The bacterial diaminopimelate (DAP) and -lysine biosynthesis pathway was elucidated by a series of biochemical studies in the 1960s and 70s (reviewed in [Patte, 1996]). Three alternative pathways leading to the synthesis of ,-DAP were identified in bacteria ([Scapin and Blanchard, 1998]). They share the first steps of aspartate (and pyruvate) conversion to the intermediate -2,3,4,5-tetrahydrodipicolinate (THDP), but vary in the subsequent reactions (Fig. 1). THDP can be converted to DAP in a single reaction which is catalyzed by an ammonium-incorporating diaminopimelate dehydrogenase, whereas the other pathway variants consist of four reaction steps involving either acetylated or succinylated intermediates. The dehydrogenase and/or acetylase variants of the -lysine biosynthesis were found among the members of the genus Bacillus ([Weinberger and Gilvarg, 1970]), whereas the succinylase pathway is present in Escherichia coli and most other bacteria ([Kindler and Gilvarg, 1960 and Scapin and Blanchard, 1998]). In the succinylase pathway, THDP is succinylated to N-succinyl-2-amino-6-ketopimelate which is the substrate of the aminotransferase DapC. The product of this reaction is then desuccinylated to give ,-DAP which is converted by the epimerase DapF to the penultimate -lysine precursor D,-DAP (Fig. 1).

 

 

Fig. 1. The branched pathway for D,-diaminopimelate and -lysine biosynthesis in C. glutamicum.

 

As a special feature, the gram-positive soil bacterium Corynebacterium glutamicum possesses the succinylase together with the dehydrogenase variant of the D,-DAP and -lysine biosynthesis ([Schrumpf et al., 1991]). Both branches operate in parallel in this organism as shown by metabolite flux analyses ( [Sonntag et al., 1993]). Currently, more than 6.0×10⁵ tons of -lysine are produced annually with C. glutamicum mutant strains. Therefore, tremendous efforts are constantly undertaken to optimize the -lysine biosynthesis with regard to higher efficiencies of such strains ([Leuchtenberger, 1996]). Genes directly involved in the synthesis of -lysine are obviously primary targets to improve the overall fermentation process. Most of these genes were already identified in C. glutamicum ([Cremer et al., 1990, Ishino et al., 1987, Kalinowski et al., 1990, Kalinowski et al., 1991, Wehrmann et al., 1994, Wehrmann et al., 1998 and Yeh et al., 1988]). However, dapC and dapF encoding N-succinyl-aminoketopimelate aminotransferase and diaminopimelate epimerase still remain to be identified (Fig. 1).

The aminotransferase DapC (EC 2.6.1.17) catalyzes the transfer of the amino group from glutamate to N-succinyl-,-diaminopimelate forming -ketoglutarate and N-N-succinyl-2,6-,-diaminopimelate. Like all known aminotransferases it uses pyridoxal 5′-phosphate (PLP) as a catalytic cofactor. The DapC enzyme of Bordetella pertussis ([Fuchs et al., 2000]) and the ArgD protein of E. coli are the sole examples of bacterial aminotransferases with this substrate specificity as disclosed in protein databases ([Ledwidge and Blanchard, 1999]). The ArgD protein of E. coli possesses both an acetylornithine and a DAP aminotransferase activity, explaining its nomination as ArgD. Interestingly, the B. pertussis DapC protein and ArgD of E. coli share no significant amino acid sequence similarity apart from the characteristic PLP-binding domain ([Fuchs et al., 2000]). Despite the availability of a large number of whole genome sequences from bacteria, no other gene encoding an N-succinyl-,-DAP aminotransferase has been characterized enzymatically so far.

In the last step of the succinylase branch, D,-DAP is generated from the corresponding ,-isomer by the dapF-encoded diaminopimelate epimerase (EC 5.1.1.7). The DapF protein is a representative of a pyridoxal phosphate-independent amino acid racemace ([Koo and Blanchard, 1999]). DapF of E. coli was purified and characterized ([Wiseman and Nichols, 1984]) and the corresponding gene was cloned and mapped in the E. coli chromosome ([Richaud et al., 1987 and Richaud and Printz, 1988]). Currently, at least 25 homologous protein sequences of other organisms have been deposited in protein databases, all containing a specific diaminopimelate epimerase signature (PROSITE PDOC01029), which clearly defines this protein family. Herein included are the corresponding DapF proteins of Mycobacterium tuberculosis and Streptomyces coelicolor which are close taxonomic relatives of C. glutamicum.

In the present paper, we report how the C. glutamicum genome sequencing project enabled us to identify and to characterize the diaminopimelate epimerase gene dapF and the N-succinyl-aminoketopimelate aminotransferase gene dapC. Both genes showed significant effects on -lysine production when overexpressed in an industrial C. glutamicum strain.

2. MATERIALS AND METHODS

2.1. Bacterial strains, plasmids and growth conditions

Relevant bacterial strains and plasmids used in this study are listed in Table 1. E. coli and C. glutamicum strains were routinely grown in Luria-Bertani medium ([Sambrook et al., 1989]) supplemented with 2 g l⁻¹ glucose (LBG) at 37 and 30 °C, respectively. Growth of C. glutamicum strains was monitored in brain–heart medium (BHI; Merck Eurolab, Darmstadt, Germany) containing 40 g l⁻¹ glucose. Complex medium CGIII ([Menkel et al., 1989]) and minimal medium MM5 (5 g l⁻¹ corn steep liquor, 20 g l⁻¹ morpholinopropanesulfonic acid, 50 g l⁻¹ glucose, 25 g l⁻¹ (NH₄)₂SO₄, 0.1 g l⁻¹ KH₂PO₄, 1 g l⁻¹ MgSO₄·7H₂O, 10 mg l⁻¹ CaCl₂·2H₂O, 10 mg l⁻¹ FeSO₄·7H₂O, 5 mg l⁻¹ MnSO₄·H₂O, 0.1 g l⁻¹ leucine, 0.3 mg l⁻¹ biotin, 0.2 mg l⁻¹ thiamin, 25 g l⁻¹ CaCO₃) were used for the production of -lysine with C. glutamicum. Antibiotics for plasmid selection were kanamycin (50 g ml⁻¹ for E. coli and 25 g ml⁻¹ for C. glutamicum) and tetracycline (5 g ml⁻¹ for E. coli and C. glutamicum).

 

 

Table 1. Relevant C. glutamicum strains and plasmids used in this study
 

 

Growth analyses with C. glutamicum strains were performed in a culture volume of 100 l in a Bioscreen C Microbial Workstation (Labsystems, Vaataa, Finland). LBG overnight cultures of C. glutamicum were washed twice in 50 mM Tris–HCl (pH 7.5), and 2×10⁵ cells were used as inoculum. The main cultures were incubated at 30 °C with intensive shaking for 48 h. The optical density (OD₅₈₀) was determined automatically every 2 h.

2.2. DNA isolation, manipulation and transfer

E. coli DH5MCR ([Grant et al., 1990]) was used for routine cloning experiments. Vector DNA was prepared from E. coli cells by alkaline lysis using the QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany). DNA restriction fragments required for cloning were purified from agarose gels by means of the QIAEX II Gel Extraction Kit (Qiagen). All recombinant DNA techniques followed standard procedures ([Sambrook et al., 1989]). E. coli and C. glutamicum cells were transformed by electroporation ([Tauch et al., 1994 and Tauch et al., 2002b]). Chromosomal DNA of C. glutamicum was prepared according to the method of [Tauch et al., 1995].

2.3. PCR techniques

PCR experiments were carried out with a PTC-100 thermocycler from MJ Research (Watertown, MA) and Pfu DNA polymerase. Initial denaturation was conducted at 94 °C for 2 min followed by denaturation for 30 s, annealing for 30 s at a primer-dependent temperature, and extension at 72 °C for 45 s. This cycle was repeated 30 times, followed by a final extension step at 72 °C for 3 min. PCR products were purified using the QIAquick PCR Purification Kit (Qiagen). Cloning of PCR products was performed in E. coli TOP10 by means of the Zero Blunt TOPO PCR Cloning Kit (Invitrogen, Karlsruhe, Germany).

2.4. Construction of deletion mutants of C. glutamicum RES167

Defined chromosomal deletions within the dapC, dapF, ddh, and argD genes of C. glutamicum RES167 were constructed with the pK18mobsacB vector system which helps to identify an allelic exchange by homologous recombination ([Schäfer et al., 1994]). Deletions were introduced into the respective genes by gene SOEing following the method of [Horton, 1995]. A defined dapF deletion was constructed by using unique EcoRI and KpnI restriction sites within the coding region. Defined deletions introduced into the chromosome of C. glutamicum were verified by PCR experiments. The length of the deleted chromosomal fragments and the corresponding amino acids of the respective proteins are given in Table 1.

2.5. Construction of dapF and dapC expression vectors

The dapF gene of C. glutamicum was amplified by PCR as 966-bp DNA fragment using Pfu DNA polymerase and the primer pair dapFex1 (5′-ATCGTACAATTGCACCGCACAA GCCTTGGAGA-3′) and dapFex2 (5′-GACGATGGATCCTAACGGACGAGCGCGCACTA-3′), carrying a MunI (dapFex1) and a BamHI (dapFex2) site within the 5′ extensions. The purified PCR product was ligated into the vector pCR-Blunt II-TOPO (Invitrogen) and the resulting plasmid was subsequently digested with MunI and BamHI. The dapF containing fragment was re-isolated from a 0.8% agarose gel and cloned into the E. coli–C. glutamicum shuttle expression vector pEC-XT99A (Table 1) which was previously digested with EcoRI and BamHI. The resulting vector used for overexpression of dapF in C. glutamicum was named pMH10.

The dapC gene was amplified as 1618-bp DNA fragment using primers dapCex1 (5′-GATCTAGAATTCGCCTCAGGCATAATCTAACG-3′) and dapCex2 (5′-GATCTATCTAGACAGAGGACAAGGCAATCGGA-3′), carrying an EcoRI (dapCex1) and an XbaI (dapCex2) site within the 5′ extensions. The PCR product was ligated into pCR-Blunt II-TOPO, re-isolated as EcoRI-XbaI DNA fragment from an agarose gel, and cloned into the shuttle expression vector pEC-XT99A, resulting in plasmid pMH12. The nucleotide sequences of the amplified dapF and dapC genes were verified by DNA sequencing (IIT, Bielefeld, Germany). Induction of dapF and dapC gene expression in C. glutamicum was carried out by addition of 1 mM isopropyl--D-thiogalactopyranoside (IPTG) to early log-phase cultures.

2.6. Determination of epimerase activity

C. glutamicum strains were grown in minimal medium CGXII ([Keilhauer et al., 1993]) to determine the ,-diaminopimelate epimerase activity. Cells were harvested after 8 h of incubation at 30 °C, washed with 20 mM Tris–HCl (pH 8.0), resuspended in buffer consisting of 20 mM Tris–HCl (pH 8.0) plus 1 mM dithiothreitol, and disrupted with a microtip-equipped sonifier. The homogenate was centrifuged for 20 min at 20000×g and the resulting extract was applied to a PD-10 column (Amersham-Pharmacia, Freiburg, Germany).

The activity was assayed with D,-diaminopimelate as a substrate in a system according to the procedure of [Wiseman and Nichols, 1984]. The assay system consisted of 200 mM Tris–HCl (pH 8.0), 40 mM hydroxylammonium chloride, 1 mM dithiothreitol, and 5 mM D,-diaminopimelate (>90% pure). Assay mixtures were incubated at 30 °C and samples (30 l) were taken at 0, 10, 20, and 30 min. Reactions were stopped by addition of 30 l stop reagent (0.75 M HClO₄ in 7 M ethanol), neutralized with 20 l neutralizing solution (0.1 M K₂CO₃, 20 mM Tris–HCl [pH 8.0]), and used for determination of D,D- and ,-DAP. This was done by automated precolumn derivatization with o-phthaldialdehyde, separation by reversed phase chromatography (LC ChemStation HP 1900), and fluorometric detection ([Jones and Gilligan, 1983]). Protein concentration was determined using the method of [Bensadoun and Weinstein, 1976].

2.7. Determination of transaminase activity

C. glutamicum strains were grown in minimal medium MM1 (MMYE without yeast extract; [Katsumata et al., 1984]) to determine the N-succinyl-,-DAP aminotransaminase activity. Cells were harvested after overnight incubation at 30 °C, washed with 0.9% NaCl, resuspended in 20 mM Tris–HCl (pH 8.0), and crude extract was prepared as described before. Determination of the aminotransferase activity was based on the succinyl-DAP-dependent formation of glutamate from -ketoglutarate. The assay system consisted of 200 mM Tris–HCl (pH 8.0), 0.25 mM PLP, 4 mM -ketoglutarate, 8 mM N-succinyl-,-DAP, 1 mM EDTA, and gel-filtered extract. Assay mixtures were incubated at 37 °C and samples (30 l) were taken and processed as described before.

2.8. Production of -lysine with recombinant C. glutamicum strains

C. glutamicum strains were precultured in CGIII medium ([Menkel et al., 1989]) at 33 °C for 24 h. A main culture was inoculated in MM5 medium in such a way that the initial optical density (OD₆₆₀) of the culture was 0.2. After 72 h of growth at 33 °C the concentration of -lysine in the supernatant was determined in an amino acid analyser (Eppendorf-BioTronik, Hamburg, Germany) by ion exchange chromatography and postcolumn derivatization with ninhydrin detection.

2.9. Nucleotide sequence accession numbers

The nucleotide sequences of the dapC and dapF gene loci of C. glutamicum ATCC 13032 were deposited in the GenBank database under accession numbers AY170830 (dapC) and AY170829 (dapF), respectively.

3. RESULTS

3.1. Identification of the dapF gene in the whole genome sequence of C. glutamicum

The complete genome of C. glutamicum ATCC 13032 was recently sequenced by means of cosmid and bacterial artificial chromosome libraries ([Tauch et al., 2002a]). Subsequently, the genome sequence was annotated with the GenDB database system (Zentrum für Genomforschung, Bielefeld, Germany) and inspected for coding regions representing the genes involved in DAP and -lysine biosynthesis. The already published genes of this pathway in C. glutamicum comprising lysC (cg0306), asd (cg0307), dapA (cg2161), dapB (cg2163), ddh (cg2900), dapD (cg1256), dapE (cg1260), and lysA (cg1334) were easily identified. In addition, a candidate gene for a putative dapF function (cg2129) was predicted. The deduced gene product of cg2129 revealed 27% amino acid sequence identity to the DapF protein of E. coli ([Richaud and Printz, 1988]) and 32% identity to the corresponding protein of Haemophilus influenzae ([Cirilli et al., 1998]) which is the best characterized member of the diaminopimelate epimerase family. Furthermore, the cg2129 protein showed significant amino acid sequence similarity to a large number of proteins in databases which were annotated as DapF without experimental confirmation. Especially, corresponding proteins of the taxonomically closely related species M. tuberculosis (52%) and S. coelicolor (43%) revealed high levels of amino acid sequence identity.

A PROSITE motif search within the deduced amino acid sequence of cg2129 identified the presence of a diaminopimelate epimerase signature (PROSITE PDOC01029) comprising amino acid residues 75–89 (Fig. 2). In addition, computational secondary structure predictions ([Baldi et al., 1999]) for the cg2129 protein revealed a structure of two homologous domains (amino acids 1–135 and amino acids 149–277) each containing eight -strands and two -helices. This structure prediction is in accordance with the three-dimensional structure of the DAP epimerase from H. influenzae ([Cirilli et al., 1998]). Two conserved cysteines, one in each domain, are involved in the catalytic mechanism of the DAP epimerase reaction. These conserved cysteines are also present in the deduced amino acid sequence of cg2129 (Cys₈₄ and Cys₂₂₂) and the active site cysteine Cys₈₄ is part of the diaminopimelate epimerase signature (Fig. 2). These data strongly suggested that the cg2129 protein is a member of the diaminopimelate epimerase family. Therefore, we have designated the cg2129 coding region of the C. glutamicum genome sequence as dapF gene.

 

 

Fig. 2. Diaminopimelate epimerase signature in DapF proteins. A multiple alignment of DapF protein sequences was performed with the CLUSTALW program ([Thompson et al., 1994]). Only the amino acid sequence of the diaminopimelate epimerase signature NxDGSx₄CGN[GA]xR (PROSITE PDOC01029) is shown. Conserved amino acid residues are indicated by asterisks. The conserved cysteine residue within the diaminopimelate epimerase signature is specifically marked by an arrow. Numbers in parenthesis correspond to the position of the motif with respect to the start of each protein. Protein sequences were from GenBank: M. tuberculosis (NC_000962), S. coelicolor (NC_003888), and H. influenzae (NC_000907).

 

3.2. The C. glutamicum dapF gene is indispensable for the succinylase branch of the diaminopimelate biosynthesis

To further analyze the identified dapF gene, we established a 237-bp EcoRI-KpnI deletion in the chromosomal dapF coding region of C. glutamicum RES167 by means of the sacB selection system present on plasmid pMH2 (Table 1). After electrotransformation into C. glutamicum of a recombinant sacB vector carrying a deletion construct, the vector can establish itself only by integration into the chromosome via homologous recombination. The resulting strain generally carries the modified gene and the wild type gene separated by the vector sequence. Excision of the vector can be selected for by growing the cells on LBG agar containing 10% sucrose. Cells able to grow on this medium have lost the plasmid due to a second cross-over event that either restores the wild type situation or leads to a defined mutant strain ([Schäfer et al., 1994]). Interestingly, recombinant C. glutamicum strains carrying the 237-bp chromosomal dapF deletion were only selected for on medium (LBG+10% sucrose) which was additionally supplemented with 50 mM ammonium sulfate. Without ammonium sulfate addition only revertants to the wild type gene arrangement were obtained, indicating that the selection of a dapF mutant depended on the ammonium availability. The isolated dapF mutant strain C. glutamicum MH1 was, therefore, used to analyze its growth characteristics in detail.

C. glutamicum MH1 showed no phenotypic alteration during growth in standard rich media (LBG or BHI) or in minimal medium MM1 when compared with the control strain C. glutamicum RES167 (data not shown). On the other hand, C. glutamicum MH1 showed an impaired growth when supplied with low ammonium but high concentrations of carbon (BHI+4% glucose). Under this condition the growth of C. glutamicum MH1 arrested at an obviously lower cell density than the control strain C. glutamicum RES167 (Fig. 3A, left). This phenotype was initially described for C. glutamicum strains lacking a functional dapD or dapE gene, which can, therefore, synthesize D,-DAP only via the dehydrogenase branch of the DAP biosynthesis pathway ([Wehrmann et al., 1998]). The observed growth effect was nullified by supplementation of the growth medium with 50 mM ammonium sulfate ( Fig. 3A, right). In supplemented medium, C. glutamicum MH1 showed a growth kinetic comparable to that of the control strain C. glutamicum RES167, indicating that the dehydrogenase variant obviously can compensate the loss of the epimerase activity when free ammonium is available in a sufficient high concentration ([Wehrmann et al., 1998]). This result also provided an explanation for the necessity to supplement the growth medium with ammonium sulfate during the sacB selection procedure which was performed to obtain the C. glutamicum ΔdapF mutant strain MH1.

 

Fig. 3. Growth of C. glutamicum strains carrying a chromosomal dapF or dapC deletion. The growth analyses were performed in a culture volume of 100 l in a Bioscreen C Microbial Workstation. The optical density (OD₅₈₀) was determined automatically. (A) Growth of the ΔdapF mutant C. glutamicum MH1 (&z.cirf;) was compared with C. glutamicum RES167 () in BHI+4% glucose without ammonium supplementation (left panel) and with 50 mM ammonium sulfate (right panel). (B) Growth of the ΔdapC mutant C. glutamicum MH3 (&z.cirf;) was compared with the control strain C. glutamicum RES167 () in BHI+4% glucose with limited ammonium availability (left panel) and with 50 mM ammonium sulfate (right panel).

 

In a further genetic approach, we tried to establish an additional deletion within the ddh gene of the ΔdapF mutant strain C. glutamicum MH1. The recombinant sacB vector carrying the ddh deletion construct (pMH4) was integrated into the chromosome of C. glutamicum MH1, but it was impossible to obtain a deletion in both genes by sacB selection. The reason might be that C. glutamicum is apparently unable to take up D,-DAP ([Yeh et al., 1988]), thus preventing the construction of the double mutant. This observation implies that besides the succinylase branch and the dehydrogenase variant no further pathway exists for D,-DAP synthesis in C. glutamicum.

3.3. Overexpression of the dapF gene in C. glutamicum DSM5715 resulted in an increased diaminopimelate epimerase activity

In order to verify the postulated function of the cg2129 protein as diaminopimelate epimerase, enzyme assays were performed (Table 2). Based on the rate of D,-DAP conversion to ,-DAP in the respective assays, a specific DapF activity of 0.01 mol min⁻¹ per mg of protein was determined in a control strain carrying the chromosomally encoded dapF gene, whereas no specific DapF activity was detectable in extracts of the defined dapF deletion mutant C. glutamicum MH1. In addition, the dapF gene was overexpressed by cloning its promotor-less coding region into the expression vector pEC-XT99A, resulting in plasmid pMH10 (Table 1). In such a way, the dapF gene was under control of the inducible P_trc promoter and present in approximately 30 copies per C. glutamicum chromosome ([Nesvera et al., 1997]). Plasmid pMH10 was transferred into the lysine-producing strain C. glutamicum DSM5715 and the recombinant derivative C. glutamicum MH6 was subsequently assayed for diaminopimelate epimerase activity. C. glutamicum MH6 revealed an almost 8-fold increase in the epimerase activity when compared with C. glutamicum DSM5715 carrying vector pEC-XT99A (Table 2). Consequently, the enzyme assays confirmed that the cg2129 coding region of C. glutamicum codes for the DapF protein.

 

 

Table 2. Specific diaminopimelate epimerase activity of C. glutamicum strains
 

 

3.4. Overexpression of the dapF gene in C. glutamicum DSM5715 led to increased -lysine production

To analyze whether the dapF gene promotes a positive effect on the fermentative production of -lysine, C. glutamicum MH6 was grown in MM5 medium for 72 h and the concentration of -lysine in the culture supernatant was determined. Overexpression of the dapF gene increased the -lysine concentration to 13.5 g l⁻¹, whereas the control strain (C. glutamicum DSM5715 containing pEC-XT99A) produced only 11.9 g l⁻¹. These values correspond to an increase of 13.4% -lysine within 72 h of fermentation. Therefore, the dapF gene is an attractive target for further improvement of lysine-producing C. glutamicum strains by molecular genetic engineering.

3.5. Identification of the dapC gene of C. glutamicum encoding a protein with N-succinyl-aminoketopimelate aminotransferase activity

Besides the already identified dapF gene, the C. glutamicum genome sequencing project revealed one more interesting coding region (cg1253) in the context of -lysine and D,-DAP biosynthesis in this organism. The deduced cg1253 protein contains a conserved PLP-binding site (amino acids 224–237) and shares 29% identical amino acids with the N-succinyl-aminoketopimelate aminotransferase DapC of B. pertussis, which is the only enzymatically characterized example of DapC proteins ([Fuchs et al., 2000]), and 52–61% identical amino acids with putative aminotransferases from M. tuberculosis and S. coelicolor. Furthermore, cg1253 is of particular interest as it is located in close vicinity to the dapD and dapE genes in the genome of C. glutamicum. In B. pertussis the dapCDE genes are located in one locus and appear to constitute an operon in this organism ([Fuchs et al., 2000]). The vicinity of cg1253 to dapD (cg1256) and dapE (cg1260) in C. glutamicum might indicate an involvement of cg1253 in the DAP biosynthesis pathway. The cg1253 gene of C. glutamicum was, therefore, tentatively named dapC.

To verify the proposed dapC function of cg1253, an enzyme assay for determination of N-succinyl-aminoketopimelate aminotransferase activity was performed. For this purpose, we constructed a defined deletion in the dapC gene of C. glutamicum RES167 using plasmid pMH6 (Table 1). The resulting deletion mutant C. glutamicum MH3 was used in the assay along with the control strain C. glutamicum RES167 (Table 3). Based on the amount of protein used in the respective assay of C. glutamicum RES167, a specific DapC activity of 0.113 mol min⁻¹ per mg of protein was determined. In contrast to the control strain, no DapC activity was detectable in extracts of C. glutamicum MH3 carrying the dapC deletion. In addition, the dapC gene was overexpressed by cloning its promoter-less coding region into the expression vector pEC-XT99A. The resulting plasmid pMH12 (Table 1) was transferred into the lysine-producing strain C. glutamicum DSM5715 and the recombinant derivative C. glutamicum MH7 was subsequently assayed for aminotransferase activity. Compared with C. glutamicum DSM5715 carrying the empty vector pEC-XT99A, C. glutamicum MH7 revealed an almost 9-fold increase in succinyl-DAP-dependent -glutamate accumulation (Table 3). In conclusion, both enzyme assays confirmed unequivocally that the cg1253 coding region of the C. glutamicum genome codes for a protein with N-succinyl-aminoketopimelate aminotransferase activity.

 

 

Table 3. Specific N-succinyl-aminoketopimelate transaminase activity of C. glutamicum strains
 

 

3.6. Overexpression of the C. glutamicum dapC gene led to improved -lysine production

To analyze whether genetic engineering of the dapC gene can positively influence fermentative -lysine production, C. glutamicum MH7 was used in a production assay. The -lysine concentration in the supernatant of a C. glutamicum MH7 culture was determined after growth in MM5 medium for 72 h. Overexpression of dapC in C. glutamicum MH7 increased the -lysine concentration to 14.7 g l⁻¹, whereas the control strain (C. glutamicum DSM5715 carrying pEC-XT99A) produced only 13.7 g l⁻¹, corresponding to an increase of 7.3%. Therefore, the dapC gene obviously represents a further interesting target for strain development in C. glutamicum.

3.7. The dapC gene is dispensable for the synthesis of D,-diaminopimelate via the succinylase branch

For a further characterization of the properties of DapC in the D,-DAP biosynthesis pathway, we tested the ability of the ΔdapC mutant strain C. glutamicum MH3 to grow on different media. As expected, C. glutamicum MH3 showed no phenotypic alterations when grown in complex media (LBG or BHI) or in minimal medium MM1 (data not shown). Surprisingly, there was also no growth reduction of C. glutamicum MH3 on rich medium (BHI+4% glucose) with excess carbon and limited ammonium availability (Fig. 3B, left). However, a reduced growth of the ΔdapC mutant C. glutamicum MH3 was expected under this growth condition, since a non-functional dapD, dapE ([Wehrmann et al., 1998]) or dapF gene (this work) was shown to cause this phenotype.

To explore this finding, we transformed C. glutamicum strain MH3, carrying a defined deletion in the dapC gene, with plasmid pMH4 (Table 1). Clones with integrated plasmid pMH4 were subsequently selected for a deleted ddh gene by means of the sacB marker system. PCR experiments confirmed that deletions in the dapC and ddh genes were present in the chromosome of C. glutamicum MH4. Since both branches of the D,-DAP and -lysine biosynthesis would be interrupted in this case, this genotype was expected to be lethal for C. glutamicum cells as it was already concluded from the experiment to delete the dapF and ddh gene. Therefore, it can be assumed that at least one additional enzyme is encoded in the C. glutamicum genome, which is able to catalyze the transamination of N-succinyl-aminoketopimelate to N-succinyl-diaminopimelate thus providing sufficient amounts of D,-DAP for normal growth of ΔdapC-Δddh cells, such as C. glutamicum MH4. The expected activity appears to be either too low for a detection in the enzyme assay or is not detectable with the applied test system at all.

In a further experimental approach we examined whether the acetylornithine transaminase ArgD is able to substitute for the DapC function in C. glutamicum. It was considered as a potential candidate for a DapC bypass-reaction since a corresponding enzymatic function was described for the ArgD protein in E. coli ([Ledwidge and Blanchard, 1999]). The E. coli ArgD protein possesses N-acetylornithine and N-succinyl-aminoketopimelate transaminase activities and exhibits a similar catalytic efficiency for both substrates. Starting from C. glutamicum strain MH4, carrying deletions in the dapC and ddh gene, we constructed an additional deletion in the argD coding region (cg1583) by applying plasmid pMH8 (Table 1). After analyzing selected C. glutamicum mutants by PCR experiments (Fig. 4), it became evident that it was possible to delete the dapC, ddh and argD genes in a single C. glutamicum strain which was designated MH5 (Table 1). This set of defined mutations did not affect the viability of C. glutamicum MH5 on standard rich medium indicating that the ArgD protein is not, or not alone, capable to substitute for DapC activity. Accordingly, another yet unknown enzyme present in C. glutamicum must be capable to catalyze the transamination reaction in the D,-DAP and -lysine biosynthesis pathway in addition to the enzymatically characterized DapC protein.

 

 

Fig. 4. Experimental confirmation of the deletions introduced into the dapC, ddh and argD gene of mutant strain C. glutamicum MH5. (A) PCR amplification of the dapC, argD and ddh coding regions of C. glutamicum RES167 (control) and of C. glutamicum MH5. PCR products were obtained with primer pairs located outside of each coding region. Lane 1: DNA marker X (Roche Diagnostics, Mannheim, Germany); lane 2: argD in RES167; lane 3: dapC in RES167; lane 4: ddh in RES167; lane 5: ΔargD in MH5; lane 6: ΔdapC in MH5; lane 7: Δddh in MH5. (B) Schematic illustration of chromosomal deletions in C. glutamicum MH5 and location of primers used for experimental confirmation. Approximate primer positions are indicated by arrows. Open arrows represent the corresponding genes in the control strain C. glutamicum RES167, grey bars symbolize the deleted regions (Δ) in C. glutamicum MH5. The length of the deleted DNA region is given in base pairs. The deletion within the ddh gene comprehends amino acid residues of the Ddh protein, which were a substantial part of the dimerization domain and the substrate-binding domain of this enzyme ([Scapin et al., 1996]).

 

4. DISCUSSION

During decades of industrial strain design, C. glutamicum -lysine producers were classically engineered by random mutagenesis and subsequent selection for desired abilities ([Leuchtenberger, 1996]). Nowadays, C. glutamicum strains with overexpressed pyruvate carboxylase gene pyc ([Peters-Wendisch et al., 2001]), mutated lysC ([Kalinowski et al., 1990 and Onishi et al., 2002]) or overexpressed dapA gene ([de Graaf et al., 2001]) are examples for lysine-producing strains which were obtained by genetic engineering techniques. Identification of the lysine exporter of C. glutamicum provided another interesting target to improve amino acid production ([Vrljic et al., 1996]). Furthermore, genetic work focussed on gene targets encoding functions in the central metabolism of C. glutamicum (reviewed in [Sahm et al., 2000]), but it became more and more difficult to identify suitable target genes and to achieve further strain improvements. A new perspective on C. glutamicum genetic engineering was obtained by establishing the complete genome sequence of the type strain ATCC 13032 ([Tauch et al., 2002a]). Genome sequencing projects provide the complete genetic information of the investigated organisms and allow along with new bioinformatics tools the identification of so far not recognized target genes for genetic engineering ( [Hodgson, 1998]). Annotation of the complete genome sequence of C. glutamicum led to the identification of two candidate coding regions for the last unknown genes, dapC and dapF, involved in the branched pathway for D,-DAP and -lysine biosynthesis of this organism. Genetic and enzymatic analyses enabled us to verify the postulated function in both cases. Cloning and overexpression of dapC and dapF resulted in increased -lysine production in C. glutamicum showing that the knowledge of the complete biochemical pathway for -lysine biosynthesis is of great relevance for biotechnological strain design.

A C. glutamicum mutant strain carrying a deleted dapF gene (MH1) showed the expected phenotype of significantly reduced growth on complex medium with high carbon content and limited availability of ammonium ions. A corresponding phenotype was initially shown for strains lacking a functional dapD or dapE gene ([Wehrmann et al., 1998]). These mutant strains can synthesize the essential cell wall compound D,-DAP only via the one-step dehydrogenase branch of the pathway. The diaminopimelate dehydrogenase has a low affinity towards ammonium ions which are directly used for the reductive conversion of THDP to D,-DAP ([Misono and Soda, 1980]). For C. glutamicum it has been shown in earlier examinations that the dehydrogenase variant is not efficiently used at an external ammonium concentration below 38 mM ([Sonntag et al., 1993]). This indicates that the -lysine biosynthesis pathway is directly influenced by the free ammonium availability, probably via the kinetic characteristics of the Ddh protein. Consistent with this view is the reduced growth of C. glutamicum strains lacking an active succinylase variant of the pathway, as it is the case with a dapD, dapE or a dapF mutant.

Surprisingly, C. glutamicum MH3, carrying a defined dapC deletion, did not show the characteristic growth behavior of mutant strains with a non-functional succinylase branch. Even a simultaneous deletion of the dapC gene together with the ddh gene could be established in C. glutamicum. This is a strong evidence that C. glutamicum possesses at least one additional aminotransferase, which is able to substitute for the mutated dapC gene function of C. glutamicum MH3. In E. coli, the ArgD protein fulfills transamination reactions in both the arginine and D,-DAP biosynthesis pathways ([Ledwidge and Blanchard, 1999]). However, growth analyses of C. glutamicum MH5, deleted in the dapC, ddh, and argD gene, indicates that another yet unknown enzyme is able to substitute the DapC function in C. glutamicum. To further investigate this phenomenon, the C. glutamicum genome sequencing project provides a very promising basis, because it allows a systematic approach for analyzing the entirety of aminotransferases encoded in the chromosome. In conclusion, the C. glutamicum genome sequencing project provided great benefit with regard to the identification of potential target genes useful for industrial strain design. The identification of the dapC and dapF gene in C. glutamicum, described here, is considered to be an impressive example of the application of genome research for strain optimization.

ACKNOWLEDGEMENTS

This work was granted in part by Hermann Schlosser Stiftung, Frankfurt. The authors thank the Degussa AG for providing nucleotide sequence data and financial support. Plasmid pMH4 was kindly provided by C. Rückert (University of Bielefeld).

REFERENCES

Baldi et al., 1999. P. Baldi, S. Brunak, P. Fiasconi and G. Soda, Exploiting the past and future in protein secondary structure prediction. Bioinformatics 15 (1999), pp. 937–946.

Bensadoun and Weinstein, 1976. A. Bensadoun and D. Weinstein, Assay of proteins in the presence of interfering materials. Anal. Biochem. 70 (1976), pp. 241–250.

Cirilli et al., 1998. M. Cirilli, R. Zheng, G. Scapin and S. Blanchard, Structural symmetry: the three-dimensional structure of Haemophilus influenzae diaminopimelate epimerase. Biochemistry 37 (1998), pp. 16452–16458.

Cremer et al., 1990. J. Cremer, L. Eggeling and H. Sahm, Cloning of the dapA dapB cluster of the lysine-secreting bacterium Corynebacterium glutamicum. Mol. Gen. Genet. 220 (1990), pp. 478–480.

de Graaf et al., 2001. A.A. de Graaf, L. Eggeling and H. Sahm, Metabolic engineering for -lysine production by Corynebacterium glutamicum. Adv. Biochem. Eng. Biotechnol. 73 (2001), pp. 9–29.

Fuchs et al., 2000. T.M. Fuchs, B. Schneider, K. Krumbach, L. Eggeling and R. Gross, Characterization of a Bordetella pertussis diaminopimelate (DAP) biosynthesis locus identifies dapC, a novel gene coding for an N-Succinyl-,-DAP aminotransferase. J. Bacteriol. 182 (2000), pp. 3626–3631.

Grant et al., 1990. S. Grant, G.J. Jessee, F.R. Bloom and D. Hanahan, Differential plasmid rescue from transgenic mouse DNAs into Escherichia coli methylation-restriction mutants. Proc. Natl. Acad. Sci. USA 87 (1990), pp. 4645–4649.

Hodgson, 1998. J. Hodgson, LION and Degussa apply genomics to fermentation. Nat. Biotechnol. 16 (1998), p. 715.

Horton, 1995. R.M. Horton, PCR-mediated recombination and mutagenesis-SOEing together tailor-made genes. Mol. Biotechnol. 3 (1995), pp. 93–99.

Ishino et al., 1987. S. Ishino, Z. Mizukami, K. Yamaguchi, R. Katsumata and K. Araki, Nucleotide sequence of the meso-diaminopimelate -dehydrogenase gene from Corynebacterium glutamicum. Nucleic Acids Res. 15 (1987), p. 3917.

Jones and Gilligan, 1983. B.N. Jones and J.P. Gilligan, o-Phthaldialdehyde precolumn derivatization and reversed-phase high-performance liquid chromatography of polypeptide hydrolysates and physiological fluids. J. Chromatogr. 266 (1983), pp. 471–482.

Kalinowski et al., 1990. J. Kalinowski, B. Bachmann, G. Thierbach and A. Pühler, Aspartokinase genes lysC and lysC overlap and are adjacent to the aspartate -semialdehyde dehydrogenase gene asd in Corynebacterium glutamicum. Mol. Gen. Genet. 224 (1990), pp. 317–324.

Kalinowski et al., 1991. J. Kalinowski, J. Cremer, B. Bachmann, L. Eggeling, H. Sahm and A. Pühler, Genetic and biochemical analysis of the aspartokinase from Corynebacterium glutamicum. Mol. Microbiol. 5 (1991), pp. 1197–1204.

Katsumata et al., 1984. R. Katsumata, A. Ozaki, T. Oka and A. Furuya, Protoplast transformation of glutamate-producing bacteria with plasmid DNA. J. Bacteriol. 159 (1984), pp. 306–311.

Keilhauer et al., 1993. C. Keilhauer, L. Eggeling and H. Sahm, Isoleucine synthesis in Corynebacterium glutamicum: molecular analysis of the ilvB-ilvN-ilvC operon. J. Bacteriol. 175 (1993), pp. 5595–5603.

Kindler and Gilvarg, 1960. S.H. Kindler and C. Gilvarg, N-succinyl--,-diaminopimelic acid deacylase. J. Biol. Chem. 235 (1960), pp. 3532–3535.

Kirchner and Tauch, 2003. Kirchner, O., Tauch, A., 2003. Tools for genetic engineering in the amino acid-producing bacterium Corynebacterium glutamicu. J. Biotechnol. 104, in press.

Koo and Blanchard, 1999. C.W. Koo and J.S. Blanchard, Chemical mechanism of Haemophilus influenzae diaminopimelate epimerase. Biochemistry 38 (1999), pp. 4416–4422.

Ledwidge and Blanchard, 1999. R. Ledwidge and J.S. Blanchard, The dual capability of N-acetylornithine aminotransferase in arginine and lysine biosynthesis. Biochemistry 38 (1999), pp. 3019–3024.

Leuchtenberger, 1996. W. Leuchtenberger, Amino acids—technical production and use. In: H.J. Rehm, G. Reed, A. Puehler and P. Stadler, Editors, Biotechnology 6, VCH, Weinheim, Germany (1996), pp. 465–502.

Menkel et al., 1989. E. Menkel, G. Thierbach, L. Eggeling and H. Sahm, Influence of aspartate availability on lysine formation by a recombinant strain of Corynebacterium glutamicum and utilization of fumarate. Appl. Environ. Microbiol. 55 (1989), pp. 684–688.

Misono and Soda, 1980. H. Misono and K. Soda, Properties of meso-diaminopimelate -dehydrogenase from Bacillus sphaericus. J. Biol. Chem. 255 (1980), pp. 10599–10605.

Nesvera et al., 1997. J. Nesvera, M. Patek, J. Hochmannova, Z. Abrhamova, V. Becvarova, M. Jelinkova and J. Vohradsky, Plasmid pGA1 from Corynebacterium glutamicum codes for a gene product that positively influences plasmid copy number. J. Bacteriol. 179 (1997), pp. 1525–1532.

Onishi et al., 2002. J. Onishi, S. Mitsuhashi, M. Hayashi, S. Ando, H. Yokoi, K. Ochiai and M. Ikeda, A novel methodology employing Corynebacterium glutamicum genome information to generate a new -lysine producing mutant. Appl. Microbiol. Biotechnol. 58 (2002), pp. 217–223.

Patte, 1996. J.P. Patte, Biosynthesis of threonine and lysine. In: F.C. Neidhardt, R. Curtiss, III, J.L. Ingraham, E.C.C. Lin, K.B. Low, B. Magasanik, W.S. Reznikoff, M. Riley, M. Schaechter and H.E. Umbarger, Editors, Escherichia coli and Salmonella—Cellular and Molecular Biology 1, ASM Press, Washington, DC (1996), pp. 528–541.

Peters-Wendisch et al., 2001. P.G. Peters-Wendisch, B. Schiel, V.F. Wendisch, E. Katsoulidis, B. Möckel, H. Sahm and B.J. Eikmanns, Pyruvate carboxylase is a major bottleneck for glutamate and lysine production by Corynebacterium glutamicum. J. Mol. Microbiol. Biotechnol. 3 (2001), pp. 295–300.

Richaud and Printz, 1988. C. Richaud and C. Printz, Nucleotide sequence of the dapF gene and flanking regions from Escherichia coli K12. Nucleic Acids Res. 16 (1988), p. 10367.

Richaud et al., 1987. C. Richaud, W. Higgins, D. Mengin-Lecreulx and P. Stragier, Molecular cloning, characterization and chromosomal localization of dapF, the Escherichia coli gene for diaminopimelate epimerase. J. Bacteriol. 169 (1987), pp. 1454–1459.

Sahm et al., 2000. H. Sahm, L. Eggeling and A.A. de Graaf, Pathway analysis and metabolic engineering in Corynebacterium glutamicum. Biol. Chem. 381 (2000), pp. 899–910.

Sambrook et al., 1989. J. Sambrook, E.F. Fritsch and T. Maniatis. Molecular Cloning: A Laboratory Manual (second ed ed.),, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989).

Scapin and Blanchard, 1998. G. Scapin and J.S. Blanchard, Enzymology of bacterial lysine biosynthesis. Adv. Enzymol. Relat. Areas Mol. Biol. 72 (1998), pp. 279–324.

Scapin et al., 1996. G. Scapin, G.R. Sreelatha and J.S. Blanchard, Three-dimensional structure of meso-diaminopimelate dehydrogenase from Corynebacterium glutamicum. Biochemistry 35 (1996), pp. 13540–13551.

Schäfer et al., 1994. A. Schäfer, A. Tauch, W. Jäger, J. Kalinowski, G. Thierbach and A. Pühler, 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 (1994), pp. 69–73.

Schrumpf et al., 1991. B. Schrumpf, A. Schwarzer, J. Kalinowski, A. Pühler, L. Eggeling and H. Sahm, A functionally split pathway for lysine synthesis in Corynebacterium glutamicum. J. Bacteriol. 173 (1991), pp. 4510–4516.

Sonntag et al., 1993. K. Sonntag, L. Eggeling, A.A. de Graaf and H. Sahm, Flux partitioning in the split pathway of lysine synthesis in Corynebacterium glutamicum. Eur. J. Biochem. 123 (1993), pp. 1325–1331.

Tauch et al., 1994. A. Tauch, O. Kirchner, L. Wehmeier, J. Kalinowski and A. Pühler, Corynebacterium glutamicum DNA is subjected to methylation-restriction in Escherichia coli. FEMS Microbiol. Lett. 123 (1994), pp. 343–348.

Tauch et al., 1995. A. Tauch, F. Kassing, J. Kalinowski and A. Pühler, The Corynebacterium xerosis composite transposon Tn5432 consists of two identical insertion sequences, designated IS1249, flanking the erythromycin resistance gene ermCX. Plasmid 34 (1995), pp. 119–131.

Tauch et al., 2002a. A. Tauch, I. Homann, S. Mormann, S. Rüberg, A. Billault, B. Bathe, S. Brand, O. Brockmann-Gretza, C. Rückert, N. Schischka, C. Wrenger, J. Hoheisel, B. Möckel, K. Huthmacher, W. Pfefferle, A. Pühler and J. Kalinowski, Strategy to sequence the genome of Corynebacterium glutamicum ATCC 13032: use of a cosmid and a bacterial artificial chromosome library. J. Biotechnol. 95 (2002), pp. 25–38.

Tauch et al., 2002b. A. Tauch, O. Kirchner, B. Löffler, S. Götker, A. Pühler and J. Kalinowski, Efficient electrotransformation of Corynebacterium diphtheriae with a mini-replicon derived from the Corynebacterium glutamicum plasmid pGA1. Curr. Microbiol. 45 (2002), pp. 362–367.

Thompson et al., 1994. J.D. Thompson, D.G. Higgins and T.J. Gibson, CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22 (1994), pp. 4673–4680.

Vrljic et al., 1996. M. Vrljic, H. Sahm and L. Eggeling, A new type of transporter with a new type of cellular function: -lysine export from Corynebacterium glutamicum. Mol. Microbiol. 22 (1996), pp. 815–826.

Wehrmann et al., 1994. A. Wehrmann, L. Eggeling and H. Sahm, Analysis of different DNA fragments of Corynebacterium glutamicum complementing dapE of Escherichia coli. Microbiology 140 (1994), pp. 3349–3356.

Wehrmann et al., 1998. A. Wehrmann, B. Phillipp, H. Sahm and L. Eggeling, Different modes of diaminopimelate synthesis and their role in cell wall integrity: a study with Corynebacterium glutamicum. J. Bacteriol. 180 (1998), pp. 3159–3165.

Weinberger and Gilvarg, 1970. S. Weinberger and C. Gilvarg, Bacterial distribution of the use of succinyl and acetyl blocking groups in diaminopimelic acid biosynthesis. J. Bacteriol. 101 (1970), pp. 323–324.

Wiseman and Nichols, 1984. J.S. Wiseman and J.S. Nichols, Purification and properties of diaminopimelic acid epimerase from Escherichia coli. J. Biol. Chem. 259 (1984), pp. 8907–8914.

Yeh et al., 1988. P. Yeh, A. Sicard and A. Sinskey, General organization of the genes specifically involved in the diaminopimelate-lysine biosynthetic pathway of Corynebacterium glutamicum. Mol. Gen. Genet. 212 (1988), pp. 105–111.

(order Full Text from publisher)