Red pepper powder is a crucial factor that influences the ontogeny of Weissella cibaria during kimchi fermentation

Weissella cibaria has been found in Korean kimchi and other sources, including fermented foods, Greek salami, Spanish sausages, and animal and human excrement. W. cibaria was recently reported to show anticancer, immunomodulatory, anti-inflammatory and antioxidant properties. Nevertheless, fundamental ecological succession studies are required to scientifically confirm the probiotic action of W. cibaria under various conditions, such as fermentation. Therefore, in the present study, we mined the W. cibaria KACC11862 genome in search of species-specific genes to use as new PCR targets for the detection and quantification of W. cibaria in kimchi. The sensitivity and specificity of the identified primer set from the putative outer membrane protein gene for the detection of W. cibaria KACC11862 in kimchi were analysed. Primer set specificity was evaluated using genomic DNA from eight W. cibaria isolates, 10 different species of Weissella and 13 other reference lactic acid bacteria (LAB) strains. Interestingly, by using the qPCR assay developed herein, we found that red pepper powder markedly affects the ontogeny of W. cibaria during kimchi fermentation.

a small number of strains. While there are currently 20 recognised species within this genus, the genomic data examined here represent only six species 12 . Therefore, we established an accurate, sensitive and practical SYBR Green quantitative polymerase chain reaction (qPCR) method to specifically monitor and quantify W. cibaria in diverse fermented foods. Currently, species-, subspecies-or strain-specific DNA probes are widely used to screen, detect, quantify and identify fungi, bacteria, yeast and viruses 13 . Numerous molecular assays that are based on the use of 16S rDNA together with a well-characterised metabolic gene are commonly used for the detection and identification of Weissella species. However, serious drawbacks have been observed for the specific identification and detection of W. cibaria isolates, as these assays also detect other Weissella species 14 . Additionally, many multiplex PCR and chromogenic DNA microarray systems have been developed for the simultaneous amplification of several genes in a single assay. However, these approaches exhibit some limitations, as the detection of target cells in food samples or in mixtures that vary widely with respect to their ratios of bacterial species is difficult 14 . Consequently, detection specificity, which can be influenced both by the uniqueness of the sequence within the targeted microbial genome of interest and by the precise annealing of the primers and probe to their target, is crucial to the efficiency of any PCR detection method.
Over the past decade, considerable efforts have been made to sequence various LAB strains. The growing number of LAB genome sequences available in public databases, combined with diverse bioinformatics tools, has promoted the emergence of methods that are more reliable, faster and more cost-effective for bacterial identification in an extensive range of samples. Particularly, LAB genome databases have made it possible to distinguish specific LAB strains from closely related lineages and species groups. However, despite recent scientific progress with respect to LAB strains in the microbiology industry, current methods for detecting, identifying and quantifying specific LAB strains are still substantially limited. Consequently, in this study, we mined publicly available genome sequence information (ftp://ftp.ncbi.nlm.nih.gov/genbank/) to develop a SYBR Green real-time PCR assay for the precise detection and quantitation of W. cibaria. A pair of species-specific primers based on a single gene from the W. cibaria KACC 11862 genome sequence was designed.
Bacterial membranes are known to have diverse functions in transport and mitochondrial activities and in the biosynthesis of membranes, cell walls and cellular capsules; the specific functions of membranes depend on whether they are specialised or are simply non-specialised cytoplasmic membranes. Membrane fusion proteins are found only in the prokaryotic world and function in conjunction with different types of transport systems in Gram-positive and Gram-negative bacteria. Bacterial membrane proteins are diverse in structure and function and show a wide size distribution 15 .
In the present study, we establish a reliable and efficient procedure for the quantitative detection of W. cibaria in kimchi samples using the SYBR Green PCR technique. Our results reveal that this method can be used for the specific detection and quantification of W. cibaria in various products. Remarkably, using this real-time PCR assay, we find that red pepper powder critically affects the ontogeny of W. cibaria during kimchi fermentation.

Results
Genomic analysis and in silico specificity test of the designed primer set. Specific primers (termed WC207F and WC207R) were designed from the genomic sequence of corresponding to a putative outer membrane protein-encoding gene of W. cibaria using the Lasergene PrimerSelect tool (Table 1). Because the specificity of the WC207F/WC207R primer pair had not been previously evaluated, thorough two-level specificity testing was undertaken. First, the specificity of the WC207F/R primer set was confirmed by using the BLAST and e-PCR analysis programs (http://www.ncbi.nlm.nih.gov/). The BLAST nucleotide program was set to the nr parameter, in which all the GenBank, RefSeq Nucleotide, EMBL, DDBJ and PDB sequences are queried. Furthermore, to increase the range of our specificity analysis, the databases that we searched for unintended gene sequence matches were not limited by organism type. The BLASTn searches returned no significant match to the known reference sequences. The BLASTx results using the predicted protein sequence revealed that the most similar protein to our hypothetical protein was a W. confusa protein [identity = 28%, score = 277 bits (709), and expected = 7e-74]. In addition, the WC207F/R primer set specificity was tested using conventional PCR techniques. For this analysis, DNA samples from 31 different strains were used ( Table 2). The expected PCR product, with a length of 207 bp, was observed in all PCR assays performed using genomic DNA from W. cibaria strains. In contrast, the WC207F/R primer pair did not amplify any product when genomic DNA from other species was used as the template (Table 2). These findings show that the primers designed in this study are specific for W. cibaria strains and are therefore suitable for their identification.
Standard curves and melting temperature. We used SYBR Green real-time PCR analysis to generate a standard curve with W. cibaria KACC 11862 by plotting the mean threshold cycle (Ct) (n = 3) based on the logarithmic concentrations of genomic DNA (ranging from 5 to 5 × 10 −5 ng/μ l), cloned DNA (ranging from 1.44 × 10 9 to 1.44 × 10 3 copies/μ l) and on the density of the cell suspension (ranging from 1.95 × 10 8 to 1.95 × 10 5 CFU/ml) ( Fig. 1a and Table 3). The limit of quantitation (LOQ) assay showed a good linear response and a high correlation coefficient (R 2 = 0.999). A standard curve analysis of the linear portion of the slope resulted in a coefficient of − 3.275, which yielded a PCR efficiency of 102.0% and a y-intercept value of 33.121 (Fig. 1b). Melting analysis (curve, temperature and peaks) with W. cibaria KACC 11862 produced from real-time PCR reactions reproducibly showed a melting temperature of 83.0 °C and specific peaks (Fig. 1c,d). The input genomic DNA concentration (R 2 = 0.994, slope = − 3.470) and the bacterial cell suspension (R 2 = 0.975, slope = − 2.902) standard curves exhibited linear correlations with their respective Ct values. The detection limits of the real-time PCR were 50 fg/μ l and 1.95 × 10 5 CFU/ml of reaction mix for the input genomic DNA concentration and the bacterial cell suspension of W. cibaria KACC11862, respectively ( Table 3). The real-time PCR assay of W. cibaria KACC11862 displayed excellent quantification and accurate detection characteristics.
Quantification of W. cibaria in kimchi samples using real-rime PCR. Whole kimchi (Chinese cabbage fermented with salt and red pepper) and white kimchi (without red pepper powder) samples were purchased from the local market and evaluated. Two samples were stored at three different temperatures (4 °C, 15 °C and 25 °C) during the fermentation period. Each sample was examined for the presence of W. cibaria by SYBR Green real-time PCR. Importantly, regardless of the temperature and fermentation period, we observed a large difference in W. cibaria density between the two types of kimchi samples with respect to the presence or absence of red pepper powder. All of the samples taken from white kimchi and stored at 4 °C, 15 °C and 25 °C produced no fluorescence signal to indicate the Ct value, in contrast to the corresponding samples from whole kimchi (Fig. 2a). W. cibaria was also confirmed as a predominant psychrophilic species during the early fermentation stage of whole kimchi, but only at 4 °C; the lowest Ct values (reflecting the highest concentrations of starting DNA material) were measured between weeks 0 and 7, after which the Ct values increased gradually (Fig. 2b). In addition, the whole-kimchi samples that were stored at 15 °C and 25 °C showed very high Ct values and only minor changes in the density of W. cibaria (Fig. 2b). Consequently, we concluded that red pepper powder drastically influences the density of W. cibaria during kimchi fermentation, regardless of the temperature or fermentation period (Fig. 2).

Discussion
Because of their ability to reach the gastrointestinal tract and to interact with the immune system of their host, probiotic bacteria are receiving substantial attention as vehicles for the delivery of biotherapeutics 16 . However, the study of probiotic communities currently faces a difficult roadblock 17 . Genome sequencing and functional genomics studies on a variety of LAB strains are now rapidly providing new insights into microbial community dynamics in diverse environmental samples, revealing the molecular basis for important traits such as sugar metabolism, stress responses, adaptation and interactions 18 . However, there is no clear evidence of the use of species-specific molecular probes to accurately quantify the proportion of a specific LAB in different ecosystems, preventing a complete understanding of how the abundance of a particular microbe that is associated with different types of food changes over time and how these changes affect the final quality of the fermented products.   Therefore, establishing a reliable and efficient process for the quantitative detection of specific LAB used in various lactic acid products is tremendously important, as it would enable the detection of specific species and overall community profile changes over time, at different temperatures, and under different conditions.
Early studies on kimchi LAB communities generally employed traditional, but often unsuccessful, methods based on the morphological and phenotypic identification of LAB species grown on agar culture media 6 . Indeed, this system has major drawbacks, such as a long assay time of 10 days and the detection of cultivable cells only. Consequently, molecular methods using the sequence of 16S ribosomal RNA and other genes for the identification of isolated strains have drawn the attention of many researchers 19 . However, these approaches are not appropriate for monitoring the succession of a targeted bacterial species during kimchi fermentation 19 . Therefore, the value of molecular detection and quantification methods for studying LAB is enormous.
Heretofore, all types of kimchi were thought to behave similarly to the kimchi samples that were used in previous studies. Specifically, an extensive diversity of bacteria was present at the beginning of the process; as fermentation progressed, the sample became dominated by three main genera, namely Lactobacillus, Leuconostoc, and Weissella 20 .
Red pepper powder in particular was shown to delay the kimchi fermentation progress, including LAB growth and metabolite production and especially during the early fermentation period. At the genus level, the proportion of Weissella was 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 20 . Recently, W. koreensis, which shows anti-obesity effects, was identified as the dominant species; it exhibits relatively good psychrophilic growth at 4 °C or colder. The density of W. koreensis is usually greater in whole kimchi with red pepper powder than in white kimchi without red pepper powder, regardless of the temperature and fermentation period. In addition, red pepper powder showed little influence on the cell proportions of Lactobacillus plantarum and Leuconostoc mesenteroides in kimchi, although temperature greatly impacted the density of L. plantarum (data not shown). Consequently, red pepper powder is estimated to greatly influence the proportion, or density of particular kimchi LAB strains 7 .
According to phylogenetic analysis based on 16S rRNA gene sequences, the genus Weissella was one of the predominant bacterial groups during kimchi fermentation 21 . Pyrosequencing analysis of commercial kimchi samples resulted in the genus Weissella being predominant at 44.4%. W. koreensis was 27.2% and W. cibaria was 8.7% 22 . However, the draft genome of W. paramesenteroides was a poor match to the metagenome sequence 21 . These results indicated that all Weissella species are not influenced by red pepper powder.
Generally, among the many bioactive materials found in kimchi, red pepper powder-derived capsaicin has been proposed as an effective source of anticancer, fat-digestion and diabetes-prevention properties 23 . However, little information exists on the correlation between red pepper powder and W. cibaria with respect to anticancer or fat-digestion effects.
Therefore, we identified species-specific genes using BLAST searches and developed a primer set to investigate these correlations. The species-specific primer set was designed using the W. cibaria KACC11862 (GenBank accession no. ZP_08416035.1, gi|332429607:82527-85805) whole-genome sequence. The outer membrane protein gene (GenBank accession no. ZP_08416035.1) was also confirmed to be highly variable among Weissella species.
Interestingly, in the present study, the presence of W. cibaria was clearly detected in whole kimchi with red pepper powder, but it was not detected at all in white kimchi without red pepper powder regardless of the temperature or fermentation period. In order to provide additional evidence that red pepper powder as a crucial factor that influences the ontogeny of W. cibaria during kimchi fermentation, we designed a new marker to detect novel gene of W. paramesenteroides. As a result, W. paramesenteroides was not detected in the whole kimchi or white kimchi regardless of red pepper powder (data not shown). Moreover, different kinds of kimchi were used to determine if the function of red pepper powder exists in other kimchi. W. cibaria was identified from the chonggak kimchi (with red pepper powder) during the early fermentation period, while it was not detected in the dongchimi and watery kimchi (without red pepper powder) (data not shown).
According to the previous report on kimchi metabolites to investigate the effect of red pepper on kimchi fermentation, the metabolite concentrations in kimchi with red pepper powder were a little higher than those in kimchi without red pepper powder, which was presumably due to the concentration effects of metabolites in kimchi supernatants influenced by the addition of red pepper powder. Therefore, it is thought that metabolites Scientific RepoRts | 6:28232 | DOI: 10.1038/srep28232 in kimchi were influenced by the addition of red pepper powder, hence having an influence on ontogeny of W. cibaria in whole kimchi 20 .
In addition, W. cibaria was more abundant at 4 °C than at 15 °C or 25 °C (Fig. 2). W. cibaria was confirmed as a relatively psychrophilic strain that predominates at 4 °C. However, the population of W. cibaria decreased rapidly from the middle of the fermentation period onwards. In general, the proportion of W. cibaria, whether incubated at 4 °C, 15 °C and 25 °C, was strikingly lower than that of W. koreensis 7 .
To control the ripening process in quality-controlled kimchi production, it is essential to accurately monitor changes in the microbial community in situ during the fermentation period 24 . The most important advantages of this newly developed assay are its specificity and rapidity. Indeed, it allows the species-specific identification and quantification of W. cibaria strains within an hour, without any prior cultivation step. The many replicates used for obtaining the standard curves and the small standard errors among these replicates demonstrated that the assay is reproducible and highly robust, even with DNA isolated from kimchi.
In conclusion, whether using DNA isolated from pure cultures or DNA mixtures extracted from kimchi, we show that this newly developed real-time PCR assay detects W. cibaria with high specificity and sensitivity and without any false-positive signals caused by other LAB strains. This real-time PCR assay may be a convenient method for the detection and quantification of W. cibaria in food for quality-management purposes.

Methods
Bacterial strains and DNA extraction. In total, 31 LAB strains were used to validate the specificity of the species-specific primers developed in this study ( Table 2) Kimchi sampling. Two types of kimchi were purchased from a commercial factory in Korea. The 26 batches (13 batches each for two kinds of kimchi) were stored at 4 °C. In case of storage at 15 °C and 25 °C, 30 batches (15 batches each for two kinds of kimchi) were used, respectively. The kimchi samples used in the experiments were obtained on the same day that they were produced and were stored at different incubation temperatures (4 °C, 15 °C and 25 °C) until the sampling time. Periodically during the fermentation period, aliquots were filtered through four layers of coarse gauze (Daehan, Daejeon, Korea) to remove large kimchi particles, and the filtrates were pooled. The filtrates were centrifuged (13,000 rpm for 10 min at 4 °C) to harvest the microorganisms. One microliter of each sample (5 ng/μ l) was centrifuged separately for quantitative analysis by real-time PCR. Total genomic DNA was extracted from the kimchi sample pellets according to a previously described procedure combining a bead-beating method with the Fast DNA Spin kit (MPbio) 25 .
Genome analysis and primer design. The genome sequences from W. cibaria KACC 11862 and the other LAB strains used in this study were downloaded from the NCBI bacterial genome database (ftp://ftp.ncbi.nlm. nih.gov/genomes/bacteria/). We used bioinformatics sequence analysis technology to integrate the different steps and used a modified Chen and Lang tool for the candidate gene selection pipeline via computational clustering 26,27 . On the basis of these comparative outputs, target genes showing no significant concordance with the other LAB strains were selected as PCR targets. The primers (termed WC207F/R) used for W. cibaria KACC11862 (GenBank accession No. ZP_08416035.1) were designed using Lasergene PrimerSelect software (version 7.2.1; DNASTAR Inc., USA) on the putative outer membrane protein-encoding gene. These primers amplify a DNA fragment of 207 bp. The nucleotide sequences of the primers were assessed for specificity by using NCBI BLAST searches (http://blast.ncbi.nlm.nih.gov). The primers were synthesised by Bioneer Corporation (Daejeon, Korea) ( Table 1) The amplified products were stained with LoadingStar (DYNEBIO, Seoul, Korea) and analysed by electrophoresis using 1.5% (w/v) agarose gels. The bands were visualised using a VersaDoc 1000 gel imaging system (Bio-Rad Laboratories, Hercules, CA, USA). SYBR Green qPCR. The SYBR Green real-time PCR amplifications were performed using a CFX96 real-time PCR system (Bio-Rad Laboratories, Hercules, CA, USA) in a final reaction volume of 20 μ l containing 5 ng of purified DNA from each sample, the abovementioned primers (0.5 μ M final concentration) and iQ TM SYBR ® Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA). The real-time PCR amplifications were performed using the following cycling conditions: initial denaturation of 30 s at 95 °C, 40 cycles of 5 s at 95 °C and 30 s at 62 °C, and a melting curve analysis from 65 to 95 °C with an increment of 0.5 °C per 5 s. Analysis of the absolute quantification data was performed automatically using the Bio-Rad CFX Manager TM Version 1.6 software package.
The LOQ and limit of detection (LOD) were determined as the amounts in 10-fold serial dilutions of plasmid DNA, genomic DNA or a bacterial cell suspension of W. cibaria KACC 11862 that corresponded to the threshold cycles at which the sum of the specificity and sensitivity of the assay was maximised. The quality of the SYBR Green real-time assay was determined by the absence of stochastic effects in the reaction efficiency and by whether or not nonspecific reaction products were formed.