2018 Volume 24 Issue 5 Pages 861-868
Lactic acid (LA) has been used for wide range of food processing and industrial applications, for example, as a raw material of biodegradable plastics, poly(lactic acid)s (PLAs). Thus, there is a demand to incorporate acid-resistance for effective LA production. Acid-tolerant lactic acid bacteria (LAB), Lactobacillus acetotolerans HT, has been isolated from rice vinegar (pH 2.9, 6% acetic acid). In this study, genes for the D- and L-lactate dehydrogenases (D-LDH, L-LDH1, and L-LDH2) of Lb. acetotolerans HT, which constitute the key enzymes in cell growth and lactic acid production, were cloned and identified. Through heterologous expression of LDH genes in Escherichia coli DH5α, recombinant E. coli DH5α harboring the D-LDH (ldhD) or L-LDH1 (ldhL1) genes were found to produce D-LA or L-LA, respectively, whereas the strain harboring the L-LDH2 (ldhL2) gene did not produce L-LA. This finding strongly suggests that the translational product of the ldhL2 gene does not exhibit L-LDH activity in vivo.
Lactic acid bacteria (LAB) have been exploited in various industries with lactic acid (LA), the end product of saccharide fermentation by LAB, having been effectively utilized in the food, medical, and chemical industries. For example, LA is used as a food acidulant, preservative, pickling agent, flavoring, and pH buffering agent, as well as in topical ointments, lotions, and parenteral solutions (Datta et al., 1995). Currently, LA is also used as a raw material of biodegradable plastics, poly (lactic acid)s (PLAs) such as PDLA (poly-D-lactic acid) and PLLA (poly-L-lactic acid), which exhibit good transparency and have been widely utilized. In addition, the properties of stereocomplex PLA, composed of both PDLA and PLLA, are superior to those of the respective single and racemic PLAs (Ikada et al., 1987; Tsuji and Fukui, 2003; Fukushima et al., 2005; Fukushima et al., 2007). Stereocomplex-type polymers show a melting point (230 °C) that is approximately 50 °C higher than that of the respective single polymers (Ikada et al. 1987). However, as industrial chemical synthesis leads to the production of racemic LA (Datta et al., 1995), the demand for optically pure isomers produced by fermentation has increased. There are considerably fewer reports on D-LA production than on L-LA production because of the small number of D-LA producers (Zhou et al., 2003; Singhvi et al., 2013; Jambunathan and Zhang, 2016). As the cell growth of bacteria, even LAB, is inhibited by the LA produced, most bacteria cannot produce lactic acid at pH below 4 owing to having low tolerance to acid (Adachi et al., 1998; Jambunathan and Zhang, 2016). Accordingly, it is preferable to employ acid-resistant bacterial strains for effective production of LA.
An acid-tolerant LAB, Lactobacillus acetotolerans HT, was previously isolated from fermented rice vinegar (pH 2.9, 6% acetic acid) (Tanaka et al., 2015, 2017). The HT strain can grow vigorously in medium supplemented with 2% (v/w) acetic acid, which inhibits the growth of general LAB. Lb. acetotolerans, which is resistant to high concentrations of acetic acid, was first isolated from fermented vinegar broth by Entani et al. (1986). This species was also detected as being dominant in long-aged Nukadoko, a fermented rice bran bed traditionally used for pickling vegetables in Japan (Nakayama et al., 2007; Sakamoto et al., 2011). In particular, Lb. acetotolerans-dominated microbiota were observed in long-aged (>40 years) Nukadoko samples whereas 1-year-old Nukadoko samples were dominated by the other Lactobacillus species. This supports that Lb. acetotolerans is readily tolerant to lower pH conditions and continues to grow with further lactate production even after other Lactobacillus spp. have ceased to proliferate (Sakamoto et al., 2011). Although the growth of strain HT is considerably slower than that of other LAB, it can produce LA in high concentration (59.5 g/L) without maintaining the culture pH (Tanaka et al., 2015), suggesting it to be a candidate as a powerful LA producer. However, few reports are available on this species, possibly owing to its very low growth rate. In particular, there are no reports on the D- and L-lactate dehydrogenases (LDHs), which constitute the most important enzymes for cell growth and LA production, although their existence is predicted by recently reported genomic information (Toh et al., 2015; accession no. AP014808). The aims of the present study are to clone and identify the genes of D- and L-LDHs (ldhD and ldhL) in Lb. acetotolerans HT for further development, e.g., improvement of strains that produce only D- or L-lactic acid by genetic modification of HT strain.
Bacterial strains, plasmids, and growth conditions Bacterial strains and plasmids used in this study are listed in Table 1. E. coli DH5α was used as a host to clone LDH genes. Transformation of E. coli DH5α was performed by standard procedures and bacteria were grown at 37 °C in 1.5 mL lysogeny broth (LB) medium (Sambrook et al., 1989). When needed, ampicillin (100 mg/L) was added to the medium. Lb. acetotolerans HT was statically grown at 30 °C in 10 mL of MRS medium (Oxoid, Hampshire, UK) containing 1% (w/v) acetic acid.
| Bacterial strains or plasmids | Characteristics | Reference or source |
|---|---|---|
| Bacterial strains | ||
| Lactobacillus acetotolerans HT | Wild strain | Tanaka et al. (2015) |
| Escherichia coli DH5α | Host for cloning and plasmid propagation, deoR, endA1, gyrA96, hsdR17 (rK−mK+), recA1, relA1, supE44, thi-1, Δ(lacZYA-argFV169), Φ80ΔlacZΔM15, F | Clontech |
| Plasmids | ||
| pT7Blue T-vector | E. coli cloning vector, Apr, lacPOZ, T7 promoter | Novagen |
| T-vector pMD20 | E. coli cloning vector, Apr, lacPOZ, SP promoter | TaKaRa |
| pUC118 HincII/BAP | E. coli cloning vector, Apr, lac POZ, lac promoter | TaKaRa |
| pBluescript II KS+ | E. coli cloning vector, Apr, lacPOZ, T7 and T3 promoter | Stratagene |
| pBS-ldhD | pBluescript II KS+ derivative; ldhD gene of Lb. acetotolerans HT | This study |
| pBS-ldhL1 | pBluescript II KS+ derivative; ldhL1 gene of Lb. acetotolerans HT | This study |
| pBS-ldhL2 | pBluescript II KS+ derivative; ldhL2 gene of Lb. acetotolerans HT | This study |
Apr, ampicillin resistance gene
Cloning of LDH genes from Lb. acetotolerans HT Total genomic DNA of Lb. acetotolerans HT was extracted according to methods in previous reports with some modifications as follows (Marmur, 1961; Berns and Thomas, 1965; Sato et al., 2000; Goh et al., 2006). For isolation of genomic DNA from Lb. acetotolerans HT, 10 mL of culture broth was statically grown in MRS medium supplemented with 1% (v/v) acetic acid at 30 °C for 60 h and was transferred to 100 mL MRS medium with 1% acetic acid and 20 mM DL-threonine. After growing at 30 °C for 24 h, the cells were collected by centrifugation (5,720 ×g, 5 min, 4 °C). The cell pellets were washed twice with 50 mL of 50 mM EDTA (pH 8.0) and resuspended in 1 mL of the same buffer. Cell lysis was initiated by addition of 5–10 mg/mL lysozyme (Wako, Osaka, Japan). After incubation at 37 °C for 2 h, 5 mg/mL labiase (Seikagaku, Tokyo, Japan) was added and gently mixed, and the mixture was further incubated at 37 °C for 3 h. The cell lysate was treated with 2% sodium dodecyl sulfate (SDS) (addition of 0.2 mL of 10% SDS) at 60 °C for 10 min. Chromosomal DNA was sequentially extracted with phenol, chloroform, and isoamyl alcohol (25:24:1) (PCI), precipitated with ethanol, washed, and dissolved in TE buffer (10 mM Tris-HCl and 1 mM EDTA, pH 8.0). Proteinase K was subsequently added to the DNA solution to a final concentration of 0.1 mg/mL and incubated at 37 °C for 2 h, followed by PCI treatment, ethanol precipitation, drying under vacuum, and dissolution in TE buffer.
RNase A-treated genomic DNA was used as a template for PCR. The sequences of the primers used in this study are listed in Table 2. PCR was performed using Ex Taq HS DNA Polymerase, LA Taq DNA Polymerase, PrimeSTAR HS DNA Polymerase, and PrimeSTAR GXL DNA Polymerase (all from TaKaRa Bio, Otsu, Japan). PCR products were amplified and purified using a QIAquick PCR Purification Kit (QIAGEN, Hilden, Germany) or isolated from agarose gels using a QIAGEX II Agarose Gel Extraction Kit (QIAGEN). The DNA fragments were ligated with pT7Blue-T vector (Novagen, San Diego, CA, USA) or pMD20-T vector with Mighty TA-cloning Reagent Set for PrimeSTAR (TaKaRa Bio) and introduced into E. coli DH5α. The constructed plasmids were sequenced using NEN Global Edition IR2 System, LIC4200L (LI-COR, Lincoln, NE, USA), and GenomeLab GeXP/CEQ System (BECKMAN COULTER, Brea, CA, USA) DNA sequencers according to the manufacturer instructions. The degenerate PCR primers (D-LDH-f1, D-LDH-r1, L-LDH-f1.2, and L-LDH-r1) were synthesized based on the highly conserved regions of D-LDHs and L-LDHs of closely related species among LAB and the internal fragments of ldhD and ldhL genes were amplified by PCR from the genomic DNA of Lb. acetotolerans HT. Subsequently, inverse PCR was performed using primers (D-LDH-f2, D-LDH-r2, L-LDH-f2, and L-LDH-r2) synthesized based on sequences of the amplified products to obtain upstream and downstream regions of the ldhD and ldhL genes. Finally, complete coding sequences of ldhD and the ldhL genes were obtained by PCR using primer pairs D-LDH-f3 and D-LDH-r3 (HincII), and L-LDH-f3 (XbaI) and L-LDH-r3 (XbaI), respectively, designed based on sequences upstream and downstream of the putative LDH genes, respectively. The 1.8-kb DraI-HincII region and the 1.7-kb XbaI fragment containing the ldhD and the ldhL (ldhL1) genes, respectively, were completely sequenced.
| Primer | Sequence |
|---|---|
| D-LDH-f1 | 5′-YTIMGIAAYGTIGGIGTIGA-3′ |
| D-LDH-r1 | 5′-ACIGCRTGIGTIGTRTARAA-3′ |
| D-LDH-f2 | 5′-AAGAATTCCCAGACAAGCGTTTAGC-3′ |
| D-LDH-r2 | 5′-AGGAACGTTGGTAATTTCGAAGC-3′ |
| D-LDH-f3 | 5′-CCAGATCTTCTGGCTGCCCACATCAGTATC-3′ |
| D-LDH-r3(HincII) | 5′-GAGTCAACGAAGTTATCTCAACCTTGTCAT-3′ |
| L-LDH-f1.2 | 5′-YTIGTIGGIGAYGGIGCIGTIGG-3′ |
| L-LDH-r1 | 5′-CCRTARAAIGTIGCICCYTT-3′ |
| L-LDH-f2 | 5′-AAGACAAGCTTGATGAAATTCATAAGAGTG-3′ |
| L-LDH-r2 | 5′-ACCCAATTCTTGGGCAATACCTTGG-3′ |
| L-LDH-f3(XbaI) | 5′-GAAAGTAGGCACCATTTCTAGAATTGCCGG-3′ |
| L-LDH-r3(XbaI) | 5′-CAGTTTCGTCGTCTAGAATAGGAAC-3′ |
| L-LDH2-f1 | 5′-CTTTTGCAGAATACTAATGTAGATGA-3′ |
| L-LDH2-r1 | 5′-TAAAACTTCTTGCATCTTATTGGCTGA-3′ |
| L-LDH2-f3 (inverse) | 5′-CAGATGTACGTAAAAAGGGTGGAAA-3′ |
| L-LDH2-r3 (inverse) | 5′-CCAGTCTTGTTTCACCTGGTTTACG-3′ |
| L-LDH2-f4 (inverse) | 5′-GTCACTTGATTCAGCCCGTCTTCTTCGC-3′ |
| L-LDH2-r4 (inverse) | 5′-GCAAAAGTTGAACCTACGGCACCATC-3′ |
| L-LDH2-f6 (UP) | 5′-AAGATGATCGGGAGTTTAGTAGAACAA-3′ |
| ldhD-SD(ApaI)-f | 5′-GGGCCCa)TTGGGAGGTTGAATTTATGACAAA-3′ |
| ldhD-XhoI(TAA)-r | 5′-CTCGAGb)TTATGCATCTAATTTAACTGGGCT-3′ |
| ldhL-SD(ApaI)-f | 5′-GGGCCCa)CAAAAGGAGACATATTATGGTAAA-3′ |
| ldhL-XhoI(TAA)-r | 5′-CTCGAGb)TTATTGACGAACCTTAACGCCAGT-3′ |
| ldhL2-SD(ApaI)-f | 5′-GGGCCCa)TGAAAGGAAATTAAATTATGAGTA-3′ |
| ldhL2-XhoI(TAA)-r | 5′-CTCGAGb)TTAATTCAAATCAATTCCGTCTAA-3′ |
The putative translational products of the ldhD and the ldhL1 genes of strain HT showed high homologies with those of Lb. acetotolerans RIB 9124 (NBRC 13120) (Toh et al., 2015; accession no. AP014808). As strain RIB 9124 has been expected to encode another L-LDH (ldhL2) based on the genomic information, cloning of the ldhL2 gene of strain HT was further attempted. The internal region of the ldhL2 gene was amplified by PCR with the genomic DNA of Lb. acetotolerans HT using primers L-LDH2-f1 and L-LDH2-r1, which were designed from the putative L-LDH2 sequence of Lb. acetotolerans RIB 9124. Inverse PCR was conducted using primer pairs L-LDH2-f3 (inverse) and L-LDH2-r3 (inverse), and L-LDH2-f4 (inverse) and L-LDH2-r4 (inverse) in a similar manner. The complete coding sequence of the ldhL2 gene was obtained by additional PCR with primer pair L-LDH2-f6 (UP) and L-LDH2-r4 (inverse) as the sequence upstream of the putative ldhL2 gene of strain HT was not obtained by inverse PCR. The resulting nucleotide sequences were analyzed using SDC-GENETYX genetic information processing software (Software Development Co., Tokyo, Japan).
Heterologous expression of LDH genes in E. coli DH5α The ldhD, ldhL1, and ldhL2 genes were amplified by PCR from the genomic DNA of Lb. acetotolerans HT using the following primer pairs with ApaI and XhoI sites (Table 2): ldhD-SD(ApaI)-f and ldhD-XhoI(TAA)-r, ldhL-SD(ApaI)-f and ldhL-XhoI(TAA)-r, and ldhL2-SD(ApaI)-f and ldhL2-XhoI(TAA)-r. The amplified fragments were cloned in T-vector pMD20 or pUC118 HincII/BAP (TaKaRa Bio). The LDH genes were excised from plasmids by digestion with ApaI and XhoI, and the fragments were subsequently introduced into pBluescript II KS+ at the same restriction sites to produce pBS-ldhD, pBS-ldhL1, and pBS-ldhL2. The plasmids were then introduced into E. coli DH5α. Recombinant strains of E. coli DH5α were grown in 1.5 mL LB medium at 37 °C for 15 h and then 1 mL of the culture broth was transferred to 300-mL conical flasks with 100 mL LB medium. Cells were cultivated at 37 °C in static or shaking (100 strokes/min) culture until OD600 reached 0.6–0.8. Thereafter, 1 mL of 0.1 M isopropyl-β-D-thiogalactopyranoside (IPTG) was added to the medium (final concentration of 1 mM) and induced for 3 h. Subsequently, filter-sterilized glucose solution was added to the medium at a concentration of 2% (w/v). Cell density and pH were measured in the culture broth after 24 h of total cultivation. Cell density was measured in terms of OD at 600 nm using a spectrophotometer. Thereafter, culture broth was centrifuged (7,780×g, 10 min, 4 °C) to separate cells and culture supernatant. The concentration of LA produced and residual glucose in the supernatant was analyzed. The D- and L-LA concentrations were determined using a D-Lactic acid/L-Lactic acid F-kit (Roche, Basel, Switzerland) according to the manufacturer instruction. The total LA concentration was calculated by adding together the D-LA and L-LA concentrations. The residual glucose concentration in the culture supernatant was determined using a biosensor (portable glucose meter GF-501, Tanita, Tokyo, Japan). Then, the LA conversion rate (%) was calculated as
LA conversion rate (%) = C LA/C Glc × 100,
where C LA is the total LA concentration in the supernatant and C Glc is the concentration of the glucose consumed.
The culture runs were performed in at least triplicate.
Nucleotide sequence accession numbers The nucleotide sequence data determined here will appear in the EMBL, GenBank, and DDBJ databases under accession nos. LC378394, LC378395, and LC378396.
Cloning of LDH genes from Lb. acetotolerans HT To identify genes for D-LDH and L-LDH1 in Lb. acetotolerans HT, degenerate and inverse PCR were performed. The complete nucleotide sequences of the 1.8-kb DraI-HincII region (ldhD) and the 1.7-kb XbaI fragment (ldhL1) were determined in both strands (Figs. 1A and 1B). Open reading frames (ORFs) of 1005 bp (ldhD) and 972 bp (ldhL1) encoding putative proteins of 334 and 323 amino acid residues were identified, respectively. The putative molecular weight of D-LDH and L-LDH1 calculated from the deduced amino acid sequence was 37 and 35 kDa, respectively. The deduced amino acid sequence of the putative ldhD gene showed 100% and 99% identities with D-LDHs of Lb. acetotolerans DSM 20749 (JCM 3825) (GenBank accession number: KRN36956.1) and Lb. acetotolerans RIB 9124 (NBRC 13120) (BAQ56456.1). The putative D-LDH of Lb. acetotolerans HT has a lysine residue at position 312, whereas the D-LDH of Lb. acetotolerans RIB 9124 has an arginine residue at the same position. The deduced amino acid sequence of the putative ldhL1 gene showed 100% identity with L-LDHs of Lb. acetotolerans DSM 20749 (KRN40783.1) and Lb. acetotolerans RIB 9124 (BAQ56724.1). In the region upstream of the ldhD gene, an ORF was oriented in the opposite direction. The ORF was similar to the 5′-terminal region of the aminoglycoside phosphotransferase fructosamine kinase gene of Lb. acetotolerans DSM 20749 (KRN36955.1) and Lb. acetotolerans RIB 9124 (BAQ56457.1) (Fig. 1A). In the region downstream of the ldhD gene, an ORF similar to the 3′-terminal region of the transposase of Lactobacillus helveticus (PAW05974.1) (84% identity) was oriented in the opposite direction. In the regions upstream and downstream of the ldhL1 gene, the nucleotide sequences revealed homologies to partial genes encoding the transposase of Lactobacillus jensenii JV-V16 (EFH30272.1) and the CBS domain-containing protein of Lb. acetotolerans DSM 20749 (KRN40782.1) and Lb. acetotolerans RIB 9124 (BAQ56723.1), respectively, based on a BLAST search (i) (Fig. 1B). However, stop codons were found at amino acid positions 6 and 21 of the putative transposase, suggesting that the gene would not function as a transposase.
Schematic representations and restriction map of the three different genes for LDHs, ldhD (A), ldhL1 (B) and ldhL2 (C), in Lb. acetotolerans HT.
According to the genomic information of Lb. acetotolerans RIB 9124 (BAQ57206.1), one gene encoding D-LDH and two genes encoding L-LDHs have been annotated. Therefore, strain HT may also have another L-LDH (ldhL2). The primers L-LDH2-f1 and L-LDH2-r1 for PCR were thus designed from another L-LDH sequence of Lb. acetotolerans RIB 9124. Using the genomic DNA of Lb. acetotolerans HT as the template, the complete nucleotide sequence of the 1.5-kb EcoRI-HindIII region (ldhL2) was determined in both strands (Fig. 1C). An ORF of 930 bp (ldhL2) encoding a putative protein of 309 amino acid residues was found. The molecular weight of the putative L-LDH2 calculated from the deduced amino acid sequence was 33 kDa. The putative ldhL2 gene of Lb. acetotolerans HT showed 100% and 99% identities with other L-LDHs of Lb. acetotolerans DSM 20749 (KRN40946.1) and Lb. acetotolerans RIB 9124 (BAQ57206.1), respectively. The putative L-LDH2 of Lb. acetotolerans strains HT and DSM 20749 have a glutamine residue at position 135 whereas another L-LDH of Lb. acetotolerans RIB 9124 has a proline residue at the same position, as do several L-LDHs of Lactobacillus that exhibit high identity (>84%) to the L-LDH2 of Lb. acetotolerans HT. In the region upstream and downstream of the putative ldhL2 gene, the nucleotide sequences revealed 100% identities to the genes encoding the f lavocytochrome c and the dimethylmenaquinone methyltransferase of Lb. acetotolerans DSM 20749 (KRN40945.1 and KRN40947.1) and RIB 9124 (BAQ57205.1 and BAQ57207.1), respectively.
The identity between the D-LDH and L-LDH1 of Lb. acetotolerans HT was 13%. The identities of the L-LDH2 with the D-LDH and L-LDH1 were 10% and 29%, respectively. This suggests that Lb. acetotolerans HT might have three distinct LDHs.
Heterologous expression of the LDH genes of Lb. acetotolerans HT in E. coli DH5α E. coli DH5α was transformed with plasmids pBluescript II KS+, pBS-ldhD, pBS-ldhL1, or pBS-ldhL2. To investigate LA production, the recombinant strains were cultivated in static or shaking culture in LB medium with 2% (w/v) glucose at 37 °C for 24 h; the resultant LA production profiles are shown in Fig. 2 and Table 3. The total LA concentration in the culture supernatant was almost the same in all recombinant strains in static culture (0.67–0.81 g/L) (Fig. 2A and Table 3), whereas the proportion of D-LA and L-LA produced in each was different among all recombinant strains. E. coli DH5α harboring pBS-ldhD (DH5α/pBS-ldhD) showed only D-LA production (0.81 g/L), whereas DH5α/pBS-ldhL1 exhibited higher production of L-LA (0.50 g/L) than D-LA (0.18 g/L). In comparison, DH5α/pBluescript II KS+ (control) and DH5α/pBS-ldhL2 resulted in the nearly equal amounts of D-LA (0.67 and 0.70 g/L) and L-LA (<0.01 g/L). In shaking culture, only DH5α/pBS-ldhD produced D-LA (0.33 g/L), whereas L-LA (0.59 g/L) was produced only by DH5α/pBS-ldhL1 (Fig. 2B). Conversely, DH5α/pBS-ldhL2 produced little LA compared to the control.
LA concentration in culture supernatant and conversion rate of glucose to LA. LA production by the recombinant strains of E. coli DH5α harboring pBluescript II KS+ (control), pBS-ldhD (ldhD), pBS-ldhL1 (ldhL1) and pBS-ldhL2 (ldhL2) in static culture (anaerobically) (A) and shaking culture (microaerobically) (B). Cells were cultivated at 37 °C for 24 h in a 300-mL conical flask containing 100 mL of LB medium with 2% (w/v) glucose as described in the Materials and Methods section. White bar, D-LA concentration; gray bar, L-LA concentration; diamond, conversion rate of glucose to D- and L-LAs. N.D., Not Detected (<0.01 g/L).
| Culture condition | Plasmid | OD600 | Glucose consumption (g/L) a) | LA conc. (g/L) b) | Final pH |
|---|---|---|---|---|---|
| Static culture (anaerobically) | pBluescript II KS+ | 1.70±0.16 | 4.7±2.2 | 0.67±0.19 | 4.83±0.09 |
| pBS-ldhD | 1.00±0.04 | 3.2±1.5 | 0.81±0.20 | 4.88±0.05 | |
| pBS-ldhL1 | 1.96±0.35 | 4.3±1.6 | 0.68±0.10 | 4.90±0.06 | |
| pBS-ldhL2 | 1.55±0.05 | 5.4±2.2 | 0.70±0.21 | 4.90±0.04 | |
| Shaking culture (microaerobically) | pBluescript II KS+ | 3.18±0.32 | 6.0±0.9 | N.D. | 4.84±0.06 |
| pBS-ldhD | 2.70±0.19 | 6.0±1.2 | 0.33±0.04 | 4.87±0.04 | |
| pBS-ldhL1 | 4.07±0.35 | 5.9±0.7 | 0.59±0.10 | 4.86±0.08 | |
| pBS-ldhL2 | 3.40±0.34 | 6.5±1.0 | N.D. | 4.95±0.05 |
Cells were cultivated at 37°C for 24 h in a 300-mL conical flask containing 100 mL LB medium with 2% (w/v) glucose as described in the Materials and Methods section.
N.D., Not Detected (<0.01 g/L).
The cell growth (OD600), glucose consumption, total LA production, and final pH in the culture broth after 24 h of cultivation in static or shaking culture are shown in Table 3. The LA conversion rate (%) calculated using the values of the glucose consumption and the total LA concentration is shown in Fig. 2. DH5α/pBS-ldhD showed the highest LA conversion rate (25.3%) in static culture (Fig. 2A and Table 3). In shaking culture, the LA conversion rate (10.1%) of DH5α/pBS-ldhL1 was relatively higher than for other strains (Fig. 2B). Under both static and shaking culture conditions, the total LA concentrations produced by DH5α/pBS-ldhL2 were almost the same as the control (Fig. 2 and Table 3). DH5α/pBS-ldhL1 showed higher cell density than those of strains harboring other plasmids, especially in shaking culture (Table 3).
LAB, the major producer of LA, constitutes an important industrial microorganism. In LA fermentation, LAB is required to have high acid tolerance because low pH due to LA production inhibits the cell growth of bacteria (Jambunathan and Zhang, 2016). Lb. acetotolerans HT can grow even at pH 3.2, a pH at which the growth of general bacteria is inhibited (Tanaka et al., 2015, 2017). Although several studies have focused on Lb. acetotolerans, there are no reports on LDHs, which play a key role in LA production. In the present study, we cloned and identified the LDH genes of Lb. acetotolerans HT, as this strain produces racemic DL-LA, indicating the presence of LDH isomers (D-LDH, L-LDH1, and L-LDH2) (Tanaka et al., 2015). Based on sequencing of PCR products amplified using the genomic DNA of Lb. acetotolerans HT, the nucleotide sequences of the LDH genes (ldhD, ldhL1, and ldhL2) and the surrounding regions of these genes were determined. To confirm the function the LDHs encoded by the ldhD, ldhL1, and ldhL2 genes, heterologous expression in E. coli DH5α was performed and the optical activities of LA produced by the recombinant strains were investigated. The production of D-LA and L-LA were observed in E. coli DH5α upon the introduction of ldhD and ldhL1 genes, respectively, whereas L-LA was not produced by E. coli DH5α with introduction of the ldhL2 gene (Fig. 2), with almost the same amount of LA (D-LA) being produced as in the control (Table 3). Thus, L-LDH2 did not show L-LDH activity in vivo although the Pfam program showed that the L-LDH2 comp r ised two doma ins, Ldh 1 N (lact a te/malate dehydrogenase, NAD binding domain, amino acid positions 3 to 141) and Ldh 1 C (lactate/malate dehydrogenase, alpha/beta C-terminal domain, amino acid positions 144 to 306), as did D-LDH and L-LDH1 in Lb. acetotolerans HT. To our knowledge, there are no reports on the activities of L-LDHs (>67% identities) of other Lactobacillus strains that are highly homologous to the L-LDH2 of Lb. acetotolerans HT. This is the first report on LDHs of Lb. acetotolerans, and there are very few reports on this species.
The recombinant E. coli strains into which the ldhD gene of Lb. acetotolerans HT were introduced produced higher D-LA than the control under both static and shaking culture conditions (Fig. 2). Particularly, in shaking culture, the strain carrying the ldhD gene produced D-LA (0.33 g/L) unlike the control (<0.01 g/L). During sugar fermentation, E. coli produces only D-LA because E. coli carries the ldhA gene encoding the fermentative D-LDH (Bunch et al., 1997; Chang et al., 1999; Zhou et al., 2003). The ldhA gene is highly expressed under anaerobic condition and at low pH (Mat-Jan et al., 1989; Bunch et al., 1997). The production of D-LA would thus be caused by the expression of the ldhD gene from Lb. acetotolerans HT together with the ldhA gene in static culture (anaerobically) (Fig. 2A), whereas only the ldhD gene was considered to be expressed in shaking culture (microaerobically) (Fig. 2B). In the strain with the introduced ldhL1 gene, L-LA production in shaking culture (0.59 g/L) was higher than that in static culture (0.50 g/L) (Fig. 2), whereas little D-LA was produced in shaking culture although 0.18 g/L D-LA was generated in static culture. This indicates that the ldhA gene of E. coli was not expressed in shaking culture, with pyruvate being converted to L-LA by the enzyme expressed by the ldhL1 gene of Lb. acetotolerans HT. Together, these findings confirmed the function of LDHs of Lb. acetotolerans HT.
When Lb. acetotolerans HT was statically cultivated at 30°C for 2 days in a test tube containing 10 mL MRS medium with 1% (v/v) acetic acid, D-LA and L-LA were produced at the proportion of 3:1 (data not shown). Therefore, we expect that the ldhL1-disrupted strain of the acid-tolerant Lb. acetotolerans HT would produce a large amount of D-LA only and might thus constitute a promising candidate as the host for D-LA production. This may further demonstrate that L-LDH2 does not have L-LDH activity.
We succeeded in cloning the genes for LDHs (D-LDH, L-LDH1, and L-LDH2) of Lb. acetotolerans HT. The recombinant E. coli strains carrying the ldhD and the ldhL1 genes produced D-LA and L-LA, respectively, whereas the strain carrying the ldhL2 gene produced no L-LA. This suggests that L-LDH2 of Lb. acetotolerans HT likely does not have L-LDH activity. This is the first phase of research for the future industrial application of the HT strain.
Acknowledgements We are grateful to Ms. Misaki Sato for her invaluable technical assistance.
