The bacterial strains and plasmids used in this study are listed in Table 1. Bt strain HD73 (accession no. CP004069) was used in this study (Liu et al., 2013). The Bt strains were transformed by electroporation, as previously described (Lereclus et al., 1989). Escherichia coli and the Bt strains were cultured in Luria-Bertani (LB) medium or Schaeffer's sporulation medium (SSM, 8 g of nutrition broth, 0.12% MgSO₄ [m/v], 0.1% KCl [m/v], 0.01 M NaOH, 0.1 M MnCl₂, 0.01 M Ca(NO₃)₂, and 0.01 M FeSO₄ in 1 L of H₂O; Schaeffer et al., 1965) with shaking (220 rpm) at 37 and 30°C, respectively. The antibiotic concentrations used for bacterial selection were 100 μg/ml kanamycin and 10 μg/ml erythromycin for Bt and 100 μg/ml ampicillin for E. coli. DNA manipulation as previously described (Peng et al., 2014). Oligonucleotide primers were listed in Table 2.

DNA fragments corresponding to the downstream and upstream regions of the lytSR genes (HD73_5856 and HD73_5855) were amplified by PCR using chromosomal DNA from Bt HD73 as the template and the lytSR-1F/lytSR-1R and lytSR-2F/lytSR-2R primer pairs, respectively. The corresponding DNA fragments were fused with overlapping PCR using primers lytSR-1F and lytSR-2R, and the PCR product was digested with BamHI and EcoRI. The fragments were purified and ligated with the temperature-sensitive suicide plasmid pMAD (Arnaud et al., 2004) digested with the same enzymes, to yield the recombinant plasmid pMAD-ΔlytSR, which was used to transformed into host strains with electroporation. The confirmed transformants were incubated at 39–41°C. Colonies lacking erythromycin resistance were selected and one mutant strain, HDΔlytSR, was verified with PCR.

The upstream (562-bp) and downstream (561-bp) fragments of lrgAB (HD73_5854 and HD73_5853) were PCR amplified with the primer pairs lrgAB-1F/lrgAB-1R and lrgAB-2F/lrgAB-2R, respectively, and using Bt HD73 genomic DNA as the template. The kanamycin (Kan)-resistance gene (1,473 bp) was amplified using primers Kan-R and Kan-F. The deletion-insertion mutant cassette was amplified with overlapping PCR using the upstream and downstream fragments and the Kan-resistance gene as the templates, with primers lrgAB-1F and lrgAB-2R. The lrgAB deletion-insertion mutant cassette was inserted into the BamHI and EcoRI restriction sites of the pMAD plasmid to generate the recombinant plasmid pMAD-ΔlrgAB, which was then used to transform Bt HD73 cells with electroporation. Transformants were grown at 30°C in LB plate containing erythromycin and kanamycin, and then transferred to liquid LB containing kanamycin at 39°C. The cells were then plated on LB agar plates. Colonies with kanamycin resistance but lacking erythromycin resistance were selected, and one mutant strain, HDΔlrgAB, was verified with PCR.

The oligonucleotide primer pairs lrgABhf-F/lrgABhf-R and lytSRhf-F/lytSRhf-R were used to amplify the lrgAB gene with its own promoter PlrgAB, and the lytSR gene with its promoter PlytSR. The resultant fragments were digested with PstI/BamHI and SalI/EcoRI, respectively, and then integrated into the shuttle vector pHT315 (Arantes and Lereclus, 1991) to generate pHTlrgAB and pHTlytSR, respectively. The genetically complemented mutant strains ΔlrgAB(lrgAB) and ΔlytSR(lytSR) were generated by introducing pHTlrgAB and pHTlytSR into HDΔlrgAB and -HDΔlytSR, respectively.

spoIID (HD73_5692), spoIIM (HD73_4392), and spoIIP (HD73_2232) mutants were constructed similar to lrgAB as described above, but using the primer pairs spoIID-1F/spoIID-1R, spoIID-2F/spoIID-2R, kanD-F/kanD-R, spoIIM-1F/spoIIM-1R, spoIIM-2F/spoIIM-2R, kanM-F/kanM-R, spoIIP-1F/spoIIP-1R, spoIIP-2F/spoIIP-2R, and kanP-F/kanP-R, respectively. The recombinant plasmids pMAD-ΔspoIID, pMAD-ΔspoIIM, and pMAD-ΔspoIIP were electroporated into Bt HD73 cells. Colonies with kanamycin resistance but lacking erythromycin resistance were selected, and mutant strains, HDΔspoIID, HDΔspoIIM, and HDΔspoIIP, were verified with PCR.

Overnight cultures of each strain grown in LB medium were used as starters for growth curve analyses. The exponential growth phase cells were washed in phosphate-buffered saline and then inoculated into SSM or M9 medium supplemented with tryptophan (50 μg/ml) and pyruvate (6 g/l) to an optical density at 600 nm (OD₆₀₀) of 0.1. The cultures were incubated at 30°C with shaking at 220 rpm, and growth was monitored by measuring the absorbance at 600 nm at different timepoints. Values represent the means of at least three independent replicates. Error bars represent standard deviations.

The HD73, HDΔlrgAB, ΔlrgAB(lrgAB), HDΔlytSR, and ΔlytSR(lytSR) strains were grown in SSM to T₂₈ (T₀ is the end of the exponential phase, and Tn is n hours after T₀) at 30°C with vigorous shaking. The number of viable cells was counted as the total colony-forming units (CFU) on the LB plates. The number of spores was determined as the number of heat-resistant (65°C for 30 min) CFU on the LB plates. Sporulation efficiency was defined as the ratio of the number of spores to the number of viable cells, multiplied by 100. Values represent the means of at least three independent replicates. The data were analyzed with SPSS (version 19.0) using a t-test. Error bars represent standard deviations. P-values are indicated in the figure legend.

The vital membrane dye FM4-64 (Molecular Probes, Inc., Eugene, OR, USA) was dissolved in dimethyl sulfoxide to a final concentration of 100 μM. The cells were stained with FM4-64 (100 μM) for 1 min on ice (Yang J. et al., 2013). To assess engulfment, 0.5 ml of cells cultured to T₁₂ were pelleted and resuspended in 0.1 ml of H₂O. An aliquot (2 μl) of this cell suspension was placed on a slide and stained with FM4-64 (100 μM) and MitoTracker Green FM (MTG, 100 nM; from Molecular Probes) for 1 min, and then scanned (476–490 nm excitation and 510–667 nm emission) with a confocal laser scanning microscope (Leica TCS SL; Leica Microsystems, Wetzlar, Germany). Each strain was scanned independently at least three times and each scan was then viewed in at least five fields. The rate of incomplete engulfment was defined as the ratio of the number of incompletely engulfed cells (stained with FM4-64 in the mother cell) to the total number of cells. The values given are the means of at least three independent replicates.

To assess the transcriptional activity of PlrgAB and PlytSR promoters, putative promoter fragments (633 and 845 bp, respectively) were cloned from Bt HD73 genomic DNA using the primer pairs PlrgAB-F/PlrgAB-R and PlytSR-F/PlytSR-R, respectively. The PstI/BamHI fragments of PlrgAB and PlytSR were separately integrated into vector pHT304-18Z, which is the Bt-E. coli shuttle harboring a promoterless lacZ gene (Agaisse and Lereclus, 1994) to generate plasmids pHT304PlrgAB and pHT304PlytSR, respectively. The former was introduced into Bt strain HD73 and the HDΔlytSR mutant, whereas the latter was introduced into Bt strain HD73 and the HDΔsigE mutant (Du et al., 2012). Resultant HD(PlrgAB), ΔlytSR(PlrgAB), HD(PlytSR), and ΔsigE(PlytSR) strains were selected with erythromycin and verified with PCR.

The constructions of PspoIID, PspoIIM, and PspoIIP (650, 437, and 668 bp, respectively) with lacZ fusions are similar to PlrgAB as described above, but using the primer pairs PspoIID-F/PspoIID-R, PspoIIM-F/PspoIIM-R, and PspoIIP-F/PspoIIP-R, respectively. The recombinant plasmids pHT304PspoIID, pHT304PspoIIM, and pHT304PspoIIP were introduced into Bt strain HD73 and the HDΔlytSR mutant. The resultant strains ΔlytSR(PspoIID), ΔlytSR(PspoIIM), and ΔlytSR(PspoIIP) were selected with erythromycin and verified with PCR.

Bt strains carrying lacZ transcriptional fusions were cultured in liquid SSM and 2-ml samples were collected at 1-h intervals. The cells were pelleted and resuspended in 0.5 ml of Z buffer (Peng et al., 2014) at 4°C, then lysed with a Mini-Beadbeater cell disrupter (BioSpec, Bartlesville, OK, USA) and centrifuged at 10,000 × g for 7 min at 4°C. β-Galactosidase activity was determined as previously described (Perchat et al., 2011). The reported values are the means of at least three independent assays. The data were analyzed with SPSS (version 19.0) using a t-test. Error bars represent standard deviations.

lrgAB is located downstream from the lytSR genes in Bt HD73 (Figure 1A). The Bt lytS (HD73_5856, sensor histidine kinase), lytR (HD73_5855, response regulator), lrgA (HD73_5854, holin-like protein), and lrgB (HD73_5853, holin-like protein) genes encode proteins that share 50, 44, 44, and 54% amino acid sequence identity, respectively, with homologs in Staphylococcus aureus (Patel and Golemi-Kotra, 2015), and 66, 65, 62, and 78% amino acid sequence identity with homologs in B. subtilis (van den Esker et al., 2017). Alignments of these proteins from Bt, S. aureus, and B. subtilis are shown in Supplementary Figure 1. The two-component system LytSR/LytST contained the conserved His_kinase domain and the response regulator receiver domain in Bt, S. aureus, and B. subtilis (Supplementary Figure 1).

To investigate the transcription from and regulation of the PlrgAB promoter in Bt, Bt strain HD73, and the lytSR mutant HDΔlytSR were transformed with a PlrgAB-lacZ fusion construct. The results of the β-galactosidase assay showed that the transcriptional activity of PlrgAB increased from T₄ to T₈ in the HD73 strain in SSM, whereas it did not increase dramatically in the HDΔlytSR mutant (Figure 1B), suggesting that the transcription of the lrgAB genes is positively regulated by LytSR during the late sporulation process.

Previous studies have shown that LytSR/LytST is involved in pyruvate utilization (Zhu et al., 2010; van den Esker et al., 2017). We also compared the growth of the lytSR mutant with that of wild-type strain HD73 in the presence of pyruvate. Results showed that ΔlytSR was unable to grow in M9 medium supplemented with pyruvate, whereas the wild-type reached an OD₆₀₀ of 0.9 after 20 h of incubation (Figure 2A), suggesting that LytSR is involved in pyruvate utilization in Bt. However, no differences in the growth curves of ΔlytSR and the wild-type were observed in SSM (Figure 2B). Thus, in order to eliminate the effects of growth medium, we selected SSM for further analyses of the sporulation efficiency.

Because the lrgAB genes have a dramatic impact on sporulation in B. anthracis (Chandramohan et al., 2009), we predicted that the lytSR or lrgAB mutation would affect the ability of the Bt cells to undergo sporulation. Therefore, the abilities of the lrgAB and lytSR mutants to sporulate were assessed. The wild-type strain HD73 had a sporulation efficiency of 85 ± 4% after growth to T₂₈ in SSM (Figure 3). The sporulation efficiency was not significantly different between HD73 and either HDΔlrgAB (72 ± 11%) or ΔlrgAB(lrgAB) (76 ± 9%), whereas it was significantly reduced in both HDΔlytSR (47 ± 3%, P ≤ 0.001) and, the genetically complemented strain ΔlytSR(lytSR) (54 ±4%, P ≤ 0.01). Based on the P-values (P ≤ 0.05, Figure 3) between ΔlytSR and ΔlytSR(lytSR), ΔlytSR(lytSR) showed a partly restored sporulation function. These results indicate that lytSR affects spore formation and the regulation of the genes involved in sporulation.

To determine the effect of LytSR on sporulation in Bt HD73, the cell membranes of Bt HD73 and its mutants were stained with the vital dye FM4-64, which labels the plasma membranes of living cells, and the process of spore formation was visualized with confocal microscopy. In cells grown to T₃ in SSM, the polar septum was curved in the wild-type and mutant cells, whereas some cells of HDΔlytSR had an incomplete septum at the distal pole (Figure 4). At T₁₂, the process of engulfment was completed in the forespores of the wild-type (Figure 5, arrow 1) and HDΔlrgAB cells. In these cases, the spores were not labeled with FM4-64, but were stained with MTG, and only the outer membranous outline of the living cells could be observed. In the mutant HDΔlytSR, a proportion of the cells had completed the process of engulfment, but 52 ± 3% cells were arrested in forespore engulfment (Figure 5, arrow 2), and a bipolar septum phenotype was observed.

HDΔlytSR cells were unable to initiate engulfment or form bipolar septa. The mother–cell-specific sigma factor SigE plays a critical role in the formation of an asymmetric septum and in forespore engulfment (Errington, 2003). Therefore, we predicted that SigE would also affect the transcription of lytSR. The results of the β-galactosidase assay indicate that the transcriptional activity of PlytSR increased rapidly from T₄ to T₁₀ in wild-type HD73, whereas it increased much more slowly in the HDΔsigE mutant grown in SSM (Figure 6), suggesting that the transcription of lytSR is controlled by the mother-cell-specific sigma factor SigE.

In B. subtilis, the sporulation genes spoIID, spoIIM, and spoIIP are controlled by SigE (Eichenberger et al., 2001) and may also be involved in suppressing septum formation at the distal pole of the sporangium (Chastanet and Losick, 2007). To determine whether LytSR affects the process of spore engulfment by regulating the expression of spoIID, spoIIM, and spoIIP, the promoters of these genes were fused to lacZ and the β-galactosidase activity was assessed in wild-type HD73 cells, lytSR and sigE mutants. The results showed that the transcriptional activities of spoIID, spoIIM, and spoIIP were sharply reduced or abolished in the sigE mutant grown in SSM (Figure 7), suggesting that the transcription of spoIID, spoIIM, and spoIIP is directly controlled by SigE in Bt. The transcription of PspoIID and PspoIIM transcription did not differ between the wild-type and mutants grown in SSM (Figures 7A,B). However, PspoIIP activity was dramatically reduced in the lytSR mutant grown in SSM (Figure 7C). These results suggest that the transcription of spoIIP is affected by LytSR.

To determine whether LytSR modulates Bt forespore engulfment by regulating spoIIP expression, we observed the phenotypes of the spoIID, spoIIM, and spoIIP mutants in SSM. In cells grown to T₃ or T₁₂ in SSM, the polar septum was curved or had completed the process of engulfment in the wild-type, whereas some HDΔspoIIM, and HDΔspoIID cells displayed an incomplete septum at the distal pole (Figure 4). At T₁₂, the phenotype of HDΔspoIIP was similar to that of HDΔlytSR, and the only difference was that more HDΔspoIIP cells (68 ± 5%) than HDΔlytSR cells (48 ± 3%) had completed the process of engulfment (Figure 5). In contrast, almost all the HDΔspoIID and HDΔspoIIM cells arrested in forespore engulfment, and bipolar septa were also observed (Figure 4), so this phenotype is similar to that of the spoIID and spoIIM mutants of B. subtilis (Pogliano et al., 1999). The β-galactosidase activity assay also revealed that PspoIIP was dramatically reduced in the lytSR mutant grown in SSM (Figure 7). All these results indicate that LytSR modulates Bt forespore engulfment, mainly by affecting spoIIP expression.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.