Abstract
Genetic markers currently used for the discrimination of Lactobacillus delbrueckii subspecies have low efficiency for identification at subspecies level. Therefore, our objective in this study was to select novel genetic markers for accurate identification and discrimination of six L. delbrueckii subspecies based on pangenome analysis. We evaluated L. delbrueckii genomes to avoid making incorrect conclusions in the process of selecting genetic markers due to mislabeled genomes. Genome analysis showed that two genomes of L. delbrueckii subspecies deposited at NCBI were misidentified. Based on these results, subspecies-specific genetic markers were selected by comparing the core and pangenomes. Genetic markers were confirmed to be specific for 59,196,562 genome sequences via in silico analysis. They were found in all strains of the same subspecies, but not in other subspecies or bacterial strains. These genetic markers also could be used to accurately identify genomes at the subspecies level for genomes known at the species level. A real-time PCR method for detecting three main subspecies (L. delbrueckii subsp. delbrueckii, lactis, and bulgaricus) was developed to cost-effectively identify them using genetic markers. Results showed 100% specificity for each subspecies. These genetic markers could differentiate each subspecies from 44 other lactic acid bacteria. This real-time PCR method was then applied to monitor 26 probiotics and dairy products. It was also used to identify 64 unknown strains isolated from raw milk samples and dairy products. Results confirmed that unknown isolates and subspecies contained in the product could be accurately identified using this real-time PCR method.
Keywords: Lactobacillus delbrueckii subspecies, pangenome, genetic marker, identification, real-time PCR, probiotic product
Introduction
Lactobacillus delbrueckii comprises six subspecies, namely delbrueckii, lactis, bulgaricus, indicus, jakobsenii, and sunkii, all of which have historically been differentiated based on their ability to metabolize different carbohydrates [1]. Among these subspecies, L. delbrueckii subsp. lactis and bulgaricus are usually associated with the manufacture of dairy products such as cheeses and yogurt [2]. L. delbrueckii subsp. bulgaricus is one of the starter culture components for the production of yogurt [1, 3]. This subspecies displays probiotic properties [4]. On the other hand, L. delbrueckii subsp. lactis is traditionally used in cheese production and can be distinguished from L. delbrueckii subsp. bulgaricus by its extensive carbohydrate metabolizing capabilities [1, 5]. L. delbrueckii subsp. delbrueckii cannot ferment lactose. It is typically associated with fermented vegetables [2]. L. delbrueckii subsp. indicus, jakobsenii, and sunkii are relatively minor subspecies isolated from Indian dairy products, fermented alcoholic beverages, and non-salted pickle as a traditional Japanese food, respectively [6-9].
Accurate identification of L. delbrueckii subspecies in food samples is an important issue to confirm probiotic properties and perform product quality assessment [4]. Genetic markers and molecular-based methods have been used to efficiently identify and detect lactic acid bacteria commonly used in commercial probiotic and dairy products. Molecular-based methods for the identification and typing of lactic acid bacteria have been reported, including amplified fragment length polymorphism (AFLP), DNA–DNA hybridization (DDH), multi-locus sequence analysis (MLST), and restriction fragment length polymorphism (RFLP) [6,10-12]. However, these techniques are labor-intensive, expensive, and time-consuming with low reproducibility whereas PCR-based methods are rapid, sensitive, and reliable for identifying lactic acid bacteria [4]. Of these methods, genetic markers such as the 16S rRNA gene and 16S–23S rRNA intergenic spacer region have been used to distinguish L. delbrueckii used in PCR-based methods [13]. Although genetic markers described above are useful for identifying L. delbrueckii at the species level, they cannot be applied to distinguish L. delbrueckii at the subspecies level [4].
Recently, the development of whole-genome sequencing (WGS) and the increase in genome sequences have made it possible to rapidly and freely process large-scale sequence data on microorganisms in public repositories [14]. Pangenome analysis based on WGS has a wide range of applications, including prediction of lifestyles of microorganisms, pathogenicities, resistome, and taxonomy [15]. Pangenome analysis also allows reclassification of bacterial species and/or subspecies, improving and clarifying criteria previously presented [16]. In the present study, we selected six L. delbrueckii subspecies-specific genetic markers by pangenome analysis to develop a real-time PCR method for rapid identification of bacterial strains. The real-time PCR method we developed was then applied to bacterial strains isolated from raw milk, probiotic products, and dairy products to identify and differentiate three L. delbrueckii subspecies.
Materials and Methods
Pangenome Analysis and Selection of Genetic Markers
The in silico scheme for selecting the genetic markers of six L. delbrueckii subspecies is illustrated in Fig. 1. A total of 41 genomes belonging to the subspecies L. delbrueckii subsp. delbrueckii, lactis, bulgaricus, indicus, jakobsenii, and sunkii were obtained from the National Center for Biotechnology Information (NCBI) (Table 1). Phylogenetic analysis based on the pangenome was performed using microbial pangenomics in Anvi’o v6 software [17]. According to the developer’s recommendation, a genome database for pangenome analysis was constructed using Anvi’o genome storage. The pangenome was then analyzed using the NCBI BLASTp and MCL algorithm. Subsequently, a phylogenetic tree was constructed based on pangenome cluster frequencies.
Fig. 1.
The in silico approach for genetic marker selection of six L. delbrueckii subspecies.
Table 1.
Summary in genome features of 41 L. delbrueckii subspecies.
| Organism name | Strain | Size (Mb) | GC% | CDS | Assembly | Accession no. |
|---|---|---|---|---|---|---|
| L. delbrueckii subsp. bulgaricus | ATCC BAA-365 | 1.85695 | 49.7 | 1579 | Complete | CP000412.1 |
| L. delbrueckii subsp. bulgaricus | ATCC 11842 | 1.865 | 49.7 | 1561 | Complete | CR954253.1 |
| L. delbrueckii subsp. bulgaricus | 2038 | 1.87292 | 49.7 | 1562 | Complete | CP000156.1 |
| L. delbrueckii subsp. bulgaricus | CNCM I-1519 | 1.79654 | 49.9 | 1630 | Contig | AGHW01 |
| L. delbrueckii subsp. bulgaricus | INRA-MIG | 1.85324 | 49.8 | 1692 | Scaffold | CCET01 |
| L. delbrueckii subsp. bulgaricus | DSM 20081 | 1.75853 | 49.9 | 1533 | Scaffold | JQAV01 |
| L. delbrueckii subsp. bulgaricus | MN-BM-F01 | 1.87507 | 49.7 | 1585 | Complete | CP013610.1 |
| L. delbrueckii subsp. bulgaricus | CFL1 | 1.75792 | 49.8 | 1539 | Contig | CZPS01 |
| L. delbrueckii subsp. bulgaricus | LBB.B5 | 1.77788 | 49.8 | 1558 | Contig | LUGK01 |
| L. delbrueckii subsp. bulgaricus | DSM 20080 | 1.86818 | 49.8 | 1564 | Complete | CP019120.1 |
| L. delbrueckii subsp. bulgaricus | ND04 | 1.86175 | 49.6 | 1538 | Complete | CP016393.1 |
| L. delbrueckii subsp. bulgaricus | MBT 92059 | 1.83117 | 49.8 | 1648 | Scaffold | QOVB01 |
| L. delbrueckii subsp. bulgaricus | L99 | 1.84811 | 49.7 | 1575 | Complete | CP017235.1 |
| L. delbrueckii subsp. bulgaricus | KLDS1.0207 | 1.86918 | 49.8 | 1620 | Complete | CP032451.1 |
| L. delbrueckii subsp. bulgaricus | FAM 21277 | 2.01984 | 49.2 | 1830 | Contig | VBSR01 |
| L. delbrueckii subsp. bulgaricus | NBRC 13953 | 1.72582 | 50.0 | 1519 | Contig | BJMY01 |
| L. delbrueckii subsp. bulgaricus | KLDS1.1011 | 1.88749 | 49.8 | 1634 | Complete | CP041280.1 |
| L. delbrueckii subsp. bulgaricus | LJJ | 1.89109 | 49.5 | 1604 | Complete | CP049052.1 |
| L. delbrueckii subsp. bulgaricus | ACA-DC 87 | 1.856 | 49.8 | 1579 | Complete | LT899687.1 |
| L. delbrueckii subsp. delbrueckii | KACC 13439 | 1.76619 | 50.0 | 1485 | Contig | LHPL01 |
| L. delbrueckii subsp. delbrueckii | KCTC 13731 | 1.91051 | 50.0 | 1600 | Complete | CP018216.1 |
| L. delbrueckii subsp. delbrueckii | DSM 20074 | 1.95372 | 49.6 | 1577 | Complete | CP018615.1 |
| L. delbrueckii subsp. delbrueckii | TUA4408L | 2.01244 | 49.9 | 1718 | Complete | CP021136.1 |
| L. delbrueckii subsp. delbrueckii | NBRC 3534 | 1.78742 | 50.3 | 1588 | Contig | BJLM01 |
| L. delbrueckii subsp. delbrueckii | NBRC 3202 | 1.91031 | 50.1 | 1653 | Complete | AP019750.1 |
| L. delbrueckii subsp. indicus | JCM 15610 | 1.87741 | 49.5 | 1627 | Contig | LGAS01 |
| L. delbrueckii subsp. indicus | DSM 15996 | 1.86357 | 49.6 | 1621 | Scaffold | AZFL01 |
| L. delbrueckii subsp. indicus | JCM 15610 | 2.02186 | 49.4 | 1694 | Complete | CP018614.1 |
| L. delbrueckii subsp. jakobsenii | ZN7a-9 | 1.73081 | 50.2 | 1552 | Contig | ALPY01 |
| L. delbrueckii subsp. jakobsenii | DSM 26046 | 1.74924 | 50.3 | 1568 | Scaffold | JQCG01 |
| L. delbrueckii subsp. jakobsenii | DSM 26046 | 1.8918 | 50.1 | 1614 | Complete | CP018218.1 |
| L. delbrueckii subsp. jakobsenii | DSM 26046 | 1.78119 | 50.1 | 1585 | Scaffold | PUFG01 |
| L. delbrueckii subsp. lactis | CRL581 | 2.13682 | 49.6 | 1639 | Scaffold | ATBQ01 |
| L. delbrueckii subsp. lactis | KCCM 34717 | 2.26338 | 49.1 | 1905 | Complete | CP018215.1 |
| L. delbrueckii subsp. lactis | DSM 20072 | 2.16598 | 49.0 | 1793 | Complete | CP022988.1 |
| L. delbrueckii subsp. lactis | KCTC 3034 | 2.23761 | 49.0 | 1889 | Complete | CP023139.1 |
| L. delbrueckii subsp. lactis | NBRC 3734 | 1.81291 | 50.2 | 1653 | Contig | BJLO01 |
| L. delbrueckii subsp. lactis | lactis1 | 2.05032 | 49.6 | 1694 | Complete | LS991409.1 |
| L. delbrueckii subsp. lactis | NWC_2_2 | 2.269179 | 48.7 | 1934 | Complete | CP031023.1 |
| L. delbrueckii subsp. sunkii | JCM 17838 | 1.94526 | 50.1 | 1713 | Contig | LGHR01 |
| L. delbrueckii subsp. sunkii | JCM 17838 | 2.00434 | 50.1 | 1726 | Complete | CP018217.1 |
The pangenome of L. delbrueckii subspecies was calculated using Bacterial Pan Genome Analysis (BPGA) pipeline ver. 1.3 (identity cut off = 50%) [18]. The pangenome was formatted into two local databases: a pangenome database and a core-genome database for each subspecies. Candidate genetic markers were selected by comparing the pangenome database composed of protein-coding genes, present in all genomes except for the target subspecies, and the core-genome database composed of protein-coding genes present in all genomes of target subspecies. Candidate genetic markers were then aligned with 59,196,562 sequences using BLASTN. Genetic markers only present in target subspecies but not present in other bacterial genomes were selected.
In Silico Specificity Confirmation and Development of Subspecies-Specific Primer
A total of four genome sequences registered at the species level were obtained from the NCBI. Genetic marker specificity was aligned with 45 genome sequences, including genomes known to the species level using USEARCH ver. 9.0 [19]. Alignment results are presented as a heatmap using Seaborn python library in Matplotlib. Subspecies-specific primer pairs for L. delbrueckii subsp. lactis, bulgaricus, and delbrueckii were designed from their genetic markers using Primer Designer (Scientific and Education Software, USA).
DNA Extractions from L. delbrueckii Subspecies and Lactic Acid Bacteria
For specificity testing of primer pairs developed in this study, the reference strains of lactic acid bacteria mainly isolated from probiotic and dairy products were used. A total of 54 strains of lactic acid bacteria including L. delbrueckii subspecies were obtained from the Korean Agricultural Culture Collection (KACC, Korea), the Korean Collection for Type Cultures (KCTC, Korea), the Korean Culture Center of Microorganisms (KCCM, Korea), the NITE Biological Resource Center (NBRC, Japan), and the Laboratory Isolates (LI, Korea) (Table 2). All reference strains were grown in MRS broth (Difco, Becton & Dickinson, USA) for extraction of genomic DNA. L. delbrueckii and other bacterial strains were cultured for 48 h at 42°C and 37°C under anaerobic condition, respectively. Bacterial cells were centrifuged at 13,600 ×g for 5 min and the supernatant was removed. Genomic DNA of reference strains was extracted using a DNeasy Blood & Tissue Kit (Qiagen, Germany) following the protocol described previously [13, 20]. DNA concentration and purity were confirmed using a MaestroNano spectrophotometer (Maestrogen, USA).
Table 2.
List of reference strains used in this study.
| Species | Strain no. |
|---|---|
| Lactobacillus delbrueckii subsp. bulgaricus | KACCa 12420 |
| Lactobacillus delbrueckii subsp. bulgaricus | LIb 00010 |
| Lactobacillus delbrueckii subsp. bulgaricus | LI 00011 |
| Lactobacillus delbrueckii subsp. bulgaricus | LI 00012 |
| Lactobacillus delbrueckii subsp. bulgaricus | LI 00013 |
| Lactobacillus delbrueckii subsp. bulgaricus | LI 00014 |
| Lactobacillus delbrueckii subsp. lactis | KACC 12417 |
| Lactobacillus delbrueckii subsp. lactis | LI 00015 |
| Lactobacillus delbrueckii subsp. delbrueckii | KACC 13439 |
| Lactobacillus delbrueckii subsp. delbrueckii | KCTC 13730 |
| Lactobacillus acidipiscis | KACC 12394 |
| Lactobacillus acidophilus | KACC 12419 |
| Lactobacillus agilis | KACC 12433 |
| Lactobacillus amylolyticus | KACC 12374 |
| Lactobacillus amylophilus | KACC 11430 |
| Lactobacillus amylovorus | KACC 12435 |
| Lactobacillus brevis | KCTCc 3498 |
| Lactobacillus buchneri | KACC 12416 |
| Lactobacillus casei | KACC 12413 |
| Lactobacillus chiayiensis | NBRCd 112906 |
| Lactobacillus coryniformis | KACC 12411 |
| Lactobacillus crustorum | KACC 16344 |
| Lactobacillus curvatus | KACC 12415 |
| Lactobacillus farciminis | KACC 12423 |
| Lactobacillus fermentum | KACC 11441 |
| Lactobacillus gallinarum | KACC 12370 |
| Lactobacillus gasseri | KCTC 3163 |
| Lactobacillus heilongjiangensis | KACC 18741 |
| Lactobacillus helveticus | KACC 12418 |
| Lactobacillus jensenii | KCTC 5194 |
| Lactobacillus johnsonii | KCTC 3801 |
| Lactobacillus kunkeei | KACC 19371 |
| Lactobacillus lindneri | KACC 12445 |
| Lactobacillus mucosae | KACC 12381 |
| Lactobacillus parabuchneri | KACC 12363 |
| Lactobacillus paracasei | KCTC 3165 |
| Lactobacillus paraplantarum | KACC 12373 |
| Lactobacillus paraplantarum | KCTC 5045 |
| Lactobacillus pentosus | KACC 12428 |
| Lactobacillus pentosus | KCCMe 40997 |
| Lactobacillus plantarum subsp. argentoratensis | KACC 12404 |
| Lactobacillus plantarum subsp. plantarum | KACC 11451 |
| Lactobacillus reuteri | KCTC 3594 |
| Lactobacillus rhamnosus | KCTC 3237 |
| Lactobacillus ruminis | KACC 12429 |
| Lactobacillus sakei | KCTC 3603 |
| Lactobacillus salivarius | KCTC 3600 |
| Lactobacillus sanfranciscensis | KACC 12431 |
| Lactobacillus zymae | KACC 16349 |
| Bifidobacterium animalis subsp. lactis | KACC 16638 |
| Bifidobacterium bifidum | KCTC 3418 |
| Bifidobacterium breve | KACC 16639 |
| Bifidobacterium longum subsp. infantis | KCTC 3249 |
| Bifidobacterium longum subsp. longum | KCCM 11953 |
aKACC, the Korean Agricultural Culture Collection
bLI, the Laboratory Isolate
cKCTC, the Korean Collection for Type Cultures
dNBRC, the NITE Biological Resource Center
eKCCM, the Korean Culture Center of Microorganisms
Specificity and Accuracy of Specific Primer Pairs
Real-time PCR assay was conducted to determine the specificity and accuracy of primer pairs using a 7500 Real-time PCR system. Each reaction contained 20 ng of genomic DNA, 10 μl of 2X Thunderbird SYBR qPCR Mix (Toyobo, Japan), 500 nM of primer pairs, and distilled water up to 20 μl total volume. Real-time PCR conditions consisted of initiation at 95°C for 2 min followed by 30 amplification cycles of 95°C for 5 s and 60°C for 30 s. Melting curves were obtained at 95°C for 15 s, 60°C for 1 min, 95°C for 30 s, and 60°C for 15 s. Specificity of primer pairs was tested against a total of 10 strains of L. delbrueckii subspecies and 44 other lactic acid bacteria. For the accuracy test, genomic DNA from each reference strain was serially diluted and used for real-time PCR.
Application of Real-Time PCR Method
To test the developed real-time PCR method, 64 isolates, 15 probiotic products, and 11 dairy products were used. L. delbrueckii subspecies were isolated from three raw milk samples and three dairy products. Serially diluted samples were spread onto MRS agar plates (Difco, Becton & Dickinson, USA) and incubated at 42°C for 48 h under anaerobic conditions. Probiotic and dairy products were obtained from markets around the world. Genomic DNAs were extracted from isolates and products under the same conditions as described in section 2.3. DNA Extraction of L. delbrueckii Subspecies and Lactic Acid Bacteria. For the application test, genomic DNAs of isolates or products were added to wells of 96-well plates containing 2X qPCR mix (Toyobo) and subspecies-specific primer pairs. The real-time PCR condition was the same as that described in section 2.4. Specificity and Accuracy for Specific Primer Pairs.
Results and Discussion
Pangenome Analysis
Many studies have previously reported a mislabeling issue regarding subspecies or closely related species in the NCBI genome database [20, 21]. In these studies, the majority of the mislabeled genomes were closely related species [20, 22, 23]. Such genomes should therefore be evaluated to avoid reaching incorrect conclusions in a comparative genomic analysis. In the present study, for the first time, we evaluated the genomes of L. delbrueckii subspecies by phylogenetic analysis based on the pangenome before specific genetic markers were selected. Phylogenetic analysis results based on pangenome frequencies were displayed along with the distribution of subspecies’ specific regions. Each bar represents L. delbrueckii subspecies genomes and each layer presents pangenome distribution (Fig. 2). Most genomes clustered according to the subspecies. However, some genomes of L. delbrueckii subsp. bulgaricus and delbrueckii clustered with different subspecies. Genomes of L. delbrueckii subsp. bulgaricus FAM 21277 and delbrueckii TUA4408L clustered with L. delbrueckii subsp. lactis and sunkii, respectively. Based on these results, L. delbrueckii subsp. bulgaricus FAM 21277 and delbrueckii TUA4408L were determined as L. delbrueckii subsp. lactis and sunkii, respectively. These genomes should be indicated correctly in the genome database to avoid further misidentification. We also suggest implementing measures to prevent and correct taxonomic errors in the NCBI database to avoid confusion in future research.
Fig. 2. Pangenome distribution of the 41 L. delbrueckii subspecies genomes.
The color bar of black, yellow, red, blue, green, and green represents L. delbrueckii subsp. bulgaricus, jakobsenii, delbrueckii, sunkii, indicus, and lactis genomes, respectively. The dark color and tinted bright of the bar indicate core-genome presence and absence, respectively. The phylogenetic tree on the right is based on gene cluster frequencies.
Closely related strains in phylogenetic analysis can be distinguished using efficient and customized mining methods for genome sequences [20, 24, 25]. Conventional methods can be used to successfully distinguish pathogenic bacteria that are difficult to differentiate, although these methods only focus on pathogenic bacteria. Studies on lactic acid bacteria are still lacking. Here, we employed a pangenome approach to identify novel genetic markers for specific identification and detection of L. delbrueckii subspecies.
As a result of pangenome analysis, a total of 67,178 genes from 41 L. delbrueckii subspecies yielded a pangenome size of 3,456 genes. The core-genome, accessory-genome, and unique-genome had 749, 2,071, and 636 genes, respectively. Six subspecies-specific genetic markers were then obtained by pangenome analysis. Genetic markers were found to be protein-coding genes present in the same subspecies but absent in other subspecies or bacterial strains. By comparing genomes of the same subspecies, 995 to 1,628 protein-coding genes were found in common in the genomes of each subspecies and considered as the core-genome for each subspecies. After comparing each core-genome with pangenome for protein-coding genes present in all genomes except for target genomes, 5 to 50 protein-coding genes were selected as candidate genetic markers for each subspecies. These candidate genetic markers were aligned with 59,196,562 genome sequences. Genes not present in other bacterial strains except target subspecies were finally selected as genetic markers. Selected genetic markers of L. delbrueckii subsp. bulgaricus, lactis, delbrueckii, indicus, jakobsenii, and sunkii were identified as YcaO-like family protein (Accession No. ABJ57813.1), Ser/Thr protein kinase (Accession No. EGD27260.1), choline kinase (Accession No. KNZ37552.1), DNA methyltransferase (Accession No. KNE31255.1), RpoD family RNA polymerase sigma factor (Accession No. EOD03403.1), and hypothetical protein (Accession No. APG74821.1), respectively.
Genetic Marker Specificity Test
The specificity of genetic markers was tested using 45 genomes including genomes registered at species level by in silico analysis. The heatmap for identities of genetic markers in genomes is shown with color codes, ranging from blue (region with high identity) to sky blue (region with low identity) (Fig. 3). Each genetic marker shared more than 95% sequence identity with genomes of most corresponding subspecies. In contrast, a genetic marker for L. delbrueckii subsp. bulgaricus was present in 19 bulgaricus genomes (95–100% identity), but one genome had the genetic marker for L. delbrueckii subsp. lactis instead of bulgaricus (99% identity). A genetic marker for L. delbrueckii subsp. delbrueckii was present in six delbrueckii genomes (99–100% identity), but one genome had the genetic marker for L. delbrueckii subsp. sunkii instead of delbrueckii (100% identity). These results were the same as those of pangenome analysis. Genetic markers were aligned with their genomes to determine the subspecies of genomes registered at the species level. L. delbrueckii AVK, TJA31, and 328M contained the genetic marker for L. delbrueckii subsp. bulgaricus (96–97% sequence identity). L. delbrueckii LDELB18P1 contained the genetic marker for L. delbrueckii subsp. lactis (100% sequence identity).
Fig. 3. Heatmap shows the presence/absence of genetic markers in 45 genomes.
The heatmap shows that the similarity of genetic markers present in each genome is visualized with a color bar of blue (high identity) to sky blue (low identity). The bottom of the heatmap presents the six subspecies-specific genetic markers; the right of the heatmap presents the 45 genomes of L. delbrueckii species or subspecies.
In previous studies, genes such as 16S rRNA, 16S–23S rRNA intergenic spacer region, and the elongation factor Tu (tuf) gene have been used to distinguish microorganisms at the species or subspecies level [4,9,26-29]. However, some studies have reported that these genes share high sequence similarities without showing sufficient variabilities to allow for the differentiation between L. delbrueckii subspecies [4, 13]. In contrast, we selected genetic markers specific to the genomes of each subspecies using pangenome analysis. The markers selected in this study were specific to L. delbrueckii subspecies and other bacterial strains. They were able to accurately identify subspecies level for unknown genomes.
Specificity and Accuracy for Specific Primer Pairs
The method to identify L. delbrueckii subspecies with genetic markers selected in this study requires WGS and bioinformatics analysis to confirm the presence of their markers. This method can accurately identify L. delbrueckii subspecies. However, the cost associated with WGS and its informational capacities must be considered. In addition, specialized researchers are needed to handle bioinformatics analysis [14, 30]. Therefore, we developed a real-time PCR method to cost-effectively identify many L. delbrueckii isolates using relatively simple procedures. This real-time PCR method is designed to identify three main subspecies, L. delbrueckii subsp. delbrueckii, lactis, and bulgaricus [2] that are mainly isolated from food or used for fermenting dairy products.
Subspecies-specific primer pairs were designed from selected genetic markers. Information for primer pairs is shown in Table 3. The specificity test for these designed subspecies-specific primer pairs was performed using 54 reference strains of lactic acid bacteria. The genomic DNA of each subspecies generated a positive signal for corresponding primer pairs, whereas genomic DNAs from other L. delbrueckii subspecies and lactic acid bacteria did not generate any signal (Fig. 4). The Ct value ranged from 12.72 to 16.94 for each subspecies-specific primer pair. Genomic DNAs of three subspecies were used to confirm the accuracy of primer pairs. Standard curves were generated using serial diluted genomic DNA at an amount ranging from 0.002 ng to 20 ng. Slopes for standard curves of L. delbrueckii subsp. bulgaricus, lactis, and delbrueckii were −3.44, −3.46, and −3.34, respectively. All correlation coefficient values (R2) were greater than 0.998 and all amplification efficiencies were more than 94%(Fig. 5). All of these values met real-time PCR conditions indicating a high efficiency [31]. Thus, the method we developed in the present study shows high accuracies. Our real-time PCR method targeting specific genetic markers enables rapid and accurate identification of three L. delbrueckii subspecies.
Table 3.
Primer pairs designed in this study.
| Target species | Primer name | Sequence (5'-3') | Size (bp) | Target gene | Accession no. |
|---|---|---|---|---|---|
| L. delbrueckii subsp. bulgaricus | Bulgaricus_F | TAC CGC TGT TCT GTC TCA AGG | 102 | YcaO-like family protein | ABJ57813.1 |
| Bulgaricus_R | TAT GCC TCC GTG AGC GAT CT | ||||
| L. delbrueckii subsp. lactis | Lactis_F | TTG TGC AAG AGC CAG CTG AA | 113 | Ser/Thr protein kinase | EGD27206.1 |
| Lactis_R | GCC GCC ATT ACT GAA GTG GA | ||||
| L. delbrueckii subsp. delbrueckii | Delbrueckii_F | CAT GGA ACT TCT GCG AAG GT | 110 | Choline kinase | KNZ37552.1 |
| Delbrueckii_R | TAG ATC CGG AGC TGT TCC AC |
Fig. 4. The specificity of subspecies-specific primer pairs against 54 lactic acid bacteria.
(A) Specificity of L. delbrueckii subsp. bulgaricus primer pair, amplification curve: L. delbrueckii subsp. bulgaricus KACC 12420, LI 00010, LI 00011, LI 00012, LI 00013, and LI 00014; (B) Specificity of L. delbrueckii subsp. lactis primer pair, amplification curve: L. delbrueckii subsp. lactis KACC 12417 and LI 00015; (C) Specificity of L. delbrueckii subsp. delbrueckii primer pair, amplification curve: L. delbrueckii subsp. delbrueckii KACC 13439 and KCTC 13730.
Fig. 5. Real-time PCR amplification plot, melt curve, and standard curve.
(A) L. delbrueckii subsp. bulgaricus amplification plot (left), melt curve (middle), and standard curve (right); (B) L. delbrueckii subsp. lactis amplification plot (left), melt curve (middle), and standard curve (right); (C) L. delbrueckii subsp. delbrueckii amplification plot (left), melt curve (middle), and standard curve (right).
Application of Real-Time PCR Method
Fifty-one isolates, 15 probiotic products, and 11 dairy products were used to perform the application test of the developed real-time PCR method. Results of its application to probiotic and dairy products were compared with their label claims. A total of 26 products were detected with the same subspecies as their label claims (Table 4). However, for 16 products, subspecies was incorrectly claimed on the label. According to the nomenclature of subspecies, these should be labeled as “L. delbrueckii subsp. bulgaricus,” not “L. bulgaricus” [12]. Dairy products labeled only with lactic acid bacteria were confirmed to contain L. delbrueckii subsp. bulgaricus by real-time PCR. As a result of the application of our method to different isolates, a total of 64 strains isolated from raw milk and dairy products were identified as L. delbrueckii subsp. lactis (n = 17) and bulgaricus (n = 47) (Table 4). These results confirmed that the real-time PCR developed in this study could accurately identify strains present in probiotic and dairy products and bacterial isolates to the subspecies level.
Table 4.
Application test of real-time PCR method to probiotic and dairy products.
| Products | Type | Label claims | Detected subspecies |
|---|---|---|---|
| Products monitoring | |||
| A1 | Probiotic product (powder, Korea) | L. delbrueckii subsp. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A2 | Probiotic product (powder, Korea) | L. delbrueckii subsp. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A3 | Probiotic product (capsules, Canada) | L. delbrueckii subsp. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A4 | Probiotic product (capsules, Canada) | L. delbrueckii subsp. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A5 | Probiotic product (powder, Korea) | L. delbrueckii subsp. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A6 | Probiotic product (capsules, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A7 | Probiotic product (capsules, USA) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A8 | Probiotic product (capsules, USA) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A9 | Probiotic product (powder, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A10 | Probiotic product (powder, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A11 | Probiotic product (capsules, Canada) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A12 | Probiotic product (powder, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A13 | Probiotic product (capsules, Canada) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A14 | Probiotic product (powder, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| A15 | Probiotic product (powder, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B1 | Dairy product (yogurt, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B2 | Dairy product (yogurt, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B3 | Dairy product (yogurt, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B4 | Dairy product (yogurt, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B5 | Dairy product (yogurt, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B6 | Dairy product (yogurt, Korea) | L. bulgaricus | L. delbrueckii subsp. bulgaricus |
| B7 | Dairy product (yogurt, Korea) | Lactic acid bacteria | L. delbrueckii subsp. bulgaricus |
| B8 | Dairy product (yogurt, Korea) | Lactic acid bacteria | L. delbrueckii subsp. bulgaricus |
| B9 | Dairy product (yogurt, Korea) | Lactic acid bacteria | L. delbrueckii subsp. bulgaricus |
| B10 | Dairy product (yogurt, Korea) | Lactic acid bacteria | L. delbrueckii subsp. bulgaricus |
| B11 | Dairy product (yogurt, Korea) | Lactic acid bacteria | L. delbrueckii subsp. bulgaricus |
| Identification of isolates | |||
| I1~I4 | Raw milk (cow's milk, Korea) | Unknown isolates | L. delbrueckii subsp. lactis |
| I5~I8 | Raw milk (cow's milk, Korea) | Unknown isolates | L. delbrueckii subsp. lactis |
| I9~I17 | Raw milk (cow's milk, Korea) | Unknown isolates | L. delbrueckii subsp. lactis |
| I18~I21 | Dairy product (powder, Korea) | Unknown isolates | L. delbrueckii subsp. bulgaricus |
| I22~I57 | Dairy product (yogurt, Korea) | Unknown isolates | L. delbrueckii subsp. bulgaricus |
| I58~I64 | Dairy product (yogurt, Korea) | Unknown isolates | L. delbrueckii subsp. bulgaricus |
Conclusion
In conclusion, pangenome analysis was performed to select genetic markers for six L. delbrueckii subspecies. These genetic markers were present in all genomes of the same subspecies but absent in genomes of other subspecies and bacterial strains. To rapidly and cost-effectively identify L. delbrueckii subspecies, subspecies-specific primer pairs for three subspecies mainly isolated from food samples were designed. The real-time PCR method using these genes could accurately identify L. delbrueckii subspecies and other lactic acid bacteria with high specificity. The developed real-time PCR method was able to successfully monitor probiotic and dairy products and identify various isolates. Thus, our method can be used to accurately identify L. delbrueckii subspecies and determine the nomenclature of these subspecies. Furthermore, it can contribute to safety in the food industry by ensuring products are labeled to show the correct strain.
Acknowledgments
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (grant no. 2020R1A6A3A01100168).
REFERENCES
- 1.El Kafsi H, Binesse J, Loux V, Buratti J, Boudebbouze S, Dervyn R, et al. Lactobacillus delbrueckii ssp. lactis and ssp. bulgaricus: A chronicle of evolution in action. BMC. Genomics. 2014;15:407. doi: 10.1186/1471-2164-15-407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zanni E, Schifano E, Motta S, Sciubba F, Palleschi C, Mauri P, et al. Combination of metabolomic and proteomic analysis revealed different features among Lactobacillus delbrueckii subspecies bulgaricus and lactis strains while in vivo testing in the model organism Caenorhabditis elegans highlighted probiotic properties. Front. Microbiol. 2017;8:1206. doi: 10.3389/fmicb.2017.01206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.van De Guchte M, Penaud S, Grimaldi C, Barbe V, Bryson K, Nicolas P, et al. The complete genome sequence of Lactobacillus bulgaricus reveals extensive and ongoing reductive evolution. Proc. Natl. Acad. Sci. USA. 2006;103:9274–9279. doi: 10.1073/pnas.0603024103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nikolaou A, Saxami G, Kourkoutas Y, Galanis A. A new methodology for rapid detection of Lactobacillus delbrueckii subsp. bulgaricus based on multiplex PCR. J. Microbiol. Methods. 2011;84:362–364. doi: 10.1016/j.mimet.2010.12.010. [DOI] [PubMed] [Google Scholar]
- 5.Weiss N, Schillinger U, Kandler O. Lactobacillus lactis, Lactobacillus leichmannii and Lactobacillus bulgaricus, subjective synonyms of Lactobacillus delbrueckii, and description of Lactobacillus delbrueckii subsp. lactis comb. nov. and Lactobacillus delbrueckii subsp. bulgaricus comb. nov. Syst. Appl. Microbiol. 1983;4:552–557. doi: 10.1016/S0723-2020(83)80012-5. [DOI] [PubMed] [Google Scholar]
- 6.Dellaglio F, Felis GE, Castioni A, Torriani S, Germond JE. Lactobacillus delbrueckii subsp. indicus subsp. nov., isolated from Indian dairy products. Int. J. Syst. Evol. Microbiol. 2005;55:401–404. doi: 10.1099/ijs.0.63067-0. [DOI] [PubMed] [Google Scholar]
- 7.Watanabe K. Lactobacillus delbrueckii subsp. sunkii subsp. nov., isolated from sunki, a traditional Japanese pickle. Int. J. Syst. Evol. Microbiol. 2012;62:2643–2649. doi: 10.1099/ijs.0.037051-0. [DOI] [PubMed] [Google Scholar]
- 8.Adimpong DB, Nielsen DS, Sørensen KI, Vogensen FK, Sawadogo-Lingani H, Derkx PMF, et al. Lactobacillus delbrueckii subsp. jakobsenii subsp. nov., isolated from dolo wort, an alcoholic fermented beverage in Burkina Faso. Int. J. Syst. Evol. Microbiol. 2013;63:3720–3726. doi: 10.1099/ijs.0.048769-0. [DOI] [PubMed] [Google Scholar]
- 9.Song Y, Sun Z, Guo C, Wu Y, Liu W, Yu J, et al. Genetic diversity and population structure of Lactobacillus delbrueckii subspecies bulgaricus isolated from naturally fermented dairy foods. Sci. Rep. 2016;6:22704. doi: 10.1038/srep22704. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Giraffa G, Lazzi C, Gatti M, Rossetti L, Mora D, Neviani E. Molecular typing of Lactobacillus delbrueckii of dairy origin by PCRRFLP of protein-coding genes. Int. J. Food Microbiol. 2003;82:163–172. doi: 10.1016/S0168-1605(02)00262-3. [DOI] [PubMed] [Google Scholar]
- 11.Cebeci A, Gürakan GC. Comparative typing of L. delbrueckii subsp. bulgaricus strains using multilocus sequence typing and RAPD-PCR. Eur. Food Res. Technol. 2011;233:377–385. doi: 10.1007/s00217-011-1526-5. [DOI] [Google Scholar]
- 12.Zheng J, Wittouck S, Salvetti E, Franz CMAP, Harris HMB, Mattarelli P, et al. A taxonomic note on the genus Lactobacillus: description of 23 novel genera, emended description of the genus Lactobacillus Beijerinck 1901, and union of Lactobacillaceae and Leuconostocaceae. Int. J. Syst. Evol. Microbiol. 2020;70:2782–2858. doi: 10.1099/ijsem.0.004107. [DOI] [PubMed] [Google Scholar]
- 13.Kim E, Yang SM, Lim B, Park SH, Rackerby B, Kim HY. Design of PCR assays to specifically detect and identify 37 Lactobacillus species in a single 96 well plate. BMC Microbiol. 2020;20:96. doi: 10.1186/s12866-020-01781-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ibrahim GM, Morin PM. Salmonella serotyping using whole genome sequencing. Front. Microbiol. 2018;9:2993. doi: 10.3389/fmicb.2018.02993. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Caputo A, Fournier PE, Raoult D. Genome and pan-genome analysis to classify emerging bacteria. Biol. Direct. 2019;14:5. doi: 10.1186/s13062-019-0234-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Caputo A, Merhej V, Georgiades K, Fournier PE, Croce O, Robert C, et al. Pan-genomic analysis to redefine species and subspecies based on quantum discontinuous variation: the Klebsiella paradigm. Biol. Direct. 2015;10:55. doi: 10.1186/s13062-015-0085-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Eren AM, Esen Ö, Quince C, Vineis J, Sogin M, Delmont T. Anvi'o: an advanced analysis and visualization platform for 'omics data. PeerJ. 2015;3:e1319. doi: 10.7717/peerj.1319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chaudhari NM, Gupta VK, Dutta C. BPGA- an ultra-fast pan-genome analysis pipeline. Sci. Rep. 2016;6:24373. doi: 10.1038/srep24373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26:2460–2461. doi: 10.1093/bioinformatics/btq461. [DOI] [PubMed] [Google Scholar]
- 20.Kim E, Yang SM, Cho EJ, Kim HY. Novel real-time PCR assay for Lactobacillus casei group species using comparative genomics. Food Microbiol. 2020;90:103485. doi: 10.1016/j.fm.2020.103485. [DOI] [PubMed] [Google Scholar]
- 21.Wuyts S, Wittouck S, De Boeck I, Allonsius CN, Pasolli E, Segata N, et al. Large-scale phylogenomics of the Lactobacillus casei group highlights taxonomic inconsistencies and reveals novel clade-associated features. mSystems. 2017;2:e00061–17. doi: 10.1128/mSystems.00061-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Beaz-Hidalgo R, Hossain MJ, Liles MR, Figueras MJ. Strategies to avoid wrongly labelled genomes using as example the detected wrong taxonomic affiliation for Aeromonas genomes in the genbank database. PLoS One. 2015;10:e0115813. doi: 10.1371/journal.pone.0115813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tran PN, Savka MA, Gan HM. In-silico taxonomic classification of 373 genomes reveals species misidentification and new genospecies within the genus Pseudomonas. Front. Microbiol. 2017;8:1296. doi: 10.3389/fmicb.2017.01296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Laing CR, Whiteside MD, Gannon VPJ. Pan-genome analyses of the species Salmonella enterica, and identification of genomic markers predictive for species, subspecies, and serovar. Front. Microbiol. 2017;8:1345. doi: 10.3389/fmicb.2017.01345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Velsko IM, Perez MS, Richards VP. Resolving phylogenetic relationships for Streptococcus mitis and Streptococcus oralis through core- and pan-genome analyses. Genome Biol. Evol. 2019;11:1077–1087. doi: 10.1093/gbe/evz049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lu W, Kong W, Yang P, Kong J. A one-step PCR-based method for specific identification of 10 common lactic acid bacteria and Bifidobacterium in fermented milk. Int. Dairy J. 2015;41:7–12. doi: 10.1016/j.idairyj.2014.08.020. [DOI] [Google Scholar]
- 27.Stachelska MA. Identification of Lactobacillus delbrueckii and Streptococcus thermophilus strains present in artisanal raw cow milk cheese using real-Time PCR and classic plate count methods. Polish. J. Microbiol. 2017;66:491–499. doi: 10.5604/01.3001.0010.7041. [DOI] [PubMed] [Google Scholar]
- 28.Kim E, Kim HJ, Yang SM, Kim CG, Choo DW, Kim HY. Rapid identification of Staphylococcus species isolated from food samples by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. J. Microbiol. Biotechnol. 2019;29:548–557. doi: 10.4014/jmb.1901.01046. [DOI] [PubMed] [Google Scholar]
- 29.Ha M, Jo HJ, Choi EK, Kim Y, Kim J, Cho HJ. Reliable identification of Bacillus cereus group species using low mass biomarkers by MALDI-TOF MS. J. Microbiol. Biotechnol. 2019;29:887–896. doi: 10.4014/jmb.1903.03033. [DOI] [PubMed] [Google Scholar]
- 30.Huang CH, Huang L. Rapid species- and subspecies-specific level classification and identification of Lactobacillus casei group members using MALDI Biotyper combined with ClinProTools. J. Dairy Sci. 2018;101:979–991. doi: 10.3168/jds.2017-13642. [DOI] [PubMed] [Google Scholar]
- 31.Broeders S, Huber I, Grohmann L, Berben G, Taverniers I, Mazzara M, et al. Guidelines for validation of qualitative real-time PCR methods. Trends Food Sci. Technol. 2014;37:115–126. doi: 10.1016/j.tifs.2014.03.008. [DOI] [Google Scholar]