The Bacillales and Escherichia coli strains and the plasmids used in this study are listed in Table 1. B. subtilis strains were derivatives of strain 168, and those newly constructed in this study were generated via plasmid DNA transformation and confirmed by polymerase chain reaction (PCR) (Kuwana et al., 2003, 2004, 2006; Takamatsu et al., 2009). E. coli strain JM109 was used for plasmid propagation. The oligonucleotides used for PCR amplification are listed in Table 2. To amplify the cgeA fragment from the B. subtilis 168 chromosome, primers CGEAM490 and CGEA398R were used. The resultant PCR product was digested at the BamHI and XhoI sites introduced by the primers and ligated into the BamHI/XhoI-digested pGFP7CA vector to create the plasmid pCGEA8GA (Imamura et al., 2011; Table 1). This plasmid was introduced into the amyE locus of strain 168 via transformation. Single-crossover transformants were identified by chloramphenicol (5 μg/mL) selection, yielding the CGEA8GA construct (Table 1). Recombinants were confirmed via PCR. Green fluorescent protein (GFP)-fused strains were constructed using overlap extension PCR (Nishikawa and Kobayashi, 2021). To construct strain G-0140, a gfp-cat cassette was PCR-amplified from plasmid pCBGFPUV4 using UV4-F and cat-T-R primers (Table 2). The cat-T-R primer contained transcription terminators to prevent read-through. Similarly, to construct strains G-0399 and G-0550, a gfp-cat cassette was PCR-amplified from plasmid pCBGFPUV4 using UV4-F and cat-R primers (Table 2). The coding and downstream regions of the target gene were amplified using the primer pairs F1/R1 and F2/R2 (Table 2). The 5′ sequences of R1 and F2 were complementary to those of UV4-F and cat-R or cat-T-R, respectively. The three PCR fragments were fused and used as templates for a second round of PCR with primers F1 and R2. The final PCR products were transformed into strain 168, generating gfp-cat strains through double-crossover recombination. Transformants were selected on Luria–Bertani (LB; BD Japan Co. Tokyo, Japan) agar plates containing chloramphenicol (5 μg/mL).

B. subtilis and other Bacillales species were grown in LB broth and Schaeffer's medium (Schaeffer et al., 1965). The conditions for the sporulation of B. subtilis have been described previously (Kuwana et al., 2002).

In addition to B. subtilis, the following Bacillales spore-forming bacteria were used: Niallia circulans NBRC 13629, B. licheniformis ATCC 14580, Brevibacillus brevis NBRC 100599, Lysinibacillus sphaericus NBRC 3526, and Paenibacillus polymyxa NBRC15309. The strains were obtained from the NITE Biological Resource Center (NBRC, Tokyo, Japan) and the American Type Culture Collection (ATCC, Manassas, VA, USA). Growth conditions were as follows: B. licheniformis ATCC 14580 was incubated on Schaeffer's agar medium at 24°C for 36 h. N. circulans NBRC 13629, L. sphaericus NBRC 3526, and P. polymyxa NBRC15309 were incubated on LB agar at 37°C for 24 h, followed by 25°C for additional 24 h to facilitate efficient spore release from mother cells.

Recombinant DNA preparations were carried out following standard protocols (Sambrook et al., 1989). Preparations of competent cells, transformation, and extraction of chromosomal DNA for B. subtilis were performed as described previously (Cutting and Vander Horn, 1990). When required, antibiotics were added at the following concentrations: chloramphenicol: 5 μg/mL and erythromycin: 0.5 μg/mL.

Bacterial cells were cultured in LB or Schaeffer's medium, and aliquots were transferred to tubes for staining. Six fluorescence reagents were used in this study. Cells were stained with different combinations of the following dyes in 10 mM Tris-HCl (pH 7.6) at 25°C for 10 min as described previously (Kuwana et al., 2023): APBT (0.01 mg/mL; Tokyo Chemical Industry, Tokyo, Japan), which stains the cell membrane and/or cell wall (Kuwana et al., 2023); safranin O (0.01 mg/mL; Waldeck GmbH & Co. KG, Münster, Germany), which is commonly used as a counterstain in gram staining (Bartholomew and Mittwer, 1952) and also for measuring mitochondrial membrane potential due to its excitation/emission wavelengths (495/587 nm). In plant tissues, it binds to acidic lignin polymers, and its fluorescence peak shifts depending on lignin content (Baldacci-Cresp et al., 2020); thioflavin T (0.01 mg/mL; Sigma-Aldrich, Tokyo, Japan) is typically used to detect amyloid fibrils and to analyze the cell structure of B. subtilis (Kuwana et al., 2023). Following staining, cells were transferred to microscope slides for imaging.

Phase contrast and fluorescence microscopy images of B. subtilis, B. licheniformis, N. circulans, B. brevis, L. sphaericus, and P. polymyxa cells were acquired using an Olympus BX51 phase-contrast microscope equipped with additional fluorescence attachments and the following mirror cube units (Olympus, Tokyo, Japan): green fluorescence (U-MGFPHQ) for GFP, auramine O, safranin O, and thioflavin T; red fluorescence (U-MWG2) for congo red, rhodamine B, and safranin O; and blue fluorescence (U-MNUA2) for APBT. Imaging was performed using a UPlanApo 100X oil iris 3 pH objective lens and a U-TV1X-2 camera adapter (Olympus, Tokyo, Japan). Images were captured using an ORCA-SPARK digital CMOS camera C11440-36U (Hamamatsu Photonics Inc. Shizuoka, Japan) and analyzed using CellSens imaging software (Olympus, Tokyo, Japan). The exposure time for fluorescence imaging ranged from 0.25 s to 4.0 s. Contrast and tone balance were adjusted using CellSens. For each strain, the images were scaled to the same intensity range.

Mature spores, characterized by high refractive brightness, exhibit a dark outer edge under phase-contrast microscopy. To determine the precise location of fluorescence within the spores, we identified the negative peak of phase-contrast intensity at the boundary between the spore's interior and exterior, following the protocol described by Imamura et al. (2010). Spore length was determined by measuring the distance between the negative peaks observed at both poles of the spore under a phase-contrast microscope. Measurements were performed using the CellSens imaging software (Olympus, Tokyo, Japan). Fluorescence microscopy of spore coat and cortex proteins fused with GFP showed a ring-shaped fluorescence pattern along the periphery of the spores. Based on the method of Imamura et al. (2010), fluorescence localization was quantified by identifying the positive fluorescence peaks of the GFP fusion protein at both poles of the spore's long axis. The distance between the two positive peaks was calculated using CellSens software. For each sample, 10 spores were analyzed, and we measured the following distances: negative peak distance (spore length from phase-contrast microscopy images) and positive peak distance (fluorescence intensity peaks from fluorescence microscopy images by fluorescent reagents and/or GFP fusion proteins). The relative fluorescence localization in spores was computed as the average of these measured distances.

To determine statistical significance, we used a two-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test to compare the means of multiple groups and identify significant differences. Results with P ≤ 0.05 were considered statistically significant, and all statistical analyses were performed using Microsoft Excel.

B. subtilis 168, B. licheniformis ATCC 14580, N. circulans NBRC 13629, B. brevis NBRC 100599, L. sphaericus NBRC 3526, and P. polymyxa NBRC 15309, all belonging to Bacillales, were grown on the Schaeffer's or LB plates at 37°C (see Methods). Harvested cells were stained with a mixture of APBT and safranin O and observed under a phase-contrast and fluorescence microscope, identifying oval-shaped mature spores for most species with notable variations in size and shape under phase-contrast microscopy (Figure 1). L. sphaericus NBRC 3526 spores were the smallest, while those of P. polymyxa NBRC 15309 were the largest among the six species. L. sphaericus NBRC3526 spores exhibited a mix of circular and oval shapes with variations in size. APBT fluorescence was detected along the periphery of vegetative cells, mother cells, forespores, and mature spores in all strains (Figure 1). This finding is comparable with previous reports, which suggested that APBT stains the cell membrane (Kuwana et al., 2023, 2024). Safranin O fluorescence was detected using green and red mirror cube units (see Methods). Though both green and red fluorescence signals of safranin O were detected in all cell types, the intensity was stronger in the forespores and mature spores than in the vegetative and mother cells (Figure 1). Additionally, safranin O fluorescence was detected only at the periphery of the forespores or mature spores and not in the core of the spores. The green fluorescence of safranin O was similar to that of APBT in B. subtilis cells. The red fluorescence of safranin O was detected in the mother cells of B. brevis NBRC 100599 and L. sphaericus NBRC 3526, and the cores of some forespores of N. circulans NBRC 13629 and L. sphaericus NBRC 3526.

To evaluate whether the observed fluorescence signals were due to intrinsic autofluorescence of the spores, we examined unstained B. subtilis 168 spores using the same filter sets and a range of exposure times. Autofluorescence signals were only detectable with exposure times of >1 s for blue and > 8 s for green and red filters (Supplementary Figure S1). Fluorescence microscopy experiments were therefore conducted with exposure times below these thresholds. Similar background levels were confirmed in other Bacillales species. These findings indicate that the observed fluorescence signal primarily results from the added fluorescent dyes rather than intrinsic autofluorescence.

We performed a supplemental experiment to assess whether safranin O alone could visualize spores by brightfield microscopy. As shown in Supplementary Figure S2, vegetative and mother cells were stained with higher concentrations of safranin O (2.5 and 1 mg/mL), whereas the mature spores did not stain clearly. In contrast, no cells were visibly stained at a lower concentration (0.01 mg/mL).

Ten spores with similar morphology were selected from each sample for analysis. We measured the distance between the negative peaks along the long axis of the spores in the phase-contrast microscopy images, and the distance between the positive fluorescence signal peaks from the applied fluorescent reagents (Figure 2). We observed that the average length of the long axis of L. sphaericus NBRC 3526 spores was the smallest, and that of P. polymyxa NBRC15309 was the largest, with a more than two-fold size difference. In mature spores of B. subtilis 168, B. licheniformis ATCC 14580, N. circulans NBRC 13629, B. brevis NBRC 100599, L. sphaericus NBRC 3526, and P. polymyxa NBRC 15309, the distance between the positive peaks of APBT, green safranin O and red safranin O fluorescence was smaller than the length of the long axis of each mature spore. These results revealed that all the fluorescent reagents selectively stained specific structures of Bacillales mature spores.

Phase contrast microscopy results of the distance between the negative peaks indicated that B. subtilis 168 and B. brevis NBRC 100599 spores had similar sizes; P. polymyxa NBRC 15309 spores were the largest, while the spore size of L. sphaericus NBRC 3526 was the smallest. Consequently, despite variations in spore size, we observed that the distance between the positive peaks of red fluorescence from safranin O consistently exceeded that of the positive peaks from the red fluorescence of APBT and the green fluorescence of safranin O across these four strains. The fluorescence of APBT and safranin O in B. licheniformis ATCC 14580 and N. circulans NBRC 13629 spores were unique. In B. licheniformis ATCC 14580 and N. circulans NBRC 13629 spores, fluorescence was detected primarily at the spore periphery, but their staining profiles differed. In B. licheniformis, APBT fluorescence peaks were spaced greater than those of green safranin O, whereas red safranin O matched APBT. In contrast, in N. circulans, all three dyes produced similar peak distances. These findings highlight species-specific variations in spore size and staining patterns, reflecting structural diversity within Bacillales spores.

In mature B. subtilis spores, GFP fusion strains of spore proteins were used to compare the detectable fluorescence of GFP with that of APBT. We previously reported the localization of spore proteins using GFP fusion strains (Imamura et al., 2010, 2011). CgeA-GFP is located in the crust, YeeK-GFP is located in the inner spore coat, and YhcN-GFP is located in the cortex and inner spore membrane (Takamatsu et al., 2009; Imamura et al., 2010; Zheng et al., 2016; Liu et al., 2022). To determine the proportion of APBT staining in the spores, we used CgeA-GFP, YeeK-GFP, and YhcN-GFP strains as marker proteins for the crust, inner coat, and cortex, respectively. The atpC and qoxB genes encode components of the ATP synthase and cytochrome aa3 quinol oxidase (subunit I), respectively, and AtpC and QoxB localize to the cell membrane (Meile et al., 2006; Hahne et al., 2008). The lonB gene encodes a Lon-like ATP-dependent protease, whose expression is dependent on sigF during early sporulation; LonB localizes to the inner spore membrane (Serrano et al., 2001; Simmons et al., 2008). We constructed the GFP fusion strains AtpC-GFP, LonB-GFP, and QoxB-GFP to mark the inner spore membrane. The GFP fusion strains were grown on Schaeffer's plates at 37°C for 24 h, after which the cells were collected and stained with APBT and observed under phase-contrast and fluorescence microscopy (Figure 3). We measured the distance between the negative peaks of the mature spores in the phase-contrast images to define the length of the long axis of the mature spore. We also measured the distance between the positive APBT and GFP fluorescence peaks in fluorescence images (Figures 3, 4). We found that the average distance between the positive peaks of APBT and GFP fluorescence was smaller than the length of the long axis of mature spores. The distance between the positive peaks of APBT fluorescence of APBT was also smaller than that of CgeA-GFP and YeeK-GFP (Figures 3B–E, 4) and closer to that of YhcN-GFP (Figures 3F, 4). Previous studies have reported that YhcN localizes in the cortex and inner spore membrane (Takamatsu et al., 2009; Imamura et al., 2010; Liu et al., 2022). We compared the fluorescence of these GFP strains with that of APBT (Figures 3A, C, D, 4). The fluorescence of AtpC-GFP, LonB-GFP, and QoxB-GFP were detected in the inner spore membrane. The distance between the positive peaks of APBT fluorescence peaks was larger than that of AtpC-GFP, LonB-GFP, and QoxB-GFP. Based on these results, we concluded that APBT stains around the inner portion of the inner spore coat and the outer portion of the inner spore membrane, similar to YhcN-GFP.

Fluorescent reagents such as auramine O, rhodamine B, thioflavin T, and congo red have been used to analyze spore structures (Kuwana et al., 2023, 2024). In this study, we determined the regions of mature spores stained by these fluorescent reagents with APBT fluorescence as a reference (Figure 5). B. subtilis 168 cells were grown on Schaeffer's plates at 37°C for 24 h, collected, and stained with a mixture of fluorescent reagents. We then observed the cells using phase contrast and fluorescence microscopy (Figure 5). We measured the distance between the negative peaks of the mature spores in the phase-contrast images to define the length of the long axis of the mature spore. We also measured the distance between the positive fluorescence peaks of the reagents in the fluorescence images (Figure 6). The distance between the positive fluorescence peaks of APBT, green safranin O, red safranin O, auramine O, rhodamine B, thioflavin T, and congo red were all shorter than the length of the long axis of the mature spore. These results indicated these fluorescent reagents stained specific structures in B. subtilis mature spores. The distance between the red fluorescence peaks of safranin O was greater than that of APBT, whereas the distance between the green fluorescence peaks of safranin O was slightly smaller than that of APBT, although the difference was not significant. The distance between the positive rhodamine B and auramine O fluorescence peaks was similar to that of the positive APBT fluorescence peaks. These results show that safranin O stains different regions of B. subtilis spores with red and green fluorescence.

To analyze how sporulation influences safranin O staining, we used wild-type, sigG, sigK, and spoIVA mutant strains deficient in the development of the cortex and spore coat (Driks, 1999; McKenney et al., 2013). We also used cotE and gerE mutant strains that are deficient in inner and outer spore coat development, respectively (Driks, 1999; McKenney et al., 2013). The mutant strains were grown on Schaeffer's plates at 37°C for 24 h. Cells were collected and stained with a mixture of APBT and safranin O and observed using phase contrast and fluorescence microscopy (Figure 7). No phase-bright forespores were detected in the sigG, sigK, and spoIVA mutant strains. However, APBT fluorescence revealed the engulfed forespore, which allowed for a distinction between the mother and vegetative cells (Figures 7B–D). Safranin O green fluorescence was detected at the periphery of the forespore in sigG and sigK mutant strain cells, similar to wild-type cells and APBT. In contrast, the red fluorescence of safranin O was less distinct in the sigG and sigK mutant strain cells than in the wild-type cells (Figures 7A–C). In sigK mutants, green and red safranin O fluorescence appeared as an oval structure within lysed mother cells, co-localizing with APBT fluorescence (Figure 7C, white arrowheads). A previous study showed an abnormal mother cell structure in spoIVA mutants (Roels et al., 1992). The APBT fluorescence was detected at the periphery of the spoIVA vegetative cell, mother cell, and forespore, and it was especially strong at the abnormal structure in the mother cell (Figure 7D, black arrowheads). The green and red fluorescence of safranin O was hardly detected at the periphery of the spoIVA forespore but was strongly detected in the abnormal structure of the spoIVA mother cells (Figure 7D, black arrowheads).

We observed the phase-bright forespores and mature spores in the cotE and gerE mutant cells (Figures 7E, F). We could distinguish the vegetative cell, mother cell, forespore, and mature spore via APBT fluorescence. Green and red safranin O fluorescence was detected at the periphery of the forespore in cotE mutant cells, with the green fluorescence being weaker than the red fluorescence. Additionally, both APBT and red safranin O fluorescence were detected in some abnormal structures as strong spots in cotE mutant cells (Figure 7E, black arrowheads). In contrast, the APBT and safranin O fluorescence staining pattern in gerE mutant cells was similar to that in wild-type cells (Figures 7A, F). We concluded that coat and cortex development influenced safranin O staining in sporulating cells.