Molecular Characterization of Lactobacillus plantarum DMDL 9010, a Strain with Efficient Nitrite Degradation Capacity

Nitrites commonly found in food, especially in fermented vegetables, are potential carcinogens. Therefore, limiting nitrites in food is critically important for food safety. A Lactobacillus strain (Lactobacillus sp. DMDL 9010) was previously isolated from fermented vegetables by our group, and is not yet fully characterized. A number of phenotypical and genotypical approaches were employed to characterize Lactobacillus sp. DMDL 9010. Its nitrite degradation capacity was compared with four other Lactobacillus strains, including Lactobacillus casei subsp. rhamnosus 719, Lactobacillus delbrueckii subsp. bulgaricu 1.83, Streptococcus thermophilus 1.204, and lactobacillus plantarum 8140, on MRS medium. Compared to these four Lactobacillus strains, Lactobacillus sp. DMDL 9010 had a significantly higher nitrite degradation capacity (P<0.001). Based on 16S rDNA sequencing and sequence comparison, Lactobacillus sp. DMDL 9010 was identified as either Lactobacillus plantarum or Lactobacillus pentosus. To further identify this strain, the flanking regions (922 bp and 806 bp upstream and downstream, respectively) of the L-lactate dehydrogenase 1 (L-ldh1) gene were amplified and sequenced. Lactobacillus sp. DMDL 9010 had 98.92 and 76.98% sequence identity in the upstream region with L. plantarum WCFS1 and L. pentosus IG1, respectively, suggesting that Lactobacillu sp. DMDL 9010 is an L. plantarum strain. It was therefore named L. plantarum DMDL 9010. Our study provides a platform for genetic engineering of L. plantarum DMDL 9010, in order to further improve its nitrite degradation capacity.


Introduction
Amongst all processed vegetables, fermented vegetables have the highest productivity and are important Asian cuisine. While fermentation has been widely used in food processing for over 2,000 years, vegetable fermentation has experienced low levels of industrialization. Additionally, the presence of high levels of salt and nitrites in fermented vegetables is a major health concern. Excessive intake of salt is harmful to human health and nitrites are potential carcinogens [1][2]. Therefore, understanding the degradation of salt and nitrites during the process of vegetable fermentation is critically important to food safety.
The Lactobacillus genus consists of over 180 bacterial species that are rodshaped, gram-positive, and facultative anaerobic or microaerophilic bacteria [3]. Certain salt-tolerant Lactobacillus species are widely used in vegetable fermentation [4]. Our previous study demonstrated that addition of Lactobacillus casei subsp. rhamnosus 719 significantly reduced the concentration of nitrites [5], as well as salt [6], in fermented vegetables. We also isolated a Lactobacillus strain (strain DMDL 9010) from naturally fermented vegetables. Preliminary results suggested that the DMDL 9010 strain can inhibit the nitrite accumulation in vegetable fermentation and that the fermentation can be completed within 24 hours without the addition of salt. Detailed characterization of this strain is important to understand the underlying mechanism of the nitrite degradation, and may also facilitate its utilization in vegetable fermentation with genetic engineering. Thus, we aimed to characterize DMDL 9010, using a number of phenotypical and genotypical approaches.
With the rapid development of molecular biology, particularly DNA sequencing technologies, genotypical characterization of bacterial strains is widely utilized in research. For species identification and genotyping of bacteria, 16S rDNA sequencing, repetitive sequencing-based PCR (REP-PCR), PCR-restriction fragment length polymorphism (PCR-RFLP), and DNA-DNA hybridization are the most commonly used [7][8][9][10][11]. Over 97% and 99% of sequence identity of 16S rDNA is widely accepted to be the criteria for the determination of genus and species, respectively [12]. DNA sequencing of 16S rDNA is mostly common method for bacterial species identification, because universal primers are suitable for amplifying 16S rDNA from unknown bacterial strains and a large number of DNA sequences of 16S rDNA are available for almost all known bacterial species in GenBank. Search and sequence comparison of 16S rDNA with GenBank is very useful for species identification of bacteria. However, using DNA-DNA hybridization, Fox, et al., found that three Bacillaceae strains that shared 99.5% sequence identity to 16S rDNA belonged to different species, suggesting that 16S rDNA sequence identity may not be sufficient to guarantee species identity [13]. PCR-RFLP is usually used in genus identification of bacteria [7] and is not distinguishable when the sequence identity of 16S rDNA is higher than 96% [14]. REP-PCR has the advantages of being easy to conduct and producing a high resolution that is useful for discriminating strain sharing of over 99.5% sequence identity of 16S rDNA; however, REP-PCR has low reproducibility and is easily contaminated [15]. DNA-DNA hybridization is considered to be the golden standard for species identification of bacteria, and a 70% cut-off is most often used to place organisms into different species [16]. However, the DNA-DNA hybridization protocol is time-consuming and a large amount of genomic DNA is required, which limits its application in fastidious bacteria [17]. DNA sequences with high sequence variations, such as non-coding intergenic regions, have recently been used in bacterial strain typing and molecular identification. In this study, we used both highly conserved 16S rDNA sequences and two non-coding fragments, in order to characterize DMDL 9010.

Measurement of nitrites
According to the GB/T5009.33-2010 reference on "measurement of food nitrite and nitrate", a naphthyl ethylenediamine hydrochloride assay was used to measure nitrites, albeit with slight modifications. Briefly, without the protein precipitation step, 2 ml sulfanilic acid was added into the solution, mixed, and left to stand 3,5 min. Then, 1 ml naphthyl ethylenediamine solution (2 g/L) was added to reach to designated volume, mixed, and left to stand for 15 min. The absorbance was then measured, in order to determine the concentration of nitrites.

Degradation of nitrites by Lactobacillus sp. DMDL 9010
10 ml sterilized MRS medium (Guangdong Haikou Microbiology Biotech Inc., Haikou, China), containing 10.00 mg/L NaNO 2 was added into 15 ml sterilized test tube. Next, 5% (v/v) Lactobacillus sp. DMDL 9010 starter was added to the test tube and sealed. The solution was cultured at 37˚C for 24 h. 1 ml fermented sample was collected, sterilized, and tested, as mentioned above. Each fermentation sample was measured for nitrites three times. The concentration of nitrites is presented as mean value ¡ standard deviation.

Molecular identification and characterization of Lactobacillus sp. DMDL9010
Lactobacillus sp. DMDL 9010 was first identified by sequencing 16S rDNA. Overnight cultured bacterial strains (0.1-1.5 ml) were centrifuged at 12000 (r/ min) at room temperature for 1 min to precipitate cell pellets. Next, 0.6 ml bacterial lysozyme was added, mixed, and kept at 37˚C for 40 min to break down bacterial cells. Genomic DNA was isolated using the DNA isolation kit (Takara, Dalian, China), according to the manufacturer's protocol. Isolated genomic DNA was used as a template in PCR amplification of 16S rDNA. Primers used for amplification of 16S rDNA are F8 (forward): 59-AGA GTT TGA TCC TGG CTC AG-39 and R1492 (reverse): 59-TAC GGT TAC CTT GTT ACG ACT-39. PCRs were carried out in a PTC-200 automated thermal cycler (Bio-Rad CO., LTD). One nanomolar concentration of each DNA preparation was amplified in a 25-ml reaction mixture containing 50 pM of each primer, 200 mM (each) dATP, dCTP, dGTP, and dTTP (Takara, Dalian, China), as well as 0.125 ml Taq polymerase (Takara, Dalian, China) and an appropriate volume of distilled water. PCR conditions consisted of an initial denaturation at 95˚C for 5 min and 30 cycles of 30 s at 94˚C, 30 s at 56˚C, and 2 min at 72˚C, with a final extension at 72˚C for 7 min. PCR products were analyzed by electrophoresis in a 1.5% agarose gel and purified using a purification DNA fragment kit Ver 2.0 (Takara, Dalian, China). The PCR products were then sent to Takara Biotechnology (Takara, Dalian, China) for DNA sequencing. The determined DNA sequences of 16S rDNA were searched in the GenBank database using the BLAST program. Top hits with known representative species were selected and sequences were downloaded and aligned using the Clustal W program (www.ebi.ac.uk/Tools/msa/clustalw2). The alignment was trimmed and imported into the Mega 5.1 program to construct the phylogenetic organization using the Neighbor-Joining method.
Since the 16S rDNA is highly conserved between closely related species, we used another marker of high sequence diversity to further characterize Lactobacillus sp. DMDL9010. The gene encoding L-lactate dehydrogenase 1 (L-ldh1) is a housekeeping gene that is highly conserved in bacteria. We chose two fragments flanking L-ldh1 as molecular markers to identify this strain. Primers used to amplify these two markers were F1 (forward): 59-TATCCGTACTGTGTTTCCTC-39, R1 (reverse): 59-ACTAGAACCAACAGCGCCGT-39, F2 (forward): 59-TAGGTGGCCTTTTCGGTAGC-39, and R2 (reverse): 59-CTCGTCTATAGC-AGACGGGC-39. The PCR reaction system was the same for amplifying 16S rDNA. To amplify the up-stream fragment using F1 and R1 primers, PCR conditions consisted of an initial denaturation at 95˚C for 5 min and 30 cycles of 30 s at 94˚C, 30 s at 58˚C, and 1.5 min at 72˚C, with a final extension of 72˚C for 7 min. In order to amplify the down-stream fragment using the F1 and R1 primers, PCR conditions consisted of an initial denaturation at 95˚C for 5 min and 30 cycles of 30 s at 94˚C, 30 s at 59˚C, and 1.5 min at 72˚C, with a final extension of 72˚C for 7 min. Electrophoresis of PCR products, as well as their DNA sequencing, were conducted in the same way as for the 16S rDNA mentioned above. The determined DNA sequences of these two fragments were searched in the GenBank database using the BLAST program, to identity the most similar sequences and species. The results are combined with phenotypical results and 16S rDNA sequence similarity to determine the species of Lactobacillus sp. DMDL 9010.

Results and Discussion
Lactobacillus sp. DMDL9010 has the highest capacity of nitrite degradation As shown in Figure 1 Lactobacillus sp. DMDL9010 is Lactobacillus plantarum or Lactobacillus pentosus, based on the 16S rDNA sequence A PCR amplicon of 1500 bp was successfully amplified from Lactobacillus sp. DMDL 9010 (Figure 2). Finally, a 1441 bp fragment was obtained from DNA sequencing and was deposited in the GenBank database (accession number: KJ 917253). Based BLASTn search in GenBank using this 1441 bp fragment as the query, we found that the 16S rDNA of Lactobacillus sp. DMDL 9010 share 99% sequence identity with Lactobacillus plantarum and Lactobacillus pentosus. We then retrieved 22 DNA sequences of 16S rDNA from different Lactobacillus species from GenBank for phylogenetic analysis. The phylogenetic analysis based on these 16S rDNA suggested that Lactobacillus sp. DMDL9010 is closely related to Lactobacillus plantarum and Lactobacillus pentosus (Figure 3). Therefore, 16S rDNA sequences are not suitable for discriminating L. pentosus and L. plantarum species, because of high sequence identity. A number of genes and non-coding sequences have been used for species identification of bacteria. For example, rpoB sequences have been shown to be powerful for characterization of Corynebacterium at the species level (PMID: 15364970). Non-coding sequences are typically more variable than housekeeping genes and can be used for discrimination of closely related strains and species. Therefore, in order to further classify Lactobacillus sp. DMDL 9010, we sequenced the flanking regions of the L-ldh1gene.

Carbohydrate fermentation of Lactobacillus sp. DMDL 9010
Based on the physiological and biochemical characteristics [18][19] and the sequence comparison of the 16S rRNA gene, Lactobacillus sp. DMDL 9010 has been determined to be of the Lactobacillaceae phylum, Lactobacillales class, Bacilli order, and Firmicutes family, according to the ''Berger Bacterial Identification Handbook" and "Bergey's Manual of Systematic Bacteriology [20].'' To further phenotypically characterize this strain, a carbohydrate fermentation of Lactobacillus sp. DMDL9010 was performed. Carbohydrate fermentation was conducted using 17 types of carbohydrates, including arabinose, cellobiose, fructose, galactose, glucose, maltose, mannose, rhamnose, ribose, sorbose, sucrose, trehalose, xylose, raffinose, and esculin ( Table 1). The carbohydrate fermentation results showed that Lactobacillus sp. DMDL 9010 is capable of utilizing galactose, mannose, glucose, maltose, sucrose, trehalose, xylose, raffinose, and esculin (Table 1). However, the application to other carbohydrates was insufficient to further discriminate L. pentosus and L. plantarum, and we therefore decided to sequence more polymorphic DNA, in order to discriminate the two.
The sequence flanking the L-ldh1 gene suggests that Lactobacillus sp. DMDL 9010 is Lactobacillus plantarum Two fragments consisting of 922 and 806 bp were amplified from Lactobacillus sp. DMDL 9010, using L-ldh1 upstream primers (59-TATCCGTACTGTGTTTCCTC-39 and 59-ACTAGAACCAACAGCGCCGT-39) and downstream primers   (59-TAGGTGGCCTTTTCGGTAGC-39 and 59-CTCGTCTATAGCAGACGGGC-39) (Figure 3). These two fragments were sequenced and deposited in the GenBank database (accession numbers: KM 017408 and KM 017409, respectively). The top hits of the fragments were from L. plantarum, suggesting that Lactobacillus sp. DMDL 9010 is closely related to L. plantarum. Based on sequence alignment using Clustal W, identity of the upstream sequence was 98.92% between L. plantarum WCFS1 and Lactobacillus sp. DMDL 9010, but was 76.98% between L. pentosus IG1 and Lactobacillus sp. DMDL 9010. The significantly higher sequence identity of upstream of L-ldh1 between Lactobacillus sp. DMDL 9010 and L. plantarum than L. pentosus suggests that the upstream region of L-ldh1 can be used for discrimination between L. plantarum and L. pentosus when 16S rDNA fails to identify between the two species.
No DNA sequences downstream of L. pentosus L-ldh1 are available in GenBank. Genomic sequences of L. pentosus strains in GenBank are incomplete and gaps were found around in the downstream of L-ldh1. Therefore, sequence comparison of the downstream region of L-ldh1 was not conducted between L. pentosus and our strain of Lactobacillus sp. DMDL 9010. The sequence identity downstream from L-ldh1 was 99.54% between L. plantarum WCFS1 and Lactobacillus sp. DMDL 9010. This dramatically high sequence identity of the downstream region from L-ldh1 suggests that Lactobacillus sp. DMDL 9010 is closely related to L. plantarum WCFS1. Taken together, the sequence similarity of the upstream region of L-ldh1 between Lactobacillus sp. DMDL 9010 and L. plantarum was much high than that between Lactobacillus sp. DMDL 9010 and L. pentosus, suggesting Lactobacillus sp. DMDL9010 is more closely related to L. plantarum. Combined with the results of 16S rDNA sequencing and phenotypical characterization, Lactobacillus sp. DMDL 9010 is considered to be an L. plantarum strain and was named L. plantarum DMDL 9010.

Conclusion
In this study, we characterized a Lactobacillus strain that was previously isolated from fermented vegetables and exhibited significant nitrite degradation capability, using a number of phenotypical and genotypical approaches, including physiological and biochemical characterization, 16S rDNA sequencing, and DNA sequencing of flanking regions of the L-ldh1 gene. Our results demonstrated that Lactobacillus sp. DMDL 9010 has the higher nitrite degradation capability than other four Lactobacillus strains we examined, by degrading nitrites in the MRS fermentation medium to an undetectable level. Based on sequence analysis of 16S rDNA and the flanking regions of the L-ldh1 gene, this strain was determined to be L. plantarum DMDL 9010. We plan to sequence the entire genome of L. plantarum DMDL 9010, which will certainly improve our understanding of its evolution and metabolic pathways, including nitrite degradation. The genomic sequence may also facilitate genetic engineering of this strain to further utilize its role in vegetable fermentation.