Introduction

Weissella koreensis is a Gram-positive, non-spore-forming, heterofermentative and nonmotile short-rod bacterium that belongs to the family Leuconostocaceae1. W. koreensis is the predominant lactic acid bacterium (LAB) isolated from kimchi, a traditional Korean food composed of fermented vegetables2. As a psychrophilic bacterium, W. koreensis produces D-(-)-lactic acid and metabolites from glucose, which contribute to kimchi’s taste and flavor3 and to sourdough fermentation during bread-making. An earlier study has also reported that W. koreensis inhibits the germination of target microorganism spores during food fermentation and exhibits an anti-obesity effect by producing the non-protein amino acid (a.a.) ornithine3. Kimchi is a traditional fermented vegetable food emblematic of Korean culture that is fermented from vegetables such as Chinese cabbage and radish. Currently, kimchi is industrially produced via fermentation and is now consumed as a side dish worldwide. The most common type of whole kimchi (baechu-kimchi) is made by mixing salted white cabbage with a kimchi paste made of red pepper powder (Capsicum annuum), garlic, spring onion, Korean radish ginger, fish sauce (salted seafood), starch paste (made of rice or wheat starch) and other ingredients, such as fresh seafood. White kimchi (baek-kimchi made from Chinese cabbage) and watery kimchi (mul-kimchi, which is made from Chinese cabbage and radish and dongchimi, which is made from radish) are made without the use of a red pepper powder. These types of kimchi are characterized by many fresh flavours and are extremely refreshing4. To date, more than 100 species of LAB and several yeast strains have been identified in kimchi, including Lactobacillus, Leuconostoc and Weissella species5. In particular, W. koreensis, a Weissella species, has been reported to be the most important microorganism in kimchi and has been used effectively in making whole-wheat bread, together with baker’s yeast3.

Thus, establishing an accurate, rapid, sensitive and practical method based on quantitative polymerase chain reaction (qPCR) to detect and quantify of W. koreensis in various fermented foods is necessary. Recently, species, subspecies and strain-specific deoxyribonucleic acid (DNA) probes have been used extensively to screen, detect, quantify and identify strains of bacteria, yeast and viruses6. Many molecular assays based on 16S ribosomal RNA (rRNA) and a well-characterized gene that encodes a function relevant for a specific microorganism’s metabolism have been used to detect and identify Weissella species, but serious problems with the identification and detection of W. koreensis isolates have been identified: these assays also detect other Weissella species or do not produce amplicons from W. koreensis strains7. In addition, many multiplex PCR and chromogenic DNA macroarray systems for simultaneous amplification of several genes in a single assay have been developed. Nevertheless, these methods exhibit limitations: detecting target cells in mixtures with significantly different bacteria ratios or in food samples remains a challenge7. Consequently, the detection specificity, which depends on both the uniqueness of the sequence to a bacterium of interest and the specific binding of the primers and probe to the target sequence, is crucial for the efficacy of any PCR detection method.

Over the past 10 years, many efforts have been made to sequence numerous strains of LABs. The increasing number of available LAB genome sequences in databases, together with various bioinformatics tools, provides a resource for the development of more reliable, fast and cost-effective methods for bacterial identification in a wide range of samples. In particular, the genomic information for LABs available in public databases makes it possible to distinguish a target LAB from closely related lineages between species groups. However, despite the major advances in LAB bioinformatics over the last few years by the microorganism industry, methods for detecting, identifying and quantifying specific LABs remain limited. Therefore, in the present study, we exploited the genome sequence information available in public databases (ftp://ftp.ncbi.nlm.nih.gov/genbank/) to develop a real-time PCR assay for accurate detection and identification of W. koreensis. A pair of species-specific primers based on a membrane protein from the genomic sequence of W. koreensis KACC15510 was designed.

Bacterial membranes have been reported to perform diverse functions dependent on whether the membrane is specialized or cytoplasmic; the latter exhibit transport, mitochondrial activities and biosynthetic functions that are crucial for the assembly of membranes, walls and capsules. Membrane fusion proteins are found only in the prokaryotic world and function in conjunction with a variety of transport systems in Gram-positive and Gram-negative bacteria. These proteins are functional subunits of multi-component transporters that perform diverse physiological functions in both Gram-positive and Gram-negative bacteria. Bacterial membrane proteins are diverse in structure and function and vary significantly in size, with residue lengths that range from 200 to 650 a.a.8.

In this study, we established a reliable and efficient procedure for quantitative detection of W. koreensis in kimchi samples via SYBR Green PCR. Our results revealed that this SYBR Green qPCR-based method can be used for the specific detection and quantification of W. koreensis in various products. Using this real-time quantitative PCR assay, we found that red pepper powder greatly influences the density of W. koreensis during kimchi fermentation.

Results

Specificity of the designed primer set

A species-specific primer set was designed based on sequences of a membrane protein-encoding gene of W. koreensis KACC15510 (GenBank accession No. WP_013989464.1). The specificity of the primer sequences was tested using Basic Local Alignment Search Tool (BLAST) searches and electronic PCR (e-PCR) analysis (http://www.ncbi.nlm.nih.gov/). The BLASTn search exhibited no significant matches with previously reported sequences, with the exception of genomic DNA from W. koreensis KACC15510. The BLASTx searches for the predicted protein sequence revealed that W. confusa (GenBank accession No. WP_003607668.1) exhibited the closest similarity to the membrane protein. However, the PCR-amplified DNA sequence exhibited no significant matches in either the Nucleotide BLAST (BLASTn) or Align Sequences Translated BLAST (BLASTx) searches.

The species-specific molecular marker, a membrane protein-encoding gene, was amplified using the WK201F/R primers. Their specificity was validated against 11 different species of Weissella and 13 other reference LAB strains. All 8 W. koreensis strains consistently tested positive, regardless of the presence of other species of the bacterium in the sample and only W. koreensis strains produced a single amplified product of 201 bp. In contrast, a PCR assay that targeted maltose phosphorylase did not distinguish W. koreensis from the other tested bacterial species listed in Table 1 (see Fig. 1).

Table 1 Bacterial strains used in the PCR specificity test.
Full size table
Figure 1
figure 1

PCR amplification of maltose phosphorylase and the membrane protein.

Lane M is the size marker (1 kb DNA plus ladder; Gibco BRL), lanes 1 to 8 are W. koreensis strains, lanes 9 to 31 are included strains from other Weissella species along with strains from species of Lactobacillus and Leuconostoc, as specified in Table 1 and lane 32 is a negative control (distilled water). (a) Maltose phosphorylase gene7. (b) The membrane protein-encoding gene used in this study.

Full size image

Standard curves and melting temperature

We used SYBR Green real-time PCR analysis of W. koreensis to generate a standard curve by plotting the mean cycle threshold (Ct; n = 3) versus the logarithmic concentration of genomic DNA, cloned DNA and the density of the cell suspension (ranges of 5 × 100 to 5 × 10–6 ng/μl, 1.42 × 109 to 1.42 × 103 copies/μl and 1.28 × 109 to 1.28 × 104 CFU/ml, respectively; Fig. 2a and Table 2). The limit of quantitation (LOQ) assay exhibited a good linear response and high correlation coefficient (R2 = 0.994). A standard curve analysis of the linear portion of the slope resulted in a coefficient of −3.102, which yielded a PCR efficiency of 110.1% and a y-intercept value of 31.057 (Fig. 2b). The melting curve derived from the amplification plot is shown in Fig. 2c and the analysis of the melting temperature and melting peaks of W. koreensis using SYBR Green qPCR revealed a reproducible melting temperature of 77.0 °C and specific peaks (Fig. 2d). A standard curve of genomic DNA and the bacterial cell suspension exhibited a linear correlation between the Ct values and the concentrations of input DNA (R2 = 0.999, slope = −3.302) and the bacterial cell suspension (R2 = 0.996, slope = −3.414). In addition, an analysis of genomic DNA and the bacterial cell suspension indicated that the detection limit of SYBR Green qPCR was 5 fg/μl (fg per μl reaction mix), which corresponds to 1.28 × 104 CFU/ml (CFU per ml reaction mix) of W. koreensis (Table 2).

Table 2 Mean Ct end-point fluorescence of 10-fold serial dilutions of Weissella koreensis KACC11853 cloned DNA, genomic DNA and a cell suspension determined with a real-time PCR assay.
Full size table
Figure 2
figure 2

Specificity, melting peak and standard curve of the WK201F/R primer set with SYBR Green qPCR.

(a) The fluorescence intensity as a function of the concentration of template. For each assay, a series of 10-fold dilutions of cloned DNA (range of 1.42 × 103 to 1.42 × 109 copies/μl) was used as the template for PCR (1–7, sample dilutions; 8, no-template control). (b) Standard curve derived from the amplification plot. (c) Melting curve analysis (1–7, sample dilutions; 8, no-template control). (d) Melting peak analysis (1–7, sample dilutions; 8, no-template control). The amplified products’ derivative of relative fluorescence units [-d(RFU)/dT] is plotted as a function of temperature (amplified product, 77.0 °C). The high peak indicates the amplified product and the low peak is the no-template control.

Full size image

Variation in the density of Weissella koreensis during kimchi fermentation

Two types of salted Chinese cabbage kimchi - whole kimchi with red pepper powder and white kimchi without red pepper powder - were obtained from a kimchi company in Korea and samples of each were stored at 4, 15 and 25 °C. The density of W. koreensis in the kimchi samples preserved at 4, 15 and 25 °C varied. The samples taken from whole kimchi stored at 4 °C exhibited the lowest Ct value between weeks 1 and 3 compared with those from white kimchi under the same conditions (Fig. 3a). However, whole kimchi stored at 15 or 25 °C presented the lowest Ct value between days 1 and 2 (Fig. 3b). Thus, red pepper powder was observed to strongly influence the density of W. koreensis during kimchi fermentation, regardless of the temperature and fermentation period (Fig. 3).

Figure 3
figure 3

Changes in the Ct value of real-time PCR for the quantification of W. koreensis in total DNA from two types of salted Chinese cabbage kimchi fermented at 4 °C (a), 15 °C and 25 °C (b).

Full size image

Discussion

Probiotic bacteria have been historically considered to hold great promise for the treatment of gastrointestinal disorders. However, further studies are required to create a more scientific basis for the action of probiotics. Although unprecedented levels of scientific evidence supporting the health benefits of LABs and their products have been accumulated, no clear evidence describing the role of a specific bacterium has been presented. Hence, establishing reliable and efficient species-specific molecular probes for quantitative detection of targeted LABs used in various lactic acid products is critical because such probes would enable detection of individual species and overall profiling of changes in the community structure in response to changes in variables such as time and temperature.

In the initial studies of kimchi LAB communities, traditional approaches based on morphological and phenotypic identification of LAB species grown on agar media were used, but such studies were often unsuccessful9. This method has major disadvantages, such as the long assay time of 10 days and the possibility of detecting cultivable cells only. For these reasons, molecular methods that use sequences of 16S ribosomal RNA genes and other genes for the identification of isolated strains have attracted the attention of many researchers10. However, these approaches are not suitable for monitoring the succession of targeted LAB during kimchi fermentation10. Thus, the value of molecular detection and quantification methods for studying LAB is immense.

Fortunately, the number of microbial genome sequences available has increased dramatically. In particular, the availability of complete or draft LAB genome sequences presents a great opportunity for improving existing molecular detection and quantification tools by identifying new targets for more sensitive and specific detection.

Recently, among the LABs found in kimchi - including L. plantarum, L. sake, Leu. mesenteroides, Leu. lactis, Leu. citreum, Pediococcus pentosaceus, W. cibaria and W. confusa - W. koreensis has been reported to be associated with L-Ornithine production from arginine and to play a crucial role in preventing intracellular lipid accumulation by down-regulating the expression of adipocyte-specific genes. L-Ornithine is a medicinal, non-protein a.a. that has the potential to combat obesity by promoting hormone release and accelerating the rate of basal metabolism2.

In addition, of the many bioactive materials found in kimchi, capsaicin derived from red pepper powder has been proposed as an effective agent for fat digestion. However, little information regarding the correlation between red pepper powder and W. koreensis exists. Therefore, we explored species-specific genes using BLAST searches and developed a primer set to investigate this correlation. The species-specific primer set was derived from the whole-genome sequence of W. koreensis KACC15510 (GenBank accession no. WP_013989464.1). We focused on a membrane protein-encoding gene that was confirmed to be highly variable among Weissella species (Fig. 1).

In a previous report11, the proportion of Weissella was found to be higher in kimchi with red pepper powder than in kimchi without red pepper powder, whereas the proportions of Leuconostoc and Lactobacillus were lower in kimchi with red pepper powder. However, red pepper powder had little influence on the cell density of L. plantarum in kimchi, whereas temperature greatly impacted the proportion of L. plantarum (data not shown). Consequently, red pepper powder is estimated to greatly influence the growth of particular LABs in kimchi. As shown in Fig. 3, in this study, the proportion of W. koreensis was much greater in whole kimchi with red pepper powder than in white kimchi without red pepper powder, regardless of temperature and fermentation period. At 4 °C, the differences in proportion of W. koreensis were more apparent than at 15 and 25 °C.

W. koreensis was verified to be the dominant species and could ferment kimchi at temperatures as low as −1 °C12. In addition, W. koreensis was confirmed to exhibit relatively good psychrophilic growth, which predominated at 4 °C or colder. However, Leuconostoc species incubated at 15 or 25 °C did not delay the growth of W. koreensis (data not shown). As shown in Fig. 3, the patterns for the changing populations of W. koreensis were similar at 4, 15 and 25 °C. The populations of W. koreensis exhibited a rapid increase during the early fermentation period and then remained constant regardless of temperature.

Consequently, to standardize the ripening process during quality-controlled kimchi production, accurate monitoring of changes in the microbial community in situ during the fermentation period is essential13.

The key advantages of this newly developed assay are its specificity and rapidity: it allows species-specific identification and quantification of W. koreensis strains within 1 h without any prior cultivation. The number of replicates used to calculate standard curves and the small standard error among these replicates ensure that the assay is reproducible and highly robust, even with DNA isolated from kimchi.

In conclusion, this newly developed real-time PCR assay detects W. koreensis with high specificity and sensitivity and without false-positive signals from other LABs in pure cultures and in DNA mixtures extracted from kimchi. This real-time PCR assay may be a useful method for detection and quantification of W. koreensis in food for quality management purposes.

Methods

Bacterial strains, growth conditions and DNA preparation

The bacterial strains were obtained from the Korean Agricultural Culture Collection (KACC) in the Republic of Korea and the Belgian Co-ordinated Collections of Micro-organisms (BCCMTM) (Table 1). All strains were grown on de Man Rogosa Sharpe (MRS) agar (Oxoid, Basingstoke, Hampshire, U.K.) plates at 25 to 30 °C for 48 h. Total genomic DNA was extracted using a bacterial genomic DNA extraction kit from Qiagen (Hilden, Germany). To measure the quantity and purity of genomic DNA, a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) was used.

Kimchi sample preparation for PCR

To test the quantitative analysis of kimchi samples, 20 ml of kimchi soup was periodically taken from the two types of kimchi. These were also used for the conventional or real-time PCR analyses. Each kimchi soup sample was filtered through four layers of sterile coarse gauze (Daehan, Daejeon, Korea) to remove large slices and was spun down at 13,000 rpm at 4 °C for 10 min. DNA was extracted using a Fast DNA® spin kit for soil (MP Biomedicals, Solon, OH, USA).

Genome analysis and primer design

The genome sequences of W. koreensis KACC15510 and the other LABs used in this study were downloaded from the NCBI bacterial genome database (ftp://ftp.ncbi.nlm.nih.gov/genomes/bacteria/) and compared using a modified computational pipeline14,15. On the basis of these comparative outputs, target genes that had no significant similarity in nucleotide sequence among other LABs were selected as PCR targets. A species-specific primer set based on a membrane protein-encoding gene of W. koreensis KACC15510 (GenBank accession No. WP_013989464.1) with a predicted PCR product of 201 bp was designed. Amplification primers for conventional PCR and SYBR Green qPCR were synthesized by Bioneer Corporation (Daejeon, Korea) (Table 3).

Table 3 PCR primers and their target and annealing temperatures of the W. koreensis PCR screenings used in this study.
Full size table

Conventional PCR

The PCR amplifications were performed with the above primers (0.2 μM final concentration) and GoTaq® Flexi DNA polymerase (1 × buffer, 4.0 mM MgCl2, 0.2 mM of each dNTP, GoTaq® DNA polymerase 1.25 U final concentration; Promega, Madison, WI, USA) in a final volume of 25 μl according to the manufacturer’s instructions; 25 ng of genomic DNA from a given bacterial strain were used. Amplifications were performed in a PTC-225 thermocycler (MJ Research, Watertown, MA, USA) with the following cycling conditions: initial denaturation of 5 min at 95 °C; 35 cycles of 1 min at 95 °C, 30 s at 61 °C and 1 min at 72 °C; and a final cycle extension of 7 min at 72 °C. After the PCR reaction, each amplified PCR product was electrophoresed through a 1.5% (wt/vol) agarose gel stained with ethidium bromide (EtBr), was visualized on an ultraviolet (UV) transilluminator and was imaged using a VersaDoc 1000 gel imaging system (Bio-Rad laboratories, Hercules, CA, USA).

SYBR Green qPCR

The SYBR Green qPCR assay was performed in a 20 μl reaction. All amplifications were performed with the WK201F/R primers (0.5 μM final concentration), iQTM SYBR® Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA) and approximately 5 ng of purified DNA from each sample. Real-time PCR amplifications were performed using a CFX96 real-time PCR system (Bio-Rad Laboratories, Hercules, CA, USA) with the following cycling conditions: initial denaturation of 30 s at 95 °C; 40 cycles of 5 s at 95 °C and 30 s at 61 °C; and a melting curve analysis from 65 to 95 °C with an increment of 0.5 °C per 5 s. The determination of the Ct and the data analysis were performed automatically using the CFX ManagerTM Version 1.6 software package (Bio-Rad Laboratories, Hercules, CA, USA).

The LOQ and limit of detection (LOD) of the SYBR Green qPCR assay were determined using 10-fold dilutions of plasmid DNA, genomic DNA and a bacterial cell suspension of W. koreensis KACC11853 in a 20 μl reaction mixture containing 10 μl of iQTM SYBR® Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA) and 5 pM each of WK201F/R primers. This LOQ and LOD were reproducible with serial dilutions and SYBR Green qPCR testing in triplicate. The copy number of the plasmid DNA was calculated using the following equation16:

Additional Information

How to cite this article: Kang, B. K. et al. The influence of red pepper powder on the density of Weissella koreensis during kimchi fermentation. Sci. Rep. 5, 15445; doi: 10.1038/srep15445 (2015).