Development of a rapid comQ-targeted quantitative polymerase chain reaction assay for specific identification and quantification of Bacillus subtilis subsp. natto

Mayu Takizawa; Daisuke Kyoui; Rika Horiguchi; Ayano Sakurai; Taketo Kawarai; Naoki Narisawa

doi:10.1371/journal.pone.0355394

Abstract

Bacillus subtilis subsp. natto is a specialized functional group essential for fermented soybean production. Despite its industrial importance, genetic differentiation of this lineage remains challenging because of its high genomic similarity within the B. subtilis species complex. Here, we developed and validated a high-precision quantitative real-time polymerase chain reaction (qPCR) assay targeting the comQ gene—a core component of the quorum-sensing system. Comparative genomic analysis of 70 bacterial genomes, including 9 B. subtilis strains newly sequenced in this study, revealed that the comQ gene is under intense purifying selection (dN/dS = 0.10) and exhibits near-absolute conservation within the natto lineage (42 strains). While screening of these 70 genomes revealed a total of 15 atypical isolates (such as Miz-8 and PN1236) with reduced sequence identity (≤99%) or deletions in traditional metabolic (bio and fliF) and insertion sequence (IS) markers, their comQ loci remained 100% structurally intact. Phylogenetic analysis further confirmed that comQ forms a distinct monophyletic cluster for subsp. natto, providing superior diagnostic reliability over conventional markers, such as bioF or IS elements. By leveraging this highly conserved target, the established qPCR platform achieved an exceptionally wide, linear quantification framework (R² = 0.9973, E = 94.2%) that completely bypasses peripheral genomic instability, maintaining high fidelity even under a 10,000-fold excess of non-target DNA. Direct validation on commercial fermented soybean matrices demonstrated a seamless correlation between molecular comQ copy number and viable cell count, underscoring its resilience against complex sample-derived enzymatic inhibition without the need for culture expansion. This assay transcends application in conventional industrial quality control—by providing a useful diagnostic and quantitative proxy, it provides a practical tool for tracking potentially opportunistic natto lineages linked to severe clinical bacteremia. Overall, we have established a reliable molecular surveillance method applicable to both industrial biotechnology and high-resolution clinical epidemiology.

Citation: Takizawa M, Kyoui D, Horiguchi R, Sakurai A, Kawarai T, Narisawa N (2026) Development of a rapid comQ-targeted quantitative polymerase chain reaction assay for specific identification and quantification of Bacillus subtilis subsp. natto. PLoS One 21(8): e0355394. https://doi.org/10.1371/journal.pone.0355394

Editor: Ajaya Bhattarai, Tribhuvan University, NEPAL

Received: February 20, 2026; Accepted: July 21, 2026; Published: August 3, 2026

Copyright: © 2026 Takizawa et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript, its Supporting Information files, and the public repositories mentioned in the text.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Bacillus subtilis subsp. natto is an industrially and clinically relevant lineage within the B. subtilis species complex. Beyond its essential role in traditional soybean fermentation, this subspecies serves as a versatile biofactory for diverse bioactive compounds, including nattokinase, poly-γ-glutamic acid (γ-PGA), and vitamin K2 (MK-7), which offer multifaceted health benefits [1–4]. Recent studies have highlighted its broader functional repertoire, including antioxidant activities, angiotensin-converting enzyme-inhibitory effects for blood pressure regulation, and resilient probiotic properties that modulate the gastrointestinal health. Moreover, metabolites of this subspecies exhibit broad-spectrum antimicrobial [5], antibiofilm [6], and antiviral activities [7]. Although predominantly recognized for these industrial and health-promoting benefits, recent clinical data have also noted rare instances of opportunistic bacteremia associated with specific strains in clinical settings [8], underscoring the relevance of precise strain-level monitoring. Consequently, there is a growing need for a high-resolution tool capable of rapidly distinguishing these lineages to ensure both the quality and comprehensive safety in functional food applications.

Genetically, subsp. natto is nearly indistinguishable from other members of the species complex, such as the model strain B. subtilis subsp. subtilis 168, and cannot be resolved using standard phylogenetic markers like 16S rRNA, gyrA, or recQ [1,9–12]. Although multilocus sequence typing (MLST) provides higher resolution, it remains too labor-intensive and cost-inefficient for large-scale screening. Conventionally, molecular identification has relied on specific genetic signatures, including the biotin biosynthetic operon (bioF and bioW), flagellar structural genes (fliF), and various insertion sequence (IS) elements [9,13–15]. However, these approaches encounter significant technical and quantitative limitations: bioF and fliF require precise discrimination of small fragments; the single-nucleotide nonsense mutation in bioW cannot be differentiated via conventional PCR size-fractionation; and IS elements exhibit highly unstable copy number. Moreover, because these conventional markers are related to non-essential peripheral pathways, they are inherently prone to sporadic genomic rearrangements or deletions in atypical wild isolates, presenting a critical risk of false-negative diagnostic results [16].

To achieve rapid, one-step identification, we sought to exploit the unique evolutionary signatures embedded within the comQXPA quorum-sensing (QS) operon, which coordinates strain-specific social communication [15,17–22]. This locus encodes the signaling machinery consisting of the modifier ComQ, signaling peptide ComX, and sensor ComP. Because the comQXP module evolves as a tightly coupled functional unit to ensure signal-receptor fidelity, it undergoes intense purifying selection within the natto lineage, maintaining near-perfect sequence conservation to preserve QS integrity. Conversely, this co-evolutionary constraint drives extreme interlineage divergence from other subsp. subtilis populations that utilize different pherotypes. For instance, the ComQ protein in the natto lineage exhibits low amino acid sequence identity with that of the protein in the laboratory model strain 168, reflecting extreme interlineage variation driven by pherotype diversification [21]. This highly stable genetic signature—strong intralineage conservation coupled with profound interlineage variation—makes the comQ gene the most appropriate diagnostic window to establish a highly specific and reliable qPCR platform that is inherently more robust than the conventional markers.

In this study, we aimed to validate the potential of the comQXPA locus as a lineage-specific marker and to develop a robust diagnostic tool for B. subtilis subsp. natto. We performed a large-scale comparative genomic analysis of 70 bacterial genomes, including 9 B. subtilis strains newly sequenced in this study, to rigorously assess the intralineage conservation and inter-subspecies divergence of the operon. Leveraging these genomic insights, we targeted the comQ gene to design highly specific primers and established a quantitative real-time PCR (qPCR) assay. The specificity, sensitivity, and matrix tolerance of the assay were rigorously evaluated using a panel of diverse Bacillus species and various commercial natto products, providing a rapid, accurate, and high-throughput solution that effectively overcomes the technical limitations of the conventional molecular markers.

Materials and methods

Bacterial strains and culture conditions

A total of 37 strains were used to evaluate the specificity and sensitivity of the comQ-targeted assay (Table 1). They included 18 Bacillus subtilis subsp. natto strains, comprising commercial industrial starters (Miyagino and Naruse), type strains, three honey isolates (HK1−1, HK3−1, and HK11−1) previously characterized by Kimijima et al. [13], and five additional honey isolates (MH3−1, MH6−2, MH18−2, MH32−1, and MH33−2). For specificity testing, 19 non-target strains were included, focusing primarily on members of the closely related B. subtilis species complex. All strains were cultured overnight in Brain Heart Infusion (BHI) broth (Becton, Dickinson and Co., Franklin Lakes, NJ, USA) or Luria–Bertani (LB) broth at 37 °C, with shaking at 160 rpm, depending on their specific growth requirements.

Download:

Table 1. PCR-based identification of Bacillus subtilis subsp. natto and related Bacillus strains using various genetic markers.

https://doi.org/10.1371/journal.pone.0355394.t001

DNA extraction and quality control

Genomic DNA was extracted using the ISOPLANT kit (Nippon Gene Co., Ltd., Tokyo, Japan) according to the manufacturer’s instructions. DNA concentration and purity (A₂₆₀/A₂₈₀ ratio) were initially assessed using a BioSpec-nano spectrophotometer (Shimadzu Corp., Kyoto, Japan) for preliminary screening and conventional PCR assays. For qPCR applications, DNA concentrations were determined with higher precision using a Qubit 4 Fluorometer with the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA). This dual-quantification strategy ensured both high purity—required for specificity assays—and stringent precision—necessary for the construction of standard curves and sensitivity analysis.

Genome sequencing and assembly

Shotgun genome sequencing libraries were constructed for the nine B. subtilis strains newly sequenced in this study using the MGIEasy Fast FS Library Prep Set V2.0 (MGI Tech Co., Ltd., Shenzhen, China) and sequenced on a DNBSEQ-T7 platform as 150-bp paired-end reads. Raw reads were quality-filtered and adapter-trimmed using fastp (version 1.0.1) with default parameters, including a qualified quality Phred score threshold of 15 and an unfiltered base limit of 40% [23]. The quality-filtered reads were assembled de novo using SPAdes assembler (version 4.2.0) with the careful option enabled to minimize mismatches and short indels, utilizing a multi-k-mer strategy with k-mer lengths of 21, 33, 55, and 77 [24]. The draft genome sequences were annotated by DFAST [25]. The draft genome sequences were deposited in the EMBL/GenBank/DDBJ database under BioProject ID PRJDB398070. The detailed genomic metrics and accession numbers for each strain are listed in S1 Table.

Genome-wide similarity assessment and phylogenetic analysis of the comQXPA locus

To evaluate the genome-wide similarity across the species complex, average nucleotide identity (ANI) was calculated for all 70 bacterial genomes—encompassing the 9 B. subtilis strains newly sequenced in this study, 49 other B. subtilis group strains, and 12 closely related Bacillus species—using the FastANI tool within the Galaxy platform (https://usegalaxy.org/) with default parameters, employing the representative subsp. natto strain BEST195 as the reference genome. The resulting pairwise ANI matrix was subjected to hierarchical clustering analysis, and the genome-wide phylogenetic stratification was visualized as a clustered heatmap (Fig 1) using Python with the scipy.cluster.hierarchy and seaborn libraries.

Download:

Fig 1. Genome-wide average nucleotide identity (ANI) heatmap and dendrogram analysis of the Bacillus subtilis group and closely related Bacillus species.

The dendrograms on the top and left sides illustrate the hierarchical phylogenetic clustering based on genomic distances, driving the matrix organization. The color scale indicates pairwise ANI percentages ranging from 75% (blue) to 100% (red) across 70 bacterial genomes. Strains exhibiting 100% comQ sequence identity with the reference strain BEST195 are highlighted in dark green text (38 strains), whereas those exhibiting 98% or higher identity are indicated in light green text (3 strains). The reference strain BEST195 itself is indicated in red text with yellow shading.

https://doi.org/10.1371/journal.pone.0355394.g001

Four distinct phylogenetic trees were constructed independently, each based on the amino acid sequence of an individual component of the comQXPA locus (ComQ, ComX, ComP, and ComA, respectively) from subsp. natto and its closely related species. In particular, these separate alignments were used to evaluate the phylogenetic relationships between subsp. natto BEST195 and other members of the B. subtilis species complex. Multiple sequence alignment was performed using MAFFT (version 7) with default parameters. Phylogenetic reconstruction was subsequently conducted in MEGA12 (version 12) [26] using the neighbor-joining (NJ) method based on the Poisson correction model. The reliability of the tree topology was evaluated via bootstrap analysis with 1,000 replicates. Accession numbers for all analyzed sequences are provided in S1 Table.

Primer design

Specific primers were designed to target the comQ gene of B. subtilis subsp. natto BEST195. Two distinct primer sets were optimized: comQ-F241/R831 for general lineage validation via conventional PCR, and comQ-qF/qR for lineage-specific qPCR assays (S2 Table). To verify the primer binding sites and evaluate amplification specificity in silico, nucleotide sequence alignments were performed using MEGA12, encompassing the comprehensive dataset of 70 bacterial genomes assembled in this study. The primer design was optimized to ensure full complementarity across the target subsp. natto cohort while avoiding cross-reactivity with non-target reference sequences, such as B. subtilis subsp. subtilis 168 and B. subtilis subsp. spizizenii NBRC 101239.

Conventional PCR amplification and specificity testing

The 16S rRNA gene, comQ, bioF, and IS elements (IS4Bsu1 and IS256Bsu1) were amplified using PCR. Detailed information about the primers used for the PCR is provided in S2 Table. As described above, the primer set comQ-F241/R831 was utilized for the direct detection and validation of the comQ gene. To evaluate conventional markers, we used the bioF- and IS-specific primers described by Kimijima et al. [13]. The 16S rRNA gene served as an internal control, with universal primers 27F/1512R used for general specificity assays and 341F/907R for sensitivity assays. Reactions were prepared in a 20 µL total volume using EmeraldAmp MAX PCR Master Mix (Takara Bio Inc., Kusatsu, Japan). The thermal cycling profile consisted of 30 cycles of denaturation at 98 °C for 10 s, annealing at 60 °C for 30 s, and extension at 72 °C for 1 min. Each assay was performed in triplicate to ensure reproducibility. PCR products were analyzed via electrophoresis on a 1.5% (w/v) agarose gel and visualized after ethidium bromide staining under UV illumination. Gel images were captured and converted to grayscale using Adobe Photoshop 2020.

qPCR and construction of a universal standard curve

qPCR was performed using the StepOne Real-Time PCR System (Thermo Fisher Scientific). Each 10 µL reaction mixture consisted of 5 µL of TB Green Premix Ex Taq II (Tli RNaseH Plus) (Takara Bio), 0.4 µM of each primer (comQ-qF/qR), and 1 µL of template DNA. To evaluate the selectivity of the assay against non-target lineages, 10 ng of genomic DNA from B. subtilis subsp. subtilis 168 or B. subtilis subsp. spizizenii NBRC 101239 was used per reaction. The fast-cycling protocol was as follows: initial denaturation at 95 °C for 30 s, followed by 40 cycles of denaturation at 95 °C for 5 s and annealing/extension at 60 °C for 30 s. A melting curve analysis (from 60 °C to 95 °C) was conducted to confirm amplicon specificity.

DNA extraction from commercial natto matrices and viable cell enumeration

For real-time validation of the assay, three distinct brands of commercial natto (fermented soybean) products were purchased from a local supermarket. For each sample, 10 grains of natto were suspended in 5 mL of phosphate-buffered saline (PBS) to prepare the crude matrix extracts. Total genomic DNA was extracted from the resulting suspension using ISOPLANT as described previously. The isolated DNA samples were subjected to 10-fold serial dilutions (1⁰⁰ to 10⁻³) and used as templates for qPCR evaluation. Concurrently, for enumerating viable cells, crude natto suspensions were serially diluted in PBS, plated onto BHI agar in triplicate, and incubated overnight at 37 °C. The resulting colonies were counted, and the viable cell counts (log₁₀ CFU/mL) were expressed as the mean ± standard deviation (SD).

Preparation of standard plasmid DNA

To construct the standard plasmid framework for absolute quantification, the comQ gene fragment was amplified from B. subtilis subsp. natto NBRC 16449 genomic DNA using the specific primer set comQ-F241/R831. The resulting PCR amplicon was cloned into the pMD20-T vector (Takara Bio) via TA cloning according to the manufacturer’s protocol. The concentration of the purified recombinant plasmid DNA was determined spectrophotometrically. The corresponding molecular copy number was subsequently calculated based on the total size of the plasmid construct (vector plus insert) using the following formula:

This plasmid stock was used to generate a 10-fold serial dilution series, ranging from 2.86 × 10³ to 2.86 × 10⁶ copies per reaction, to construct the absolute quantification standard curve (R² = 0.9973; efficiency = 94.2%). To determine the absolute comQ copy numbers in the natto extracts, the experimental quantification cycle (Cq) values obtained from the samples were plugged into the linear regression equation derived from this plasmid standard curve (Cq = slope × log₁₀(copy number) + intercept), thereby converting the raw Cq metrics into absolute log copies per reaction tube.

Data visualization and statistical analysis

Based on the phylogenetic relationships, established as described in section “Genome-wide similarity assessment and phylogenetic analysis of the comQXPA locus,” we statistically evaluated the evolutionary congruence within the comQXPA locus. Mantel tests were performed using the skbio.stats.distance library to calculate correlations between the amino acid distance matrices of the four proteins (ComQ, ComX, ComP, and ComA). The statistical significance of the correlation coefficient (r) was determined using a permutation test with 1,000 random shufflings.

To evaluate the evolutionary selective pressure acting on the quorum-sensing components, the rates of synonymous (dS) and nonsynonymous (dN) substitutions were calculated using MEGA12. This analysis encompassed 42 strains, consisting of the reference strain BEST195 and 41 strains exhibiting 98.0% or higher comQ sequence identity with BEST195. The dN/dS ratios were determined independently for each of the comQ, comX, and comP genes using the Nei–Gojobori method with the Jukes–Cantor correction. Furthermore, a codon-based Z-test for purifying selection (alternative hypothesis: dN < dS) was conducted in MEGA12 to establish the statistical significance of the selective constraints, and the corresponding Z-statistics and p-values were determined for each signaling component.

For qPCR validation, linear regression was conducted using scipy.stats [27] and scikit-learn [28] to determine the slope, y-intercept, and R². Amplification efficiency (E) was calculated using the formula E = 10^(−1/slope) − 1.

The robustness of the assay against non-target DNA interference was evaluated via a two-way analysis of variance (two-way ANOVA) using the statsmodels library [29]. This model evaluated the statistical significance of the main effects (log-transformed DNA concentrations and background groups) as well as their interaction effect on the Cq values. Data visualization was performed using matplotlib [30] and seaborn [31]. Cq values are presented as mean ± SD from three independent replicates, except for the data of strain NBRC 16449 in S3 Fig, which are based on five independent replicates. A p-value <0.05 was considered statistically significant.

Results

High genomic similarity within the B. subtilis group

To establish a robust genomic foundation, we conducted a genome-wide ANI analysis encompassing 70 bacterial genomes: 9 B. subtilis strains newly sequenced in this study, 49 other B. subtilis group strains, and 12 closely related Bacillus species. Strains within the B. subtilis group, including both subsp. natto and subsp. subtilis, exhibited extremely high genomic similarity, with ANI values exceeding 98% relative to the representative natto strain BEST195 (Fig 1). This confirmed a high degree of genomic uniformity, consistent with their classification within the B. subtilis species complex.

Phylogenetic stratification and co-evolutionary analysis of the comQXPA locus

In contrast to the overall genomic homogeneity, the comQXPA locus displayed profound sequence divergence. Bioinformatic analysis revealed a distinct stratification of the B. subtilis group based on the amino acid sequence identity of each signaling component (S1 Table). Beyond the reference strain BEST195, a primary cluster of 38 strains exhibited 100% identity for the QS activator ComQ, and three additional strains shared over 98.0% identity. Conversely, ComQ homology with subsp. subtilis strains (e.g., strain 168) dropped to approximately 35.5%. Similarly, the signaling peptide ComX and the sensor kinase ComP maintained high conservation within these 42 natto-lineage strains (100.0% and 99.0% or greater identity, respectively) but the homology dropped to 16.0% and 39.8% relative to strain 168. In sharp contrast, the response regulator ComA remained highly conserved across all subspecies (S1 Table).

This stratification was visually confirmed using neighbor-joining phylogenetic trees, where ComQ, ComX, and ComP from strain BEST195 formed a distinct monophyletic cluster that was well-separated from other subspecies (Fig 2A-C), whereas ComA showed no lineage-specific clustering (Fig 2D).

Download:

Fig 2. Phylogenetic analysis of the comQXPA quorum-sensing locus components.

Neighbor-joining phylogenetic trees were constructed based on the amino acid sequences of (A) ComQ, (B) ComX, (C) ComP, and (D) ComA from various Bacillus subtilis groups. The evolutionary history was inferred using the Neighbor-Joining method based on the Poisson correction model, and the reliability of the tree topology was evaluated via bootstrap analysis with 1,000 replicates. The subsp. natto BEST195 is highlighted in red to represent the target functional lineage. Numbers in parentheses indicate NCBI accession numbers. Scale bars represent the number of amino acid substitutions per site.

https://doi.org/10.1371/journal.pone.0355394.g002

To evaluate whether these components function as an integrated system, we assessed their evolutionary coupling. Mantel tests confirmed strong and statistically significant correlations among ComQ, ComX, and ComP (Mantel r = 0.526–0.751, p < 0.01; Fig 3), indicating tight co-evolution as a functional unit to maintain signaling specificity. In contrast, ComA showed no evolutionary correlation with the other components (Fig 3), confirming its role as a universal, independent regulator.

Download:

Fig 3. Correlation heatmap of the Mantel test among the comQXPA quorum-sensing locus components.

The heatmap displays Mantel correlation coefficients (r) derived from the amino acid sequence distance matrices of ComQ, ComX, ComP, and ComA. The color gradient (blues) represents the strength of the correlation, with r values ranging from 0.0 to 1.0. Statistical significance was determined based on p-values calculated using the permutation tests (**, p < 0.01; *, p < 0.05; and n.s., not significant).

https://doi.org/10.1371/journal.pone.0355394.g003

Co-occurrence of comQ with known natto-specific genetic markers

We next evaluated the diagnostic distribution of comQ against established genetic signatures of the natto lineage (S1 Table). Among the 38 core strains (100% comQ identity), we observed a high rate of co-occurrence, with the majority harboring BEST195-type sequences for bioF, bioW, fliF, and IS elements. However, this co-occurrence was incomplete in 15 atypical isolates (e.g., Miz-8 and PN1236), which exhibited discrepancies or deletions in one or more conventional markers (S1 Table). Remarkably, even in these variants where traditional markers were absent or diverged, both comQ and comX remained 100% conserved, indicating that comQ provides an evolutionary robustness that eliminates the risk of false-negative identification.

To investigate the evolutionary drivers underlying this conservation, we calculated the dN/dS ratio across the 42 natto-lineage strains (Table 2). The comQ gene exhibited a significantly low dN/dS ratio (0.10) and a significant negative Z-statistic (−1.68, p < 0.05). This indicates that comQ is under strong purifying selection. While comX remained sequence-invariant and comP reflected neutral evolution (dN/dS = 0.36), the intense evolutionary constraint on comQ suppresses non-synonymous mutations to preserve quorum-sensing integrity. Consequently, comQ serves as an exceptionally stable genomic anchor, which renders it the highly reliable molecular target for diagnostic assay development.

Download:

Table 2. Evolutionary constraints on the comQXP quorum-sensing locus in subsp. natto.

https://doi.org/10.1371/journal.pone.0355394.t002

Design and in silico PCR validation of comQ-targeted primers

We designed two primer sets targeting the conserved regions of comQ in strain BEST195: comQ-F241/R831 for conventional PCR and comQ-qF/qR for qPCR. Alignment analysis revealed multiple nucleotide substitutions in non-natto subspecies (subsp. subtilis 168 and subsp. spizizenii NBRC 101239), particularly at the 3′ terminus, which is critical for polymerase extension (S1 Fig). In silico PCR simulations achieved 100% accuracy across the 70 genomes (S3 Table); all 42 natto-lineage strains were predicted to yield positive amplicons due to identical primer-binding sites, whereas all non-natto strains were predicted to be negative in silico.

Specificity, inclusivity, and sensitivity of the comQ-targeted PCR assay

Laboratory experiments using 32 bacterial strains empirically validated the bioinformatic predictions (Table 1, Fig 4A). The comQ-F241/R831 primer set demonstrated 100% inclusivity, yielding positive results for all 18 subsp. natto strains (Table 1), with specific amplification visually confirmed via agarose gel electrophoresis for 13 representative strains (S2 Fig). In contrast, conventional bioF and IS-element markers caused cross-reactivity or non-specific bands in several non-target strains, including subsp. subtilis 168 and NBRC 14140 (Fig 4A). The comQ assay showed complete exclusivity (no detection) across all 19 non-target species, providing a clear presence/absence result. The conventional PCR assay also showed a reliable limit of detection (LOD) of 0.01 ng of template DNA per reaction (approximately 2.2 × 10³ genomic copies; Fig 4B).

Download:

Fig 4. Specificity and sensitivity of the PCR assays using multiple primer sets.

(A) Comparison of specificity between the comQ and conventional genetic markers. PCR amplification was performed targeting the 16S rRNA gene, comQ, IS4Bsu1, IS256Bsu1, and bioF. The tested strains included subsp. natto Miyagino (Mi), Bacillus subtilis subsp. subtilis 168, and B. subtilis subsp. subtilis NBRC 14140. Lane M: Gene-Ladder 100 (Nippon Gene Co., Ltd., Tokyo, Japan); black triangles indicate 2,000 and 500 bp. (B) Detection sensitivity of the comQ PCR assay. A 10-fold serial dilution of genomic DNA from subsp. natto Miyagino (1–0.001 ng per reaction) was used as a template. The 16S rRNA gene was used as a control. Gel images were converted to grayscale and adjusted for contrast using Adobe Photoshop 2020. All lanes in (B) originated from the same gel. The images shown are representative results from three independent experiments.

https://doi.org/10.1371/journal.pone.0355394.g004

Quantitative performance, specificity, and robustness of the qPCR assay

The comQ-qF/qR qPCR assay demonstrated high specificity and quantitative performance. When evaluated against the closest phylogenetic neighbors, the target strain NBRC 16449 yielded a mean Cq of 19.53 (S3A Fig), whereas non-target strains 168 and NBRC 101239 exhibited delayed Cq values of 33.18 and 32.41, respectively (S4 Table). This ΔCq of approximately 13 cycles represents a > 8,000-fold discrimination efficiency against non-target orthologs. Melting curve analysis confirmed a single, sharp target dissociation peak (Tm = 74.95 °C), distinctly separated from non-target profiles (S3 Fig, S4 Table).

To establish the universal applicability and analytical sensitivity of the comQ-targeted qPCR assay, we constructed standard curves using genomic DNA from three representative strains, Miyagino, Naruse, and NBRC 16449. The assay employed a 10-fold serial dilution series ranging from 1.0 ng to 0.0001 ng of template DNA per reaction.

The assay exhibited excellent linearity and high reproducibility across four orders of magnitude (1.0–0.001 ng) for all three strains (Fig 5). The composite standard curve showed a high coefficient of determination (R² = 0.994) and an optimal amplification efficiency (E) of 101.3%. Within this linear range, Cq values were nearly identical across the three strains at each concentration (S4 Table). This minimal interstrain variance confirms that the comQ marker provides consistent quantitative results regardless of the specific subsp. natto strain analyzed.

Download:

Fig 5. Standard curve and melting curve analysis of the comQ qPCR assay.

(A) Universal standard curve generated from three Bacillus subtilis subsp. natto strains. The quantification cycle (Cq) values are plotted against the log–log-transformed genomic DNA concentration (ranging from 1.0 ng to 0.0001 ng per reaction) for subsp. natto Miyagino, Naruse, and NBRC 16449. The dashed line represents the linear regression calculated from the combined data from all three strains. (B) Melting curve profiles for the sensitivity assay. The derivative melting curve plot (-dF/dT vs. temperature) shows the Tm profiles of the amplicons generated from the three natto strains at varying template concentrations. All data represent the mean values from three independent replicates. Error bars in (A) indicate the standard deviation (SD).

https://doi.org/10.1371/journal.pone.0355394.g005

To test matrix robustness, genomic DNA from strain Miyagino (1.0–0.001 ng) was quantified in the presence of a 10-fold excess (10 ng) of non-target strain 168 background DNA (Fig 6). At the lowest detection limit (0.001 ng), this background represents a 10,000-fold excess of non-target genetic material. Nonspecific background amplification remained negligible (Cq = 32.76 ± 1.32). For instance, 1.0 ng of target DNA yielded a Cq of 20.15 ± 0.40 with dH₂O and 19.68 ± 0.54 with the 168 background DNA. Two-way ANOVA confirmed no significant background-induced differences in Cq values (Table 3). The regression slope was stable at −3.44, corresponding to an amplification efficiency (E) of 95.3%, with a consistent Tm peak (74.8–74.9 °C).

Download:

Table 3. Two-way ANOVA for the effects of subsp. natto Miyagino DNA concentration and Bacillus subtilis 168 DNA background on quantification cycle (Cq) values.

https://doi.org/10.1371/journal.pone.0355394.t003

Download:

Fig 6. Robustness of the comQ-targeted qPCR assay against background genomic DNA.

(A) Standard curves for Bacillus subtilis subsp. natto Miyagino genomic DNA (0.001–1.0 ng per reaction) in the absence (blue circles) and presence (red square) of 10 ng of genomic DNA from the nontarget strain B. subtilis subsp. subtilis 168. Lines represent the linear regression models for each condition. (B) Melting curve analysis of the qPCR products obtained from the samples in (A). For (A), each data point represents the mean ± standard deviation for triplicate reactions (n = 3).

https://doi.org/10.1371/journal.pone.0355394.g006

Absolute quantification of comQ in commercial natto extracts

A standard curve for recombinant plasmid carrying the comQ fragment demonstrated high linearity (R² = 0.9973) and efficiency (E = 94.2%, S5 Fig) of the assay. This assay was applied to total DNA extracted from three commercial natto brands (Table 4) with viable cell counts ranging from 8.73 to 8.82 log₁₀ CFU/mL. Across a 10-fold serial dilution series (1⁰⁰ to 10⁻³), mean Cq values increased in a predictable, dilution-dependent manner. Stable mean Tm values (74.51–75.10 °C) confirmed the absence of matrix-derived inhibition. At 1⁰⁰ dilution, the calculated copy number for comQ ranged from 6.46 to 6.63 log₁₀ copies/tube and decreased systematically with each dilution step. These metrics validate the utility of this plasmid-standardized qPCR assay for precise, culture-independent monitoring of the natto lineage within complex commercial matrices.

Download:

Table 4. Quantitative performance and matrix tolerance of the comQ-targeted qPCR assay in commercial natto extracts.

https://doi.org/10.1371/journal.pone.0355394.t004

Discussion

In this study, we developed and validated a qPCR assay targeting the comQ gene for the rapid, culture-independent, and specific identification of B. subtilis subsp. natto. Although tracking this lineage is essential to the food biotechnology industry, universal markers often lack sufficient resolution. This limitation stems from the high genomic similarity within the B. subtilis species complex [1,9,10], which frequently compromises standard housekeeping or phylogenetic loci [11,12]. By leveraging an expanded comparative dataset of 70 genomes and absolute quantification metrics, we demonstrate the robustness of comQ as a genomic anchor in resolving subspecies-level stratification despite genome-wide homogeneity.

Our analysis revealed a remarkable conservation of the comQXPA locus across diverse ecological origins. Notably, eight analyzed subsp. natto genomes originated from honey [13], yet their comQXPA sequences exhibited 100% identity with those isolated from fermented soybeans. This high stability suggests that the comQ-mediated QS system is a core, nonredundant component of the natto genetic identity, which has remained evolutionarily static across disparate isolation sources, ensuring reliable detection without false negatives.

While genome-wide ANI analysis confirmed high uniformity (>98%) across the B. subtilis complex, the comQXPA operon stands out as a hotspot of diversification [20,21], where comQ determines “pherotype” specificity [20,32]. Multi-locus dN/dS analysis yielded a remarkably low ratio (0.10) and a significant negative Z-statistic (−1.68, Table 2), providing empirical evidence of intense purifying selection. This evolutionary constraint is likely driven by the ecological requirement to maintain a functional pherotype system that coordinates cell density-dependent traits—such as γ-PGA synthesis, nattokinase production, extracellular enzyme secretion, and biofilm formation—that collectively define the natto phenotype [1,32,33]. Furthermore, Mantel tests confirmed strong correlations among ComQ, ComX, and ComP (Mantel r = 0.526–0.751, p < 0.01; Fig 3), indicating tight co-evolution to preserve signaling fidelity. Thus, comQ copy number serves as a functional proxy for the fermentation potential of the population within a given matrix [18].

To establish a genetic framework for quantifying subsp. natto, traditional loci, such as the bio operon (bioF and bioW) and fliF, are common candidate targets [13–15]. However, genomic screening of 42 strains with >98.0% comQ identity revealed the failure of 15 strains to meet the 99.0% identity threshold for at least one traditional locus. Cross-referencing these with public pangenomic resources showed that 10 belonged to the primary “natto-related group,” whereas strain PRO120 was positioned outside this cluster based on macro-genomic distance metrics [16]. This demonstrates that the comQ assay provides a complementary functional perspective, effectively capturing such strains that, despite their macro-genomic divergence, retain the essential QS architecture for fermentation.

In contrast to the robustness of the comQ locus, relying on bio or fliF introduces technical risks of false negatives or cell-equivalent underestimation due to localized truncations [16]. Similarly, alternative IS-based markers, such as IS4Bsu1 and IS256Bsu1, lack subspecies exclusivity and vary inconsistently among strains [15]. Notably, these mobile elements can induce genetic instability in key industrial traits [34,35], further highlighting the limitations of conventional region-specific targeting.

Despite the phylogenetic proximity and shared pherotype trends between subsp. natto and Bacillus spizizenii [20,21], our platform clearly differentiates the natto clade from B. spizizenii NBRC 101239, its closest taxonomic neighbor [36,37]. By targeting sublineage-specific variations, our assay achieved a significant ΔCq of approximately 13 cycles against strain NBRC 101239 (S3A Fig). Traditional frameworks require precise fragment size discrimination (bioF) [14,15] or rely on volatile elements (IS) [1,15]. In contrast, comQ offers a highly specific detection window rooted in profound sequence divergence [1, 38], ensuring comprehensive coverage and high quantification fidelity even from its closest taxonomic relatives.

Compared with conventional PCR, the qPCR assay developed in this study demonstrated enhanced analytical capability. While the conventional method exhibited a gel-based LOD of 0.01 ng (Fig 4B), the qPCR platform established a strict limit of quantification (LOQ) of 0.001 ng of genomic DNA based on statistical reproducibility (Fig 5A and S4 Table). Although these metrics are derived from different criteria, the fact that the qPCR LOQ is already 10-fold lower than the conventional LOD logically underscores the vastly superior analytical sensitivity and dynamic range of the qPCR framework. To bridge this sensitivity with practical applications, a standard curve for recombinant plasmid carrying the comQ fragment was introduced (R² = 0.9973, y = −3.468x + 38.811, S5 Fig), utilizing robust statistical and visualization environments. When applied to three commercial natto brands (Table 4), the calculated comQ copies systematically decreased from 6.46 to 6.63 log₁₀ copies/tube at the undiluted level across a 10-fold dilution series (1⁰⁰ to 10⁻³), demonstrating robust quantification at trace levels.

The assay also addressed genomic interference under rigorous conditions involving up to a 10,000-fold excess of non-target DNA. Two-way ANOVA confirmed that background DNA from strain 168 did not significantly shift the Cq values (p = 0.293, Table 3), demonstrating over 8,000-fold selectivity. This resistance to competitive inhibition justifies the use of SYBR Green I over probe-based methods because selectivity is inherently established during primer extension. Furthermore, a stable Tm profile (approximately 74.5–75.1 °C) serves as an indispensable secondary validation of product purity, offering a level of assay transparency unavailable for the binary fluorescent signals from probes, as highlighted by the MIQE guidelines [39] and established melting curve principles [40].

Beyond industrial quality control, this platform is applicable to clinical microbiology. Although generally recognized as safe, subsp. natto has recently been implicated in opportunistic bacteremia and sepsis [8]. Our assay enables direct, culture-independent verification of subsp. natto in complex clinical matrices, supporting both fermentation integrity and public health surveillance.

From a practical perspective, our findings confirm that comQ serves as a stable, single-copy marker across the vast majority of the natto group [16], with its high sequence identity and consistent standard curves accurately reflecting actual cell equivalents without the copy-number variations often associated with alternative markers [15,34]. Our genomic screening identified only a single structural exception in strain AKPS2, which harbors two identical copies of comQ (S3 Table, S4 Fig). Because both copies exhibit 100% primer-binding identity, amplification is highly efficient, preventing false negatives. Although such variants could slightly overestimate cell equivalents in rare instances, this structural regularity generally offers a rapid, cost-effective alternative to conventional, labor-intensive MLST [9] or whole-genome sequencing approaches [15,16].

Because DNA detection alone does not necessarily prove the intactness of the full quorum-sensing pathway or actual fermentation performance, further phenotypic validation is necessary to clarify the correlation between genotypic abundance and industrial capacity in solid-state matrices. Furthermore, a previous study reported that IS4Bsu1-mediated insertion into comP can occur in B. subtilis subsp. natto, leading to genetic instability of γ-PGA production [41]. This indicates that while the comQXPA system is highly conserved within the sampled strains, it is not completely resistant to insertional disruption. Therefore, comQ should be regarded as a highly useful diagnostic and quantitative proxy within the analyzed dataset, rather than a definitive lineage-defining taxonomic marker. Finally, although our optimized standard curve achieved a low LOQ, tracking trace amounts in specific environmental or complex processing settings may face limitations. For such applications, integrating pre-enrichment protocols, probe-based TaqMan chemistries, or digital PCR platforms could further enhance the analytical sensitivity and robustness of the assay [39, 40].

Supporting information

S1 Table. Comprehensive genomic comparison, comQXPA locus conservation, and molecular marker profiles among Bacillus subtilis group and related species.

Detailed bioinformatic profiles of the 70 analyzed bacterial genomes, including Bacillus subtilis group strains and closely related Bacillus species. Average nucleotide identity (ANI) values and aligned fragment counts were calculated relative to the representative B. subtilis subsp. natto strain BEST195. Amino acid sequence identities (%) and sequence lengths (aa) are provided for each component of the quorum-sensing locus (ComQ, ComX, ComP, and ComA). The groupings indicated in the column “Analyzed by Seki & Nagano (2025)” reflect the taxonomic and functional classifications (such as Narrow-sense natto group, natto-related group, and Not natto group) adapted from the comparative genomic framework established by Seki and Nagano [16]. The rightmost columns indicate the presence (+) or absence (−) of conventional natto-specific genetic markers (bioF, bioW, fliF, IS4Bsu1, and IS256Bsu1), defined by a sequence identity threshold of ≥99% relative to strain BEST195. Gray shading visually highlights the presence (+) of these target markers to demonstrate their co-occurrence and discrepancy patterns across the lineages.

https://doi.org/10.1371/journal.pone.0355394.s001

(XLSX)

S2 Table. List of primers used in this study.

https://doi.org/10.1371/journal.pone.0355394.s002

(XLSX)

S3 Table. In silico copy number analysis and PCR validation of the comQ gene across various Bacillus strains.

This table presents the predicted copy numbers of the comQ gene and the in silico PCR electronic amplification profiles using two diagnostic primer sets (comQ-F241/R831 and comQ-qF/qR). The copy number of comQ represents the number of genetic loci exhibiting ≥98% sequence identity relative to the representative Bacillus subtilis subsp. natto strain BEST195. For the in silico PCR analysis, a plus sign (+) denotes that both the forward and reverse primers share a 100% perfect match with their respective target binding sites, predicting positive electronic amplification. A zero (0) indicates a lack of predicted amplification due to nucleotide mismatches or the absence of the target locus.

https://doi.org/10.1371/journal.pone.0355394.s003

(XLSX)

S4 Table. Quantitative performance and melting temperature analysis of the comQ-targeted qPCR assay for Bacillus subtilis subsp. natto and related strains.

https://doi.org/10.1371/journal.pone.0355394.s004

(XLSX)

S1 Fig. Alignment of the comQ gene sequences from representative Bacillus subtilis strains for primer designing.

Multiple sequence alignment of the comQ gene from B. subtilis subsp. natto BEST195, B. subtilis subsp. subtilis 168, and B. subtilis subsp. subtilis NBRC 101239. Red letters indicate mismatched nucleotides relative to the BEST195 sequence. Arrows indicate the primer binding sites and directions.

https://doi.org/10.1371/journal.pone.0355394.s005

(TIF)

S2 Fig. PCR amplification of the comQ gene from various Bacillus subtilis (natto) strains.

PCR products were amplified using the primer set comQ-241F and comQ-831R and analyzed via agarose gel electrophoresis. Lanes 1–13, NBRC 3013, Miyagino, Takahashi, Naruse, HK1−1, HK11−1, HK3−1, NBRC 3336, NBRC 16449, NBRC 3009, NBRC 3335, NBRC 3936, and NBRC 13169. Lane M, Gene-Ladder 100 (Nippon Gene Co., Ltd., Tokyo, Japan); black triangles indicate 2,000 and 500 bp. Gel images were converted to grayscale and adjusted for contrast using Adobe Photoshop 2020. The images shown are representative results from three independent experiments.

https://doi.org/10.1371/journal.pone.0355394.s006

(TIF)

S3 Fig. Specificity of the qPCR assay for the Bacillus subtilis subsp. natto-specific target.

(A) Difference in Cq values (ΔCq) between the target and nontarget strains. (B) Derivative melting curve analysis (−dF/dT) showing the distinct melting peaks for the target and nontarget strains. The green, blue, and red lines represent the subsp. natto-specific target (NBRC 16449), B. subtilis 168, and B. subtilis NBRC 101239, respectively. Both (A) and (B) were generated from the mean values from three independent experiments, with the exception of the data for strain NBRC 16449, which were based on five independent experiments.

https://doi.org/10.1371/journal.pone.0355394.s007

(TIF)

S4 Fig. Genomic organization of the comQXPA quorum-sensing locus and its structural duplication in Bacillus subtilis subsp. natto.

Comparative schematic representation of the com gene clusters among the reference strain BEST195 and the two duplicated loci in strain AKPS2 (designated as AKPS2−1 and AKPS2−2). Arrows indicate the position, length, and transcriptional direction of each gene. Red arrows represent the core quorum-sensing components (comA, comP, comX, comQ) and degQ. Blue and gray arrows indicate the tlpA pseudogene and the uncharacterized gene, respectively. Genomic coordinates (kb) are indicated above the horizontal line for each locus.

https://doi.org/10.1371/journal.pone.0355394.s008

(TIF)

S5 Fig. Standard curve of the comQ-targeted qPCR assay generated using a recombinant plasmid standard.

The quantification cycle (Cq) values are plotted against the log₁₀-transformed copy numbers of the comQ gene fragment per reaction tube. The dotted red line represents the linear regression model derived from the experimental data. Each data point represents the mean value obtained from three independent replicates (n = 3), with error bars indicating the standard deviation (SD).

https://doi.org/10.1371/journal.pone.0355394.s009

(TIF)

Acknowledgments

B. subtilis subsp. subtilis 168 was provided by the National BioResource Project (NIG, Japan).

References

1. Nishito Y, Osana Y, Hachiya T, Popendorf K, Toyoda A, Fujiyama A, et al. Whole genome assembly of a natto production strain Bacillus subtilis natto from very short read data. BMC Genomics. 2010;11:243. pmid:20398357
- View Article
- PubMed/NCBI
- Google Scholar
2. Inatsu Y, Nakamura N, Yuriko Y, Fushimi T, Watanasiritum L, Kawamoto S. Characterization of Bacillus subtilis strains in Thua nao, a traditional fermented soybean food in northern Thailand. Lett Appl Microbiol. 2006;43(3):237–42. pmid:16910925
- View Article
- PubMed/NCBI
- Google Scholar
3. Elbanna K, Alsulami FS, Neyaz LA, Abulreesh HH. Poly (γ) glutamic acid: a unique microbial biopolymer with diverse commercial applicability. Front Microbiol. 2024;15:1348411. pmid:38414762
- View Article
- PubMed/NCBI
- Google Scholar
4. Afzaal M, Saeed F, Islam F, Ateeq H, Asghar A, Shah YA, et al. Nutritional Health Perspective of Natto: A Critical Review. Biochem Res Int. 2022;2022:5863887. pmid:36312453
- View Article
- PubMed/NCBI
- Google Scholar
5. Zhang J, Bilal M, Liu S, Zhang J, Lu H, Luo H, et al. Isolation, Identification and Antimicrobial Evaluation of Bactericides Secreting Bacillus subtilis Natto as a Biocontrol Agent. Processes. 2020;8(3):259.
- View Article
- Google Scholar
6. Yokoyama T, Nakamura T, Kimijima M, Mandokoro K, Tokumaru M, Takatsuka A, et al. Subtilisin NAT, a subtilisin-like serine protease present in fermented soybean “natto” extract, inhibits <i>Streptococcus mutans</i> biofilm formation. FSTR. 2021;27(3):537–42.
- View Article
- Google Scholar
7. Fujii K, Kubo Y, Noguchi T, Tobita K. Effects of Bacillus subtilis Natto Strains on Antiviral Responses in Resiquimod-Stimulated Human M1-Phenotype Macrophages. Foods. 2023;12(2):313. pmid:36673407
- View Article
- PubMed/NCBI
- Google Scholar
8. Sada RM, Yamamoto G, Hamaguchi S, Kuroda E, Okura A, Cai M, et al. The majority of Bacillus subtilis strains isolated from blood cultures were derived from traditional Japanese fermented soybeans natto: A single-center retrospective study. Open Forum Infect Dis. 2025;12:ofaf574.
- View Article
- Google Scholar
9. Kubo Y, Rooney AP, Tsukakoshi Y, Nakagawa R, Hasegawa H, Kimura K. Phylogenetic analysis of Bacillus subtilis strains applicable to natto (fermented soybean) production. Appl Environ Microbiol. 2011;77(18):6463–9. pmid:21764950
- View Article
- PubMed/NCBI
- Google Scholar
10. Kamada M, Hase S, Sato K, Toyoda A, Fujiyama A, Sakakibara Y. Whole genome complete resequencing of Bacillus subtilis natto by combining long reads with high-quality short reads. PLoS One. 2014;9(10):e109999. pmid:25329997
- View Article
- PubMed/NCBI
- Google Scholar
11. Heo J, Kim J-S, Hong S-B, Park B-Y, Kim S-J, Kwon S-W. Genetic marker gene, recQ, differentiating Bacillus subtilis and the closely related Bacillus species. FEMS Microbiol Lett. 2019;366(16):fnz172. pmid:31675066
- View Article
- PubMed/NCBI
- Google Scholar
12. Liu Y, Štefanič P, Miao Y, Xue Y, Xun W, Zhang N, et al. Housekeeping gene gyrA, a potential molecular marker for Bacillus ecology study. AMB Express. 2022;12(1):133. pmid:36287351
- View Article
- PubMed/NCBI
- Google Scholar
13. Kimijima M, Mandokoro K, Ichikawa Y, Tokumaru M, Narisawa N, Takenaga F. Isolation and characterization of <i>Bacillus subtilis</i> from commercially available honey and its application in natto fermentation. FSTR. 2022;28(3):267–73.
- View Article
- Google Scholar
14. Sasaki M, Kawamura F, Kurusu Y. Genetic analysis of an incomplete bio operon in a biotin auxotrophic strain of Bacillus subtilis natto OK2. Biosci Biotechnol Biochem. 2004;68(3):739–42. pmid:15056910
- View Article
- PubMed/NCBI
- Google Scholar
15. Kamada M, Hase S, Fujii K, Miyake M, Sato K, Kimura K, et al. Whole-Genome Sequencing and Comparative Genome Analysis of Bacillus subtilis Strains Isolated from Non-Salted Fermented Soybean Foods. PLoS One. 2015;10(10):e0141369. pmid:26505996
- View Article
- PubMed/NCBI
- Google Scholar
16. Seki K, Nagano Y. Conserved accessory genes link a phylogenetically distinct Bacillus subtilis strain from Indian bekang to the Japanese natto clade. Sci Rep. 2025;15(1):43097. pmid:41407778
- View Article
- PubMed/NCBI
- Google Scholar
17. Xu X, Kovács ÁT. How to identify and quantify the members of the Bacillus genus? Environ Microbiol. 2024;26: e16593.
- View Article
- Google Scholar
18. Na H-E, Heo S, Kim T, Lee G, Lee J-H, Jeong D-W. ComQXPA quorum-sensing systems contribute to enhancing the protease activity of Bacillus velezensis DMB05 from fermented soybeans. Int J Food Microbiol. 2023;401:110294. pmid:37336024
- View Article
- PubMed/NCBI
- Google Scholar
19. Hoffmann K, Wollherr A, Larsen M, Rachinger M, Liesegang H, Ehrenreich A, et al. Facilitation of direct conditional knockout of essential genes in Bacillus licheniformis DSM13 by comparative genetic analysis and manipulation of genetic competence. Appl Environ Microbiol. 2010;76(15):5046–57. pmid:20543043
- View Article
- PubMed/NCBI
- Google Scholar
20. Stefanic P, Mandic-Mulec I. Social interactions and distribution of Bacillus subtilis pherotypes at microscale. J Bacteriol. 2009;191(6):1756–64. pmid:19114482
- View Article
- PubMed/NCBI
- Google Scholar
21. Tortosa P, Logsdon L, Kraigher B, Itoh Y, Mandic-Mulec I, Dubnau D. Specificity and genetic polymorphism of the Bacillus competence quorum-sensing system. J Bacteriol. 2001;183(2):451–60. pmid:11133937
- View Article
- PubMed/NCBI
- Google Scholar
22. Dogsa I, Choudhary KS, Marsetic Z, Hudaiberdiev S, Vera R, Pongor S, et al. ComQXPA quorum sensing systems may not be unique to Bacillus subtilis: a census in prokaryotic genomes. PLoS One. 2014;9(5):e96122. pmid:24788106
- View Article
- PubMed/NCBI
- Google Scholar
23. Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884–90. pmid:30423086
- View Article
- PubMed/NCBI
- Google Scholar
24. Prjibelski A, Antipov D, Meleshko D, Lapidus A, Korobeynikov A. Using SPAdes De Novo Assembler. Curr Protoc Bioinformatics. 2020;70(1):e102. pmid:32559359
- View Article
- PubMed/NCBI
- Google Scholar
25. Tanizawa Y, Fujisawa T, Nakamura Y. DFAST: a flexible prokaryotic genome annotation pipeline for faster genome publication. Bioinformatics. 2018;34(6):1037–9. pmid:29106469
- View Article
- PubMed/NCBI
- Google Scholar
26. Kumar S, Stecher G, Suleski M, Sanderford M, Sharma S, Tamura K. MEGA12: Molecular Evolutionary Genetic Analysis Version 12 for Adaptive and Green Computing. Mol Biol Evol. 2024;41(12):msae263. pmid:39708372
- View Article
- PubMed/NCBI
- Google Scholar
27. Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020;17(3):261–72. pmid:32015543
- View Article
- PubMed/NCBI
- Google Scholar
28. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825–30.
- View Article
- Google Scholar
29. Seabold S, Perktold J. Statsmodels: Econometric and Statistical Modeling with Python. In: Proceedings of the Python in Science Conference, 2010. 92–6. https://doi.org/10.25080/majora-92bf1922-011
30. Hunter JD. Matplotlib: A 2D graphics environment. Comput Sci Eng. 2007;9: 90–5.
- View Article
- Google Scholar
31. Waskom M. seaborn: statistical data visualization. JOSS. 2021;6(60):3021.
- View Article
- Google Scholar
32. Hirooka K, Shioda S, Okada M. Identification of critical residues for the catalytic activity of ComQ, a Bacillus prenylation enzyme for quorum sensing, by using a simple bioassay system. Biosci Biotechnol Biochem. 2020;84(2):347–57. pmid:31670609
- View Article
- PubMed/NCBI
- Google Scholar
33. Kada S, Ishikawa A, Ohshima Y, Yoshida K. Alkaline serine protease AprE plays an essential role in poly-γ-glutamate production during natto fermentation. Biosci Biotechnol Biochem. 2013;77(4):802–9. pmid:23563567
- View Article
- PubMed/NCBI
- Google Scholar
34. Kimura K, Itoh Y. Determination and characterization of IS4Bsu1-insertion loci and identification of a new insertion sequence element of the IS256 family in a natto starter. Biosci Biotechnol Biochem. 2007;71(10):2458–64. pmid:17928718
- View Article
- PubMed/NCBI
- Google Scholar
35. Špacapan M, Danevčič T, Štefanic P, Porter M, Stanley-Wall NR, Mandic-Mulec I. The ComX Quorum Sensing Peptide of Bacillus subtilis Affects Biofilm Formation Negatively and Sporulation Positively. Microorganisms. 2020;8(8):1131. pmid:32727033
- View Article
- PubMed/NCBI
- Google Scholar
36. Rooney AP, Price NPJ, Ehrhardt C, Swezey JL, Bannan JD. Phylogeny and molecular taxonomy of the Bacillus subtilis species complex and description of Bacillus subtilis subsp. inaquosorum subsp. nov. Int J Syst Evol Microbiol. 2009;59(Pt 10):2429–36. pmid:19622642
- View Article
- PubMed/NCBI
- Google Scholar
37. Dobrzyński J, Wróbel B, Górska EB. Taxonomy, Ecology, and Cellulolytic Properties of the Genus Bacillus and Related Genera. Agriculture. 2023;13(10):1979.
- View Article
- Google Scholar
38. Tran LS, Nagai T, Itoh Y. Divergent structure of the ComQXPA quorum-sensing components: molecular basis of strain-specific communication mechanism in Bacillus subtilis. Mol Microbiol. 2000;37(5):1159–71. pmid:10972833
- View Article
- PubMed/NCBI
- Google Scholar
39. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55(4):611–22. pmid:19246619
- View Article
- PubMed/NCBI
- Google Scholar
40. Tajadini M, Panjehpour M, Javanmard SH. Comparison of SYBR Green and TaqMan methods in quantitative real-time polymerase chain reaction analysis of four adenosine receptor subtypes. Adv Biomed Res. 2014;3:85. pmid:24761393
- View Article
- PubMed/NCBI
- Google Scholar
41. Nagai T, Tran LS, Inatsu Y, Itoh Y. A new IS4 family insertion sequence, IS4Bsu1, responsible for genetic instability of poly-gamma-glutamic acid production in Bacillus subtilis. J Bacteriol. 2000;182(9):2387–92. pmid:10762236
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Nishito Y, Osana Y, Hachiya T, Popendorf K, Toyoda A, Fujiyama A, et al. Whole genome assembly of a natto production strain Bacillus subtilis natto from very short read data. BMC Genomics. 2010;11:243. pmid:20398357
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Inatsu Y, Nakamura N, Yuriko Y, Fushimi T, Watanasiritum L, Kawamoto S. Characterization of Bacillus subtilis strains in Thua nao, a traditional fermented soybean food in northern Thailand. Lett Appl Microbiol. 2006;43(3):237–42. pmid:16910925
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Elbanna K, Alsulami FS, Neyaz LA, Abulreesh HH. Poly (γ) glutamic acid: a unique microbial biopolymer with diverse commercial applicability. Front Microbiol. 2024;15:1348411. pmid:38414762
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. Afzaal M, Saeed F, Islam F, Ateeq H, Asghar A, Shah YA, et al. Nutritional Health Perspective of Natto: A Critical Review. Biochem Res Int. 2022;2022:5863887. pmid:36312453
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Zhang J, Bilal M, Liu S, Zhang J, Lu H, Luo H, et al. Isolation, Identification and Antimicrobial Evaluation of Bactericides Secreting Bacillus subtilis Natto as a Biocontrol Agent. Processes. 2020;8(3):259.
View Article
Google Scholar

[18] View Article

[19] Google Scholar

[ref6] 6. Yokoyama T, Nakamura T, Kimijima M, Mandokoro K, Tokumaru M, Takatsuka A, et al. Subtilisin NAT, a subtilisin-like serine protease present in fermented soybean “natto” extract, inhibits <i>Streptococcus mutans</i> biofilm formation. FSTR. 2021;27(3):537–42.
View Article
Google Scholar

[21] View Article

[22] Google Scholar

[ref7] 7. Fujii K, Kubo Y, Noguchi T, Tobita K. Effects of Bacillus subtilis Natto Strains on Antiviral Responses in Resiquimod-Stimulated Human M1-Phenotype Macrophages. Foods. 2023;12(2):313. pmid:36673407
View Article
PubMed/NCBI
Google Scholar

[24] View Article

[25] PubMed/NCBI

[26] Google Scholar

[ref8] 8. Sada RM, Yamamoto G, Hamaguchi S, Kuroda E, Okura A, Cai M, et al. The majority of Bacillus subtilis strains isolated from blood cultures were derived from traditional Japanese fermented soybeans natto: A single-center retrospective study. Open Forum Infect Dis. 2025;12:ofaf574.
View Article
Google Scholar

[28] View Article

[29] Google Scholar

[ref9] 9. Kubo Y, Rooney AP, Tsukakoshi Y, Nakagawa R, Hasegawa H, Kimura K. Phylogenetic analysis of Bacillus subtilis strains applicable to natto (fermented soybean) production. Appl Environ Microbiol. 2011;77(18):6463–9. pmid:21764950
View Article
PubMed/NCBI
Google Scholar

[31] View Article

[32] PubMed/NCBI

[33] Google Scholar

[ref10] 10. Kamada M, Hase S, Sato K, Toyoda A, Fujiyama A, Sakakibara Y. Whole genome complete resequencing of Bacillus subtilis natto by combining long reads with high-quality short reads. PLoS One. 2014;9(10):e109999. pmid:25329997
View Article
PubMed/NCBI
Google Scholar

[35] View Article

[36] PubMed/NCBI

[37] Google Scholar

[ref11] 11. Heo J, Kim J-S, Hong S-B, Park B-Y, Kim S-J, Kwon S-W. Genetic marker gene, recQ, differentiating Bacillus subtilis and the closely related Bacillus species. FEMS Microbiol Lett. 2019;366(16):fnz172. pmid:31675066
View Article
PubMed/NCBI
Google Scholar

[39] View Article

[40] PubMed/NCBI

[41] Google Scholar

[ref12] 12. Liu Y, Štefanič P, Miao Y, Xue Y, Xun W, Zhang N, et al. Housekeeping gene gyrA, a potential molecular marker for Bacillus ecology study. AMB Express. 2022;12(1):133. pmid:36287351
View Article
PubMed/NCBI
Google Scholar

[43] View Article

[44] PubMed/NCBI

[45] Google Scholar

[ref13] 13. Kimijima M, Mandokoro K, Ichikawa Y, Tokumaru M, Narisawa N, Takenaga F. Isolation and characterization of <i>Bacillus subtilis</i> from commercially available honey and its application in natto fermentation. FSTR. 2022;28(3):267–73.
View Article
Google Scholar

[47] View Article

[48] Google Scholar

[ref14] 14. Sasaki M, Kawamura F, Kurusu Y. Genetic analysis of an incomplete bio operon in a biotin auxotrophic strain of Bacillus subtilis natto OK2. Biosci Biotechnol Biochem. 2004;68(3):739–42. pmid:15056910
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref15] 15. Kamada M, Hase S, Fujii K, Miyake M, Sato K, Kimura K, et al. Whole-Genome Sequencing and Comparative Genome Analysis of Bacillus subtilis Strains Isolated from Non-Salted Fermented Soybean Foods. PLoS One. 2015;10(10):e0141369. pmid:26505996
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref16] 16. Seki K, Nagano Y. Conserved accessory genes link a phylogenetically distinct Bacillus subtilis strain from Indian bekang to the Japanese natto clade. Sci Rep. 2025;15(1):43097. pmid:41407778
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref17] 17. Xu X, Kovács ÁT. How to identify and quantify the members of the Bacillus genus? Environ Microbiol. 2024;26: e16593.
View Article
Google Scholar

[62] View Article

[63] Google Scholar

[ref18] 18. Na H-E, Heo S, Kim T, Lee G, Lee J-H, Jeong D-W. ComQXPA quorum-sensing systems contribute to enhancing the protease activity of Bacillus velezensis DMB05 from fermented soybeans. Int J Food Microbiol. 2023;401:110294. pmid:37336024
View Article
PubMed/NCBI
Google Scholar

[65] View Article

[66] PubMed/NCBI

[67] Google Scholar

[ref19] 19. Hoffmann K, Wollherr A, Larsen M, Rachinger M, Liesegang H, Ehrenreich A, et al. Facilitation of direct conditional knockout of essential genes in Bacillus licheniformis DSM13 by comparative genetic analysis and manipulation of genetic competence. Appl Environ Microbiol. 2010;76(15):5046–57. pmid:20543043
View Article
PubMed/NCBI
Google Scholar

[69] View Article

[70] PubMed/NCBI

[71] Google Scholar

[ref20] 20. Stefanic P, Mandic-Mulec I. Social interactions and distribution of Bacillus subtilis pherotypes at microscale. J Bacteriol. 2009;191(6):1756–64. pmid:19114482
View Article
PubMed/NCBI
Google Scholar

[73] View Article

[74] PubMed/NCBI

[75] Google Scholar

[ref21] 21. Tortosa P, Logsdon L, Kraigher B, Itoh Y, Mandic-Mulec I, Dubnau D. Specificity and genetic polymorphism of the Bacillus competence quorum-sensing system. J Bacteriol. 2001;183(2):451–60. pmid:11133937
View Article
PubMed/NCBI
Google Scholar

[77] View Article

[78] PubMed/NCBI

[79] Google Scholar

[ref22] 22. Dogsa I, Choudhary KS, Marsetic Z, Hudaiberdiev S, Vera R, Pongor S, et al. ComQXPA quorum sensing systems may not be unique to Bacillus subtilis: a census in prokaryotic genomes. PLoS One. 2014;9(5):e96122. pmid:24788106
View Article
PubMed/NCBI
Google Scholar

[81] View Article

[82] PubMed/NCBI

[83] Google Scholar

[ref23] 23. Chen S, Zhou Y, Chen Y, Gu J. fastp: an ultra-fast all-in-one FASTQ preprocessor. Bioinformatics. 2018;34(17):i884–90. pmid:30423086
View Article
PubMed/NCBI
Google Scholar

[85] View Article

[86] PubMed/NCBI

[87] Google Scholar

[ref24] 24. Prjibelski A, Antipov D, Meleshko D, Lapidus A, Korobeynikov A. Using SPAdes De Novo Assembler. Curr Protoc Bioinformatics. 2020;70(1):e102. pmid:32559359
View Article
PubMed/NCBI
Google Scholar

[89] View Article

[90] PubMed/NCBI

[91] Google Scholar

[ref25] 25. Tanizawa Y, Fujisawa T, Nakamura Y. DFAST: a flexible prokaryotic genome annotation pipeline for faster genome publication. Bioinformatics. 2018;34(6):1037–9. pmid:29106469
View Article
PubMed/NCBI
Google Scholar

[93] View Article

[94] PubMed/NCBI

[95] Google Scholar

[ref26] 26. Kumar S, Stecher G, Suleski M, Sanderford M, Sharma S, Tamura K. MEGA12: Molecular Evolutionary Genetic Analysis Version 12 for Adaptive and Green Computing. Mol Biol Evol. 2024;41(12):msae263. pmid:39708372
View Article
PubMed/NCBI
Google Scholar

[97] View Article

[98] PubMed/NCBI

[99] Google Scholar

[ref27] 27. Virtanen P, Gommers R, Oliphant TE, Haberland M, Reddy T, Cournapeau D, et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods. 2020;17(3):261–72. pmid:32015543
View Article
PubMed/NCBI
Google Scholar

[101] View Article

[102] PubMed/NCBI

[103] Google Scholar

[ref28] 28. Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, et al. Scikit-learn: machine learning in Python. J Mach Learn Res. 2011;12:2825–30.
View Article
Google Scholar

[105] View Article

[106] Google Scholar

[ref29] 29. Seabold S, Perktold J. Statsmodels: Econometric and Statistical Modeling with Python. In: Proceedings of the Python in Science Conference, 2010. 92–6. https://doi.org/10.25080/majora-92bf1922-011

[ref30] 30. Hunter JD. Matplotlib: A 2D graphics environment. Comput Sci Eng. 2007;9: 90–5.
View Article
Google Scholar

[109] View Article

[110] Google Scholar

[ref31] 31. Waskom M. seaborn: statistical data visualization. JOSS. 2021;6(60):3021.
View Article
Google Scholar

[112] View Article

[113] Google Scholar

[ref32] 32. Hirooka K, Shioda S, Okada M. Identification of critical residues for the catalytic activity of ComQ, a Bacillus prenylation enzyme for quorum sensing, by using a simple bioassay system. Biosci Biotechnol Biochem. 2020;84(2):347–57. pmid:31670609
View Article
PubMed/NCBI
Google Scholar

[115] View Article

[116] PubMed/NCBI

[117] Google Scholar

[ref33] 33. Kada S, Ishikawa A, Ohshima Y, Yoshida K. Alkaline serine protease AprE plays an essential role in poly-γ-glutamate production during natto fermentation. Biosci Biotechnol Biochem. 2013;77(4):802–9. pmid:23563567
View Article
PubMed/NCBI
Google Scholar

[119] View Article

[120] PubMed/NCBI

[121] Google Scholar

[ref34] 34. Kimura K, Itoh Y. Determination and characterization of IS4Bsu1-insertion loci and identification of a new insertion sequence element of the IS256 family in a natto starter. Biosci Biotechnol Biochem. 2007;71(10):2458–64. pmid:17928718
View Article
PubMed/NCBI
Google Scholar

[123] View Article

[124] PubMed/NCBI

[125] Google Scholar

[ref35] 35. Špacapan M, Danevčič T, Štefanic P, Porter M, Stanley-Wall NR, Mandic-Mulec I. The ComX Quorum Sensing Peptide of Bacillus subtilis Affects Biofilm Formation Negatively and Sporulation Positively. Microorganisms. 2020;8(8):1131. pmid:32727033
View Article
PubMed/NCBI
Google Scholar

[127] View Article

[128] PubMed/NCBI

[129] Google Scholar

[ref36] 36. Rooney AP, Price NPJ, Ehrhardt C, Swezey JL, Bannan JD. Phylogeny and molecular taxonomy of the Bacillus subtilis species complex and description of Bacillus subtilis subsp. inaquosorum subsp. nov. Int J Syst Evol Microbiol. 2009;59(Pt 10):2429–36. pmid:19622642
View Article
PubMed/NCBI
Google Scholar

[131] View Article

[132] PubMed/NCBI

[133] Google Scholar

[ref37] 37. Dobrzyński J, Wróbel B, Górska EB. Taxonomy, Ecology, and Cellulolytic Properties of the Genus Bacillus and Related Genera. Agriculture. 2023;13(10):1979.
View Article
Google Scholar

[135] View Article

[136] Google Scholar

[ref38] 38. Tran LS, Nagai T, Itoh Y. Divergent structure of the ComQXPA quorum-sensing components: molecular basis of strain-specific communication mechanism in Bacillus subtilis. Mol Microbiol. 2000;37(5):1159–71. pmid:10972833
View Article
PubMed/NCBI
Google Scholar

[138] View Article

[139] PubMed/NCBI

[140] Google Scholar

[ref39] 39. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem. 2009;55(4):611–22. pmid:19246619
View Article
PubMed/NCBI
Google Scholar

[142] View Article

[143] PubMed/NCBI

[144] Google Scholar

[ref40] 40. Tajadini M, Panjehpour M, Javanmard SH. Comparison of SYBR Green and TaqMan methods in quantitative real-time polymerase chain reaction analysis of four adenosine receptor subtypes. Adv Biomed Res. 2014;3:85. pmid:24761393
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref41] 41. Nagai T, Tran LS, Inatsu Y, Itoh Y. A new IS4 family insertion sequence, IS4Bsu1, responsible for genetic instability of poly-gamma-glutamic acid production in Bacillus subtilis. J Bacteriol. 2000;182(9):2387–92. pmid:10762236
View Article
PubMed/NCBI
Google Scholar

[150] View Article

[151] PubMed/NCBI

[152] Google Scholar

Figures

Abstract

Introduction

Materials and methods

Bacterial strains and culture conditions

DNA extraction and quality control

Genome sequencing and assembly

Genome-wide similarity assessment and phylogenetic analysis of the comQXPA locus

Primer design

Conventional PCR amplification and specificity testing

qPCR and construction of a universal standard curve

DNA extraction from commercial natto matrices and viable cell enumeration

Preparation of standard plasmid DNA

Data visualization and statistical analysis

Results

High genomic similarity within the B. subtilis group

Phylogenetic stratification and co-evolutionary analysis of the comQXPA locus

Co-occurrence of comQ with known natto-specific genetic markers

Design and in silico PCR validation of comQ-targeted primers

Specificity, inclusivity, and sensitivity of the comQ-targeted PCR assay

Quantitative performance, specificity, and robustness of the qPCR assay

Absolute quantification of comQ in commercial natto extracts

Discussion

Supporting information

S1 Table. Comprehensive genomic comparison, comQXPA locus conservation, and molecular marker profiles among Bacillus subtilis group and related species.

S2 Table. List of primers used in this study.

S3 Table. In silico copy number analysis and PCR validation of the comQ gene across various Bacillus strains.

S4 Table. Quantitative performance and melting temperature analysis of the comQ-targeted qPCR assay for Bacillus subtilis subsp. natto and related strains.

S1 Fig. Alignment of the comQ gene sequences from representative Bacillus subtilis strains for primer designing.

S2 Fig. PCR amplification of the comQ gene from various Bacillus subtilis (natto) strains.

S3 Fig. Specificity of the qPCR assay for the Bacillus subtilis subsp. natto-specific target.

S4 Fig. Genomic organization of the comQXPA quorum-sensing locus and its structural duplication in Bacillus subtilis subsp. natto.

S5 Fig. Standard curve of the comQ-targeted qPCR assay generated using a recombinant plasmid standard.

Acknowledgments

References