_007164 (Tauch et al., 2005)], were searched for using the protein–protein blast program (Altschul et al., 1990) with Best Reciprocal Hit method.

    A search in the whole C. glutamicum genome for DNA motifs similar to the SugR-binding motif found in the ldh promoter was performed using the Scan For Matches software (Dsouza et al., 1997) and specific Perl scripts to select motifs located in intergenic regions and oriented like the one in the ldh promoter.

    Promoter regions of putative target genes of SugR in C. glutamicum (ptsG, sugR-cg2116, ptsI, cg2118-ptsF, ptsH, ptsS, scrB, ldh), C. efficiens (CE1824, CE1827, ldh) and C. diphtheriae (DIP1427, DIP1429, ldh) were compared with the meme software (Bailey & Elkan, 1994) in order to determine a consensus motif for the SugR-binding site.

    RESULTS AND DISCUSSION

    Batch-culture characteristics during biotin-limited growth on glucose

    The growth of C. glutamicum RES167 (Dusch et al., 1999) on glucose was investigated in biotin-limited batch cultures. A typical batch fermentation profile is shown in Fig. 1⇓. The fermentation profile was very different from that observed during growth with excess biotin, and showed three distinct phases. In all these phases, growth appeared to be arithmetic rather than exponential, as the increase in biomass concentration was a linear function of time. During the initial phase of growth, glucose was consumed at a specific rate (qglc) of 1.53±0.05 mmol (g dry cells)−1 h−1 and a growth rate of 0.558±0.03 g dry cells h−1 l−1 was maintained. This corresponded to a rate of carbon assimilation into biomass (YX/C) of 12.485±0.96 g dry cells (mol carbon consumed)−1. Acetic acid [γace=0.678±0.04 mmol (g dry cells)−1 h−1], l-lactic acid [γlac=0.382±0.01 mmol (g dry cells)−1 h−1] and trace amounts of pyruvic acid were produced during this first phase. No glutamate was detected in the culture medium. This was not surprising because the culture conditions used in this study were significantly different from those used for glutamate production under biotin limitation (Choi et al., 2004). The most noticeable phenomenon observed was the overflow of organic acids during the growing phase without oxygen limitation and no heat or nutritional stress. To date, organic acid production by C. glutamicum has been mainly observed under oxygen deprivation, at the onset of the stationary phase or after heat shock during continuous cultivation (Dominguez et al., 1993; Okino et al., 2005; Inui et al., 2004, Uy et al., 2003; Stansen et al., 2005). During growth on glucose with excess biotin, qglc was 4.44 mmol (g dry cells)−1 h−1 and YX/C was 11.02 g dry cells (mol carbon consumed)−1, as recalculated from data published by Gerstmeir et al. (2003). In biotin excess, the specific rate of glucose consumption was significantly higher than that in biotin limitation. As the specific activity of the glucose-specific PTS was shown to be constant under various conditions with glucose as a carbon source (Cocaign-Bousquet et al., 1996), and no inhibition of glucose PTS enzymes was reported in C. glutamicum, the low qglc value in our study was attributed to a low glycolytic flux due to the bottleneck created by biotin limitation. Acetate and lactate excretion was the probable means to avoid intracellular accumulation of pyruvic acid, a direct consequence of biotin limitation, and ensure reoxidation of NADH produced by glyceraldehyde-3-phosphate dehydrogenase (EC 1.2.1.12). Acetate and lactate were assumed to be the products of pyruvate metabolism by pyruvate oxidase (EC 1.2.2.2) and lactate dehydrogenase respectively, as described for Lactobacillus plantarum when cultivated with carbon sources supporting low growth rates under aerobic growth conditions (Sedewitz et al., 1984). As described for C. glutamicum R under oxygen-deprived conditions, phosphotransacetylase (EC 2.3.1.8), acetate kinase (EC 2.7.2.1) and CoA transferase (EC 2.8.3.–) could also be involved in acetate production (Yasuda et al., 2007). The second phase extended from 7.5 to 11 h of cultivation. Growth rate decreased (0.194±0.02 g dry cells h−1 l−1), while qglc slightly increased [2.98±0.04 mmol (g dry cells)−1 h−1] and acetic acid was consumed and exhausted at a high specific rate [qace=8.38±0.1 mmol (g dry cells)−1 h−1]. l-Lactic acid was also consumed but at a low specific rate [qlac=0.815±0.03 mmol (g dry cells)−1 h−1]. Consumption of acetic acid and l-lactic acid to a lesser extent occurred when significant amounts of glucose were still present in the culture medium (10 g l−1). The slower rate of growth was attributable to the decreased rate of carbon assimilation into biomass [YX/C=7.73±0.37 g dry cells (mol carbon consumed)−1] due to the use of acetate as the prefered carbon source. During this phase, qglc and qace were similar to the corresponding values determined by Gerstmeir et al. (2003) with a mixture of glucose and acetate as carbon sources and biotin excess [qglc=2.16 mmol (g dry cells−1) h−1 and qace=8.1 mmol (g dry cells−1) h−1]. Acetate consumption was then probably mediated by the standard acetate utilization pathway that bypasses the pyruvate bottleneck (Gerstmeir et al., 2003).

    Figure image not available in archive
    Fig. 1.

    Features of C. glutamicum RES : : pAG1022 growth during biotin-limited batch cultures on glucose. (a) Biomass (□, g dry cells), glucose concentration (▪, % w/v), lactate concentration (▴, g l−1), acetate concentration (•, g l−1). (b) l-Lactate dehydrogenase specific activity [▵, mmol min−1 (mg protein)−1], β-Galactosidase activity [○, 10 U (mg protein)−1], lactate concentration (▴, g l−1), intracellular fructose 1,6-bisphosphate concentration (◊, mM). Each point corresponds to the mean value from nine measurements. Standard deviations are indicated by error bars.

    During the third phase, from 11 to 16.5 h of cultivation, growth rate decreased (0.113±0.02 g dry cells h−1 l−1) and glucose and l-lactic acid were progressively consumed with specific rates of 3.87±0.03 mmol (g dry cells)−1 h−1 and 2.12±0.4 mmol (g dry cells)−1 h−1 respectively. Acidification of the medium occurred and pH fell to about 5. During this phase, YX/C was slightly higher than during the second phase [9.98±0.05 g dry cells (mol carbon consumed)−1]. After 25 h cultivation, glucose and l-lactic acid were exhausted from the medium and no more growth occurred. l-Lactic acid consumption was essentially catalysed by Lld, the membrane-bound and FMN-dependent lactate dehydrogenase (EC 1.1.2.3) encoded by cg3227 (data not shown).

    l-Lactate dehydrogenase in C. glutamicum

    Of the phenomena seen during biotin-limited growth on glucose, l-lactic acid overflow was the most intriguing to us. We therefore focused our work on the identification and characterization of the C. glutamicum gene and the corresponding protein involved in l-lactic acid production under these conditions. Whole-genome screening enabled us to focus on the annotated lactate dehydrogenase-encoding genes cg0763, cg1027, cg3219 and cg3227. Of these genes, cg3219 was the best candidate as the corresponding protein YP_227149 exhibited all the features expected for a cytoplasmic, l-lactic acid-producing, l-lactate dehydrogenase (Clarke et al., 1989). The cg3219 gene was disrupted and the behaviour of the corresponding mutant (RES167 : : pMM29) was investigated in biotin-limited batch cultures as previously described. The growth profile of the Ldh-deficient mutant was similar to that of the wild-type strain. l-Lactic acid was no longer synthesized but acetic acid production was doubled (2.9±0.53 g l−1). In addition, no Ldh activity was detected in crude extracts obtained from the mutant strain. This confirmed the conclusion of Inui et al. (2004) that cg3219 was the sole l-lactate dehydrogenase-encoding ldh gene and was not essential for growth on glucose. Growth of this mutant and of the parental wild-type strain were compared in biotin-limited batch cultures with l-lactate as a carbon source. No difference was seen between the two strains, indicating that ldh was also not essential for growth on l-lactate.

    Enzymic characteristics of the C. glutamicum Ldh were analysed. Substrate specificity assays showed that the enzyme was specific for the reduction of pyruvic acid to l-lactic acid, with NADH as the exclusive cofactor. The optimal pH for the reaction was 6.6. No reaction was detected when the enzyme was assayed for l-lactic acid oxidation, irrespective of the pH tested and the cofactor used. Moreover, no signal was detected when activity staining was performed (Selander et al., 1986). These results indicated that Ldh operated only in the direction of pyruvic acid reduction to l-lactic acid. To our knowledge, only the Clostridium acetobutylicum Ldh has previously been experimentally demonstrated to be irreversible (Contag et al., 1990). The Km value was calculated as 7.3 mM for pyruvic acid and the Vmax value was calculated as 120 μmol min−1. Out of the different effectors tested, only fructose 1,6-bisphosphate had a marked effect on the reaction. In the presence of 10 mM fructose 1,6-bisphosphate, Vmax increased to 450 μmol min−1. This activation of Ldh by fructose 1,6-bisphosphate has been described for the enzymes of a great number of Gram-positive bacteria (Garvie, 1980) and was not specific to the C. glutamicum enzyme.

    Time-course of expression of ldh

    The time-course expression of ldh was followed during biotin-limited growth (Fig. 1⇑). As Muffler et al. (2002) suggested an upregulation of ldh during growth on glucose when compared to acetate, strain RES167 : : pAG1022 was cultivated in biotin-limited conditions with either glucose or acetate. Both Ldh and β-Gal specific activities were measured. During growth on glucose, Ldh specific activity increased from 0.3±0.02 mmol min−1 (mg protein)−1 to about 3.1±0.04 mmol min−1 (mg protein)−1 during the first 9 h and progressively decreased after to a value of 2.6±0.05 mmol min−1 (mg protein)−1. Simultaneously, β-Gal specific activity increased from 23.7±1.3 U (mg protein)−1 to 88.3±2.7 U (mg protein)−1 during the first 9 h and then progressively decreased to a value of 64±0.5 U (mg protein)−1. The maxima of both Ldh and β-Gal specific activities corresponded to the moment when l-lactic acid consumption began. These results indicated a transcriptional induction of ldh during the first 9 h followed by a partial transcriptional repression. Even if a putative post-translational regulation of Ldh was possible, as the C. glutamicum Ldh was shown to be susceptible to serine phosphorylation under certain conditions (Bendt et al., 2003), this phenomenon was unlikely to be involved in this case. During growth on acetate, the levels of Ldh [0.4±0.01 mmol min−1 (mg protein)−1] and β-Gal [15.3±0.8 U (mg protein)−1] remained low throughout the cultivation time and no l-lactic acid was detected in the culture medium. Glutamate dehydrogenase (Gdh) and isocitrate dehydrogenase (Icd) activities were measured as internal controls because both activities were constant as a function of time and almost identical irrespective of the use of acetate or glucose as a carbon source (Börmann et al., 1992; Eikmanns et al., 1995). The values determined were 2.05±0.3 and 2.1±0.2 mmol NADPH oxidized min−1 (mg protein)−1 for Gdh during growth on glucose and acetate respectively and 1.87±0.1 and 1.86±0.1 mmol NADP reduced min−1 (mg protein)−1 for Icd during growth on glucose and acetate respectively. It was thus concluded that ldh regulation was specific and not the result of a general change in the transcriptional profile.

    The time-course regulation of ldh and the l-lactate variations in the culture medium could be explained as follows. During the first hours of cultivation, the pool of glycolytic intermediates, including fructose 1,6-bisphosphate and pyruvate, was high due to the bottleneck created at the pyruvate node by the low flux supported by pyruvate carboxylase under biotin limitation. Of the glycolytic intermediates accumulated, one was probably involved somehow in the increase of ldh expression. Biotin limitation also led to an increase of the NADH/NAD ratio, due to glyceraldehyde-3-phosphate dehydrogenase activity and a suspected diminished flux through the Krebs cycle. Both fructose 1,6-bisphosphate and a high NADH/NAD ratio have been described as activators of Ldh activity (Garvie, 1980; Garrigues et al., 1997). Increase in ldh expression and Ldh activation then allowed the efficient conversion of pyruvate into lactate and a progressive decrease of the intracellular concentration of glycolytic intermediates and the NADH/NAD ratio. This led to a partial repression of ldh expression and a diminished flux through Ldh. This hypothesis was substantiated, as the intracellular concentration of fructose 1,6-bisphosphate actually decreased progressively during the cultivation time as shown in Fig. 1(b)⇑. After 9 h cultivation, l-lactic acid consumption was essentially catalysed by Lld, the membrane-bound and FMN-dependent lactate dehydrogenase encoded by cg3227, expression of which was induced by the excreted l-lactic acid (Georgi et al., 2008). During this second phase, the reducing equivalents that had been used first for l-lactic acid synthesis were redirected to the respiratory chain to gain ATP and pyruvic acid was redirected to glycolysis.

    Identification of the ldh transcriptional regulator

    In order to explain the increase in ldh transcription during growth on glucose and its low, constitutive transcription during growth on acetate, we had to postulate the existence of a transcriptional regulator binding to the ldh promoter. A DNA-affinity approach (Gabrielsen et al., 1989; Rey et al., 2003; Engels et al., 2004) was used to purify such a protein, with the ldh–cg3220 intergenic region as a target. Protein purification was performed with dialysed crude extracts from cells grown on glucose or acetate, harvested after 7.5 h cultivation when lactate reached its maximum concentration on glucose. Two proteins from the extract of acetate-grown cells (YP_224408 and YP_226173) and three from the extract of glucose-grown cells (YP_226173, YP_227153 and YP_227238) appeared to bind efficiently to the fragment, as shown in Fig. 2⇓. It is noteworthy that all these proteins can bind the DNA fragment in the absence of added metabolites. Unequivocal protein identification was performed by MALDI-TOF mass spectrometry as described in Methods. All these proteins were predicted to be potential transcriptional repressors, protein YP_226173 being present in both extracts. As the target DNA fragment contained both the ldh promoter and the promoter of the diverging cg3220 putative gene, it was impossible to unambiguously identify the ldh regulator. Defined in-frame deletion mutants for each of the candidate genes (cg2115, cg 3224, cg0146, cg3315) were established by a gene replacement strategy (Schäfer et al., 1994). As all these genes were predicted to belong to operon structures (Baumbach et al., 2007), extreme care was taken to avoid polar effects of these deletions. After transformation of the four deleted strains with pAG1022, β-Gal (data not shown) and Ldh activities were analysed in extracts from glucose-grown or acetate-grown cells. Ldh specific activity in the Δcg2115 strain was about 12 mmol min−1 (mg protein)−1 irrespective of the carbon source, whereas Ldh specific activity was not affected in the other mutant strains (Δcg3224, Δcg0146 and Δcg3315) when compared to the wild-type one. These results clearly show that the YP_226173 protein encoded by cg2115 was the repressor of ldh. Time-course analysis of ldh expression in the Δcg2115 strain grown on glucose was investigated. The Ldh specific activity remained constant, indicating that YP_226173 was also responsible for the transcriptional repression observed after 9 h cultivation. The cg2115 gene was recently identified as sugR, SugR being the repressor of ptsG, ptsF, ptsI, ptsH and ptsS, the genes encoding the glucose-, fructose- and sucrose-specific PTSs and the general components EI and Hpr (Engels & Wendisch, 2007; Gaigalat et al., 2007; Tanaka et al., 2008a). Comparison of Ldh activities in extracts from the wild-type [3.1±0.04 mmol min−1 (mg protein)−1] and the ΔsugR [12±0.7 mmol min−1 (mg protein)−1] strains showed that ldh was not completely derepressed during growth on glucose.

    Figure image not available in archive
    Fig. 2.

    SDS-PAGE of C. glutamicum RES167 proteins eluted from a DNA-affinity chromatography experiment using the ldh promoter/operator as a probe and cell extracts from glucose-grown cells (lane 1) and acetate-grown cells (lane 3). Lane 2, molecular mass marker.

    Recently, Georgi et al. (2008) have shown that YP_227153 corresponded to LldR, the transcriptional regulator of lld (cg3227) that encodes the membrane-bound l-lactate dehydrogenase (EC 1.1.2.3) involved in l-lactate consumption in C. glutamicum (Stansen et al., 2005). Georgi et al. (2008) have shown that the sequence 5′-TGGTCTGACCA-3′ was essential for the binding of LldR on the cg3226-lld promoter. The presence of the highly similar motif 5′-TGGTCAGACCA-3′ in the ldh–cg3220 intergenic region sequence, overlapping the –35 region of Pldh, explained the binding of LldR in our experiment. As described before, the deletion of lldR (cg3224) does not affect ldh expression, indicating that LldR is not involved in ldh regulation.

    Given the presence of the genes coding for general PTSs EI (pstI) and HPR (ptsH), fructose-specific PTS EII (ptsF), 1-phosphofructokinase (pfkB), the transcriptional regulator FruR (fruR) and a putative phosphofructokinase (cg2116) in the chromosomal vicinity of sugR, fructose was further considered as an additional carbon source of interest to study ldh expression. Ldh activities were comparable in extracts from cells grown on fructose [11.2±0.5 mmol min−1 (mg protein)−1], irrespective of the genetic background. This indicated that ldh was fully derepressed when cells were grown in the presence of fructose and/or in the absence of SugR.

    As fruR was a paralogue of sugR (DeoR-like protein, 51 % identity at the protein level with SugR) and was proven to be involved in the transcriptional regulation of ptsI, ptsH and the cg2118-cg2119-cg2120 operon (Tanaka et al., 2008b), genes also under the control of SugR, we tested the effect of the FruR deficiency on the ldh expression level. C. glutamicum RES : : pAG1035 was cultivated with glucose or fructose as carbon sources and the Ldh activity was measured in both conditions. When compared to Ldh activities from the wild-type strain grown in the same conditions, no difference was seen, indicating that FruR was not involved in ldh regulation.

    Because of the common features of fructose and sucrose metabolism in C. glutamicum (Dominguez & Lindley, 1996), Ldh specific activity was also measured in extracts from cells grown on sucrose [11.5±0.6 mmol min−1 (mg protein)−1]. In these conditions, ldh was also derepressed.

    Growth of the ΔsugR mutant was similar to that of the wild-type strain on minimal medium containing acetate, glucose, fructose and sucrose, indicating that sugR was not essential for growth on these different carbon sources, as mentioned by Engels & Wendisch (2007).

    Identification of the SugRHis-binding site

    To demonstrate direct binding of the SugR protein to the ldh promoter by band-shift assays, a SugRHis protein was constructed and purified as described in Methods.

    To test the in vivo functionality of SugRHis in C. glutamicum, the RES : : pAG1034 strain was constructed as described in Methods. This strain and the wild-type strain were grown on glucose and on fructose, and Ldh specific activities were measured and compared. Ldh specific activities were 3.1±0.04 mmol min−1 (mg protein)−1 and 11.2±0.5 mmol min−1 (mg protein)−1 for RES167 : : pAG1034 grown on glucose and fructose respectively and 2.9±0.1 mmol min−1 (mg protein)−1 and 10.9±0.4 mmol min−1 (mg protein)−1 for RES grown on glucose and fructose respectively. This indicated that the recombinant SugRHis was fully functional in vivo in C. glutamicum.

    DNA band-shift assays were performed with the purified SugRHis protein from E. coli and a DNA fragment corresponding to the full-length ldh-cg3220 intergenic region. The 205 bp DNA was obtained by PCR with oligonucleotides Ldh-10 and Ldh-11, and 1 pmol of this fragment was incubated with different amounts of SugRHis protein (0, 30, 60, 90, 120 and 150 pmol). The assays were then separated by non-denaturing agarose gel electrophoresis and visualized by fluorescence after ethidium bromide staining of gels (Fig. 3a). The presence of purified SugRHis protein caused a complete mobility shift at the concentration of 90 pmol, a 90-fold molar excess that probably represented only a 10-fold molar excess of native protein, as purified DeoR-type repressors seemed to be mainly octamers (Zeng et al., 2000; Mortensen et al., 1989).

    To determine the exact location of the SugR-binding site in relation to the ldh promoter, the transcriptional start site of ldh was identified by using the 5′ RACE method. The transcriptional start site was located 88 bp upstream of the annotated start codon. The determination of the transcriptional start site allowed the identification of −35 (5′-TGACCA-3′) and −10 (5′-CATAAT-3′) regions, the sequences of which were in good accordance with the features of corynebacterial promoter sequences (Pátek et al., 2003). Interestingly, a 28 bp DNA sequence overlapping the −35 and −10 regions of the C. glutamicum ldh promoter exhibited 93 % sequence similarity to the corresponding region of the C. efficiens ldh gene, whereas the surrounding sequences shared only 23 % sequence similarity.

    To obtain a more precise location of the SugR-binding site, EMSAs were performed with three subfragments of the ldh promoter. These fragments were obtained by PCR with the oligonucleotides Ldh-26 and Ldh-11, Ldh-24 and Ldh-11 or Ldh-27 and Ldh-11. With respect to the transcriptional start site, these fragments extended from −60 to +67, from −35 to +67 and from −10 to +67 respectively. As a control, an internal fragment of ldh was amplified with oligonucleotides Ldh-1 and Ldh-2 and tested for SugR binding. The results (Fig. 3b⇓), summarized in Fig. 4(a)⇓, showed that the SugR-binding site on the ldh promoter overlapped the −35 and −10 regions of the ldh promoter, a typical location of DNA-binding motifs for repressors (Gralla & Collado-Vides, 1996).

    Figure image not available in archive
    Fig. 3.

    Binding of SugRHis to various target DNA fragments. (a) EMSA using the ldh promoter region (1 pmol) from C. glutamicum (PCR fragment obtained with oligonucleotides Ldh-10 and Ldh-11) and increasing amounts of purified His-tagged SugR protein (SugRHis). Lanes: 1, control with no added SugRHis; 2–6, increasing amounts of SugRHis (30–150 pmol, in 30 pmol steps). (b) EMSA using subfragments of the ldh promoter region (0.5 pmol), control DNA (0.5 pmol) from C. glutamicum and SugRHis (50 pmol). Lanes: 1, Ldh-27–Ldh-11 fragment (SugRHis heat inactivated); 2, Ldh-27–Ldh-11 fragment (native SugRHis); 3, Ldh-24–Ldh-11 fragment (SugRHis heat inactivated); 4, Ldh-24–Ldh-11 fragment (SugRHis heat inactivated); 5, Ldh-26–Ldh-11 fragment (native SugRHis); 6, Ldh-26–Ldh-11 fragment (SugRHis heat inactivated); 7, Ldh-1–Ldh-2 (control DNA) fragment (SugRHis heat inactivated); 8, Ldh-1–Ldh-2 fragment (native SugRHis).

    Figure image not available in archive
    Fig. 4.

    The organization of promoters of genes containing determined or predicted SugR-binding motifs, and construction of a consensus sequence of the binding motifs. The double-stranded sequence of each promoter is shown. The predicted −10 and −35 promoter regions and the transcriptional start sites are in bold. The translational start codons of the corresponding genes are underlined. The experimentally determined SugR-binding motifs are shown as grey boxes with a black border. The SugR-binding motifs predicted by sequence comparisons are shown as transparent boxes with a black border. A frequency plot of the deduced consensus sequence of all SugR-binding motifs predicted by sequence comparison constructed by means of the WebLogo tool (Crooks et al., 2004) is also shown. The overall height of each stack of letters indicates the sequence conservation at each position of the motif, whereas the height of each symbol within the stack reflects the relative frequency of the corresponding nucleotide at that position. (a) Comparisons of ldh promoters in C. glutamicum, C. efficiens and C. diphtheriae. (b) Comparisons of ldh, cg2118-ptsF, ptsS, sugR-cg2116, ptsH and ptsG promoters in C. glutamicum.

    Comparing the ldh promoter region in Corynebacterium species with meme software revealed a consensus motif of 27 bp (Fig. 4a⇑), which strictly corresponded to the motif identified from EMSAs, in the C. glutamicum ldh promoter. The C. efficiens and C. diphtheriae motifs were strongly conserved (respectively 89 % and 82 % identity) and displayed the same structure as the C. glutamicum one. For those three species the motif began between the putative −35 and −10 regions and totally overlapped the −10 region. The motif found for C. jeikeium had low identity (48 %) and its structure was not coherent with that of the C. glutamicum motif.

    Using Scan For Matches software, we found in the promoter regions of the ptsG, cg2118-ptsF and sugR-cg2116 genes of C. glutamicum motifs partially similar to the 25 bp motif experimentally identified in the ldh promoter (Fig. 4b⇑). For the ptsG promoter, only one of the motifs we found partially overlapped the motif found by Engels & Wendisch (2007). For the cg2118-ptsF promoter, both motifs we found more or less overlapped the motifs found by Gaigalat et al. (2007).

    The analysis of putative targets of SugR in C. glutamicum genome revealed a 9 bp consensus motif (5′-TCGGACATA-3′) for its binding site, which was totally or partially conserved in Pldh (identity 100 %), PptsG (100 %), PptsS (89 %), PptsH (89 %), PsugR-cg2116 (67 %), Pcg2118-ptsF (67 %), PptsI (55 %), but was not present in PscrB (Fig. 4b⇑). This 9 bp consensus motif was consistent with the SugR-binding site experimentally determined on a ptsI-fruR promoter fragment by Tanaka et al. (2008a). For Pldh, PptsG and Pcg2118-ptsF, this 9 bp motif was a part of the experimentally determined motif. Thus, it was likely that the SugR-binding sites contained this 9 bp motif.

    From these results two models of structure for SugR-binding sites were built. The first one came from the ldh, sugR-cg2116, cg2118-ptsF, ptsH and ptsS promoters, and consisted of about 20–25 bp, more or less overlapping the −35/−10 region. The second one came from the ptsG promoter and consisted of about 25 bp located upstream from the −35 region and which did not overlap the −35/−10 region of the promoter. Nevertheless, in each case, the locations of DNA-binding motifs are typical of those observed for repressors (Gralla & Collado-Vides, 1996).

    To confirm our hypothesis on the SugR-binding site on Pldh in C. glutamicum, we constructed pAG1033, a derivative of pAG1022 (lacZ under the control of the ldh promoter) in which the 9 bp consensus binding site 5′-TCGGACATA-3′ described in Fig. 4(b)⇑ was replaced by 5′-AGCCTGTAT-3′. Strains RES167 : : pAG1022 and RES167 : : pAG1033 were grown on glucose or fructose as carbon source and β-Gal activities were measured. β-Gal activities were 83.2±2.5 U (mg protein)−1 and 231±7 U (mg protein)−1 for RES167 : : pAG1022 grown on glucose and fructose respectively; 187±8 U (mg protein)−1 and 220±7.5 U (mg protein)−1 for RES167 : : pAG1033 grown on glucose and fructose respectively. This clearly indicated that the 9 bp consensus binding site 5′-TCGGACATA-3′ was involved in the in vivo binding of SugR to the ldh promoter.

    Note that in PptsS and Pcg2118-ptsF, the motifs similar to the 9 bp consensus motif we identified were quite different from those found by Engels & Wendisch (2007). The motif they mentioned for Pcg2118-ptsF (5′-TGTGCAAC-3′) was 250 bp upstream of the one we identified, and for PptsS the sequence 5′-TGTGCAAC-3′ they mentioned was not detectable.

    Another interesting feature was that the sugR and the cg2118 coding sequences exhibit a strong nucleotide similarity in a 41 bp region included in the HTH_DEOR-encoding domains of both genes. Moreover, this region contains the 9 bp consensus motif for SugR binding on Pldh. This feature has been shown to exist in various genes encoding HTH regulatory proteins (Roy et al. 2002), the internal binding motif being involved in autoregulation of some of these regulators.

    Features of sugR transcription and expression

    As the locus encompassing sugR and cg2116 was predicted to be an operon (Baumbach et al., 2007), its transcriptional organization was analysed by RT-PCR. Total RNA isolated from C. glutamicum was transcribed into cDNA by using primer Regg1-12. The resulting cDNA was used for PCR amplification with primers Regg1-12 and Regg1-3 or Regg1-12 and Regg1-11. The cDNA obtained with Regg-12, annealing in cg2116, allowed the amplification of not only a cg2116 internal fragment with Regg1-12 and Regg1-3, but also a fragment overlapping sugR and cg2116 with Regg1-12 and Regg1-11. Both fragments were at the expected size. No PCR product was obtained when reverse transcriptase was omitted. From this experiment we concluded that sugR and cg2116 were organized as an operon, an observation that was also reported by Gaigalat et al. (2007).

    The transcriptional start site of the sugR-cg2116 operon was identified by using the 5′ RACE method. The transcriptional start site was located 48 bp upstream of the annotated start codon. The sequences of the −35 (5′-TGAGGC-3′) and −10 (5′-GACAAT-3′) regions were also in accordance with the features of corynebacterial promoter sequences (Pátek et al., 2003).

    A transcriptional fusion with the sugR-cg2116 promoter and a promoterless lacZ gene was constructed. The resulting plasmid pAG1030 was introduced in C. glutamicum as described in Methods. Strain RES167 : : pAG1030 was cultivated in biotin-limited batch cultures on acetate, glucose, fructose or sucrose, and β-Gal specific activities were measured. Specific activities were 4.86±0.4 U (mg protein)−1, 4.68±0.3 U (mg protein)−1, 4.73±0.3 U (mg protein)−1 and 9.32±0.5 U (mg protein)−1 in extracts from glucose-, fructose-, sucrose- and acetate-grown cells respectively. This indicated that SugR was present in all the conditions tested, as suggested by the DNA-affinity experiment described for the isolation of SugR (Fig. 2⇑). From these measurements we also concluded that the sugR-cg2116 operon was subject to transcriptional regulation during growth on acetate. The same result was obtained by Gaigalat et al. (2007) using real-time RT-PCR.

    Identification of SugR co-inducers

    To explain the differences in ldh expression in extracts from acetate-grown, glucose-grown, sucrose-grown and fructose-grown cells, we postulated that SugR binding on Pldh was affected by inducer molecules derived from sugar metabolism.

    We therefore tested the in vitro prevention of SugR binding to the ldh promoter (PCR fragment obtained with oligonucleotides Ldh-10 and Ldh-11) with several metabolites, including fructose 1-phosphate, fructose 6-phosphate, fructose 1,6-bisphosphate and glucose 6-phosphate, that have been previously shown (Engels & Wendisch, 2007; Gaigalat et al., 2007) to prevent SugR binding to various pts promoters.

    Fructose 1-phosphate fully inhibited SugR binding to Pldh at concentrations ≥100 μM (Fig. 5a⇓), as did fructose 1,6-bisphosphate at concentrations ≥10 mM (Fig. 5b⇓). No effect was observed by using up to 20 mM of either fructose 6-phosphate (Fig. 5c⇓) or glucose 6-phosphate (Fig. 5d⇓). In addition, no influence on SugR binding was observed by using glyceraldehyde 3-phosphate, phospho(enol)pyruvate, pyruvate, glucose, fructose, sucrose, lactate or acetate (data not shown).

    Figure image not available in archive
    Fig. 5.

    Effector screening studies. EMSAs were conducted using the ldh promoter region (0.05 pmol) from C. glutamicum (PCR fragment obtained with oligonucleotides Ldh-10 and Ldh-11) incubated with 5 pmol purified SugRHis and increasing concentrations of different putative effectors. Effectors tested were: (a) fructose 1-phosphate, (b) fructose 1,6-bisphosphate, (c) fructose 6-phosphate and (d) glucose 6-phosphate. The non-complexed ldh promoter was used as a control (lane P).

    Fructose 1-phosphate could be considered the specific metabolite of fructose and sucrose metabolism in C. glutamicum for the following reasons. Fructose uptake in C. glutamicum involves two PTSs enabling phospho(enol)pyruvate-dependent transport and phosphorylation of fructose to give fructose 1-phosphate (PTSFru) and fructose 6-phosphate (PTSGlc). Fructose 1-phosphate is then converted to fructose 1,6-bisphosphate by the cg2119-encoded 1-phosphofructokinase (Moon et al., 2005). Partitioning between the two systems was estimated as follows: fructose uptake due to PTSFru=3.5 mmol (g dry cells)−1 h−1 and fructose uptake due to PTSGlc=0.9 mmol (g dry cells)−1 h−1 (Dominguez et al., 1998). Sucrose metabolism in C. glutamicum involves a phospho(enol)pyruvate-dependent sucrose PTS to give intracellular sucrose 6-phosphate, and a sucrose-6-phosphate hydrolase that yields glucose 6-phosphate and fructose (Engels et al., 2008). In the absence of fructokinase, further complete sucrose metabolism requires efflux of fructose and its reassimilation via the fructose PTS and the glucose PTS to a lesser extent as previously described (Dominguez & Lindley, 1996; Moon et al., 2005).

    Although the internal concentration of fructose 1-phosphate in C. glutamicum is unknown, micromolar concentrations are probable in cells grown on fructose or sucrose. Then, given its strong effect in vitro at low concentrations on SugR binding to the ldh promoter, plus the complete derepression of ldh on both fructose and sucrose when compared to glucose as carbon sources, fructose 1-phosphate could be considered the most effective co-inducer of SugR for ldh expression.

    To test the in vivo effect of fructose 1-phosphate on ldh expression, we constructed a cg2120 mutant (RES167 : : pAG1032) devoid of the fructose-specific EIIFru activity (EC 2.7.1.69). As fructose alone only supports weak growth of a ptsF mutant (Moon et al., 2005), the wild-type and the RES167 : : pAG1032 strains were grown on minimal medium with either glucose or sucrose, or a mixture of glucose and fructose, as carbon sources and Ldh activities were measured in cell extracts. Ldh activities were 2.27±0.1 mmol min−1 (mg protein)−1, 12.13±0.7 mmol min−1 (mg protein)−1 and 11.02+1.1 mmol min−1 (mg protein)−1 in extracts from the wild-type strain grown on glucose, sucrose and glucose/fructose respectively, and 2.76±0.2 mmol min−1 (mg protein)−1, 2.43±0.2 mmol min−1 (mg protein)−1 and 3.49±0.3 mmol min−1 (mg protein)−1 for the RES167 : : pAG1032 strain grown on glucose, sucrose and glucose/fructose respectively. This experiment confirmed that fructose 1-phosphate, the product of fructose phosphorylation by the fructose-specific EIIFru component, was actually an efficient inducer of ldh transcription.

    As fructose 1,6-bisphosphate should be present in fructose-grown or sucrose-grown cells, its contribution to ldh induction should not be neglected. However, its contribution to ldh induction is unlikely to exceed its contribution to ldh expression with glucose as a carbon source, otherwise the ldh expression would be as high with glucose as a carbon source as with sucrose in both the wild-type and the ptsF mutant strain.

    On the other hand, fructose 1,6-bisphosphate, the main activator of Ldh activity, was a good candidate to explain ldh induction and the time-course variation of ldh expresssion during biotin-limited growth on glucose (Fig. 1a⇑). This can be seen in Fig. 1(b)⇑, as the intracellular concentration of fructose 1,6-bisphosphate progressively diminished from about 10 mM, a concentration that led to a complete in vitro inhibition of the SugR binding to Pldh (Fig. 5b⇑), to about 1 mM at the end of cultivation time of the wild-type strain on glucose, a concentration that did not allow the in vitro inhibition of SugR binding to Pldh.

    As a striking difference, Engels & Wendisch (2007) identified fructose 6-phosphate as the only effector of SugR binding to the ptsG promoter, whereas Gaigalat et al. (2007) identified fructose 1-phosphate as the main effector of SugR binding to the promoters of ptsI, ptsH, ptsG, ptsS and the cg2118-fruK-ptsF operon, and fructose 1,6-bisphosphate and glucose 6-phosphate to a lesser extent. Gaigalat et al. (2007) suggested that SugR may be able to form complexes with individual binding sites with different sensitivities to sugar phosphates as described for the TrmB regulator of Pyrococcus furiosus (Lee et al., 2005, 2007), and proposed mutational analyses of the SugR-binding sites to clarify this issue. Prior to this analysis, however, SugR binding on its various targets under the same in vitro binding conditions and with a native SugR protein should be analysed. A note of caution should be exercised, as Engels & Wendisch (2007) used an amino-terminally His-tagged SugR protein, Gaigalat et al. (2007) used a recombinant SugR obtained after thio-induced self-cleavage of an intein tag, and we used a recombinant carboxyterminal-His-tagged SugR, and these differences may have a subtle, yet important impact on SugR-binding properties.

    Acknowledgments

    The authors are grateful to Dr P. Decottignies for assistance during MALDI-TOF mass spectrometry and Lee Cyster for proof-reading of the manuscript.

    References

    1. Abe, S., Takayama, K. I. & Kinoshita, S. (1967). Taxonomical studies on glutamic acid-producing bacteria. J Gen Appl Microbiol 13, 279–301.
    2. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403–410.
    3. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D., Seidman, J. G., Smith, J. A. & Struhl, K. (1987). Current Protocols in Molecular Biology. New York: Wiley-Interscience.
    4. Bailey, T. L. & Elkan, C. (1994). Fitting a mixture model by expectation maximization to discover motifs in biopolymers. In Proceedings of the Second International Conference on Intelligent Systems for Molecular Biology, pp. 28–36. Menlo Park, CA: AAAI Press.
    5. Baumbach, J., Wittkop, T., Rademacher, K., Rahmann, S., Brinkrolf, K. & Tauch, A. (2007). CoryneRegNet 3.0 – an interactive systems biology platform for the analysis of gene regulatory networks in corynebacteria and Escherichia coli. J Biotechnol 129, 279–289.
    6. Bendt, A. K., Burkovski, A., Schaffer, S., Bott, M., Farwick, M. & Hermann, T. (2003). Towards a phosphoproteome map of Corynebacterium glutamicum. Proteomics 3, 1637–1646.
    7. Bergmeyer, H. U. & Bernt, E. (1974). UV assay with pyruvate and NADH. In Methods of Enzymatic Analysis, pp. 576–579. Edited by H. U. Bergmeyer. New York: Academic Press.
    8. Bonamy, C., Guyonvarch, A., Reyes, O., David, F. & Leblon, G. (1990). Interspecies electro-transformation in corynebacteria. FEMS Microbiol Lett 54, 263–269.
    9. Börmann, E. R., Eikmanns, B. J. & Sahm, H. (1992). Molecular analysis of the Corynebacterium glutamicum gdh gene encoding glutamate dehydrogenase. Mol Microbiol 6, 317–326.
    10. Bower, S., Perkins, J. B., Yocum, R. R., Howitt, C. L., Rahaim, P. & Pero, J. (1996). Cloning, sequencing, and characterization of the Bacillus subtilis biotin biosynthetic operon. J Bacteriol 178, 4122–4130.
    11. Cerdeño-Tárraga, A. M., Efstratiou, A., Dover, L. G., Holden, M. T. G., Pallen, M., Bentley, S. D., Besra, G. S., Churcher, C., James, K. D. & other authors (2003). The complete genome sequence and analysis of Corynebacterium diphtheriae NCTC13129. Nucleic Acids Res 31, 6516–6523.
    12. Choi, S.-U., Nihira, T. & Yoshida, T. (2004). Enhanced glutamic acid production of Brevibacterium sp. with temperature shift-up cultivation. J Biosci Bioeng 98, 211–213.
    13. Clarke, A. R., Atkinson, T. & Holbrook, J. J. (1989). From analysis to synthesis: new ligand binding sites on the lactate dehydrogenase framework. Part I. Trends Biochem Sci 14, 101–105.
    14. Cocaign-Bousquet, M., Guyonvarch, A. & Lindley, N. D. (1996). Growth rate-dependent modulation of carbon flux through central metabolism and the kinetic consequences for glucose-limited chemostat cultures of Corynebacterium glutamicum. Appl Environ Microbiol 62, 429–436.
    15. Contag, P. R., Williams, M. G. & Rogers, P. (1990). Cloning of a lactate dehydrogenase gene from Clostridium acetobutylicum B643 and expression in Escherichia coli. Appl Environ Microbiol 56, 3760–3765.
    16. Crooks, G. E., Hon, G., Chandonia, J. M. & Brenner, S. E. (2004). WebLogo: a sequence logo generator. Genome Res 14, 1188–1190.
    17. Dartois, V., Phalip, V., Schmitt, P. & Diviès, C. (1995). Purification, properties and DNA sequence of the d-lactate dehydrogenase from Leuconostoc mesenteroides subsp. cremoris. Res Microbiol 146, 291–302.
    18. Dominguez, H. & Lindley, N. D. (1996). Complete sucrose metabolism requires fructose phosphotransferase activity in Corynebacterium glutamicum to ensure phosphorylation of liberated fructose. Appl Environ Microbiol 62, 3878–3880.
    19. Dominguez, H., Nezondet, C., Lindley, N. D. & Cocaign, M. (1993). Modified carbon flux during oxygen limited growth of Corynebacterium glutamicum and the consequences for amino acid overproduction. Biotechnol Lett 15, 449–454.
    20. Dominguez, H., Rollin, C., Guyonvarch, A., Guerquin-Kern, J. L., Cocaign-Bousquet, M. & Lindley, N. D. (1998). Carbon-flux distribution in the central metabolic pathways of Corynebacterium glutamicum during growth on fructose. Eur J Biochem 254, 96–102.
    21. Dsouza, M., Larsen, N. & Overbeek, R. (1997). Searching for patterns in genomic data. Trends Genet 13, 497–498.
    22. Dusch, N., Pühler, A. & Kalinowski, J. (1999). Expression of the Corynebacterium glutamicum panD gene encoding l-aspartate-α-decarboxylase leads to pantothenate overproduction in Escherichia coli. Appl Environ Microbiol 65, 1530–1539.
    23. Eikmanns, B. J., Rittmann, D. & Sahm, H. (1995). Cloning, sequence analysis, expression and inactivation of the Corynebacterium glutamicum icd gene encoding isocitrate dehydrogenase and biochemical characterization of the enzyme. J Bacteriol 177, 774–782.
    24. Eisenberg, M. A. (1987). Biosynthesis of biotin and lipoic acid. In Escherichia coli and Salmonella: Cellular and Molecular Biology, vol. 1, pp. 544–550. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.
    25. Engels, S., Schweitzer, J. E., Ludwig, C., Bott, M. & Schaffer, S. (2004). clpC and clpP1P2 gene expression in Corynebacterium glutamicum is controlled by a regulatory network involving the transcriptional regulators ClgR and HspR as well as the ECF sigma factor σH. Mol Microbiol 52, 285–302.
    26. Engels, V. & Wendisch, V. F. (2007). The DeoR-type regulator SugR represses expression of ptsG in Corynebacterium glutamicum. J Bacteriol 189, 2955–2966.
    27. Engels, V., Georgi, T. & Wendisch, V. F. (2008). ScrB (Cg2927) is a sucrose-6-phosphate hydrolase essential for sucrose utilization by Corynebacterium glutamicum. FEMS Microbiol Lett 289, 80–89.
    28. Gabrielsen, O. S., Hornes, E., Korsnes, L., Ruet, A. & Oyen, T. B. (1989). Magnetic DNA affinity purification of yeast transcription factor tau – a new purification principle for the ultrarapid isolation of near homogeneous factor. Nucleic Acids Res 17, 6253–6267.
    29. Gaigalat, L., Schlüter, J. P., Hartmann, M., Mormann, S., Tauch, A., Pühler, A. & Kalinowski, J. (2007). The DeoR-type transcriptional regulator SugR acts as a repressor for genes encoding the phosphoenolpyruvate : sugar phosphotransferase system (PTS) in Corynebacterium glutamicum. BMC Mol Biol 8, 104
    30. Garrigues, C., Loubière, P., Lindley, N. D. & Cocaign-Bousquet, M. (1997). Control of the shift from homolactic acid to mixed-acid fermentation in Lactococcus lactis: predominant role of the NADH/NAD+ ratio. J Bacteriol 179, 5282–5287.
    31. Garvie, E. I. (1980). Bacterial lactate dehydrogenases. Microbiol Rev 44, 106–139.
    32. Georgi, T., Engels, V. & Wendisch, V. F. (2008). Regulation of l-lactate utilization by the FadR-type regulator LldR of Corynebacterium glutamicum. J Bacteriol 190, 963–971.
    33. Gerstmeir, R., Wendisch, V. F., Schnicke, S., Ruan, H., Farwick, M., Reinscheid, D. & Eikmanns, B. J. (2003). Acetate metabolism and its regulation in Corynebacterium glutamicum. J Biotechnol 104, 99–122.
    34. Gerstmeir, R., Cramer, A., Dangel, P., Schaffer, S. & Eikmanns, B. J. (2004). RamB, a novel transcriptional regulator involved in acetate metabolism of Corynebacterium glutamicum. J Bacteriol 186, 2798–2809.
    35. Gralla, J. D. & Collado-Vides, J. (1996). Organization and function of transcription regulatory elements. In Escherichia coli and Salmonella: Cellular and Molecular Biology, vol. 1, pp. 1232–1245. Edited by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.
    36. Hatakeyama, K., Hohama, K., Vertès, A. A., Kobayashi, M., Kurusu, Y. & Yukawa, H. (1993a). Analysis of the biotin biosynthesis pathway in coryneform bacteria: cloning and sequencing of the bioB gene from Brevibacterium flavum. DNA Seq 4, 87–93.
    37. Hatakeyama, K., Hohama, K., Vertès, A. A., Kobayashi, M., Kurusu, Y. & Yukawa, H. (1993b). Genomic organization of the biotin biosynthetic genes of coryneform bacteria: cloning and sequencing of the bioA-bioD genes from Brevibacterium flavum. DNA Seq 4, 177–184.
    38. Hermann, T., Pfefferle, W., Baumann, C., Busker, E., Schaffer, S., Bott, M., Sahm, H., Dusch, N., Kalinowski, J. & other authors (2001). Proteome analysis of Corynebacterium glutamicum. Electrophoresis 22, 1712–1723.
    39. Hoischen, C. & Krämer, R. (1989). Evidence for an efflux carrier system involved in the secretion of glutamate by Corynebacterium glutamicum. Arch Microbiol 151, 342–347.
    40. Hüser, A. T., Chassagnole, C., Lindley, N. D., Merkamm, M., Guyonvarch, A., Elisáková, V., Pátek, M., Kalinowski, J., Brune, I. & other authors (2005). Rational design of a Corynebacterium glutamicum pantothenate production strain and its characterization by metabolic flux analysis and genome-wide transcriptional profiling. Appl Environ Microbiol 71, 3255–3268.
    41. Inui, M., Murakami, S., Okino, S., Kawaguchi, H., Vertès, A. A. & Yukawa, H. (2004). Metabolic analysis of Corynebacterium glutamicum during lactate and succinate production under oxygen deprivation conditions. J Mol Microbiol Biotechnol 7, 182–196.
    42. Ishige, T., Krause, M., Bott, M., Wendisch, V. F. & Sahm, H. (2003). The phosphate starvation stimulon of Corynebacterium glutamicum determined by DNA microarray analyses. J Bacteriol 185, 4519–4529.
    43. Jäger, W., Peters-Wendisch, P. G., Kalinowski, J. & Pühler, A. (1996). A Corynebacterium glutamicum gene encoding a two-domain protein similar to biotin carboxylases and biotin-carboxyl-carrier proteins. Arch Microbiol 166, 76–82.
    44. Kalinowski, J., Bathe, B., Bartels, D., Bischoff, N., Bott, M., Burkovski, A., Dusch, N., Eggeling, L., Eikmanns, B. J. & other authors (2003). The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of l-aspartate-derived amino acids and vitamins. J Biotechnol 104, 5–25.
    45. Keilhauer, C., Eggeling, L. & Sahm, H. (1993). Isoleucine synthesis in Corynebacterium glutamicum: molecular analysis of thr ilvB-ilvN-ilvC operon. J Bacteriol 175, 5595–5603.
    46. Kikuchi, M. & Nakao, Y. (1986). Glutamic acid. In Progress in Industrial Microbiology, vol. 24, pp. 101–106. Edited by K. Aido and others. Amsterdam, The Netherlands: Elsevier.
    47. Kinoshita, K., Udaka, S. & Shimono, M. (1957). Studies on the amino acid fermentation: I. Production of l-glutamic acid by various microorganisms. J Gen Appl Microbiol 3, 193–205.
    48. Kirchner, O. & Tauch, A. (2003). Tools for genetic engineering in the amino acid-producing bacterium Corynebacterium glutamicum. J Biotechnol 104, 287–299.
    49. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.
    50. Lee, S. J., Moulakakis, C., Koning, S. M., Hausner, W., Thomm, M. & Boos, W. (2005). TrmB, a sugar sensing regulator of ABC transporter genes in Pyrococcus furiosus exhibits dual promoter specificity and is controlled by different inducers. Mol Microbiol 57, 1797–1807.
    51. Lee, S. J., Surma, M., Seitz, S., Hausner, W., Thomm, M. & Boos, W. (2007). Differential signal transduction via TrmB, a sugar sensing transcriptional repressor of Pyrococcus furiosus. Mol Microbiol 64, 1499–1505.
    52. Meers, J. L., Tempest, D. W. & Brown, C. M. (1970). ‘Glutamine (amide) : 2-oxoglutarate amino transferase oxido-reductase (NADP)’ an enzyme involved in the synthesis of glutamate by some bacteria. J Gen Microbiol 64, 187–194.
    53. Merkamm, M. & Guyonvarch, A. (2001). Cloning of the sodA gene from Corynebacterium melassecola and role of superoxide dismutase in cellular viability. J Bacteriol 183, 1284–1295.
    54. Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
    55. Moon, M. W., Kim, H. J., Oh, T. K., Shin, C. S., Lee, J. S., Kim, S. J. & Lee, J. K. (2005). Analyses of enzyme II gene mutants for sugar transport and heterologous expression of fructokinase gene in Corynebacterium glutamicum ATCC 13032. FEMS Microbiol Lett 244, 259–266.
    56. Mortensen, L., Dandanell, G. & Hammer, K. (1989). Purification and characterization of the DeoR repressor of Escherichia coli. EMBO J 8, 325–331.
    57. Muffler, A., Bettermann, S., Haushalter, M., Hörlein, A., Neveling, U., Schramm, M. & Sorgenfrei, O. (2002). Genome-wide transcription profiling of Corynebacterium glutamicum after heat shock and during growth on acetate and glucose. J Biotechnol 98, 255–268.
    58. Nachlas, M. M., Davidson, M. B., Goldberg, J. D. & Seligman, A. M. (1963). Colorimetric method for the measurement of isocitric dehydrogenase activity. J Lab Clin Med 62, 148–158.
    59. Nishio, Y., Nakamura, Y., Kawarabayasi, Y., Usuda, Y., Kimura, E., Sugimoto, S., Matsui, K., Yamagishi, A., Kikuchi, H. & other authors (2003). Comparative complete genome sequence analysis of the amino acid replacements responsible for the thermostability of Corynebacterium efficiens. Genome Res 13, 1572–1579.
    60. Okino, S., Inui, M. & Yukawa, H. (2005). Production of organic acids by Corynebacterium glutamicum under oxygen deprivation. Appl Microbiol Biotechnol 68, 475–480.
    61. Pátek, M., Nesvera, J., Guyonvarch, A., Reyes, O. & Leblon, G. (2003). Promoters of Corynebacterium glutamicum. J Biotechnol 104, 311–323.
    62. Peters-Wendisch, P. G., Kreutzer, C., Kalinowski, J., Pátek, M., Sahm, H. & Eikmanns, B. J. (1998). Pyruvate carboxylase from Corynebacterium glutamicum: characterization, expression and inactivation of the pyc gene. Microbiology 144, 915–927.
    63. Portevin, D., de Sousa D'Auria, C., Montrozier, H., Houssin, C., Stella, A., Lanéelle, M. A., Bardou, F., Guilhot, C. & Daffé, M. (2005). The acyl-AMP ligase FadD32 and AccD4-containing acyl-CoA carboxylase are required for the synthesis of mycolic acids and essential for mycobacterial growth: identification of the carboxylation product and determination of the acyl-CoA carboxylase components. J Biol Chem 280, 8862–8874.
    64. Rey, D. A., Pühler, A. & Kalinowski, J. (2003). The putative transcriptional repressor McbR, member of the TetR-family, is involved in the regulation of the metabolic network directing the synthesis of sulfur containing amino acids in Corynebacterium glutamicum. J Biotechnol 103, 51–65.
    65. Rey, D. A., Nentwich, S. S., Koch, D. J., Rückert, C., Pühler, A., Tauch, A. & Kalinowski, J. (2005). The McbR repressor modulated by the effector substance S-adenosylhomocysteine controls directly the transcription of a regulon involved in sulfur metabolism of Corynebacterium glutamicum ATCC 13032. Mol Microbiol 56, 871–887.
    66. Rollin, C., Morgant, V., Guyonvarch, A. & Guerquin-Kern, J. L. (1995). 13C-NMR studies of Corynebacterium melassecola metabolic pathways. Eur J Biochem 227, 488–493.
    67. Rovira, A. D. & Harris, J. R. (1961). Plant root excretions in relation to the rhizosphere effect. V. The exudation of B-group vitamins. Plant Soil 14, 199–214.
    68. Roy, S., Sahu, A. & Adhya, S. (2002). Evolution of DNA binding motifs and operators. Gene 285, 169–173.
    69. Ruffert, S., Lambert, C., Peter, H., Wendisch, F. & Krämer, R. (1997). Efflux of compatible solutes in Corynebacterium glutamicum mediated by osmoregulated channel activity. Eur J Biochem 247, 572–580.
    70. Sahm, H., Eggeling, L. & de Graaf, A. A. (2000). Pathway analysis and metabolic engineering in Corynebacterium glutamicum. Biol Chem 381, 899–910.
    71. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
    72. Samols, D., Thornton, C. G., Murtif, V. L., Kumar, G. K., Haase, C. & Wood, H. G. (1988). Evolutionary conservation among biotin enzymes. J Biol Chem 263, 6461–6464.
    73. Schäfer, A., Tauch, A., Jäger, W., Kalinowski, J., Thierbach, G. & Pühler, A. (1994). Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145, 69–73.
    74. Sedewitz, B., Schleifer, K. H. & Götz, F. (1984). Physiological role of pyruvate oxidase in the aerobic metabolism of Lactobacillus plantarum. J Bacteriol 160, 462–465.
    75. Selander, R. K., Caugant, D. A., Ochman, H., Musser, J. M., Gilmour, M. N. & Whittam, T. S. (1986). Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl Environ Microbiol 51, 873–884.
    76. Silberbach, M., Schäfer, M., Hüser, A. T., Kalinowski, J., Pühler, A., Krämer, B. & Burkovski, A. (2005a). Adaptation of Corynebacterium glutamicum to ammonium limitation: a global analysis using transcriptome and proteome techniques. Appl Environ Microbiol 71, 2391–2402.
    77. Silberbach, M., Hüser, A. T., Kalinowski, J., Pühler, A., Walter, B., Krämer, B. & Burkovski, A. (2005b). DNA microarray analysis of the nitrogen starvation response of Corynebacterium glutamicum. J Biotechnol 119, 357–367.
    78. Soual-Hoebeke, E., de Sousa-d'Auria, C., Chami, M., Baucher, M. F., Guyonvarch, A., Bayan, N., Salim, K. & Leblon, G. (1999). S-layer protein production by Corynebacterium strains is dependent on the carbon source. Microbiology 145, 3399–3408.
    79. Stansen, C., Uy, D., Delaunay, S., Eggeling, L., Goergen, J. L. & Wendisch, V. F. (2005). Characterization of a Corynebacterium glutamicum lactate utilisation operon induced during temperature-triggered glutamate production. Appl Environ Microbiol 71, 5920–5928.
    80. Sullivan, J. T., Brown, S. D., Yocum, R. R. & Ronson, C. W. (2001). The bio operon on the acquired symbiosis island of Mesorhizobium sp. strain R7A includes a novel gene involved in pimeloyl-CoA synthesis. Microbiology 147, 1315–1322.
    81. Tanaka, Y., Teramoto, H., Inui, M. & Yukawa, H. (2008a). Regulation of expression of general components of the phosphoenolpyruvate : carbohydrate phosphotransferase system (PTS) by the global regulator SugR in Corynebacterium glutamicum. Appl Microbiol Biotechnol 78, 309–318.
    82. Tanaka, Y., Okai, N., Teramoto, H., Inui, M. & Yukawa, H. (2008b). Regulation of the expression of phosphoenolpyruvate : carbohydrate phosphotransferase system (PTS) genes in Corynebacterium glutamicum R. Microbiology 154, 264–274.
    83. Tauch, A., Kaiser, O., Hain, T., Goesmann, A., Weisshaar, B., Albersmeier, A., Bekel, T., Bischoff, N., Brune, I. & other authors (2005). Complete genome sequence and analysis of the multiresistant nosocomial pathogen Corynebacterium jeikeium K411, a lipid-requiring bacterium of the human skin flora. J Bacteriol 187, 4671–4682.
    84. Toh, H., Kondo, H. & Tanabe, T. (1993). Molecular evolution of biotin-dependent carboxylases. Eur J Biochem 215, 687–696.
    85. Trésaugues, L., Collinet, B., Minard, P., Henckes, G., Aufrère, R., Blondeau, K., Liger, D., Zhou, C. Z., Janin, J. & other authors (2004). Refolding strategies from inclusion bodies in a structural genomics project. J Struct Funct Genomics 5, 195–204.
    86. Uy, D., Delaunay, S., Germain, P., Engasser, J. M. & Goergen, J. L. (2003). Instability of glutamate production by Corynebacterium glutamicum 2262 in continuous culture using the temperature-triggered process. J Biotechnol 104, 173–184.
    87. Vallino, J. J. & Stephanopoulos, G. (1994). Carbon flux distributions at the pyruvate branch point in Corynebacterium glutamicum during lysine overproduction. Biotechnol Prog 10, 320–326.
    88. Wendisch, V. F. (2003). Genome-wide expression analysis in Corynebacterium glutamicum using DNA microarrays. J Biotechnol 104, 273–285.
    89. Wennerhold, J., Krug, A. & Bott, M. (2005). The AraC-type regulator RipA represses aconitase and other iron proteins from Corynebacterium under iron limitation and is itself repressed by DtxR. J Biol Chem 280, 40500–40508.
    90. Yasuda, K., Jojima, T., Suda, M., Okino, S., Inui, M. & Yukawa, H. (2007). Analyses of the acetate-producing pathways in Corynebacterium glutamicum under oxygen-deprived conditions. Appl Microbiol Biotechnol 77, 853–860.
    91. Zeng, X., Saxild, H. H. & Switzer, R. L. (2000). Purification and characterization of the DeoR repressor of Bacillus subtilis. J Bacteriol 182, 1916–1922.