Characterization of the Dicarboxylate Transporter DctA in Corynebacterium glutamicum

Bacterial strains, plasmids, and culture conditions.All strains and plasmids used in this study are listed in Table 1. Cultivation of E. coli DH5α was performed in Luria-Bertani (LB) complex medium (28) at 37°C. C. glutamicum was cultivated at 30°C in LB or in brain heart infusion (Difco) complex medium. CGXII was used as minimal medium for C. glutamicum (5). Glucose, sodium succinate, sodium fumarate, sodium l-malate, potassium acetate, magnesium citrate, sodium pyruvate, or sodium l-lactate was used as a carbon source at a concentration of 100 mM. Kanamycin (25 mg/liter) and isopropyl-β-d-thiogalactopyranoside (IPTG) (1 mM or 25 μM) were added when appropriate. The growth of C. glutamicum was observed by measuring the optical density at 600 nm (OD₆₀₀) in 500-ml baffled shake flasks with 120-rpm agitation. The biomass concentration was calculated from OD₆₀₀ values using an experimentally determined correlation factor of 0.25 g cells (dry weight [DW]) liter⁻¹ for an OD₆₀₀ of 1. For growth under oxygen-deprived conditions, the media were flushed with nitrogen for 30 min in anaerobic test tubes, and cysteine-HCl (0.5 g liter⁻¹) and reazurin (1 mg liter⁻¹) were added to the media to remove and control the presence of residual oxygen. Nitrate (30 mM) was added to the anaerobic growth media as a terminal electron acceptor. The cells were incubated for 20 h at 30°C. For succinate production, C. glutamicum was cultivated aerobically at 30°C in CGXII minimal medium with 4% glucose before being harvested and washed twice with CGXII. The washed cells were inoculated into anaerobic test tubes that were flushed with nitrogen and contained CGXII minimal medium with 200 mM glucose, cysteine-HCl (0.5 g liter⁻¹), 30 mM nitrate, and reazurin (1 mg liter⁻¹). The formation of succinate was determined by high-performance liquid chromatography.

Construction of plasmids and strains.All plasmids were constructed in E. coli DH5α by using PCR-generated fragments (KOD Hot Start DNA polymerase; Novagen). Genomic DNA from wild-type (WT) C. glutamicum, which was prepared as described previously (6), was used as a template for PCR. E. coli was treated and transformed by standard methods (28).

Homologous overexpression of dctA.In order to overexpress dctA in WT C. glutamicum, dctA was amplified by using the primer pair Ex-dctA-fw (GCGGATCCGAAAGGAGGCCCTTCAG ATGGATTCAAACACAGAATC; the BamHI site is in boldface, and the artificial ribosome binding site is in italics) and Ex-dctA-bw (GCGGTACCCTAGTGGTCTTCTTCTAACTC; the KpnI site is in boldface). The PCR fragment was cloned via BamHI and KpnI sites into the expression vector pVWEx1 (25). The constructed vector pVWEx1-dctA allows the IPTG-inducible expression of dctA in C. glutamicum.

Deletion of dccT and dctA in C. glutamicum.For construction of in-frame dctA and dccT deletion mutants of C. glutamicum, a two-step homologous-recombination procedure was applied by using the deletion vectors pK19mobsacB-Del-dctA and pK19mobsacB-Del-dccT. The upstream and downstream regions of dctA or dccT were PCR generated by using the primers Del-dctA-1 (CGGGATCCAAGTGGCGTGAGACTGCATT; the BamHI site is in boldface), Del-dctA-2 (CCCATCCACTAAACTTAAACATTCTGTGTTTGAATCCATGA), Del-dctA-3 (TGTTTAAGTTTAGTGGATGGGATCTCCATTTGTGGAGTTAG), and Del-dctA-4 (CGGGATCCGACATTCTTGCGCCACCATT; the BamHI site is in boldface) and Del-dccT-1 (CGGGATCCCCAAGTTGGTGCTCGGAATCT; the BamHI site is in boldface), Del-dccT-2 (CCCTCCACTAAACTTAAACAAATGTCAGGTGTGCTCATTG), Del-dccT-3 (TGTTTAAGTTTAGTGGATGGGTTCGTAGCGATCCCACTCTTTG), and Del-dccT-4 (CGGGATCCCTCGGCGCACGGATCGAAGTGTT; the BamHI site is in boldface), respectively. The fragment amplified by crossover PCR using the up- and downstream PCR fragments as templates was blunt-end cloned into a SmaI-restricted deletion vector, pK19mobsacB (29). The plasmid pK19mobsacB-Del-dctA was introduced into WT C. glutamicum via conjugation with E. coli S17-1. The vector pK19mobsacB-Del-dccT for in-frame deletion of dccT was introduced into C. glutamicum via electroporation. The deletion was introduced into the C. glutamicum genome as described previously (24, 29). The chromosomal deletions of dctA and dccT in C. glutamicum were confirmed via PCR with the primers Del-dctA-co-fw (GGCCGTAACATCGCTCAAC) and Del-dctA-co-bw (ATGCGGACAAGCGCACTGTA) or Del-dccT-co-fw (CCAGCGGTGTTCTCTAATCCT) and Del-dccT-co-bw (CCAGCGACGCGCCAAAGAATC), respectively. To construct a double-deletion mutant of dctA and dccT, pK19mobsacB-Del-dccT was transformed into C. glutamicum ΔdctA by electroporation.

Construction of reporter gene fusions and chloramphenicol acetyltransferase assay.To compare and to monitor the dctA promoter activity of WT C. glutamicum and the l-malate spontaneous mutant (MSM) during growth with different carbon sources, transcriptional fusion analyses were performed. The dctA promoter region (−229 to +45) was amplified with the primer pair pET2-dctA-Prom-1 (CGGGATCCCAACCGGTGTGACCGAATAA; the BamHI site is in boldface) and pET2-dctA-Prom-2 (CGGAGCTCGTGTGATCGTTGCCGACGAA; the SacI site is in boldface). The PCR-generated fragment was cloned via BamHI and SacI sites into the corynebacterial promoter probe vector pET2 (37). WT C. glutamicum was transformed with the constructed vectors pET2-dctA-WT and pET2-dctA-MSM. For the determination of the chloramphenicol acetyltransferase activity, 50 ml of exponentially growing cells was harvested by centrifugation (10 min; 3,200 × g at 4°C) and washed twice with 25 ml 0.08 mM Tris-HCl, pH 7.0. The pellets were resuspended in 1 ml 0.08 mM Tris-HCl, pH 7.0, and mechanically disrupted by using a Silamat S5 (Vivadent, Ellwangen, Germany) twice for 20 s each time with 250 mg 0.1-mm zirconium-silica beads (Roth, Karlsruhe, Germany). The supernatants were isolated by centrifugation (1 h; 14.000 × g; 4°C) and used to measure the chloramphenicol acetyltransferase activity as described previously (32).

DNA microarray analysis and 5′-RACE PCR.To compare the transcriptomes of WT C. glutamicum and the MSM, total RNA was isolated from cells growing exponentially in LB medium as described previously (38). The purified RNA was analyzed by UV spectrometry for quantity and quality and was stored at −20°C until it was used. DNA microarray analysis, synthesis of fluorescently labeled cDNA from total RNA, DNA microarray hybridization, and gene expression analysis were performed as described previously (23, 26, 38). To determine the transcription start site of dctA, a 5′ rapid amplification of cDNA ends (5′-RACE) PCR was performed as described previously using two independent labeling reactions with either adenine or cytosine (40) and using the primers RACE-dctA-1 (CCACGAGGATGGCGATGAAC) and RACE-dctA-2 (CGAGTGCGAATGTGGACATC). The generated PCR fragment was sequenced at the Klinikum Muenster, Labor für molekulare Diagnostik (Muenster, Germany).

Transport assays.Cells were grown to mid-exponential phase in minimal medium MM1 (21) supplemented with glucose as a sole carbon source and 25 μM IPTG for induction of pVWEx1-dctA expression. Subsequently, the cells were washed three times with 50 mM MES (morpholineethanesulfonic acid), 50 mM Tris, pH 8.0, 10 mM NaCl, 10 mM KCl and incubated on ice until the measurement was performed. Before the transport assay, cells were incubated for 3 min at an OD₆₀₀ of approximately 2 at 30°C with 10 mM glucose for energizing. As tracers, ¹⁴C-labeled succinate, fumarate, and malate with specific activities of 2, 0.13, and 2.04 GBq/mmol, respectively (MP-Biochemicals, Illkirch, France) were added as sodium salts (final volume, 1.2 ml), and 200-μl samples were taken every 5 s for 1 to 2 min in the case of succinate and every 30 s in the cases of malate and fumarate in order to determine initial uptake rates and to avoid saturation of transport. Cells were collected on GF55 glass fiber filters (Schleicher and Schuell, Dassel, Germany) and washed twice with 2.5 ml 0.1 M LiCl at 22°C. After resuspension of the cells in scintillation fluid (Rotiszinth, Roth, Germany), the radioactivity of the sample was determined in a scintillation counter (Beckman, Krefeld, Germany). In order to analyze the substrate specificity and driving force, uptake measurements were performed in the presence of saturating substrate concentrations (700 μM l-malate) and putative substrates in 100-fold excess or in the presence of inhibitors (17).

Isolation and transcriptome analysis of the C. glutamicum MSM able to grow with l-malate as a sole carbon source.WT C. glutamicum did not show significant growth in CGXII minimal medium containing l-malate as a sole carbon source for at least 24 h (Table 2). However, after prolonged incubation in l-malate minimal medium, a spontaneous mutant (MSM) was isolated. When cells of such a culture were cultivated intermittently on LB complex medium before being transferred to minimal medium with l-malate as the sole carbon source, rapid growth occurred without a lag phase, indicating that a mutant had been selected during the first culture on l-malate-containing minimal medium. The MSM grew without a lag phase in minimal medium with either l-malate, fumarate, or succinate when inoculated from minimal or complex medium (Table 2). In order to determine the genes that showed altered expression in the MSM compared to WT C. glutamicum, a DNA microarray experiment was performed with total RNA isolated from cells growing exponentially in LB medium, i.e., under presumed noninducing conditions. The transcriptome analysis revealed unaltered expression of dccT, the gene for the DASS family uptake system for dicarboxylates, while 10 genes showed decreased mRNA levels in the MSM, and expression of three genes was higher than in the WT (Table 3). The mRNA levels of two genes encoding putative transport systems, cg2870/dctA and cg3226, were higher in the MSM than in the WT. The lldD-cg3226 operon is derepressed during growth on l-lactate (35) and is repressed by LldR (9). While lldD-encoded quinone-dependent l-lactate dehydrogenase is required for growth of C. glutamicum with l-lactate as the sole carbon source, cg3226 is dispensable (35). The gene cg2870 encodes the putative transport protein DctA (TC 2.A.23.1.3), which belongs to the DAACS family. Consequently, we assumed that the MSM had gained the ability to grow on dicarboxylates due to high expression levels of dctA.

Identification of a promoter-up mutation of dctA in C. glutamicum MSM.The increased mRNA level of dctA could be due either to a cis mutation in the promoter region of the dctA gene or to a trans mutation affecting a regulatory gene. To localize a possible mutation of the dctA promoter region, the transcription start point was first determined by 5′-RACE PCR. Transcription was initiated 119 nucleotides upstream of the translation start site with adenine. The −10 hexamer, CATAAT, of the WT dctA promoter is similar to the −10 core hexamer consensus sequence (TAATAT) (22). The promoter sequence of dctA in C. glutamicum MSM showed a C-to-T transition mutation 12 nucleotides upstream of the transcription start site (Fig. 1). To test whether this point mutation of the promoter region is responsible for high expression of dctA, a promoter transcription-fusion analysis was performed. The dctA promoter regions of WT C. glutamicum and the MSM were cloned into the promoter probe vector pET2. The resulting plasmids, pET2-dctA-WT and pET2-dctA-MSM, were transformed into WT C. glutamicum, and the expression levels of both transcription fusions were analyzed under different growth conditions (Fig. 2). Under oxygen deprivation conditions, expression of the dctA′-′cat fusion C. glutamicum(pET2-dctA-WT) was low in glucose minimal medium (0.02 μmol mg⁻¹ min⁻¹) and in LB medium (0.04 μmol mg⁻¹ min⁻¹) (data not shown). Under aerobic conditions, expression of the dctA′-′cat fusion in C. glutamicum(pET2-dctA-WT) was low (0.02 to 0.04 μmol mg⁻¹ min⁻¹) on minimal medium with glucose, l-lactate, acetate, pyruvate, or citrate, as well as on LB complex medium with or without the addition of succinate, fumarate, or l-malate. In contrast, C. glutamicum(pET2-dctA-MSM) showed 25- to 45-fold-higher expression levels of the dctA′-′cat fusion under these conditions (Fig. 2). Thus, the mutation of the −10 promoter region in C. glutamicum MSM led to constitutive and high expression under the tested conditions.

• Open in new tab
• Download powerpoint

FIG. 1.

DNA sequence of the dctA promoter region from WT C. glutamicum. The transcriptional start site (TS) is underlined and in boldface, the putative −10 hexamer is underlined, and the start codon of dctA is highlighted in boldface. The C→T transition identified in the MSM is shown below the sequence of WT C. glutamicum and marked by an arrow. The genes adjacent to dctA encode a hypothetical protein with an unknown function (cg2875); phosphoribosylaminoimidazole-succinocarboxamide synthase (hemH); a prolyl oligopeptidase (ptrB); a putative deacetylase, COG3233 (cg2869); and an extracellular nuclease (nuc). The first and last nucleotides of the sequence shown are in italics and correspond to the indicated positions of the genome sequence BX927147 (19). nt, nucleotide.

• Open in new tab
• Download powerpoint

FIG. 2.

Comparative expression analysis of dctA′-′cat reporter gene fusions of the upstream dctA regions from WT C. glutamicum and the MSM. Expression of the dctA′-′cat reporter gene fusions was determined for WT C. glutamicum(pET2-dctA-WT) (empty columns) and WT C. glutamicum(pET2-dctA-MSM) (filled columns) by measuring the chloramphenicol acetyltransferase activity after cultivation in CGXII minimal medium containing 100 mM glucose, 100 mM l-lactate, 100 mM acetate, 100 mM pyruvate, and 100 mM MgCl₂ with 100 mM citrate or in LB complex medium without or with 100 mM fumarate, 100 mM l-malate, or 100 mM succinate. The values represent means and standard deviations from at least three independent cultivations.

Homologous overexpression of dctA in C. glutamicum enables growth with l-malate, succinate, and fumarate as sole carbon sources.In order to demonstrate that high expression of dctA is sufficient for C. glutamicum to allow growth with l-malate, succinate, and fumarate as sole carbon sources, the IPTG-inducible plasmid pVWEx1-dctA was constructed and transferred into WT C. glutamicum. The resulting strain, C. glutamicum(pVWEx1-dctA), was grown in the presence or absence of IPTG in CGXII minimal medium containing either l-malate, succinate, or fumarate (Fig. 3). Significant growth of C. glutamicum(pVWEx1-dctA) was observed only in the presence of the inducer. When the IPTG concentration was varied, optimal growth was observed with 25 μM IPTG, while higher concentrations led to slower growth, which is not unusual for overproduction of membrane proteins. C. glutamicum(pVWEx1-dctA) showed growth rates of 0.34 h⁻¹ in succinate, 0.35 h⁻¹ in fumarate, and 0.32 h⁻¹ in l-malate minimal medium. In contrast, utilization of 2-oxoglutarate was not observed for WT C. glutamicum or for C. glutamicum(pVWEx1-dctA) (data not shown).

• Open in new tab
• Download powerpoint

FIG. 3.

Growth of WT C. glutamicum(pVWEx1-dctA) in CGXII minimal medium with different carbon sources in the absence (open symbols) or presence (filled symbols) of 25 μM IPTG. Fumarate (triangles), succinate (circles), or l-malate (squares) was used as a sole carbon source at a concentration of 100 mM. Means and ranges or standard deviations of two or more independent growth experiments are shown.

DctA imports dicarboxylic acid intermediates of the citric acid cycle.The uptake of l-malate was analyzed in transport experiments monitoring the import of ¹⁴C-labeled substrates to demonstrate that DctA is active as an uptake system for dicarboxylates and that the isolated mutant exhibits increased l-malate uptake activity. As depicted in Fig. 4A, very low malate uptake rates were observed in WT C. glutamicum, corresponding to the inability of the strain to utilize malate for growth. In contrast, significant malate uptake activity was observed for the MSM, with a V _max of 41 ± 1.6 nmol min⁻¹ mg (DW)⁻¹ and a K_m value of 561 ± 46 μM determined after nonlinear regression according to Michaelis-Menten kinetics. For WT C. glutamicum(pVWEx1-dctA), a V _max of 91 ± 3 nmol min⁻¹ mg (DW)⁻¹ and a K_m value of 736 ± 40 μM were determined. The comparable K_m values for malate uptake in C. glutamicum MSM and WT C. glutamicum(pVWEx1-dctA) indicate that the same transport system is active in both strains, while the different V _max values may be explained by different dctA expression levels of the two strains.

• Open in new tab
• Download powerpoint

FIG. 4.

Uptake of l-malate (A), fumarate (B), and succinate (C) by WT C. glutamicum (triangles), WT C. glutamicum(pVWEx1-dctA) (circles), and the MSM (squares in panel A) as measured with ¹⁴C-labeled substrates. The uptake rates were determined as a function of the substrate concentration, and the data were fitted according to the Michaelis-Menten equation. The data are means and standard deviations derived from at least three independent measurements.

Either the electrochemical Na⁺ or H⁺ potential is suggested to be the driving force for DctA-type transporters. To determine the mode of energizing C. glutamicum DctA, l-malate uptake was measured in WT C. glutamicum(pVWEx1-dctA) in the presence of very low concentrations of Na⁺ ions (10 μM), but no impact on l-malate uptake was observed (Fig. 4B and C). To further elucidate the dependence of DctA activity on the H⁺ potential, the effect of collapsing the membrane potential with valinomycin in the presence of 200 mM K⁺ on malate transport was investigated in the presence or absence of a proton gradient. Assuming an internal pH of about 7.4 (7) at an external pH of 6, upon addition of valinomycin, a significant proton gradient is still present, which is not the case at an external pH of 7.5. Whereas l-malate transport was not significantly affected at pH 6, it was strongly reduced at pH 7.5, indicating that the driving force for DctA-dependent l-malate uptake in C. glutamicum is the electrochemical proton potential (Fig. 5).

• Open in new tab
• Download powerpoint

FIG. 5.

Driving force and inhibition of l-malate uptake by DctA. (A) l-Malate uptake by WT C. glutamicum(pVWEx1-dctA) was measured in terms of dependence on the external H⁺ concentration in the presence of 200 mM potassium chloride. The membrane potential was disrupted by the addition of 20 μM valinomycin (Val), and the measurements were performed at different pH values in order to maintain a pH gradient across the membrane. Transport assays were performed in the presence of 700 μM l-malate, and means and standard deviations of triplicates are shown. (B) Inhibition of l-malate uptake upon the addition of several organic acids as measured for WT C. glutamicum(pVWEx1-dctA) with 700 μM l-malate. Competing substrates were added in 100-fold excess (fumarate in 10-fold excess). Measurements were performed in triplicate and normalized to the control value of 62 ± 2, which was set to 100%.

Uptake measurements with labeled succinate and fumarate were performed in WT C. glutamicum and WT C. glutamicum(pVWEx1-dctA) to test whether these C₄-dicarboxylates are also transported by DctA. Whereas for WT cells only low uptake activities were found, with V _max values below 10 nmol min⁻¹ mg (DW)⁻¹, overexpression of the dctA gene caused a significant increase in fumarate and succinate uptake activities (results not shown). The data were fitted according to the Michaelis-Menten equation, and after subtraction of the WT values, K_m values of 218 ± 69 μM for succinate and 232 ± 28 μM for fumarate, as well as V _max values of 37 ± 4 min⁻¹ mg (DW)⁻¹ for succinate and 15 ± 0.9 min⁻¹ mg (DW)⁻¹ for fumarate could be determined. Thus, DctA mediates the uptake of l-malate, succinate, and fumarate into C. glutamicum.

In order to test for further substrates of C. glutamicum DctA, l-malate uptake assays were carried out with unlabeled putative substrates present in 100-fold excess (Fig. 5B). Succinate and fumarate blocked l-malate uptake almost completely, while minor inhibition of malate transport was found at high concentrations of the monocarboxylates l-lactate and pyruvate, as well as the amino acids l-glutamate and l-aspartate. Addition of isocitrate had a minor effect, but in the presence of citrate, uptake of l-malate was inhibited more than twofold. Moreover, in the presence of the TCA intermediates 2-oxoglutarate and oxaloacetate, uptake of l-malate was blocked completely. Interestingly, very low malate uptake activity was also found in the presence of glyoxylate. Accordingly, besides l-malate, succinate, and fumarate, the intermediates of the TCA and glyoxlate cycles 2-oxoglutarate, oxaloacetate, and glyoxylate may be substrates of DctA from C. glutamicum.

Glyoxylate inhibits utilization of l-malate by WT C. glutamicum(pVWEx1-dctA).DctA is inhibited by glyoxylate, whereas previous experiments revealed that uptake of C₄-dicarboxylates by the uptake system DccT is not inhibited by glyoxylate (40). Glyoxylate did not support the growth of C. glutamicum as a sole carbon source, and in its presence, growth in glucose minimal medium was perturbed, while growth in acetate minimal medium was less affected (data not shown), presumably due to high malate synthase activity present under the latter conditions (10). Therefore, utilization of l-malate by WT C. glutamicum(pVWEx1), WT C. glutamicum(pVWEx1-dctA), and WT C. glutamicum(pVWEx1-dccT) was tested in acetate minimal medium in the absence or presence of glyoxylate. l-Malate was utilized as a cosubstrate with acetate if either dctA or dccT was expressed (Fig. 6). Glyoxylate inhibited utilization of l-malate for biomass formation only if uptake of l-malate was mediated via DctA, but not via DccT (Fig. 6), which is commensurate with the finding that in the in vitro transport assays glyoxylate inhibited DctA, but not DccT (Fig. 5B) (40). Similar results were obtained for growth in acetate minimal medium with either succinate or fumarate in the presence or absence of glyoxylate (data not shown).

• Open in new tab
• Download powerpoint

FIG. 6.

Biomass formation of WT C. glutamicum(pVWEx1), WT C. glutamicum(pVWEx1-dctA), and WT C. glutamicum(pVWEx1-dccT) in CGXII minimal medium with 50 mM acetate (black columns), 50 mM l-malate plus 50 mM acetate (grey columns), or 50 mM acetate plus 50 mM l-malate plus 50 mM glyoxylate (open columns) as carbon sources. For induction, 1 mM IPTG was used for WT C. glutamicum(pVWEx1) and WT C. glutamicum(pVWEx1-dccT) and 25 μM IPTG for WT C. glutamicum(pVWEx1-dctA). Means and ranges or standard deviations of two or more independent growth experiments are shown.

Roles of dctA and dccT in the occurrence of spontaneous mutants able to grow on succinate, fumarate, or l-malate.The growth of C. glutamicum on l-malate, fumarate, and succinate required overexpression of either dctA (Fig. 3) or dccT (40) due to their ectopic expression or due to spontaneous promoter-up mutations. To determine if additional spontaneous mutations leading to growth on C₄-dicarboxylates might occur, defined numbers of cells of WT C. glutamicum and of the newly constructed ΔdctA, ΔdccT, and ΔdctA ΔdccT strains were plated on CGXII mineral medium agar plates containing 20 mM l-malate as the sole carbon source. After incubation for 7 days at 30°C, small colonies appeared for WT C. glutamicum with a frequency of 7.1 × 10⁻⁸ and with reduced frequencies for the ΔdctA and ΔdccT single mutants (3.6 × 10⁻⁸ each), while no colonies were obtained for the ΔdctA ΔdccT double deletion mutant (<1.7 × 10⁻⁸). On fumarate minimal medium agar plates, colonies were found for WT C. glutamicum (2.5 × 10⁻⁷) and C. glutamicum ΔdctA (1.4 × 10⁻⁷), but not for the ΔdccT and ΔdctA ΔdccT strains (<1.7 × 10⁻⁸). Similarly, on succinate minimal medium agar plates, colonies were found for WT C. glutamicum (2.7 × 10⁻⁷) and C. glutamicum ΔdctA (5.8 × 10⁻⁸), but not for the ΔdccT and ΔdctA ΔdccT strains (<1.7 × 10⁻⁸). Thus, the occurrence of mutants able to grow on succinate or fumarate primarily depended on dccT, while mutations leading to growth on l-malate depended on either dccT or dctA.

Biomass formation from C₄-dicarboxylates added to complex medium requires DccT.When added to complex medium, succinate, fumarate, and l-malate contribute to biomass formation by C. glutamicum (Fig. 7). To test whether DctA and/or DccT is involved in the uptake of these C₄-dicarboxylates during growth in LB, WT C. glutamicum and the ΔdctA and ΔdccT strains were cultivated in LB medium without or with 50 mM succinate, fumarate, or l-malate. While WT C. glutamicum and the ΔdctA strain grew to higher biomass concentrations in LB with succinate, fumarate, or l-malate than in LB alone, C. glutamicum ΔdccT grew to about the same biomass concentration regardless of the presence or absence of succinate, fumarate, or l-malate (Fig. 7). Thus, DccT is required for the utilization of the C₄-dicarboxylates succinate, fumarate, and l-malate for biomass formation in complex medium. In contrast, growth in LB or in glucose minimal medium was not affected by deletion of dccT and/or dctA, and under anaerobic conditions, WT C. glutamicum and the ΔdctA, ΔdccT, and ΔdctA ΔdccT strains accumulated comparable concentrations of succinate (data not shown).

• Open in new tab
• Download powerpoint

FIG. 7.

Biomass formation by WT C. glutamicum (black columns), C. glutamicum ΔdctA (gray columns), and C. glutamicum ΔdccT (open columns) in LB complex medium or LB containing either 50 mM succinate, 50 mM fumarate, or 50 mM l-malate. Means and ranges or standard deviations of two or more independent growth experiments are shown.

Properties of DctA from C. glutamicum.In this study, we identified a second functional uptake system for C₄-dicarboxylates in addition to DccT (40) in C. glutamicum, which is encoded by cg2870/dctA. Based on its sequence similarities to members of the DAACS transporter family, the gene has previously been annotated as dctA (19). As shown here, DctA functions as a proton motive force-dependent uptake system for the C₄-dicarboxylates l-malate, fumarate, and succinate and likely also for oxaloacetate and glyoxylate. DctA-like carriers were found in aerobic, as well as in facultatively anaerobic, bacteria, archaea, and eukaryotes. They catalyze Na⁺- or H⁺-dependent uptake of dicarboxylates during aerobic growth (16). The closest homologs of C. glutamicum dctA in the related, but pathogenic, Corynebacterium jeikeium and Mycobacterium tuberculosis (jk1314 and Rv2443) encode proteins with high sequence similarity (>59% amino acid identity) that likely also function in uptake of C₄-dicarboxylates. DctA from Sinorhizobium meliloti and E. coli exhibit affinities for succinate of approximately 50 μM (15, 42), and the affinity for fumarate was determined to be approximately 30 μM for DctA from E. coli (20). Here, the affinities for succinate and fumarate of DctA from C. glutamicum were shown to be three to five times lower than those of the previously described transporters, but the activities of the transporters were comparable. The K_m value for l-malate is comparable to the one published by Jones et al., (18), though the kinetics were determined only for mixed bacterial populations and not for a single organism. It was proposed previously that the observed Na⁺-dependent succinate uptake in C. glutamicum is mediated by DctA (16). However, we clearly showed that uptake by DctA depends on the H⁺ potential across the membrane (Fig. 5), while DccT-mediated uptake is Na⁺ dependent (40).

Though several similarities regarding the kinetics data and the energetic coupling (12) of the transport were found between DctA from C. glutamicum and the DctA transporters characterized previously, the substrate spectrum of DctA from C. glutamicum was found to be broader. Besides the TCA cycle intermediates succinate, fumarate, and l-malate, DctA from C. glutamicum likely also transports glyoxylate and oxaloacetate. The observed inhibition of the DctA-dependent dicarboxylate uptake by glyoxylate in vitro and in vivo provides indirect evidence for DctA-mediated uptake of glyoxylate into the C. glutamicum cell.

In contrast, 2-oxoglutarate inhibited DctA-mediated uptake of l-malate in vitro, but utilization of 2-oxoglutarate in growth experiments was not observed for the WT or when dctA was overexpressed ectopically. In E. coli, uptake of 2-oxoglutarate is not mediated by DctA but by KgtP, a member of the major facilitator superfamily transporter family (30, 31). It is interesting that uptake of 2-oxoglutarate by KgtP from E. coli is inhibited by the addition of succinate, fumarate, or l-malate but that transporters other than KgtP are responsible for the uptake of succinate or fumarate (15).

Expression of dctA.Neither the presence of the substrates of the DctA uptake system in the culture medium nor anaerobiosis, conditions known to induce expression of dctA in E. coli, affected the expression of the dctA reporter gene fusions in C. glutamicum (Fig. 2). In E. coli, the two-component regulatory system DcuSR is responsible for the induction of dctA in the presence of succinate or fumarate (1, 11, 14, 43). Recently, the two-component system CitA-CitB, which belongs to the CitAB/DcuSR subgroup of the two-component regulatory system family, has been described in C. glutamicum (3). While it was shown that CitAB is required for induction of the citrate permease genes citH and tctCBA and thus for utilization of citrate, dctA was not among the genes differentially expressed in the presence/absence of CitAB (3), and a typical binding motif of DcuR could not be detected upstream of dctA. The analysis of the dctA promoter of C. glutamicum MSM revealed a C→T transition upstream of dctA (position −12; TTTATAATTTTCCGA [the mutated nucleotide is underlined]) that increased the similarity to the consensus sequence of the −10 region (TATAAT) in C. glutamicum (22). Under all tested conditions, the expression of the reporter gene fusions containing the mutation (dctA-cat^MSM) was 25-fold to 45-fold higher than that containing the WT promoter (Fig. 2). Thus, the mutation may increase promoter recognition by RNA polymerase and transcription initiation and likely is a promoter-up mutation. However, as the mutation reduces the symmetry of a palindrome (GAAAAAGTTTTTC→GAAAAAGTTTTTT [the mutated nucleotide is underlined]), the possibility that the dctA promoter is repressed under all tested conditions by an unknown repressor binding to this palindrome cannot be excluded.

Physiological roles of DctA and DccT.Growth in minimal medium with succinate, fumarate, or l-malate required high expression levels of either dccT (40) or dctA (Fig. 3) as a consequence of plasmid-borne expression or due to promoter-up mutations. When succinate, fumarate, or l-malate was added to LB complex medium, their utilization by WT C. glutamicum required DccT (Fig. 7), indicating that the low constitutive dccT expression levels are sufficient for this ability. Moreover, DccT showed eight-times-higher affinities for succinate and fumarate, and the affinity for l-malate was twice as high as that of DctA (Fig. 4) (40). Accordingly, DccT was made responsible for the very slow growth of C. glutamicum strain R on dicarboxylates (36). As the conditions leading to high dctA expression are currently not known, the physiological role of DctA for WT C. glutamicum is still not fully understood.

As both DctA and DccT were identified by analysis of spontaneous mutants that had gained the ability to efficiently utilize succinate, fumarate, or l-malate, the influence of dctA and dccT on the occurrence of such spontaneous mutants was analyzed. Although eight genes are predicted to encode putative uptake systems for C₄-dicarboxylates (19); the CitMHS family transporter CitH and the TRAP family system TctABC for uptake of citrate (3); two DASS family transporters encoded by cg2072 and cg2243, in addition to DccT (40); and, besides DctA, two further DAACS-type transporters encoded by cg2810 and cg3356, spontaneous mutants able to grow with succinate, fumarate, or l-malate could not be isolated when both dctA and dccT were deleted (see above). These results suggest that under the tested conditions no other transport systems besides DctA and DccT allow the utilization of succinate, fumarate, or l-malate.