Competent E. coli OmniMax and B. subtilis cells were prepared using the procedures described (Green et al., 2012; Smith, 1991). The B. subtilis control strain HT800F (patent no. VN103078; Vietnam National Office of Intellectual Property, 2024)) was from our lab stock collection, which originated from WB800N (Nguyen et al., 2011; Jeong et al., 2018).

The vector pHT2380 was designed as a shuttle vector for double-crossover integration into B. subtilis. It contains the fusion sequence of PcotB-cotB-mntC, which was synthesized by GenScript. To generate the recombinant B. subtilis strain, we introduced the vector pHT2380 to competent cells of HT800F strain. Colonies were selected on an agar plate with neomycin added at 25 μg/mL, then confirmed by colony PCR using three pairs of primers. Primer information was listed in Table 1. B. subtilis new strain BsHT2380 was stored at −80 °C.

Bacillus subtilis colonies were cultured directly in 400 mL of Difco Sporulation Medium (DSM) at 33 °C to induce the sporulation. The biomass was collected at different time points by centrifugation at 13,000 × g for 1 min. The biomass was treated with lysozyme (10 mg/mL) at 37 °C for 1 h to eliminate vegetative cells, followed by centrifugation at 13,000 × g for 1 min and four washes with 1 × PBS to obtain purified endospores. Samples were stored in PBS 1X with 10% glycerol at −80 °C for the following experiments.

To determine the percentage of sporulation, the colony-forming unit (CFU) number of samples taken from the culture before and after lysozyme treatment was counted using serial dilution and plate spread techniques. The experiment was performed in triplicate. The sporulation rate was calculated as follows:

Percentage of sporulation=Number of endospores/Total number ofendospores andvegetative cells

The Schaeffer–Fulton method is a differential staining technique developed to visualize highly resistant bacterial endospores (Schaeffer and Fulton, 1933). A 20 μL volume of either biomass (cells and spores) or purified spores after 48 h of cultivation was smeared onto glass slides. The samples were heat-fixed and stained with 5% malachite green for 5 min, followed by cooling and rinsing with water to remove excess dye. Subsequently, 2.5% safranin O was applied for 30 s, after which the slides were rinsed with water and air-dried. The results were then examined under a light microscope.

The purified endospores were fixed with 4% formaldehyde for 20 min, washed three times with PBS 1X, and then incubated with blocking buffer (PBS-T with 5% skim milk) for 30 min. The primary antibody for MntC developed in Swiss mice by our research group was used at a ratio of 1:1000 for 1 h, gently shaking. After the washing steps, the sample was incubated with the secondary antibody Goat anti-Mouse IgG (H+L) Cross-Adsorbed Secondary Antibody, Alexa Fluor™ 488 (Invitrogen, A-11001) at the concentration of 1:1000 and washed eventually. Finally, a 5 μL sample was spread on the slide, added 5 μL mounting media, and then covered with a coverslip. The result was read by Confocal Microscopy (Nikon AX) and analyzed by NIS-Element Viewer Software.

The spore ELISA was performed as described in our previous publication (Nguyen et al., 2021). Briefly, the purified spores were resuspended in coating buffer (100 mM NaHCO₃; pH 9.6) to the OD_600nm of 4, then coated onto a 96-well plate (Thermo Scientific™ Nunc™ MicroWell™ 96-Well Microplates) at 50 μL per well. After incubation, washing, and blocking, the primary MntC-antibody, as the same used for Fluorescent Confocal Microscopy, was used at a dilution ratio of 1:10,000. The second antibody was anti-mouse IgG—peroxidase antibody produced in rabbits (whole molecule) (Sigma, A9044–2 mL), used at a ratio of 1:40,000. The signal was developed by TMB Liquid Substrate for ELISA (Sigma). Absorbance values were measured at a wavelength of 450 nm in a CLARIOstar plate reader. Each sample was tested in triplicate, with results expressed as mean ± SD. Statistical significance was determined using a t-test performed with GraphPad Prism (USA).

Purified spores were adjusted to an optical density of 5 at 600 nm, pelleted by centrifugation at 13,000 × g for 1 min, and resuspended in spore protein extraction buffer (50 mM Tris–HCl, pH 6.8, 1% SDS, and 50 mM DTT). The suspension was incubated at 65 °C for 1 h to extract proteins, and the supernatant containing the extracted proteins was collected after centrifugation. Proteins were separated by SDS-PAGE and transferred onto a nitrocellulose membrane. The membrane was blocked in PBS–0.1% Tween™ (PBS-T) containing 5% (w/v) skim milk (Oxoid™), and incubated overnight at 4 °C with a primary anti-MntC antibody, generated in Swiss mice by our group, diluted 1:20,000 in blocking buffer. After washing, the membrane was incubated for 1 h at room temperature with a rabbit anti-mouse IgG peroxidase-conjugated secondary antibody (whole molecule, Sigma A9044), diluted 1:160,000 in blocking buffer. Following final washes, protein expression was visualized using TMB substrate, and the colorimetric signal was recorded. Protein sizes were estimated using the PageRuler™ Plus Prestained Protein Ladder (Thermo, 26619).

The biomass of B. subtilis spores was washed with deionized water, collected at an OD_600nm of 5, and then mixed with 0.5 mL of nitric acid 65% (Merck) and heat-treated at 80 °C in a water bath for 10 min. When dissolved entirely, the sample was added precisely with 0.5 mL of deionized water and diluted appropriately. A flame atomic absorption spectrometer (Shimadzu AA6300, Japan) was used to determine the Mn²⁺ concentration.

Purified endospores were incubated with lysozyme at the concentration of 15 mg/mL at 37 °C for 1 h, shaking at 200 rpm. Spores were incubated with distilled H₂O at a ratio of 1:1 at 85 °C for 30 min to test the wet heat resistance. All samples were then centrifuged at 13,000 ×g for 1 min and washed with PBS 1X. The serial dilution method was applied to determine the number of survival spores.

The recombinant plasmid pHT2380 was constructed using pHT2304 (Nguyen et al., 2021), a shuttle vector backbone derived from pCotB-CL (Iwanicki et al., 2014). This plasmid contains key elements for cloning and chromosomal integration, including the E. coli ColE1 origin of replication, multiple cloning sites, ampicillin and neomycin resistance genes for selection, and homologous regions (amyE3 and amyE5) for double-crossover integration at the amyE locus. In addition, pHT2380 features two mutant loxP sites (lox66 and lox71) for Cre-lox-based elimination of antibiotic resistance markers, and a cotB-mntC fusion for spore surface display under the control of the native PcotB promoter. A schematic map of the pHT2380 construct was shown in Figure 1A.

The vector pHT2380 containing the manganese-binding domain of MntC was introduced into B. subtilis HT800F competent cells via natural transformation. Homologous recombination at the amyE locus facilitated stable chromosomal integration of the cotB-mntC cassette. This genomic integration ensures the inserted gene remains stably maintained in the chromosome without requiring continuous antibiotic selection pressure (Härtl et al., 2001). Transformants were selected on antibiotic-containing agar and screened by colony PCR using three primer pairs: two targeting the 5′ and 3′ flanking regions of the amyE locus, and one specific for the cotB-mntC fusion. The primer binding sites were illustrated in Figure 1C. PCR analysis (Figure 1B) confirmed the successful integration, yielding the expected amplicon sizes of 1,100 bp, 1,022 bp, and 1,172 bp for the respective primer sets. A confirmed colony was subsequently cultured in LB medium and stored in 15% glycerol at −80 °C. The recombinant strain was designated BsHT2380.

The BsHT2380 strain was engineered to express a CotB-MntC fusion protein during sporulation, driven by the native PcotB promoter, to facilitate surface localization of the manganese-binding domain on the spore coat. BsHT2380 and two control strains, BsHT800F (parental strain) and BsHT2304 (Nguyen et al., 2021) (harboring a similar construct without the target gene mntC), were cultured in Difco Sporulation Medium (DSM) for 48 h to induce spore formation. To confirm the successful display of MntC, spores were collected for western blot, spore ELISA and immunofluorescence microscopy to verify the surface display of MntC.

The Western blot results shown in Figure 2F revealed a distinct band in the BsHT2380 sample, whereas no band was observed in the control BsHT2304, demonstrating the specific recognition of MntC expressed in BsHT2380 by the anti-MntC antibody. The expected molecular mass of MntC alone is 32.8 kDa, while the CotB-MntC fusion protein is predicted to be 66 kDa. A visible band appeared at approximately 70 kDa, confirming the stable expression of the CotB-MntC fusion protein.

In spore ELISA, the whole endospores, suspended in a coating buffer at an OD_600nm value of 4, were coated onto a 96-well plate as an antigen factor. The next steps were basically the indirect ELISA, in which the released signal proved the positive interaction between the spores and the specific anti-MntC antibody. Results in Figure 2E demonstrated a statistically significant increase in MntC expression on the BsHT2380 sample (p ≤ 0.0001). This strong signal suggested a successful interaction between the surface proteins of BsHT2380 and the anti-MntC antibody, confirming the expression of the target protein, MntC, by the spores immobilized on the plate.

To further confirm the presence of MntC on the spore surface, we performed immunofluorescence labeling. Spores were incubated with an anti-MntC primary antibody and a fluorescent secondary antibody, followed by visualization using confocal fluorescence microscopy. Figure 2D showed that the BsHT2380 sample emitted clear and high fluorescence signals. In the picture under the transmitted light channel (TD) (Figure 2C), they appeared as intact objects with thick walls and approximately 1.5 μm in length, which is the morphological characteristics of B. subtilis endospores. Besides, BsHT2304, the control sample that expressed CotB only, was observed with similar spore morphology images through the TD channel (Figure 2A) but lacked any fluorescence signal (Figure 2B).

Taken together, the distinct fluorescence observed by confocal microscopy, along with the Western blot and spore ELISA results, conclusively demonstrated MntC expression and its anchoring by CotB on the outer spore coat of BsHT2380.

We next evaluated whether the manganese-binding activity of recombinant MntC displayed on B. subtilis spores was functionally effective by measuring manganese ion concentrations in BsHT2380 compared to the parental strain BsHT800F. Both strains were sporulated under the same conditions and harvested at two time points: 48 and 72 h. Spores were treated with lysozyme to remove residual vegetative cells, with or without the addition of EDTA, and then subjected to ion quantification. Manganese levels were normalized to calcium content (Mn/Ca ratio), as Ca-DPA (calcium-dipicolinic acid) in the spore core remains stable and unaffected by surface treatments, including EDTA, under standardized sporulation conditions. In contrast, manganese associated with the spore coat can be removed by EDTA, making calcium a reliable internal reference for spore quantity. To compare strains, the relative Mn/Ca ratio (reported as “Mn/Ca vs. Control” in Table 2) was calculated by dividing the Mn/Ca value of BsHT2380 by that of the control strain. The percentage reduction by EDTA (reported as “% Mn reduction by EDTA” in Table 2) was defined as the decrease in the Mn/Ca ratio of EDTA-treated samples relative to untreated samples, expressed as a percentage of the untreated value.

As shown in Table 2, both BsHT2380 and control spores exhibited a reduction in Mn/Ca ratio upon EDTA treatment; however, the decrease was greater in BsHT2380, 45.64 and 46.6% at 48 and 72 h, respectively, compared to 25.66 and 21.84% in the control. This reduction indicated that MntC actively bound manganese on the spore surface, and that EDTA was capable of displacing it. These findings suggested that while wild-type spores retain a modest amount of surface-bound manganese, likely through nonspecific interactions, the presence of surface-displayed MntC enhanced manganese accumulation on the spore coat, making it more susceptible to EDTA-mediated removal.

The manganese content in spore samples treated with lysozyme and water, without EDTA, also revealed differences between strains. At 48 h, the Mn/Ca ratio in BsHT2380 was 1.23-fold higher than in BsHT800F, indicating an enhanced ability of BsHT2380 to bind and accumulate Mn²⁺ at this time point. However, by 72 h, this difference diminished, and the Mn/Ca ratios between the two strains became comparable. This suggests that the manganese-binding capacity of BsHT2380 is transient, being most evident during the early phase of cultivation after sporulation onset and diminishing as cultivation progresses, possibly due to saturation of surface MntC binding sites or reduced accessibility of MntC during spore maturation.

To further investigate whether the observed increase in Mn²⁺ accumulation was associated with improved spore production efficiency, we next examined spore formation kinetics and spore resistance in BsHT2380 compared to control strains.

To evaluate the effect of MntC on sporulation progression, the biomass of B. subtilis strains BsHT800F, BsHT2304, and BsHT2380 was harvested at 48 h post-inoculation and subjected to Schaeffer–Fulton staining. This method distinguishes different cellular forms based on different staining properties: vegetative cells appear pink-red due to Safranin O uptake, mature endospores stain blue–green with Malachite Green, and prespores are observed as pink cells containing internal green-stained endospores.

Microscopy images shown in Figures 3A–C illustrate the presence of vegetative cells, prespores, and mature spores in all three strains. In these representative images, the recombinant strain BsHT2380 appeared to contain more fully stained mature endospores compared to the control strains BsHT800F and BsHT2304. This qualitative observation suggests that BsHT2380 may generate a higher number of spores.

To investigate the impact of MntC expression on sporulation kinetics, the percentage of mature spores was quantified at five distinct time points. Spores were isolated, serially diluted, and counted via colony-forming unit (CFU) analysis. As shown in Figure 3D, sporulation initiation was observed around 24 h post-inoculation across all strains. By 32 h, BsHT2380 exhibited a significantly higher proportion of mature spores (~63%) compared to the control strains BsHT800F and BsHT2304, both of which remained below 50% (p < 0.05). At 48 h, BsHT2380 reached 71% spore formation, significantly exceeding BsHT800F (53%) and BsHT2304 (55%) (p < 0.01). From 48 to 72 h, spore percentages remained relatively stable in all strains; however, BsHT2380 consistently maintained the highest spore proportion (74%), compared to BsHT800F and BsHT2304 (both 57%). The minimal difference between the two control strains indicated that the plasmid backbone in BsHT2304, lacking the mntC insert, did not impact the sporulation.

Collectively, these findings demonstrated that surface expression of MntC in BsHT2380 enhanced the kinetics and efficiency of spore production in B. subtilis, resulting in a consistently higher proportion of mature spores. The relative plateau in spore percentages beyond 48 h across all strains suggested that this time point may represent the peak of sporulation under the tested conditions.

To evaluate whether the BsHT2380 strain retained or enhanced its resistance characteristics, purified spores were subjected to lysozyme and wet heat treatments. Spore survival was quantified by CFU counting and expressed as a percentage relative to the total input spores under each treatment condition (Figure 4). The results showed that BsHT2380 exhibited significantly improved resistance compared to both the parental strain BsHT800F and the plasmid control strain BsHT2304 under the same treatment conditions. Upon exposure to lysozyme, BsHT2380 spores exhibited significantly higher survival (73%) compared to BsHT800F (50%, p ≤ 0.01) and BsHT2304 (57%, p ≤ 0.05). In response to wet heat, BsHT2380 also demonstrated enhanced resistance, with a survival rate of 70%, which was significantly higher than both BsHT800F (55%, p ≤ 0.01) and BsHT2304 (59%, p ≤ 0.0001). These results suggested that surface display of MntC enhanced spore resistance to both chemical (lysozyme) and physical (wet heat) stressors.